-------
50. Pleaae comolete the following table for the five principal hazardous waatss wnicn triia
facility used, reused, recycled, or reclaimed, on aite in greatest volume during the 1981
calendar year. [REFE3 TO FACING PAGE FOR INSTRUCTIONS FOR COMPLETING EACH CCLUHN Cf THE
TABLE] '
IF THIS FACILITY DID NOT USE, REUSE, RECYCLE, OR RECLAIM ANY HAZARDOUS
WASTES ON SITE IN 1981, CHECX' HERE F"! AN0 3KIP TD 2UESTION 52.
I—I O &*x
/1G3
COLUMN A
EPA Nureer AND
description of
waste used, re-
used, recycled
or reclaimed on
sita
a. 1 1 I I 1
Descnotion:
3.TO /9-Ci
o. 1 1 1 1 1
Description:
GToPr*3-
c. 1 ! ! 1 1
Description:
Q5"^fro3
d. I 1 ! 1 1
Description:
6> JT3 1^ 0 M,
e. ! 1 ! ! 1
Description:
G^^cr
i
COLUMN 3
How was this waste used, reused, recycled or reclamed?
[CIRCLE ALL THAT APPLY]
As feed-
stack in
manufac-
turing
pracssa
OT
QSooio)
01
QfcS'0,1
01
Ci-T^jo^
01
Q S*j iOi
01
QroGi'i'
Aa fuel
or fuel
suople-
ment
02
GuTx^C)
02-
cist^iP^
02
QS**53§s
02
Q-Ti32?r
02
as=^3?r
In manner
consti-
tuting
disposal
03
GriO3te4
03
<JC^
03
^^33^,
03
35-330 ^
03
Q.szisj'f
Reclaimed
04
Qi^lSvOl
ca
Q5^6V0i
cw.
Q^fiVo2>
oa
c^rcsvoV
M
£i**$tef
Other [S?EC:.rv]:
GLUMN C
Quantity or"
.taste jsei,
reused, rs-
cyciad or
rsclaimed
=unng "981*
05
Q^i-ii"o ,
CtS* C-C^
35
C 5~C >'~iS C • 3»
05
C^GS^rs
05
^5^eo~CM
05
^^Tdr^r
CS-c-?-
Q r: c ^=
ui5-cc-^
^^c.r
•CCIRCJ: ONE]:
Metric tonnes
English (or snort) tons
Gallons
Other [SPECIFY]:
31
02
03
32
-------
CC'.'JPN A: ENTER THE EPA WASTE NUMBERS W£ SHORT DESCRIPTIONS OF T^ FIVE WASTES, INCLUDING 7-t
NAMES OF T'-E WASTES AND CCSCS1PTIONS CF THE PROCESSES THROUGH WHICH EACH WASTE *A3
T^ERATED". [£": v,iST£ CCCES ME LISTED IN APPENDIX A OP THE 0£*£RAL INSTRUCTIONS]
t
nu,"N 9: CIRCLE *h£ .CODE OR COOES THAT DESCRIBE rOW THIS -ASTE -AS USES, REUSES, RECYCLED OR
"ECLAIJ-CD DURING :981.
CC'.'JVN C: INDICATE THE AVERAGE QUANTITY OF WASTE USED. REUSED, RECYCLED OR RECLAIMED DuflING 1981
AND. CIRCLE THE UNIT CODE AT Tr£ BOTTOM OF THE TABLE. PLEASE USE THE ;AME UNIT OF
'•CASURE FOR AIL WASTES IN THE TA3LE.
COLUMN 0: CIRCLE THE CODE OR CODES THAT INDICATE HOW EACH n'ASTE 'WAS STORED PRIOR TO USE, REUSE,
RECYCLING, OR RECLAMATION.
CSLL'HN E1: FOR EACH FACILITY, INDICATE THE AVERAGE"NUMBER- OT DAYS- EWH WASTE WAS" S7CRS5 .3
-------
51. For eacfl waste indicated :n Question 50 as Ssing jsed ;n 3 -nanner cartst lining iiscosai ['"33"
circled in Coluon 3), Biease oescriae tne scecir'i; nanner in «nicn cne -asce -as uded.
X. FINANCIAL ASSURANCE
52. How will closure and/or post closure casts be cov-rra far tnis facility? [CI
'Ca EACH COVESACE METKCO]
1
| " COVERAGE 3Y:
d c 2 /T.
*. Trust fund Sr( P.*: V ...
'" ?~0 ^
c. Letter of credit. . . . . > v •** ^"" .
i. Insurance solicy. . . '. X ^ "*""
f. Corporate guarantee . . . ^-,^.3^".
g. State guarantee .'-S^T . . . .
n. federal or state exemotion. '5?*' ? . .
i. Otner [SPSCirY]: (£ 5 A fl
^5<^ ra.
i
Closure
t
| costs
I only
1
1
1
1
1
1
Post -closure
coats only
I
*
,
2
2
2
2
2=tn
ciusurs and
| cost-closure
i
j
3
3
t
•t
3
3
closure r*or
pbst-cicaure
,
4
4
•
,'25
53.
are tfte annual aoninistratave charges for maintaining t.^e financial assurance necnaniama
Listed in Sueation 52?
Collars ;er »»ar
a. Closure $ C 5 "3
b. Post-closure $ £ -S"i
:?- —
c. TOTAL ADMINISTRATIVE CSSTS. . . $ C_ £ 2
34
'iT-55
-------
54. Old this 'acility or the company that owns this 'acility put uo collateral *for '-.nancial
assurance coverage'' [CIHCLi C.NLT 3.NE ZCDE ]
. .
[SKIP TO QUESTION 56] ............... 2
Yes [co ON ro QUESTION 55; ..... C-^'. ^ . . . . . •
,'5e
55. *hat is the value of the collateral? •
VALUE. OF- COLLATERAL: S-
56. Does this facili'ty have liaoility insurance for tnirc1 party aamages 'i.e., Bodily injurv
and property damage-) resulting from sudden or nonsudden releases af lazarcous waste7
[CIRCLE CNLY ONE CODE]
Yes [CO CN TO QUESTION 57j ........ ^ ST '& .
NO CREASE SIGN THE crsT:r:cATio.N STATEMENT
CN PAGE 37 AND RETURN 'H13 "ORM T3 E?A] ......
57. In wnat year did this facility ootain liability insurance far third part-, sanages"1 ['.'
MORE THAN ONE POLICY HAS 3EIN WRITTEN TCH THIS" TACXiTV, 3tEASc GIVE. rKE"^AR THE. MOST
•RECENT POLICY *AS OBTAINED]
YEAR LIABILITY INSURANCE OBTAINED: ^C. <~^7
58. In order to ootain liaoility insurance, »>s it necessary to ucgraoe tnis 'aciiity, ncoif
current practices at this facility, or have- a risk assessment of this facility serfomeo
[CIRCLE ONE CSCE TOR EACH ITEM]
res
a. Uograae facility .......... ^r.-^^^i
b. Modify current practices ...... y-.'O-. L> 1
^^ ^^^ .•(
-. Obtain risk assessment ...... s<.*J. -i W :
59. How -nany facilities, INC. 'JC INC this facility, coes the liaoilitv insurance caver"1
WM8ER Or FACILITIES CCVESED: _ L^^
35
-------
SO. In the tattle below, please indicate the ncmoer of policies held, the amount of coverage,
the annual coat of the policy, and the amount of the deductible for the liaoiiity insurancs
[IF *3RE THAN ONE POLICY IS HELD, INDICATE THE TOTAL COVERAGE, THE TO^AL COST, AND 'HE
AVERAGE DEDUCTIBLE FOR ALL POLICIES IN EACH CATEGORY CF THE TABLE BELOW]
IT MORE THAN ONE FACILITY IS COVERED BY THE POLICY, CHECX HERE! |ANO ENTER THE
AMOUNT OT OVERAGE, CCST AND DEDUCTIBLE FOR ALL FACILITIES COVERED.
/82
122!
Type of
Policy
a. Sudden
a. Norsudden
c. Combined policy
Numoer of
Policies
le
S^tr.A^
C-)/^ ^ /^V
&/LOP ^
a!. Coes the policy cover accidents resulting from sudoen or nonsudden releases or" na;arcoua
«aste wnicn .nay have occurred pridr to the year in «nicn the policy »«3 detained"*
[CIRCLE ONLY ONE CSC£]
Yes [CO CN TO QUESTION 62] G/. Vc ./ . . '
,-ta [SKIP TO QUESTION 63] 2
62. For how iiany ye«« prior to the year the soliev waa ootained ia accident liaoility insurance
providea?
NUMBER OF YEARS PRIOR COVERAGE:
63. Does tne policy caver legal aefera* costs? CCIRCLI CM.Y ONE C3DE]
YM
MO
/"C6
PLEASE SIGN THE CERTIFICATION STATEXNT ON ?ACE 37
RETURN THIS FORM TO EPA IN THE ENVELOPE °SOVID£D.
36
-------
CERTIFICATION STATEMENT
THE OWNER OR THE OPERATOR OF THE FACILITY, OR HIS AUTHORIZED REPRESENTATIVE,
SIGN AND DATE THE CERTIFICATION WERE INDICATED. THE PRINTED OR TYPED NAME OF THE
PERSON SIGNING THE CERTIFICATION MUST ALSO 3£ INCLUDED *HE-RE INDICATED.
CERTIFICATION;
I certify under penalty of la* that I nave personally examined ana am familiar «ith the infor-
mation suomitted in this and all attached documents, and that baaed on my inouiry of those
individuals immediately reaponsiole for obtaining :he information. I oelieve that the suomittsd
information is true, accurate, and complete. I am aware that there are significant oenalties
'or susmitting false in format ion, including the possibility of fine and imorisonment.
3RINT.OR TYPE
SIGNAruRE.
DATE
j AFTER COMPLETING THIS QUESTIONNAIRE, RETURN IT TO EPA IN THE ENVELOPE '
i ENCLOSED IN THE QUESTIONNAIRE PACXAGE. i
\
IF THIS FACILITY HAS RECEIVED MORE THAN CNE QUESTIONNAIRE, PLEASE RETURN J
ALL COMPLETED QUESTIONNAIRES IN THE SAME"ENVELOPE. ;
37
-------
OMB No: 2000-0396
Expiration Date: 6/30/84
RCRA Section 3007 Questionnaire
Organic Rubber Processing Chemicals Manufacturing Industry
Return within 45 days from date of receipt to:
Ms. Dina Villari (WH 562)
Characterization and Assessment Division
Office of Solid Waste
U.S. Environmental Protection Agency
401 M St., S.W.
Washington D.C. 20460
1. Corporate/Plant Data
A. Name of Corporation _
B. Address of Corporation Headquarters
Street
City State Zip.
C. Name of Plant
D. Address of Plant
Street
City State Zip.
Hazardous waste generator ID number:
E. Mailing Address of Plant (if different from above)
F. Namels) of personnel to be contacted for additional information pertaining to this questionnaire
Name Title Telephone
-------
2. Type of Plant Operation
A. Indicate whether the following organic rubber processing chemicals1 were manufactured at this facility in 1983:2
1. CAS No.: Manufactured: D Yes G No
Chemical Name:
2. CAS No.: Manufactured: C Yes G No
Chemical Name:
3. CAS No.: Manufactured: u Yes G No
Chemical Name:
4. CAS No.: Manufactured: G Yes D No
Chemical Name:
5. CAS No.: Manufactured: G Yes D No
Chemical Name:
6. CAS No.: Manufactured: D Yes D No
Chemical Name:
7. CAS No.: Manufactured: D Yes G No
Chemical Name:
8. CAS No.: Manufactured: D Yes G No
Chemical Name:
9. CAS No.: Manufactured: u Yes G No
Chemical Name:
10. CAS No.: Manufactured: D Yes G No
Chemical Name:
Complete this questionnaire for each chemical listed above which you manufactured. If none of these chemicals
were manufactured, return pages 1 and 2 of this questionnaire.
B. Identify as follows the chemical intermediate(s) produced at this facility in the production of the chemicals
identified above:
CAS Number Chemical Name Common Name
'Rubber processing chemicals are defined as the synthetic organic compounds that are added to natural or synthetic rubber to produce or
enhance specific properties in the final product.
2lf additional space is needed for listing products or intermediates, attach an additional sheet.
Intermediate means any chemical substance (1) which is intentionally manufactured and removed from the equipment in which it is
manufactured, and (2) which either is consumed in whole or in part in chemical reactions(s) used for the intentional manufacture of other
chemical substance(s) or mixture(s).
2
-------
C. Indicate those classes of chemical products or intermediates which were produced at this facility in 1983. Circle
appropriate code number(s).
Cod*
Number
Classes of Products
and Intermediatea
ORGANIC DYES & PIGMENTS
286 52 Organic Dyes
286 53 Organic Pigments
CYCLIC INTERMEDIATES
286 61 Aromatic Acids & Derivatives
286 62 Aromatic Acids, Anhydrides & Esters
286 63 Aromatic Ketones & Aldehydes
286 64 Aromatic Alcohols
286 65 Aromatic Hydrocarbons
286 66 Cyclic Amines
Coda Classes of Products
Number and Intermediates
286 67 Halogenated Aromatics NEC
286 68 Alicyclic Chemicals
286 69 Cyclic Intermediates NEC
CYCLIC CHEMICALS NEC
286 71 Salts of Aromatic Acids
286 72 Other Cyclic Chemicals
ACYCLIC CHEMICALS
286 81 Halogenated Hydrocarbons
286 82 Monohydric Acyclic Alcohols
286 83 Polyhydric Alcohols & Ethers
Coda Classes of Products
Number and Intermediates
286 84 Acyclic Acids, Anhydrides & Esters
286 85 Acyclic Aldehydes
286 86 Acyclic Ketones
286 87 Acyclic Nitrogens
286 88 Acyclic Compounds NEC
286 89 Acychcs NEC
ORGANIC CHEMICALS NEC
286 91 Flavor & Perfume Materials
286 93 Plasticizers
NEC — Not elsewhere classified
3. Process and Treatment Residual Information
This information will be used to address industry wide variation in type and quantity of residuals generated.
Residuals include any process stream generated during the manufacture of a product which is not used as a
raw material or principally sold as a commercial product. Treatment residuals include wastes from the treatment
of process residuals. Residuals may be solids (e.g., still bottoms), liquids (e.g., wastewater), confined gases
(e.g., gases that are containerized to facilitate disposal), and unconfined gases generated by the management
of solid or liquid residuals (e.g., incinerator stack emissions) or unconfined gases containing condensable
gases (e.g., vented light ends).
For each unit process provide a general process block flow diagram that identifies major unit operations and
treatment processes and indicates the types and points of introduction/generation of feedstocks, products,
co-products/by-products, and residuals (See Examples I and II.) Include the information requested in Questions
3-A through 3-D in the flow diagram. Provide the information requested in Questions 3-E and 3-F in an
attachment.
A. Identify the product process, intermediates, co-products and by-products produced by the process.
B. Provide a block for each major unit operation (e.g., reactor, washer, filtration, air emission control, aeration
lagoon, etc.) in the production process and in each residuals management process.
C. Identify process input such as raw materials, reagents and solvents by chemical or common name or chemical
formula, and indicate the point of introduction with arrows.
D. Assign a Residual Identification Number to each of the following types of residuals and indicate its point of
generation with an arrow:
1. Residuals generated by unit operations in the product process, including unit operations that produce/
recover co-products, by-products and solvents; and
2. Final treatment residuals (i.e., residuals generated by physical, chemical (including incineration and other
thermal treatment) or biological treatment and that are not intermediate treatment residuals generated within
a treatment train).
When more than one process block flow diagram is provided (i.e., for multiple product processes), assign
unique, sequential Residual Identification Numbers to the residuals for each flow diagram.
-------
Example I — Process Block Flow Diagram
Production
OH sue
Landlill
Treatment
VdLuum Jet Condense
to Treatment
NaOH
Spent
Dryer
Na2S Byproduct (sold off site)
Xylene Recyt la
Product: 2 Mercaptobenzothiazole and its Zinc Salt
Intermediates: None
10% Sludge lo
•^•Treated Effluent 10
NPDES Discharge
-------
E. If residuals from this product process are combined with the residuals from other product processes at this
facility prior to treatment, identify the product process residual by Residual Identification Number and specify
the source of the other residuals using the codes provided in Question 2-C.
F. Indicate the typical annual production, the 1983 production, and system capacity (specify), for each: product,
co-product and by-product.
Example II — Response to Questions 3-E and 3-F
Product Process: 2-Mercaptobenzothiazole, Zinc salt (MBT) Production
A. Mixing of MBT Production Residuals with Other Residuals
MBT Residual
Identification Number
(from Flow Diagram)
2,5
Source of Other Residuals
(Manufacturing Code from Question 2-B)
286 66, 286 65
B. Annual Production (1983)
Product .
Technical MBT
MBT, Zinc Salt
1,000,000 Ibs (1983)
1,500,000 Ibs (typical)
2,000,000 Ibs (capacity)
12,500,000 Ibs (1983, typical)
15,000,000 Ibs (capacity)
Co-product/ By-product
Sodium Sulfide
250,000 Ibs (1983)
300,000 Ibs (typical)
400,000 Ibs (capacity)
-------
4. Residuals Characterization Information
For each process used to manufacture the chemicals listed in Question 2, complete Table 1 by providing the
following information for each identified residual. An example is provided below (Example III).
A. Identify the product process.
B. List each residual by Residual Identification Number (include by-products and residuals generated from the trea
ment of process residuals.)
C. If the residual has been identified in the facility RCRA notification, indicate whether it was identified as
ignitable (I), corrosive (C), reactive (R), EP toxic (E), listed by EPA or reported by the facility as toxic (T), or
acutely hazardous (H).
0. For each residual, describe the following properties where appropnate: physical state [e.g., liquid (specify
whether aqueous or organic), solid, slurry (indicate solids content), gas]; pH; flash point; Btu content;
viscosity; toxicity.
E. List the compounds which are known by analysis to be present in the residual and specify, as known, the
concentration ranges as follows:
Code Range
A > 50%
B > 10% to 50%
C >1% to 10%
D >0.1%to1%
E > 0.01% to 0.1%
[Actual concentration]* <0.01%
F. If residual analyses are not available, list the compounds which are expected to be present in the residual anc
expected concentrations based on chemical engineering principles.
Example III — Response to Question 4
A. Product Process: MBT Production
B. C. D. E. F.
Residual RCRA Properties Known Compounds, Other Expected
Identification Identification of Residual Concentrations. Compounds
Number (I,C,R,E,T, or H) Ranges
3 Not identified Solid Sulfur- C
Aniline: C Carbon Disulfide
MBT: C
Sodium Sulfide: C
Not identified Aqueous Liquid Water: A
pH=7.8 MBT: 200 pern
-------
Table I — Response to Question 4
A. Product Process:
B.
Residual
Identification
Number
C.
RCRA
Identification
(I.C.R.E.T, or HI
D.
Properties
of Residual
E.
Known Compounds,
Concentrations,
Ranges
F.
Other Expected
Compounds
•RESIDUAL CONSTITUENT CONCENTRATION CODE
Code
A
B
C
D
E
[Actual concentration It
Range
>50%
>10% to 50%
>1% to 10%
>0.1% to 1%
>0.01% to 0.1%
t If concentration is less than 0 01%. specify, as known, the
typical concentration in ppm
-------
5. Residuals Management Information — General
For each process used to manufacture the chemicals listed in Question 2, complete Table II by providing the
following information for each identified residual. An example is provided below. (Example IV).
A. Identify the product process.
B. Specify the Residual Identification Number.
C. Specify residual category in accordance with codes provided.
Code Categories of Residuals
C1. Process precipitates or filtration residues
and process sludges
C2. Process decantates or filtrates
C3. Treatment sludges: (specify)
a. biological b. other
C4. Spent activated carbon or other adsorbent
(specify)
C5. Spent catalyst
C6. Heavy ends:
a. distillation residues b. misc. heavy ends
'"Acidic: pH < 2.0, Neutral 2.0 < pH < 12.0, Caustic, pH > 12.0
2lLight ends are condensable if primarily composed of gases which are
Code Categories of Residuals (continued)
C7. Spent solvents
C8. Untreated process wastewater:
a. acid b. caustic c. neutral"1
C9. Treated wastewater discharge
C10. Containers, liners, cleaning rags, gloves, etc.
C11. Off specification products and feedstock
C12. Other (specify)
C13. By-product
C14. Light ends:
a. condensable'2' b. noncondensable
liquid at ambient temperature and pressure.
D. Specify management methods in accordance with codes provided. If a residual is subject to a sequence of
methods (e.g., storage in a tank, incineration), list the methods in sequence. If a residual is handled alterna-
tively by more than one method (e.g., either incinerated or burned in a boiler), identify the alternate methods.
Code Management Methods
M1. Storage in:
a. tank b. container c. pile
d. surface impoundment
M2. Treatment in:
a. tank b. container c. surface
impoundment
M3. Burning in boiler
M4. Recovery/reclamation:
a. recovery b. reused same process
c. reused different process d. sales
M5. Incineration
M6. Landfill
Code Management Methods (continued)
M7. Underground injection
M8. On-site wastewater treatment in:
a. tank b. surface impoundment c. container
M9. Discharge to publicly owned wastewater
treatment works
M10. Discharge to surface water under NPDES
M11. Discharge to off-site privately owned
wastewater treatment works
M12. Other (specify)
M13. Scrubber:
a. caustic b. water c. other (specify)
M14. Flare
E. Indicate the amount of each residual managed by each method in 1983 (specify units). Include any clarification
needed to describe the waste stream.
F. Indicate whether the residual is managed on-site or off-site. If managed off-site, identify the site in the space
provided in Table III.
G. For residuals managed off-site, except for discharges to a POTW or surface water under a NPDES permit,
indicate the average management cost per unit quantity of residual in 1983.
H. Indicate planned changes in residual management methods by specifying the codes for the new management
methods, and indicate the anticipated date of change.
8
-------
Example IV — Responses to Question 5
A. Product Process: MBT Production
B.
Residual
Identification
Number
1
3
c.
Residual
Code
C8
C1
D.
Management
Code
M4-a
(Sodium sulfide
recovery)
M1-c
M-6
E.
1983 Residual
Quantities
25.000.000 oal*
400 tons
F.
On-site or
Off-site
Management
On-site
recovery
On-site storaqe
Off-site disposal
G.
Costs
for Off-site
Management
N.A.
NA
$15/wet ton
H.
Changes in
Management
Methods
None
None
None
M4 11984)
C9
M-IO
50 million gal.
Off-site
N.A.
None
•Prior to drying stage, NazS content is approximately 5% by weight of Residual /1.
Table II — Response to Question 5
A. Product Process:
B.
Residual
Identification
Number
C. D.
Residual Management
Code Code
E.
1983 Residual
Quantities
(specify units)
F.
On-site or
Off -site
Management*
G.
1983 Costs
for Off -site
Management
(cost/ton)
H.
Changes in
Management
Methods
'Identify off-site waste management or recycling /reuse facility as indicated in the following Table (Table III).
-------
Table III — Response to Question 5-F
Name of Facility: Name of Facility:
Residual Identification Numbers: Residual Identification Numbers:
Facility Mailing Address: Facility Mailing Address:
Street or P.O. Box: Street or P.O. Box:
City or Town: City or Town:
State: Zip: State: Zip:.
Facility Location (if different from above): Facility Location (if different from above):
Street, Route Number or Other Specific Identifier: Street, Route Number or Other Specific Identifier:
City or Town: City or Town:
State: Zip:__ State: Zip:
Hazardous Waste Facility I.D. Number (if any): Hazardous waste Facility I.D. Number (if any):
10
-------
6. Specific On-site Residuals Management Information
Provide specific information for residuals managed on-site using available information to complete the following sec-
tions which pertain to your residuals management operations.
A. Storage or Treatment in Tanks*
B. Storage or Treatment in Containers*
C. Storage or Treatment in Piles*
D. Burning in a Boiler
E. incineration
F. Land Treatment*
G. Surface Impoundments*
H. Landfills*
"If residuals are managed by these methods, provide the following information:
(1) Are groundwater monitoring data available?
(2) Are geologic or hydrogeologic data available?
A. Storage or Treatment in Tanks
Have identified residuals been stored or treated in on-site tanks at
any time since January 1, 1983?
If yes, provide the following information for the 10 largest tanks:
D Yes
n Yes
D No
n NO
D Yes D No
Type of
Residuals Design Storage or Treatment Avg. Length
Tank Managed1 Capacity^ Treatment Used3 of Storage
1
2
3
4
5
6
7
8
9
10
Part of Wastewater
Treatment Train
(Circle Yes or No)
Yes No
Yes No
Yes No
Yes No
Yes No
Yes No
Yes No
Yes No
Yes No
Yes No
Covered
(Circle Yes or No)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
Secondary
Containment
Provided5
(Circle Yes or No)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
1Use Residual Identification Numbers to identify residuals.
2Use the following codes to designate the tank capacity:
A <10,000 gallons
B >10,000 gallons to 100,000 gallons
C >100,000 gallons to 1,000,000 gallons
D >1,000,000 gallons
4Treatment train from which wastewater is discharged under a NPDES permit or through a sewer system to a
publicly-owned treatment works.
Secondary containment is provided when the tank is located inside a dike area where the volume of liquid
that the diked area can contain is at least equivalent to the capacity of the largest tank.
11
-------
B. Storage or Treatment in Containers*
Have identified residuals been stored or treated on-site in containers
at any time since January 1, 1983? D Yes LJ No
If yes, provide the following information (if the facility has several container storage areas, provide information
only on the primary container storage area):
1. Check typical and maximum quantity stored on any day in 1983 for each residual:
Average
Average Maximum Length of
Residual No.1 Daily Quantity2 Daily Quantity2 Storage
1Use Residual Identification Number to identify residuals
2Use the following code to designate the quantity of residual(s) in storage on any day in 1983:
A ^550 gallons
B > 550 to 5500 gallons
C >5500 to 55,000 gallons
D > 55,000 gallons
2. Identify the storage area base material:
D Concrete D Asphalt D Soil D Other (specify)
3. If liquid residuals or residuals containing free liquids* are stored, is the storage area designed and operated
to collect and contain surface runoff?
G Yes C No D Liquids are not stored
"Container means any portable device in which residuals were stored, treated or otherwise handled.
TA residual contains free liquids if liquids readily separate from the solid portion of the residual under ambient temperature
and pressure.
12
-------
C. Storage or Treatment in Piles
Have identified residuals been stored or treated in on-site piles
at any time since January 1, 1983?
If yes, provide the following information for the 10 largest piles:
Pile
1
2
3
4
5
6
7
8
9
10
Residuals
Managed1
Typical
Quantity2
Managed
Under Roofed
Structure
(Circle Yes or No)
El Yes
Containment3
Provided
(Circle Yes or No)
C No
Synthetic4
Liner Base
(Circle Yes or No)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
1Use Residual Identification Numbers to identify residuals.
2Use the following code to designate the typical quantity of residual(s) contained in the pile on any day in 1983:
A <20 cubic yards
B >20 to 200 cubic yards
C > 200 to 2,000 cubic yards
D > 2,000 to 20,000 cubic yards
E > 20,000 cubic yards
3Containment is provided when the pile base is designed, operated, and maintained to contain leachate
and run-off.
4ls a synthetic liner installed in the pile base? Waste may lie directly on synthetic liner or the liner may be
covered with clay layer.
13
-------
D. Burning in a Boiler
Have identified residuals been burned in an on-site boiler
at any time since January 1 1983?
If yes, provide the following information for each boiler:
1. Boiler and fuel type:
D Yes LJ No
Type
Fire tube
Water tube
Boiler Capacity
(Heat Input in
Million Btu/hr)
I] <10 million
I! >10 million to
100 million
H >100 million
Primary
Boiler Fuel
Oil
Gas
Coal
Wood or other
Percentage of Fuel
Replaced by Residuals
(Heat Input Basis)
> 5-10%
> 10-25%
> 25-50%
>50%
Typical Boiler Load
When Firing Residual
(% of Capacity)
Z £50%
~ > 50-75%
C >75%
2. Provide the following information for each of the residuals burned:
Use Residual Identification Numbers to identify residuals.
3. Provide the following information on the total feed mixture when residual is burned:
Feed Rate (Pounds per hour)
Typical BTU Content (BTU/lb)
Typical Total Ash Content (% by wt.)
Typical Total Halogen Content (% by wt.)
Typical Total Water Content (% by wt.)
Boiler
Temperature (°C)
Inlet Outlet
Typical BTU
Residual Feed Rate Content
No.1 (Ibs. per hour) (BTU/lb)
Typical Total
Ash Content
{% by wt.)
Typical Total
Halogen
Content
(% by wt.)
Total Water
Content
(% by wt.)
4. If the boiler is equipped with an air pollution control device, specify the type of device:
D Scrubber G Electrostatic precipitator C Other (specify)
5. Are residual-burning stack emissions data available?
Yes
G No
14
-------
E, Incineration
Have identified residuals been incinerated on-site at any time since
January 1, 1983?
If yes, provide the following information for each incinerator:
1. Incinerator type:
: Yes
Type
D Liquid injection
D Rotary kiln
D Hearth
D Other
(specify)
Incinerator Capacity
(Heat input in MMBtu/hr.)
U <10 million
D > 10 million to
100 million
Ll >100 million
2. Combustion Chamber Design Parameters:
Combustion Chamber Temp.
Location of Temp. Monitor
Residence Time
Feed Type
C Liquid — nozzle type
(specify)
— Atomizing pressure
(specify)
D Solid
D Batch charge
C Continuous charge
Primary Chamber
No
Percentage of
Auxiliary Fuel Required
(Heat Input Basis)
Secondary Chamber
(sec)
.(sec)
3. If the incinerator is equipped with an air pollution control device, specify the type(s) of device(s):
C Scrubber ~L Electrostatic precipitator G Other (specify)
4. Are incinerator stack emissions data available?
5. Provide the following information for each of the residuals burned:
Yes
No
Typical BTU
Residual Feed Rate Content
No.1 (Ibs. per hour) (BTU/lb)
Typical Total
Ash Content
(% by wt.)
Typical Total
Halogen
Content
(% by wt.)
Total Water
Content
(% by wt.)
Use Residual Identification Numbers to identify residuals.
15
-------
F. Land Treatment
Have identified residuals been managed in an on-site land treatment
operation at any time since January 1, 1983? D Yes G No
If yes, provide the following information:
1. Year land treatment initiated at site:
2. Year land treatment of identified residuals initiated:.
3. Have residuals other than identified residuals been land treated
at any time since January 1, 1983? D Yes G No
4. What was the total area actively used for land treatment in 1983? acres
5. What is the average slope of the land treatment site? percent
6. Is surface water run-off from the site collected for treatment,
re-application to the site, or analyzed prior to discharge? Q Yes D No
7. Check method(s) used to apply residuals to the land treatment site.
a. G Surface spreading or spray irrigation without plow or disc incorporation. Indicate residuals applied in
this manner using Residual Identification Numbers:
b. D Surface spreading or spray irrigation with plow or disc incorporation to a depth of (specify).
Indicate residuals applied in this manner using Residual Identification Numbers:
c. D Subsurface injection to a depth of (specify). Indicate residuals applied in this manner
using Residual Identification Numbers:
d. G Other methods (specify methods and residuals):
8. Is soil core monitoring performed? G Yes G No
9. Is soil pore water monitoring performed? G Yes G No
16
-------
G. Surface Impoundments*
Have identified residuals been stored, treated, or disposed of in an on-site
surface impoundment at any time since January 1, 1983? [H Yes [H No
If yes, complete Table IV.
*A surface impoundent is defined as holding, storage, settling, and aeration pits, ponds, or lagoons formed primarily of earthen materials.
Table IV — Response to Question 6-G
If more than 5 surface impoundments have been used since January 1, 1983 to manage identified residuals, provide
information only on the 5 impoundments with the largest capacities. Use Residual Identification Numbers to identify
residuals. If you do not know whether a liner has been installed, circle both "Yes" and "No." If you do not know
the thickness of a liner, indicate "UNK" for unknown.
Leachate Collection
Synthetic Liner Clay Liner System
Specify
Residuals Total Storage or Treatment
Disposed Capacity Treatment Type if Thickness No. of Thickness No. of Leachate
Impoundment (HINI (Gallons)1 (specify) Applicable2 Installed (mils) Liners Installed (in) Liners Installed Generated
1 Y«s
2 " YBS
3 VPS
4 Yes
5 Yes
No
No
Nn
No
No
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
1Use the following code to designate the quantity of residual(s) in storage on any day in 1983:
A < 550 gallons
B > 550 to 5,500 gallons
C > 5,500 to 55,000 gallons
D > 55,000 gallons
Use the following codes to specify treatment type:
A Neutralization
B Settling/Clarification
C Aeration
D Equalization
E Mixing
F Evaporation
G Other (specify)
17
-------
H. Landfills
1. Have identified residuals been landfilled on-site at any time that
you owned or operated this facility?
If yes, answer questions 2 and 3.
2. Has any on-site landfill (or landfill cell) that was used to dispose of
identified residuals been closed (i.e, no longer used to dispose of wastes)?
If yes, complete Table V.
3. Have any identified residuals been landfilled on-site at any time since
January 1, 1983 in a cell that has not been closed?
If yes, complete Table VI.
Table V — Response to Question 6-H-2
D Yes D No
D Yes C No
D Yes D No
Closed Landfill Cells
If more than 5 cells containing identified residuals have been closed, provide information only on the 5 cells that
were most recently closed. Use Residual Identification Numbers to identify residuals. If you do not know whether a
layer or liner was installed, circle both "Yes" and "No." If you do not know the thickness of a layer or liner,
indicate "UNK" for unknown.
A. Cap/Cover Design
Call
1
2
3
4
5
Drainage Layer
Residuals
Disposed Thickness
(RIN) Installed Material (in)
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
Cap Design
Clay Layer
Installed
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
Thickness
(in)
Synthetic Liner
Installed Material
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
Thickness
(Mils)
B. Bottom Liner Design/Leachate Collection
Synthetic Liner Clay Liner
Leachate Collection
System
Cell Number
(As Assigned Thickness
Above) Installed (mils)
1
2
3
4
5
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No. of Thickness
Liners Installed (in)
Yes
Yes
Yes
Yes
Yes
No
No
No
No
Nn
No. of
Liners Installed
Yes
YBS
Yes
Yes
Yes
No
No
No
No
No
Laachate
Generated
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
18
-------
Table VI — Response to Question 6-H-3
Landfill Cells Used to Dispose of Identified Residuals at any Time Since January 1, 1983
If more than 5 cells have been used since January 1, 1983 to dispose of identified residuals, provide information
only on the 5 containing the greatest quantities of identified residuals. Use Residual Identification Numbers to
identify residuals. If you do not know whether a liner has been installed, circle both "Yes" and "No." If you do
not know the thickness of a liner, indicate "UNK" for unknown.
Bottom Liner Design/Leachate Collection
Synthetic Liner Clay Liner
Leachate Collection
System
Residuals
Disposed Thickness No. of Thickness No. of
Cell (RIN) Installed Material (mils) Liners Installed (in) Liners Installed
1
2
3
4
5
Yes
Yes
Yes
Yes
Yes
No
Mo
No
No
No
Yes
Yes
Yes
Yes
Yes
No
Nn
No
No
No
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
Leachate
Genrated
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
19
-------
APPENDIX B
PROCESS STUDIES
-------
1.0 INTRODUCTION
t
This appendix contains 22 exploratory studies of various industrial processes and
practices known to generate (or influence the generation of) hazardous wastes.
The list of the analyzed processes and practices is given below:
Study No. Description
1 Acrylonitnle Manufacture
2 Agricultural Chemicals Formulation
3 Electroplating
4 Epichlorohydrin Manufacture
5 Inorganic Pigments Manufacture
6 Metal Surface Treatment
7 Organic Dyes 4 Pigments Manufacture
3 Paint Manufacturing
9 Petroleum Refining
10 Phenolic Resins Manufacture
11 Printed Circuit Boards Manufacture
12 Printing Operations
13 Synthetic Fiber Manufacture
14 Synthetic Rubber Manufacture
15 1,1,1-Trichloroethane Manufacture
16 Tnchloroethylene/Perchloroethylene Manufacture
17 Vinyl Chloride Monomer Manufacture
18 Wood Preserving
19 Good Operating Practices
20 Metal Parts Cleaning
21 Paint Application
22 Process Equipment Cleaning
The studies were performed in order to accomplish the following set of
objectives:
o Identify and characterize source reduction technologies currently
being used.
-------
o Identify and characterize new technological approaches which may
reduce waste generation.
o Provide an assessment of the current extent of waste reduction
achieved due to implementation of the considered methodology.
o Provide an assessment of the future extent of waste reduction due
to further implementation of both current and new technologies.
Accomplishing these objectives was approached in the manner outlined below.
2.0 APPROACH AND METHODOLOGY
It is important to note that source control measures are highly process- or site-
specific. The preparation of a truly accurate assessment of the extent to which
wastes have been and can be minimized by industry nationwide would require an
extensive site-by-site inspection, coupled with engineering and economic
analyses of many processes and large numbers of facilities. The simplified
approach used here relies on the concept of an "average" process. The
"average" process is one which represents a synthetic approximation of major
variations encountered from facility to facility. Process configurations were
evolved mainly from descriptions available in the public domain and general
literature. In most cases, the appropriate degree of resolution was judged to be
at the level of a generic block flow diagram.
The approach used to accomplish the objectives listed in Section 1 consisted of
the following steps:
(a) Definition of a manageable set of waste-producing processes and
practices to be analyzed.
(b) Data collection and engineering analysis of each process and
practice to produce a comprehensive list of source reduction
techniques, including those that are currently practiced and those
that may be practiced in the future.
(c) Review of the compiled source reduction techniques by industry and
consultants. In this review process, each technique would ideally
receive an independent rating as to its waste reduction
effectiveness, extent of current use and future application
-------
potential. Additional objectives of such review consist of
broadening the compilation to include techniques not considered
originally and to limit the necessarily conjectural nature of some
results obtained from part (b) above.
(d) Derivation of estimates for current and future extents of waste
reduction for each process, based on ratings obtained from (c), and
(e) Derivation of nationwide estimates for the current and future
extents of waste reduction, based on the composite profiles for each
industry category. The industry composites are obtained, in turn,
from the averaged extent of waste reduction achieved and potential
future reduction for the individual representative processes.
The sample set of 22 processes and practices was selected based on EPA
guidance, production volumes, the RIA Mail Survey database and on the "F" and
"K" waste stream RCRA listings.
Data collection involved inspection of the pre-compiled source reduction
bibliographies, automated bibliographic searches, consultation with standard
references (e.g. Kirk-Othmer's Encyclopedia of Chemical Technology),
government studies (e.g. U.S. EPA Effluent Guidelines development documents),
and consultations with experts and industry contacts. After the data were
compiled, a standardized format for data categorization was developed for each
process and practice. Section 9 of each individual process study was prepared
following identification of source control methods through preliminary
engineering analysis of the process and review of the compiled information.
Following their description, all of the identified source control techniques were
assembled in a summary table and preliminary ratings were generated by
project staff for each source reduction technique in the following categories:
waste reduction effectiveness, extent of current use, and future application
potential. Additionally, ratings were given to the quantity and quality of the
available data. All ratings were based on the integer scale of zero (none), one
(low), two (medium), three (high), four (all). Product substitution options were
identified mainly through database review and are described in a separate
section (Section 10) of each process study.
-------
Following the preparation of a preliminary draft, each process study was
submitted for an independent review by interested industrial firms involved in
that particular process. In other cases, the reviewers were independent
consultants or resident technical staff with experience in a particular process.
The reviewers were asked to comment on the technical substance of the report
with special emphasis on the integer ratings given to each source control
technique. Additionally, the reviewers were asked to provide information on
the actual source reduction applications known to them and to augment the
compiled list with any techniques that they felt were worth considering.
After the individual process study review comments were received and
evaluated, drafts were revised and the extent of waste reduction due to source
control was computed for each study from the ratings developed using the
computational methodology described below. Waste reduction estimates were
calculated for each wastestream and for each process overall.
For each process, two sets of numbers were generated:
1) Qualitative estimates of waste reductions which firms as a group
are currently achieving, and
2) Qualitative estimates of waste reductions achievable in the future
using techniques suggested and identified in the study.
The intent in furnishing these qualitative estimates is to provide a basis for
crude projections of the overall extent of waste reduction possible for the
entire industry. Under no circumstances should these estimates be construed as
definitive thresholds or limits on reductions achievable by the individual firms
employing the analyzed process.
It must be noted that, in many cases, both RCRA and non-RCRA wastestreams
were analyzed for their minimization potential. This was done primarily
because of concern that control of non-RCRA waste streams can be indirectly
responsible for RCRA waste generation. For example, control of solvent
-------
emissions into the air has, in the past, been frequently accomplished using
adsorption on activated carbon. The regeneration of carbon usingt steam
produced in certain cases a non-recyclable mixture of solvent and water that
had to be disposed of.
The interactions between hazardous and non-hazardous wastes are complex and
not readily visible. Therefore, it was deemed prudent to apply the analysis to
all principal wastestreams or residuals and not just to those which are
considered RCRA streams under current regulations. However, each process
study clearly differentiates between the listed RCRA "F" and "K" wastes and
other streams, and this distinction is utilized in the final analysis (see Section 3
of the main report).
3.0 COMPUTATIONAL PROCEDURES
The text below explains how qualitative ratings based on engineering judgement
were given to effectiveness, extent of current use, and future application
potential for each source control method and how they were combined to yield
qualitative estimates of the waste that has been reduced based on current
practice and the degree of reduction achievable based on potential future
practices. These estimates, or indices, were determined for each individual
method, each individual wastestream, and each individual process or study.
Following the discussion of methodology development, sample calculations are
provided.
Current Waste Reduction Index
The current waste reduction index (C) is a qualitative measure of past waste
reduction from the employment of current waste reduction measures. In other
words, it represents the reduction in waste that occurred because of measures
already implemented. The derivation of equations allowing for computation of
C from the ratings given to each waste reduction method is presented below.
-------
Consider a process which, in the past, generated waste at a specific rate Wp.
Assume that due to implementation of certain source control techniques, the
same process now generates waste at a specific rate W0. Hence, the current
overall reduction index of waste generated is:
(1) C = S (Wp-W0)/Wp
where: C = current reduction index of waste produced by the
process. Index is based on a range of 0 to S (for this
study, a range of 0 to 4 was employed).
S = the upper bound of the integer range selected (4).
Wp = past specific waste generation rate (Ibs waste/lbs
product).
V/0 = current specific waste generation rate (Ibs waste/lbs
product).
Now consider that the process has N different wastestreams and that for each
wastestream (j, where j=l,2,..N) there are M different source control methods
(i, where i=l, 2,..Mj) available. Each individual method can be characterized by
its effectiveness (ejp, its extent of current use (ujj), and its extent of further
application or application potential (pu). The first parameter, effectiveness
(ejj), is defined as the measure of volume or toxicity reduction for a specific
wastestream (j) due to full implementation of an identified source control
method (i). The second parameter, the extent of current use (ujj), is defined as
the measure of the current level of use of the particular method to control a
specific wastestream. Finally, the future application potential (pjj), is a
measure of probability that a given source control method will be implemented
to control a specific wastestream. Such a variable depends on capital and
operating cost, level of difficulty in implementation, implementation period,
technological risk, and the risk of detrimental effects on product quality.
In mathematical terms, the current reduction index of the specific wastestream
(j) by the implementation of the identified source control method (i) is given by:
(2)
-------
Now that the current reduction index for each source control method has been
calculated, the current reduction index for the wastestream (j) is given by:
(3)
c2j l-cijS + c3j
The above equation shows that the current reduction index for a wastestream is
equal to the reduction due to implementation of the first source control
measure plus the reduction of the remaining waste due to implementation of
the second source control measure and so on.
However, use of equation (3) can be unwieldly when a large number of control
methods must be analyzed. In addition, estimates of C; equal to 100 percent
reduction will occur many times after accounting for the GJJ'S of only three or
four methods. The problem is not with the equation however, but with the
relative coarseness of the rating scale used (more accurate or quantitative data
on percent effectiveness and extent of current use ratings would be required to
use this equation). Therefore, using the single most effective applied control
method (GJ; maximum) to represent the overall reduction achieved for a
wastestream greatly simplifies the estimate while keeping it conservative (the
actual reduction index should account for the most effective measure plus all of
the other source control measures). Limiting C; to a single most effective
control method is given by the equation:
(3a) C, = maximum (cji, i=l,2,...M;)
Now that C; is determined for the wastestream, the current reduction index (C)
for the process (all wastestreams) can be determined. Recalling that:
(1) C = 5 (Wp - W0)/Wp
the equation may be rewritten in terms of a single wastestream and rearranged
so that:
(4) Wpj = Woj
N
(5) Wp = Z Wpj
-------
N *
. (5a) Wp = S W0j/(S-Cj)
j=l
To obtain the current reduction index C for the whole process (all waste streams),
equations (5) and (1) are combined to yield:
N
(6) CrS-W0/E S (Woj/(S -Cj) )
j=l
Since it is usually easier to estimate the fraction of waste (Zp that a wastestream
represents as part of the total waste generated (as opposed to estimating the actual
waste generation rate), the following equations are used to calculate C:
(7) Zj = W0j/W0
(7a) Woj = W0Zj
N
(8) C = S- 1/Z (Zj/(S -Cj))
j=l
The above equation relates the estimated current extent of waste reduction for the
entire process to the current fraction Zj of each process wastestream and to the
current extent of waste reduction C; for each wastestream.
Future Waste Reduction Index
In addition to the current reduction index C, an estimation of the potential future
reduction index (F) is required. Individual future reduction indices (fjj) for each
control method can be calculated by using the equation:
(10) fjj = e-,j (S-Ujj) Pij/s2
Now that individual future reduction indices for each source control measure are
known, future reduction indices for each wastestream (F;) can be determined. As was
the case for the current reduction index, a conservative approach was used for
-------
estimating the future reduction index of each wastestream. This approach assumed
that industry would implement only a single source control method for each
wastestream and not multiple methods.
The problem faced in this calculation is that it is even more difficult to predict which
source control methods will be implemented in the future by individual facilities than
to estimate the current levels of reduction. The variable F; is subject to a high degree
of uncertainty and is dependent on a number of factors including capital and operating
costs, level of difficulty in implementation, implementation period, technological risk,
and the risk of detrimental effects on product quality. Therefore, in estimating Fj,
two different scenarios were assumed based on implementation of a single source
control measure by industry.
The first scenario (F probable) assumes that industry as a whole will implement a
range of methods. This implies that averaging of fjj's will yield a probable value of FJ:
Mj
(11) FJ (probable) = (1/Mj) I f^
j=l
This estimation of future reduction can be viewed as being a lower bound. The second
scenario (F maximum) assumes that only the single, most effective method per
wastestream will be implemented by industry. This implies that a conservative upper
bound for F; can be taken as:
(12) F; (maximum) = maximum (fjj, i = l,2,...Mj)
-------
Finally, for the entire process, the future probable and maximum waste reduction
indices were taken as weight-fraction averages: *
N
(13) Fpj-Qbabie = E Fj (probable) ZQJ
N
(14) Fmaximum = £ Fj (maximum)
Future reduction indices F, expressed by equations (13) and (14) are qualitative
measures of expected waste reduction from the process due to implementation of a
range of methods for a given wastestream (F probable) or implementation of the single
most effective control option for each wastestream (F maximum). Due to the need for
two differing scenarios, the estimate of future waste reduction is presented as a range
rather than a single value as shown for the current reduction index.
In summary, the method described above is based on an approach where the estimates
of waste reduction were derived from the qualitative ratings based on engineering
judgement given to the variables e-,;, u,;, and p;;. In actual applications however, it is
usually not possible to obtain precise fractional values of e, u, and p. Instead, the
variables were rated using an integer scale of zero to four. If this scale had to be
converted into an equivalent range of percentages, then this range would
approximately be:
Integer
0
1
2
3
4
Rating
None
Low
Medium
High
All
Ranqe, %
0-5
5-35
35-65
65-95
95-100
10
-------
Sample Calculation
<
The following steps are provided to guide the reader through the various calculations
required in determining C and F. The first step required is to list all control methods
by wastestream. Next, ratings are assigned to each method's effectiveness, extent of
current use, and future application potential using the scale previously discussed (zero
to four). In addition, fractional waste generation rates were assigned to each
wastestream. As an example, see Table 3-1.
Now that the ratings and fractional waste generation rates have been assigned,
calculations can begin. Using method #1, wastestream //I as an example, the current
reduction index for method //I is computed using equation (2):
(2) c-.j = ejjUjj/S
cn = 2 x 4/4 = 2.0
This calculation is repeated for all the methods. As a side note, the above calculation
must be slightly modified whenever a current reduction index of 4 occurs (method #2,
wastestream #2). The modification consists of restricting the maximum c value to 3.9
and not 4. This modification is required so that division by zero in equation (8) will not
occur.
Now that the individual c values have been determined, the current reduction indices
for the wastestreams are calculated using equation (3a):
(3a) C; = maximum (cji, i=l, 2,..M.)
Ci = maximum (C;Q, 021, 03^) = 2.0
G£ = maximum (cj2, C22> C32^ = '.9
Cj - maximum (0^3, 023, 033) = 1.5
11
-------
TABLE 3-1 EXAMPLE SOURCE CONTROL METHODOLOGY TABLE
1
Haste Source |
1
Hastestream II |l.
12.
13.
1
Hastestream 1? |1.
12.
13.
1
Hastestream 13 |1.
12.
13.
1
| All Sources |
L 4__
Control Methodology
Control Mthod
Control iiethod
Control «ethod
Overall
Control Mthod
Control method
Control method
Overa 1 1
Control *ethod
Control method
Control nethod
Overall
number
number
number
number
nu*ber
number
number
nurcbtr
number
1.
2.
3.
1.
2.
3.
1.
2
3.
All Methods
| Hast* | Extent of | further | Traction of | Current
| Reduction | Current Use | Application | Total Haste | Reduction
| Effectiveness j
1 2 1
1 « I
1 3 |
1 3.00 |
1 1 1
1 < 1
1 2 1
1 2.33 I
1 2 |
1 3 1
1 3 |
1 2.67 |
| Potential | |
< I
0 1
1 1
1.67 |
2 1
< 1
1 1
2.33 |
1 1
2 1
1 1
1.33 |
0 1 1
1 1 1
1 1 1
0.67 | 0.35 |
2 1 1
0 1 1
3 1 1
1.67 | 0.20 |
1 I 1
2 1 1
2 1 1
1.67 | 0.45 |
1 LOO |
ndex
2
0
0
2
0
3
0
3
0
1
0
| Future Reduction
i. .
1
| Probable
0 1
0 1
8 1
0 1
5 1
s I
5 1
9 1
5 I
5 1
8 1
1.5 |
3
» 1
0
1
0
0
0
0
1
0
0
0
1
0
0
Index |
.. i
1
| Maximum |
0 1
0 1
6 1
5 1
3 1
0 1
1 1
5 1
< 1
8 1
1 1
8 1
6 1
1
1.0 |
1
1.0 |
1
1
1.1 |
1.1 |
1
1
1.1 |
1.1 |
1.1 |
-------
and the current reduction index for the process is calculated using equation (8):
N
(8) C =S-l/Z(Zj/(S-Cp)
C =4- 1
0.35 + 0.20 + 0.45
4-2.0 4-3.9 4-1.5
~ "
0.175 * 2.0 + 0.18
To calculate the future reduction index for each method, equation (10) is used:
(10) f,j = ejj(S-u,p Pij/s2
fll = 2 x (4-4) x 0/16 = 0.0
and so on. Then, the probable and future reduction indices are calculated for each
wastestream using equations (11) and (12):
N
(11) Fj(Probable) = (1/Mj) z f,j
i = l
F1(probable) = (1/3) x (0.0 + 1.0 + 0.6) = 0.5
(12) Fj(rnaximum) = maximum (f j, i = l,2,...Mp
Fj_(maximum) = maximum (0.0, 1.0, 0.6) = 1.0
13
-------
Finally, equations (13) and (14) are used to compute the overall future reduction
indices for the process:
M
(13) F(probable) = £ Fj (probable) Z0j
j=l
F(probable) = 0.5 x 0.35 + 0.5 x 0.20 + 0.8 x 0.45 = 0.6
M
(14) F(maximum) = £ Fj (maximum) Z0j
j=l
F(maximum) = 1.0 x 0.35 + 1.1 x 0.20 + 1.1 x 0.45 = 1.1
14
-------
1. PROCESS: ACRYLONITRILE (VINYL CYANIDE) MANUFACTURE
2. SIC CODE: 2869
3. INDUSTRY DESCRIPTION
The acrylonitrile manufacturing industry comprises facilities which tend to be part of
large multiproduct corporations. In the U.S., the acrylonitrile industry is controlled
by four large producers.
3.1 Company Size Distribution
Nearly all of the world's supply of acrylonitrile is manufactured by the ammoxidation
of propylene using the Sohio Process. Plant capacities are large in order to take
advantage of economics of scale. The total U.S. production capacity of the
acrylonitrile industry in 1983 was 1,112,500 short tons. Plant capacity at the six U.S.
acrylonitrile production facilities ranges from 132,500 to 230,000 short tons.
Competition has reduced the number of acrylonitrile producers to the four companies
listed in Table 3-1.
3.2 Principal Producers
There are four major producers of acrylonitrile:
American Cyanamid Company Monsanto Co.
E.I. du Pont de Nemours & Co. Inc. Standard Oil Company of Ohio
3.3 Geographical Distribution
Of the six plant locations listed in Table 3-1, four are located in Texas. Five of the
establishments are located in EPA Region VI and one in Region V.
-------
Table 3-1 Acrylonitrile Producers in the United States
Company
American Cyanamid Co.
Avondale, LA
E.I. du Pont de Nemours &. Co.
Beaumont, TX
Monsanto Co.
Chocolate Bayou, TX
Texas City, TX
The Standard Oil Co.
Lima, OH
Green Lake, TX
TOTAL
Capacity
(short tons/yr.)
132,500
175,000
230,000
225,000
150,000
200,000
1,112,500
% of Total
11.9
15.7
20.7
20.2
13.5
18.0
100.0
Source: Chemical Economics Handbook (SRI 1982) and industry comments.
B.l-2
-------
4. PRODUCTS AND THEIR USE
The production of acrylonitrile increased rapidly during World War II due to the
demand for its use in oil and solvent resistant rubber used for tires. Later,
acrylonitrile became an integral raw material for fibers, synthetic resins, block
copolymers, elastomers, and intermediates in organic synthesis (Groet 1979). Table
4-1 contains a breakdown of the uses of acrylonitrile.
Table 4-1 1984 Acrylonitrile Products and Use Distribution
Product Percentage
Acrylic Fibers 44%
Acrylonitrile-Butadiene-Styrene (ABS) 26%
and Styrene-Acrylonitrile (SAN) resins
Adiponitrile 14%
Acrylamide 7%
Nitrile Elastomers 3.5%
Others (including materials unaccounted for) 5.5%
Source: Project staff estimates
5. RAW MATERIALS
The principal raw materials used to produce acrylonitrile are propylene, ammonia, and
air.
6. PROCESS DESCRIPTION
The Sohio ammoxidation process (see Figure 6-1) is the most significant process route
responsible for nearly 90 percent of the world's acrylonitrile production. In the Sohio
ammoxidation process, propylene, ammonia, and air are exothermically reacted in a
fluidized bed catalytic reactor at about 750-950°F and 5-30 psig. The reaction may
be represented as follows:
Bl-3
-------
PROPYLENE'
AMMONIA
[ NOTE A }
VENT 6AS
A
~ 1 ' CD
H2S04 | W
~T
r-
FLUID 1
BED |
REACTOR '
**-S j
QUENCH /
ABSORPTION
SECTION
T
AIR
*
STEAM
©@
'
VASTENATER
STRIPPIN8
( NOTE A {
CRUDE HCN
A
@
1 A
f
RECOVERY
SECTION
k
j
( NOTE A )
'
tJ tr-
ACRYLONITRILE ACRYLO -
PURIFICATION NITRILE
j. PRODUCT
T
k
}
'
1
") ,.7
^ ©
(NOTE A )
,j J
ACETONITRILE
PURIFICATION
^ ircTnuTTnn p
COPROOUCT
-». AQUEOUS
MASTE TD
UNDER8ROUND
INJECTION
NOTE A . THESE STREAMS CAN BE INCINERATED
PROCESS HASTE CATESORIESI
(T) VENT 6AS j
© AN PRODUCT COLUMN BOTTOMS
© CRUDE HCN
© SPENT CATALYST
© ABSORBER EFFLUENT
© ACE COLUMN BOTTONS
(7) MASTEVATER STRIPPER BOTTOMS
Figure 6-1 The Sohio Aaaoxidation Recess for Acrylonitrile Manufacture
Bl-4
-------
catalyst
2CH2 = CHCH3 + 2NH3 + 302 -> 2CH2 = CHCN + 6H2O + by-products
The catalyst has been undergoing continuous improvement to produce higher yields of
acrylonitrile. The early process version used a bismuth-phosphomolybdate catalyst,
followed by introduction of antimony-uranium (catalyst 21) in 1967. A ferrobismuth-
phosphomolybdate (catalyst 41) was introduced in 1972 boosting the capacity up to
35 percent. Currently, two catalysts are used: "Catalyst 49" with undisclosed
composition and antimony-tellurium catalyst manufactured by Nitto Corporation.
A once through (non-recycle) reactor operation mode with a residence time of a few
seconds is employed. Commercially recoverable quantities of acetonitrile (0.02 Ib/lb
propylene feed) and hydrocyanic acid (0.1-0.2 Ib/lb propylene feed) are produced as
by-products*. The reactor effluent is quenched and countercurrently scrubbed with a
dilute sulfuric acid solution. The vent gas, primarily consisting of nitrogen with
acrylonitrile, carbon oxides, propane, propylene, and trace amounts of acetonitrile and
ammonia, is incinerated.
The organic-rich stream from the quench/absorption section is sent to the recovery
section where crude acrylonitrile (AN) is separated and sent to the AN purification
section to yield AN product, crude HCN stream and heavy-ends (AN product column
bottoms).
Crude hydrocyanic acid (HCN) can be purified and sold or incinerated The AN product
column bottoms are typically combined with the aqueous effluent from the quench-
absorption section upstream of the wastewater stripping section (wastewater column).
In some facilities, the AN product column bottoms are incinerated**.
The crude acetonitrile from the recovery section can be further purified or incinerated
directly, depending on whether or not there is a commercial demand for it.
* E.I. du Pont de Nemours & Co. 1985: Personal communication.
**Monsanto Co. 1985: personal communication.
Rl-5
-------
The aqueous effluent from the quench-absorption section is typically comingled with
the AN product column bottoms, steam stripped in the wastewater column, sent to the
settling pond fronn which it is routed to underground injection. The wastewater
stripper overhead is recycled back to the quench-absorption sections for organics
recovery.
7. WASTE DESCRIPTION
The primary wastes evolved in the Sohio process are listed in Table 7-1. Generally,
the undesirable heavy compounds formed as the result of side reactions during
ammoxidation are separated as distillation column bottoms, combined with process
water used to quench the reactor off-gas and absorb ammonia, and disposed of via
underground injection.
Based on the information obtained, all operational facilities in the United States
dispose of streams #2, 6 and 7 via underground injection. As mentioned before,
stream #6 (ACE Column Bottoms) may not be produced by some facilities because of
incineration of the crude acetonitrile. The constituents of these streams were
reported as being not amenable to biological or chemical treatment (PRI 1977).
The disposal methods for minor wastes, such as spent catalyst, spills cleanup and
equipment cleanup wastes could not be totally ascertained, although landfilling
appears likely.
8. WASTE GENERATION RATES
The specific waste generation rates for selected waste streams from the Sohio process
were given in terms of Ib of waste/lb of product (PRI 1977) and are reported in
Table 7-1. Overall waste generation rates for the manufacturing of acrylonitrile in
the United States were unavailable. Fractional waste generation (percentage of waste
each wastestream represents as compared to all waste generated) was estimated by
the project staff based on available information and industry comments. These values
are shown in Table 9-1.
Bl-6
-------
Table 7-1 Wastes from Sohio Acrylonitrile Manufacture Process
No. Description
Process Origin
Composition (Ib/lb AN)
RCRA Codes
1.
Vent Gas
8.
9.
Quench-Absorption
Section
AN product column AN purification
Bottoms (AN Product Column)
Crude HCN
Spent Catalyst
Wastewater
Lights column
Reactor
Absorber
Acetonitrile Column ACE Purification
Bottoms Section
Wastewater Wastewater stripper
Stripper Bottoms
Spills & Leaks Process Eqpt. &. Piping
Equipment Cleaning Process Eqpt. Cleaning
Carbon Monoxide - .20
Propane - .05
Propylene - .02
Trace Organics & NH}
High rnol. wt. nitriles-
and polymers-0.002
Acrylonitrile
HCN 0.1
Light Impurities
Catalyst 49 or
Nitto
Water-1.02
Ammonium Sulfate
Acetonitrile- 0.02
Acrylomitrile
HCN, solids
Water
Acetonitrile 0.02
Heavy ends 0.003
See //2, 5 above
K012
(delisted)
K013
K014
K011
-------
9. WASTE REDUCTION THROUGH SOURCE CONTROL
This section deals with the problem of how to reduce the volumes and/or toxicities of
the Sohio acrylonitrile process waste streams. The off-gas, HCN, and crude
acetonitrile will not be considered in this report. Both HCN and crude acetonitrile are
often incinerated. The off-gases, containing substances such as acetonitrile, HCN,
propane, propylene, carbon monoxide, and acrylonitrile, are incinerated or
catalytically abated.
9.1 Description of Techniques
The summary of source control methodologies is given in Table 9-1. Sections below
deal with the description of the listed methods. In addition to the waste reduction
measures classified as being process changes or material/product substitutions, a
variety of waste reducing measures labeled as "good operating practices" has also been
included. Good operating practices are defined as being procedural or institutional
policies which result in a reduction of waste. The following items highlight the scope
of good operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
Bl-8
-------
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Aqueous Waste to Underground Injection
This principal process waste stream is mostly water which contains low concentrations
of ammonium sulfate, acrylonitrile, acetonitrile and heavier organic impurities which
are formed as a result of side reactions. Generally, waste minimization can be
addressed using four different approaches:
o Avoiding formation of undesirable byproducts in the reactor.
o Decrease of concentration of acrylonitrile (and acetonitrile) in the bottoms
of their respective purification columns.
o Segregation and destruction of concentrated wastes (AN column bottoms
and ACE column bottoms) instead of mixing them with the agueous
absorber effluent.
o Detoxification of aqueous waste.
The above measures would reduce the toxicity of the waste stream, but will not have a
large impact on its volume. The volume reductions equate to reductions in water use
and these can be addressed using the following approaches:
o Increasing concentration of sulfuric acid with the attendant decrease in
water use.
o Use of a reboiler instead of steam in the wastewater stripper.
o Evaporation of water.
While these measures may reduce the volume of the ultimate waste, their effect on
toxicity will be negative, i.e., the hazardous organics will tend to be more
concentrated thus making the entire stream more toxic. For this reason, the principal
BJ-9
-------
thrust of waste minimization should be toward toxicity reduction and the first four
approaches indicated. The discussion of the specific measures identified for each
approach is provided below.
9.1.1.1 Avoidance of Heavy Impurities Formation
This can be accomplished through the following measures:
o Increase in catalyst selectivity toward AN formation.
In the past, catalyst improvements have contributed significantly to the
increase of AN yield with attendant decrease in the yields of ACE and
heavier impurities. The research should be continued to develop more
selective and stable catalysts.
o Reduction of ammonia concentration along the reaction path.
Literature indicates that some of the impurities are formed because of the
excess ammonia present during the primary reaction (Jones 1971, Lurgi
1975). Conceptually a decrease in ammonia concentration in the reaction
zone can lower the formation rate of the undesirable byproducts. The
following avenues were identified.
staged NH3 addition e.g. using multiple ring distributors or multiple
beds.
direct quench of hot reactor offgas without heat recovery to
minimize byproduct formation in the absence of catalyst contact.
o Improved contacting between gas and catalyst in the ammoxidation
reactor.
According to present knowledge, the interchange rate of gas between the
bubble and emulsion phase is diminished, as bubble size increases due to
coalescence (Kunii & Levenspiel 1977). As a result, gas-to-catalyst
contact is decreased, along with an attendant drop in product yield and an
increase in by-product formation. It would follow that by keeping the
effective bubble size small, the ammonia residence time in the emulsion
phase (where most of the catalytic reaction takes place), will increase,
Bl-10
-------
thus reducing ammonia bypassing along with by-product formation. The
disadvantage is present in the decrease of the solid flux in the reactor with
attendant drop of heat transfer rate. This can be overcome by using more
heat transfer area or a cooler heat transfer medium. The bubble size can
be controlled by cooling coil pitch, bed aspect (length-to-diameter ratio)
and distributor design. It is suggested that consideration be given to all
means of increasing the gas-to-catalyst contact.
Rapid quench of the reactor off-gas.
The rationale behind proposing this method is that the by-product
formation rate may be curtailed by limiting the time elapsed between the
moment the reactor off-gas leaves the catalytic zone and the moment it is
cooled down sufficiently to arrest the undesirable reactions. These
reactions would occur in the interim when the gas is hot and not in the
intimate contact with selective catalyst. The following specific measures
are noted:
installation of cooling coil in the freeboard space of the reactor.
moving the cooler or venturi as close as possible to the reactor.
elimination of cyclone in favor of an integral high temperature
ceramic fiber filter to reduce freeboard space requirement.
use of cooling jacket or radiator fins on the reactor exit piping.
elimination of indirect cooling in favor of direct quench.
use of cooler with low residence time (higher heat transfer rate).
Raw material purification.
Most facilities that purchase their raw materials (as opposed to producing
them), utilize standard (99.8% purity) grade ammonia and chemical grade
propylene in the production of acrylonitrile. In principle, the generation of
certain by-products, such as tars might be reduced by using a higher grade
of feed-stock. Results from a test reactor study performed by DuPont,
however, indicated that there was no difference in tars formation between
a polymer grade and a chemical grade propylene feed*. The effect of
ammonia purity was not established, although it is doubtful that it is
significant.
* E.I. du Pont de Nemours & Co. 1985: Personal communication.
Bl-11
-------
Some of the above measures are purely suggestions which ultimately may or may not
deserve an in-depth consideration, however, it is felt that a further assessment is
warranted.
9.1.1.2 Decrease of AN and ACE concentrations in the purification column
bottoms.
This approach will reduce the amounts of both compounds discharged with the
principal waste stream. The decrease of AN or ACE concentration in the bottom
stream means an increase of concentration of heavier impurities. In principle, this can
be accomplished through additional distillation, e.g. an increase in the number of
stages in the stripping section of both AN and ACE columns or secondary recovery.
9.1.1.3 Segregation and destruction of AN and ACE column bottoms.
Presently, these streams are mixed together with the aqueous effluent from the
quench-absorption section (AN product bottoms upstream and ACE bottoms down-
stream of the wastewater column). By segregating these streams and their subsequent
incineration at high temperature, the content of hazardous compounds in the
discharged wastewater will be lessened. There are industrial precedents to
incineration of both streams.
9.1.1.4 Detoxification of the aqueous waste.
Various approaches to the detoxification problem have been reported. Alkaline
hydrolysis was in operation at Du Pont's now defunct Memphis facility, where a
biodegradable effluent was produced and routed to a municipal sewer (Lowenbach and
Schlesinger 1978). Inorganic compounds recovery from the wastewater was
considered. The process developed by Erdolchemie recovers fertilizer grade
ammonium sulfate (Groet 1979). Difficulties with metallic compound impurities
(present because of catalyst metals carryover) and with control of crystal size make
this process questionable; no one in the United States uses it.
Bl-12
-------
9.1.2 Spent and Lost Catalyst
There are two causes of catalyst loss: attrition and deactivation. Attrition involves
breakup of catalyst particles into fines which are elutriated out of the reactor ending
up as solids leaving the process via wastewater column. Deactivation appears to be
caused by loss of active metal from the catalytic sites due to volatilization (e.g. of
molybdenum).
In comparison to the wastewater, these wastestreams are minor, yet they could be
minimized by:
o Development of tougher, more attrition-resistant catalyst support.
o Lowering the frequency of high temperature upsets in the fluid bed.
9.1.3 Equipment Cleanup Wastes
Usually, the wasteloads associated with equipment cleaning are small and periodic in
nature (once every 1 or 2 years). Further reductions may be obtained through:
o Provision of more drainage time before cleanup or steamout.
o Use of non-stick (electropolished or Teflon*) heat exchanger tubes to
reduce deposit clingage (Anonymous 1985b).
o Use of in-process heat exchanger tube cleaning devices (Anonymous 1985a).
o Lower process film temperatures and increased turbulence at the heat
exchange surfaces to reduce fouling rates.
•^Registered trademark of E.I. du Pont de Nemours &. Co.
Bl-15
-------
9.1.4 Spills and Leaks
As mentioned before, spills and leaks constitute a very minor waste stream owing to
extensive implementation of preventative maintenance measures in facilities dealing
with hazardous materials. Further source reduction is possible, in principle, through
better operating practices (see practice study entitled "Good Operating Practices").
Additionally, some consideration should be given to:
o Use of bellows-sealed valves.
o Replacing single mechanical seals with double mechanical seals on pumps
or use of canned seal-less pumps,
o Use of leak detection systems and portable monitors.
o Enclosed sampling and analytical systems.
o Use of vapor-recovery systems for loading, unloading and equipment
cleaning.
9.2 Implementation Profile
Some of the identified source control options may require extensive engineering and
economic analysis before implementation. The four U.S. producers of acrylonitrile
are large organizations with excellent technical capabilities. Therefore, analyses of
technical and economic feasibility are best performed by their resident technical
staff. No process-specific source control implementation avenues have been identified
here.
9.3 Summary
Table 9-1 presents a summary of proposed source control methodologies. Each
technique described in the text was rated using the input from industrial reviewers in
three categories: effectiveness, extent of current use and future application
potential. The ratings were combined into current and future reduction indices for
each technique, waste stream and the entire process.
Bl-14
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY TOR THE ACRYLONITRILE MANUFACTURING INDUSTRY
1
| Waste Stream
1
| Aqueous Waste to
| Underground
| Injection (*)
1
1
!
1
1
1
1
| Spent and Lost
| Catalyst
1
1
I Equipment Cleanup
rjrj 1 Wastes
ij
| Spills and Leaks
1
1
1
1
1
| All Sources
1
1
1
12.
13.
l».
15.
16
IT.
18
1
It
12.
1
|2.
(3.
l<
1
1'
I?
|3
1*.
1
1
Control Methodology
Develop more selective catalyst
Reduce ammonia concentration
Improve gas/catalyst contact
Quench reactor offgas rapidly
Purify raw material feed
Decrease AN/ACE cone in col botto»s
Segregate/destroy AN/ACE col bottoms
Detoxification of aqueous waste
Overall
Develop tougher catalyst support
Minimize reactor hot zone
Overall
Increase equipment drainage time
Lower heat exchanger film temperature
Electropolish heat exchanger tubes
Use in-process H.X cleaning devices
Found Documentation ! Waste i Extent of 1 Future Fraction of | Current | Future Reduction Index
Quantity | Quality | Effectiveness 1 Potential | | Index I Probable | Maximum
21 1 | 31 2| 4 | | 1.5 | 1 5 | 15
t | 1 i t | 01 1 | I 0 0 | 03
01 0 ! 2 ! 21 3 | | 1.0 | 0 8 | |
01 0 | 2 ! 1 | 2 0.5 | 0.8 |
1 I 2 | 11 M 1 I 1 0 3 | 02 |
1 1 1 | 2 | 1 | 2 | 0.5 | 0.3 |
'I 1 | 2 | 11 3 i 051 1.1 | !
1 I 1 | 1 | 0 2| | 0.0 | 0 5 | |
0 38 | 0 83 | 1 75 1 1 00 i 2 25 | 0 90 | 1 5 | 0 7 | 1.5 I
1 | 1 | 2 | 1 1 | i 0.5 | 0.4 | 0.4 i
1 1 1 I 11 0 1 | | 00| 0.3 | |
1 00 | 1.00 | 1 50 I 0 50 | 1.00 | 0 01 | 0 5 | 0.3 | 0.4 |
'I '1 3 | 3| 2 | 1 2 3 | 0.4 | |
1 I 1 I 2 | 1 | 1 | | 0.5 | 0 4 |
21 1 | 3 | 0| 2| | 001 1 5 | 1.5 |
21 1 | 21 1 2 | 0 5 | 0.8 |
Overall I 1 50 ! t DO | 2.50 | 1 25 1 ! 75 ( 0 02 1 2 3 I 08 1.5 I
Use double mechanical seals on pumps | 1 | Ij 3 | 4 2 | 30) 00| j
Use leaK detectors | 3 | 3 I 2 | 3 | 1 | | 1.5 | 0.1 j 0 1 |
Enclosed sampling and analy systems 3 | 2| 21 3 1 | | 1 5 | 0.1 0 1 |
Use of vapor recovery systems | 2 | 2 | 4 | t | 1 | | 3.9 | 0.0 | |
Overall | 2 25 | 2 00 | 2 75 | 3.50 | 1.25 | 0 01 | 3 9 | 0 1 | 0.1 |
All Methods | 1 00 | 2 0 | 0.7 | 1.5 j
(*) These waste streams include listed T' and/or "K" RCRA wastes
-------
It appears that the currently achieved reduction of waste due to the present level of
implementation of all methods listed is characterized by the current reduction index
(CRI) of 2.0 (50 percent) on a scale of zero to four. This is a measure of how much
waste was reduced with respect to the quantity that would have been produced if none
of the methods listed were employed at their current level. High CRI value indicates
that acrylonitrile producers have reduced their waste considerably.
It also appears that the future reductions are characterized by future reduction index
of 0.7 to 1.5 on a scale of zero to four (18 and 38 percent respectively). This means
that by implementing all listed techniques to their full rated potential, the current
waste can be further reduced only to a moderate extent.
Many of the noted techniques were arrived at through engineering analysis of rather
limited available process information and do require further evaluation. Among the
techniques that appear especially worthy of further investigation (as evidenced by high
individual future reduction index) are the improvement of catalyst selectivity and
segregation and destruction of AN and ACE column bottoms.
10. PRODUCT SUBSTITUTION ALTERNATIVES
Today's major use of acrylonitrile is in the making of acrylic and modacrylic fibers.
Acrylic fibers contain about 90% acrylonitrile. These fibers are essential in the
production of various knit fabrics, carpets, blankets, draperies, upholsteries, felts
fiberfils, hairpieces, industrial filtration systems, and paint roller covers. Due to the
many uses of acrylic fiber, it's demand has been sharply increasing, thus, so has the
demand for acrylonitrile.
In 1973, limitations on propylene supplies (the raw material used for producing
acrylonitrile) began to occur. This, coupled with an increase in demand for acrylic
fibers, led to an acrylonitrile shortage. The raw material shortage affected fiber
producers in several ways. Textile mills began to place emphasis on increasing fabric
yardage by producing lighter weight fabrics and on increasing production of the more
profitable better quality fabrics (as opposed to more abundant low and medium priced
ready-to-wear garments). This precedent indicates the possibility that fabric yardage
can be extended.
Bl-16
-------
11. CONCLUSIONS
The acrylonitrile industry appears to have reduced its waste generation rates
considerably, in the range of 50 percent. The extent of future reductions of toxic
components in all wastestreams appears moderate, in the estimated range of 18 to 38
percent.
Most of the identified techniques need further evaluation. Catalyst improvement and
bottoms stream segregation and destruction appear to be most promising routes to
minimize waste. The only product substitution alternative identified was fabric
yardage extension.
12. REFERENCES
Anonymous. 1983. Hydrocarbon Processing. 62(11): 74.
. 1984. Production by the U.S. chemical industry. Chem.
Enq. News. 62(24):37.
• 1985a Chemical Engineering Progress, 81(7): 7.
. 1985b Chemical Engineering Progress 81(7): 104-5
Crumpler, G. 1980. Treatment alternatives for hazardous waste management in nine
industry groups. Washington, D.C.: U.S. Environmental Protection Agency.
Groet, L.T. 1979. Acrylonitrile. In Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed. Vol. 1, pp. 414-426. New York, N.Y.: Wiley.
Jones, H.R. 1971. Environmental control in the organic and petrochemical industries.
Park Ridge, N.J.: Noyes Data Corp.
Kunii D., Levenspicl O. 1977. Fluidization Engineering, J. Wiley and Sons Publ. Co.
Lowenbach W., Schlesinger J. 1978. Acrylonitrile manufacture: Pollutant prediction
and abatement, Mitre Technical Report MTR-7752.
Lurgi Corp. 1975. Hydrocarbon Processing. 54(11): 158-9.
MCA. 1974. Manufacturing Chemists Association. Chemical safety data sheet 50-31:
acrylonitrile. Washington, D.C.: Manufacturing Chemists Assoc.
PRI. 1977. Process Research, Inc. Alternatives for hazardous waste management in
the organic chemicals, pesticides, and explosives industries. EPA-530-SW-151C.
Cincinnati, Ohio: U.S. Environmental Protection Agency.
B1-J7
-------
Stiles, A.B. 1983. Catalyst manufacture. Vol. 14. Marcel Dekkel, Inc.
UCBSA. 1975. Maleic anhydride. Hydrocarbon Processing. 54(11): 160.
13. INDUSTRY CONTACTS
J.R. Cooper, Manager, Environmental Affairs and Occupational Health, E.I. du Pont de
Nemours & Co., Wilmington, DE.
J.M. Schroy, Fellow, Monsanto Co., St. Louis, MO.
K.R. McClain, Manager, Acrylonitrile & Field Sales, Sohio Chemical Co., Cleveland,
OH.
Bl-18
-------
1. PROCESS: AGRICULTURAL CHEMICAL FORMULATION AND APPLICATION
2. SIC CODE: 2879
3. INDUSTRY DESCRIPTION
The industry includes companies which formulate and prepare agricultural pest control
chemicals or pesticides. This includes insecticides, herbicides, and fungicides. These
products are formulated from pesticide concentrates manufactured elsewhere and are
distributed to farmers in ready-to-use form.
3.1 Company Size Distribution
The industry is comprised of roughly 330 establishments nationwide. More than 64% of
the establishments employ less than 20 people, although this group comprises only 7%
of the total work force. A small number of large companies employ the majority of
workers. Table 3-1 summarizes company size distribution.
Table 3-1 Company Size Distribution
No. of employees per facility
No.
No.
of establishments
of employees
Total
330
16,500
1-9
166
800
10-49
113
2600
50-99
23
1700
100-249
15
2200
250+
13
9400
Source: 1982 Census of Manufacturers (USDC 1985).
3.2 Principal Producers
There are clearly no major producers which dominate the industry. Some formulation
plants are owned by large pesticide manufacturers, while others are owned by
independent formulators.
3.3 Geographical Distribution
The geographical distribution of establishments in the agricultural chemicals
formulation industry is shown in Table 3-2 and Figure 3-1. Approximately 59% of
B2-1
-------
the establishrnents are located in 10 states, although no one state accounts for a major
share of the industry. Most of the establishments are located near the agricultural
areas which make use of the products.
Table 3-2 Location of Facilities by EPA Region
EPA Region
I
II
III
IV
V
VI
VII
VIII
IX
X
Number of
Establishments'3)
-
33
-
52
32
37
13
25
43
5
National
Source: 1982 Census of Manufacturers (USDC 1985)
(a) Includes only those establishments in states with 150 or more employees.
v^) The discrepancy between Table 3-1 and Table 3-2 is caused by the exclusion of
establishments in states with less than 150 employees.
4. PRODUCTS
The agricultural chemicals industry (SIC 2879) produces pesticides and other
agricultural chemicals not elsewhere classified, such as soil conditioners. This study is
mainly concerned with pesticides. In the U.S., over 600 different pesticides are
B2-2
-------
VIII
CO
1X3
I
to
o-i
2-5
6-10
11-20
21-50
Roman numerals show EPA regions
Figure 3-1 Pesticide Formulating Plants in the U.S
-------
produced (Kryeger 1983). Most pesticides can be classified as either insecticides,
herbicides, or fungicides, although many other minor classifications exist. Table 4-1
below lists the production of the major classes of pesticides. Each division is
subdivided according to chemical type.
Roughly 75% of all insecticides and herbicides, and 66% of all pesticides, are used on
agricultural cropland. The remainder are used in private homes and gardens and on
commercial and industrial property (Dillon 1981). The majority of pesticides are used
on only a few major crops. Currently, cotton, corn, and apples receive 67% of all
insecticides used in agriculture. Corn and soybeans receive 60% of the herbicides
used, and 84% of the fungicides are applied to fruits and vegetables (Dahlston 1983).
Only 48% of the total U.S. cropland is treated with pesticides (Dahlston 1983).
Table 4-1 1982 Pesticide Production in the U.S.
Product Quantity Produced
(tons per year)
Insecticidal formulations
Inorganic compounds 54,300
Organic compounds 206,750
Chlorinated hydrocarbons 18,900
Carbamates 78,400
Organophosphates 73,150
Biological (botanical, bacterial) 11,250
Other organics 25,050
Herbicide formulations
Inorganic compounds'3' N/A
Organic compounds 541,750
Phenoxy 101,400
Metal organic 9,450
Triazine 97,250
Urea, Amide, Benzoic, other organics 333,150
Fungicide formulations
Inorganic compounds^3' N/A
Organic compounds 56,250
Other pesticidal formulations
Fumigants 17,450
Defoliants and dessicants 3,500
All others^3' N/A
Source: 1982 Census of Manufacturers (USDC 1985).
(a)Data not available.
B2-4
-------
5. RAW MATERIALS
As listed below, input raw materials include the pesticide concentrates from pesticide
manufacturing plants as well as diluents and other chemical additives used in the
formulating process (Metcalf 1981):
Pesticide Concentrates
Organic/Inorganic Insecticides, herbicides, fungicides, others
Formulation and Preparation Materials
Dust Carriers Organic flours, sulfur, silicon oxide, lime, gypsum,
talc, pyrophyllite , bentonites, kaolins, attapulgite,
volcanic ash
Solvents Kerosenes, xylenes, methyl isobutyl ketone, amyl
acetate
Others Wetting and dispersing agents, masking agents,
deodorants, emulsifiers
6. PROCESS DESCRIPTION
There are two major steps in the production of pesticides for agricultural use. The
first step is the manufacturing of the pesticide concentrate from basic chemical
feedstocks including petrochemicals, inorganic acids, gases such as chlorine, and other
chemicals. This produces the pesticide, but not in a form which is ready for use. The
second major step, which is the focus of this report, is the formulation and preparation
of the pesticide for final use.
The formulation process depends on the desired form of the pesticide. Common
pesticide formulations include dusts, wettable powders, emulsions, granules, and
aerosols (USEPA 1979). Wettable powders are the most widely used formulation in
agriculture. While each of these formulations is produced differently, the major unit
B2-5
-------
operations common to them all include the dry mixing and grinding of solids, dissolving
or melting of solids into solution, and blending. A typical liquid pesticide formulation
process is shown in Figure 6-1.
The application of pesticides to agricultural crops is an entirely separate operation,
but is included here since it also results in the generation of pesticide wastes. The
three major methods of pesticide application in agriculture are spraying (of liquid
pesticides), dusting (of powders), and fumigation (of aerosols) (Metcalf 1981). Aerial
spraying accounts for roughly two-thirds of all applications of pesticide in agriculture.
Wastes are generated during the cleaning of the application equipment.
7. WASTE DESCRIPTION
The primary wastes from the formulation and application of pesticides along with their
process origin and composition are given in Table 7-1. Pesticide formulations are
generally produced in a batch mode. As a result, the same formulating equipment is
often used to produce a variety of products. This results in the need to clean the
process equipment prior to every product switch-over in order to prevent
contamination between batches. The resulting cleaning wastes account for a
substantial amount of the waste produced during pesticide formulation. The cleaning
wastes may be either aqueous or organic liquids, or solid materials such as powders or
granules, depending on the type of pesticide being formulated.
Used drums, bags, and other packaging material may contain up to several ounces of
pesticides or other raw materials, thus becoming an additional waste source. Other
wastes produced during the formulation process include spills, off-specification
batches of pesticides, and air pollution scrubber wastewater (Gruber 1975). Poor
process control often results in the production of off-specification material. This is
either disposed of as a waste, or recycled within the process and upgraded to an
acceptable quality.
The application of pesticides to crops also results in waste generation. As in the
formulation process, sprayers and other application equipment are often used to apply
more than one product. The equipment is usually cleaned between applications of
B2-6
-------
PESTICIDE
CONCENTRATE
SPECIAL
PURPOSE
ADDITIVES
MIXER
CD©©
J FILTER
1 CD
|
PACKAGING
AND
STORAGE
FINAL
PRODUCT
GO
K3
SOLVENT
STORAGE
PROCESS NASTE CATEGORIES
(T) NASTE RINSE MATER
~ HASTE CLEANING SOLVENTS
DISCARDED RAM MATERIAL CONTAINERS
PESTICIDE DUSTS
OFF- SPECIFICATION PRODUCTS
Figure 6- 1 Liquid Pesticide Formulation Process
-------
Table 7-1 Pesticide Formulating Process Wastes
No.
Waste Description
Process Origin
Composition
RCRA
Codes
03
5.
6.
Waste rinse water
Waste solvents
Leftover raw materials
Pesticide dust
Off-specification products
Scrubber water
Equipment cleaning,
spills, area washdown
Equipment cleaning,
spills
Raw material containers
Air pollution equipment
Formulating
Air pollution equipment
Dissolved organics,
suspended and dis-
solved solids.
See Section 5.
Bags, fiber drums,
steel drums with
small amounts of
residual raw material.
Pesticide dust,
inert carrier dust.
See Section 5.
See Section 5.
Dissolved organics,
suspended and dis-
solved solids.
F002
F003
-------
different chemicals to prevent cross contamination. The cleaning wastes produced are
similar to those produced in the formulation process.
Evaporation is the most popular method of wastewater disposal at formulation plants
(USEPA 1976). An evaporation system consists of a pit lined with concrete or an
impervious plastic liner, and it may be heated or have a transparent cover to keep out
rainfall. Spray irrigation has also been used to dispose of wastewater at some plants
(USEPA 1976). Spray irrigation is sometimes limited by climatic conditions,
necessitating disposal of the wastewater to a sewer during certain periods of the year.
Rinse water from the cleaning of application equipment can be applied to land. This is
a permissible practice under FIFRA and many individual farmers do it routinely.
8. WASTE GENERATION RATES
Very little data was available on specific waste generation rates from agricultural
chemical formulation and application operations. In 1975, it was estimated that
0.0033 pounds of waste per pound of pesticide formulated was produced (Gruber 1975).
This rate did not include waste from application of the pesticide. More recently, the
nationwide waste generation rate (which includes waste produced during the cleaning
of pesticide application equipment) was estimated to be over 100 million gallons per
year (Dillon 1981). These wastes were mostly very dilute (less than 500 ppm) pesticide
solutions, the large majority of which originated from the cleaning and washing of
equipment and containers. Fractional rates have been estimated by the project staff
based on best available information and engineering judgment and are shown in
Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The various wastes produced during the formulation and application of pesticides are
shown in Table 7-1. The available methods of reducing waste production through
source control can generally be grouped as either process modification methods or as
those belonging to the group known as good operating practices.
B2-9
-------
Good operating practices are defined as being procedural or institutional policies
which result in a reduction of waste. The following items highlight the scope of good
operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Equipment Cleaning Wastes
Most of the wastewater produced at pesticide formulation plants results from the
cleaning of process equipment. As noted earlier, a typical formulation plant produces
a variety of different pesticides, all on a batch basis. Between batches, the mixing
tanks and all other equipment exposed to the pesticide must be cleaned to avoid
contamination between different products.
If powders or other "dry" pesticides are formulated, then cleaning is accomplished
using a dry, inert material, such as clay. If liquid pesticides are formulated, cleaning
is normally performed by rinsing with water. More than one rinse is usually required
to adequately clean a tank. The following waste reduction methods are noted:
o Storage and reuse of rinse water and other cleaning wastes.
Many plants collect and store rinse water from equipment cleaning and
reuse it as make-up water during the next formulation of the same product
B2-10
-------
(USEPA 1976). This greatly reduces wastewater generation, in addition to
allowing for product recovery. This idea has also been used in the
formulation of solvent-based products. One plant cleaned all of their
equipment with solvent, which was then collected and reused in the next
compatible batch formulation (USEPA 1976).
If it is not practical to use the waste rinse water as make-up during a later
formulation, it can be reused as rinse water. In those instances where
more than one rinse is needed to clean the equipment, the first rinse can be
performed using old rinse water from a previous formulation. This rinse
will remove the bulk of the pesticide residue from the equipment, then a
second rinse with fresh water can be used to complete the cleaning.
o Use of high pressure spray nozzles.
High pressure spray nozzles can be used in place of the standard rinsing
hoses. According to a study of equipment cleaning in the paint industry
(USEPA 1979), water consumption can be cut by 80-90% when high
pressure rinsing systems are used.
For additional information and examples of low-waste approaches to equipment
cleaning, the reader is referred to the study of process equipment cleaning in this
appendix.
9.1.2 Spills and Area Washdowns
The cleanup of spills and area washdowns often contributes significantly to the total
waste volume produced at formulation plants. Spills are caused by the accidental
discharge of pesticides during transfer operations or from equipment failures such as
leaks. Area washdowns with water hoses are performed routinely at some formulation
plants, and are necessary in the event of contamination of the working area (USEPA
1976). Waste reduction methods available for these wastes include the use of dry
cleanup methods for spills. Rather than cleaning spills with water and discharging the
water to the sewer, many formulating plants use dry absorbents for spill cleanups
B2-11
-------
(USEPA 1976). This greatly decreases the waste volume associated with the cleanup.
In addition, floor sweeping can be used to collect spills for product reformulation.
Such practice was reportedly performed by Chevron Chemical Co. to reduce their
waste generation volume (LWVM 1985).
9.1.3 Off-Specification Products
Off-specification batches of pesticide formulations are produced as a result of poor
process control and operation. Ideally, this waste source could be eliminated totally
by making use of the following source control techniques:
o Strict quality control and process automation
The formulation of pesticides is a relatively simple process. Nevertheless,
process automation and control during formulation ensures repeatable high
quality products and avoids generation of off-spec batches due to operator
error.
o Reformulation of off-specification batches
If a batch of off-specification pesticide is produced, it should be
reformulated to an acceptable quality rather than discarded as a waste.
9.1.4 Packages and Drums
Drums or other packaging materials contribute to the waste output from a formulating
plant. After the drums, which are used to store or transport pesticides, are emptied, a
small amount of pesticide residue remains. To clean the drums, it is necessary to rinse
them with water or use an inert solid, such as clay. Thus, additional cleaning waste is
generated. If the drums and packaging material are not cleaned or decontaminated,
then they must be treated and disposed of as a waste. Some formulation plants sell
used drums to a drum reconditioner, while others reuse them internally. There is thus
a tradeoff between disposing uncleaned drums as a waste, and producing a waste by
cleaning them. From a waste reduction standpoint, both of these methods should be
considered, and the most efficient method should be chosen on a plant by plant basis.
B2-12
-------
As an example, Chevron Chemical Co. combined the use of water conservation
techniques in triple-rinsing pesticide containers to be returned to the manufacturers.
This practice reduced approximately 50% of the containers discarded. In addition,
non-returnable pesticide-containing drums were replaced by returnable bulk tote bins
to reduce the overall waste generation amount (LWVM 1985).
9.1.5 Dust From Air Pollution Equipment
Dusts generated during handling, grinding, and other formulation operations are a
potential waste source. It is common practice to install dust collection equipment,
such as hoods served by a baghouse filter, on all dust-generating operations. This
allows for the recycling of the dust, thereby reducing waste production and decreasing
worker exposure to hazardous substances (USEPA 1976). For example, Daly-Herring
Co. reportedly altered their dust collection equipment operation so that waste streams
containing organic chemicals from various production areas were collected separately
rather than mixed in a single baghouse. The uncontaminated streams collected were
recycled to the processes where they were generated (Huisingh et. al. 1985).
9.2 Implementation Profile
The major thrust of source control in the pesticide formulation process should be
towards reduction of the principal waste stream, i.e. equipment cleaning wastes. The
two methods identified are the storage and reuse of rinse water and other cleaning
wastes, and the use of high pressure spray systems. Storing cleaning wastes for future
reuse would require the establishment of a storage area with tanks, pumps, piping and
instrumentation. The installation of a high pressure spray system would require a
spraying unit, pump, and piping at each cleaning area. The space requirements for
such a system would not be large. Both of these source reduction measures would
require a capital expenditure. This expense would be offset by the savings in
wastewater treatment due to the smaller volumes of waste produced.
9.3 Summary
The waste sources and the associated source control techniques are summarized in
Table 9-1. The ratings listed in this table are based on a scale of 0 to 4 and are used
B2-13
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY FOR THE AGRICULTURAL CHEMICALS FORMULATION INDUSTRY
!
1
1
t
1
1
1
1
4
1
1
1
1
i
1
1
1
1
+
1
1
g!
si
1
+•
1
Haste Stream
Equipment Cleaning
Wastes (')
Spills and Area
Washdowis
Off-Gpec Products
Packages and Drums
Dust /Air Pollutlo
Equipment
All Sources
1
1
1
|1
12.
1
It-
12
1
I'-
|2
1
|t
1
i|t
1
1
Control Methodology
Reuse waste rlnseuater/solvent
Use high pressure sprays
Found Documentation i
Quantity I Quality |
3 I 3|
1 1 3 I
Overall i 2 00 | 3 00 I
Setter operating practices
Use of dry clean up methods
Overall
Quality control/automation
Reformulation of off-spec batches
Overall
Clean and re-use
Overall
Collect dust and reuse
Overall
All Methods
1 1 1 i
2 1 21
1.50 i 1.50 i
1 1 1 i
2 i 2 |
1 50 I 1.50 |
11 2 |
1 00 | 2 00 |
1 1 2 !
1 00 | 2 00 I
Waste 1 Extent of i Future Fra
Reduction i Current Use | Application i Tot
Effectiveness | 1 Potential
3 | 3 | 3
31 2| 3 |
3 00 | 2.50 | 3 00 I
3 | 3 i 2|
2| 21 2
2.50 | 2 50 | 2 00 |
3 | 2 i 2
31 3 i 2 |
3 00 j 2 50 | 2 00 |
3 | 1 | 3
3.00 | 1 GO | 3 00
3 | 3 | 1
3 00 | 3 00 | 1 00 |
:tion of Current future Reduction Index I
| Index Probable | Maximum 1
| 2.3 0 6 | |
1 1.5 1.1 111
0 60 1 2.3 | 08 111
i 23 0 4 | i
1 0 | 05| 051
0 05 | 2 3 | 0 4 | 051
| 1 5 | 0.8 | 03|
2 3 i 04| I
0.05 | 23| 061 001
| 08| 171 17',
0 25 | 0 8 I 1 7 | 1 7 I
| 2.3 | 0.2 | 0 2 |
0.05 2.3 | 0.2 | 021
1 00 2 0 | 1 0 | 1.2 !
(') These streams include listed "F" and/or "IT RCRA wastes
-------
to evaluate each technique for its waste reduction effectiveness, extent of current
use, and future application potential. The ratings were derived by project staff based
on a review of the available information.
It appears that the current level of waste minimization in the pesticide formulation
industry is relatively high, as evidenced by the current reduction index (CRI) of 2.0 (50
percent). CRI is the measure of waste reduction achieved due to the current
application level of all the techniques listed.
Any future reductions of waste generation appear to be moderate as characterized by
the future reduction index (FRI) of 1.0 to 1.2 (25 to 30 percent). FRI is the measure of
waste reduction achievable through implementation of the listed techniques to their
full rated potential. The most effective measures for achieving this reduction involve
reusing waste rinse water, using high pressure spray systems, using dry cleanup
methods for spills, increasing the use of automation, and reusing empty drums.
10. PRODUCT SUBSTITUTION ALTERNATIVES
The discussion has, so far, been concerned with reducing the wastes associated with
the formulation of pesticides. While this is a valid concern, this approach does not
consider the hazards associated with the pesticides themselves. The wastes produced
during the formulation process are basically dilute pesticide solutions and off-
specification pesticide formulations. These are often no more hazardous than the
pesticide formulations being produced. Therefore, since pesticides are hazardous
substances themselves, it is of major importance to evaluate safer, substitute
products.
Pesticides represent only one means by which to protect agricultural crops from
insects and other pests. Many other methods of crop protection exist which do not
rely solely on the use of man-made chemicals. Integrated Pest Management (IPM) is
an ecologically-based pest control strategy involving a variety of control methods.
Pests are defined here as organisms which interfere with the production of agricultural
crops. Pests include insects, diseases, parasites, and weeds.
82-15
-------
The main underlying principle of IPM is that no single control method can be effective
over the long term because of the adaptive powers of insects and other pests. The
goal of IPM is not to eradicate pest populations, but rather to keep them just below
the level at which the loss is deemed economically acceptable. The pest control
methods used in an IPM program include (Bottrell 1979):
o Biological control
Natural enemies such as parasites, predators, and diseases are used to
regulate and balance pest populations. For example, if no natural enemies
of a pest exist on a farm, farmers can purchase insects from insectaries
and introduce them onto the farm (Smith 1985). This form of control is of
major importance in IPM.
o Genetic control
The breeding of plant species which are resistant to certain diseases or
insects is highly effective and is used widely today. The large majority of
U.S. cropland is currently planted with crop varieties which are resistant
to, at least, one disease.
o "Cultural" control
Cultural control of pests include farming practices such as land cultivating
crop rotations, and strategic timing of planting, irrigation, and harvesting
of crops. These practices can be used to prevent the proliferation of many
pests.
o Chemical control
Chemical control using pesticides is also an important part of IPM.
Pesticides are applied only when absolutely necessary and in small amounts
in order to keep pest populations from rising above the threshold level
where they cause economic loss. The selective use of pesticides also
avoids the needless destruction of parasites, predators, and diseases that
can aid in the control of a pest population.
IPM has been used successfully on many farms to date, reducing pesticide use by
50-90% while maintaining crop yields (Metcalf 1981, Bottrell 1979). Since IPM is an
82-16
-------
ecologically-based control strategy, each farm requires the development of its own
program. While IPM has been used for the past two decades with much success, it is
not being used extensively today. Several barriers exist which have hindered its
widespread use. These include:
o Lack of knowledge
Implementing an IPM program requires extensive scientific knowledge of
the crop and its surrounding ecosystem. Growth characteristics of both the
crop and its pests, as well as their interrelationships, must be understood.
Such knowledge is often lacking and is difficult to establish.
o Lack of trained IPM personnel
Before a farmer can begin an IPM program, an IPM specialist must be
contacted. Since IPM requires extensive scientific information, the
average farmer is dependent upon the help of an expert. Such experts are
in short supply and are often unavailable. In contrast, there
o Farmers' attitudes
Farmers are often skeptical of incorporating IPM into their farming
practices. Those who have had much success with using pesticides as their
primary means of pest control see no reason to change over to IPM.
o Lack of incentives
IPM has been used successfully on many farms, drastically reducing
pesticide consumption (Metcalf 1981, Bottrell 1979). Nevertheless, it has
not yet gained widespread popularity; pesticides continue to be the primary
means of protecting many crop from pests. The major advantages of IPM
over current pesticide use are mainly environmental. There does not
appear to be a general consensus concerning the economic benefits of IPM
compared to traditional chemical methods of pest control. It is not clear
whether or not IPM is economically advantageous.
One method which has been proposed to decrease pesticide use is to lower
the appearance standards set on food (Dahlston 1983). The current
demands on such appearance qualities as color of fruits and vegetables in
B2-17
-------
the U.S. prohibits many farmers from adopting IPM, which often results in
food of lower cosmetic quality compared to food produced using pesticides.
11. CONCLUSIONS
Several methods of reducing the waste generated during the formulation of pesticides
are available, it is estimated that the total waste output could be moderately reduced
by implementing the source reduction methods discussed. The most effective methods
available for waste reduction are reusing waste rinse water and other cleaning media,
using high pressure spray systems for equipment cleaning, and incorporating dry
cleanup methods for spills.
The greatest environmental concern stemming from the use of pesticides is the
toxicity of the pesticides themselves. Concern over the wastes associated with
pesticide production is of secondary importance. This is evident from incidents in the
past concerning the contamination of water supplies with pesticides, the build up of
pesticides in the food chain, and farm workers' exposure to pesticides.
The use of pesticides on some farms has been greatly reduced by the use of Integrated
Pest Management (IPM). IPM does not eliminate the use of pesticides, but reduces
their consumption rate by placing more emphasis on natural and biological pest control
methods.
12. REFERENCES
Atkins, P.R., 1972. The pesticide manufacturing industry - current waste treatment
and disposal practices. Texas University, EPA-12020-FYE-01/72. Washington, D.C.:
U.S. Environmental Protection Agency.
Bottrell, D.R., 1979. Integrated pest management. Washington, D.C.: Council on
Environmental Quality.
Dahlston, D.L., 1983. Pesticides in an era of IPM. Environment. 25(10):45-54.
Dillon, A.P., ed., 1981. Pesticide disposal and detoxification processes and techniques.
New Jersey: Noyes Data Corp.
82-18
-------
Gruber, G.I., 1975. Assessment of industrial hazardous waste practices, organic
chemicals, pesticides and explosives industries. TRW Systems Group. EPA-530-SW-
118c. Washington, D.C.: U.S. Environmental Protection Agency.
Huisingh, D., Martin, L., Hilge, H., et al. 1985. Proven profit from pollution
prevention. Washington, D.C.: The Institute for Local Self-reliance.
Johnson, H., 1973. A study of hazardous waste materials, hazardous effects and
disposal methods. Vol. 2. Booz-Allen Applied Research, Inc. EPA-670-2-73-15.
Washington, D.C.: U.S. Environmental Protection Agency.
Kryeger, R., ed., 1983. Treatment and disposal of pesticide wastes. In ACS
Symposium Series 259.
LWVM, 1985. League of Women Voters of Massachusetts. "Waste reduction, The
untold story". Conference at the National Academy of Sciences, Conference Center
on June 19-21. Woods Hole, Mass.: Conference materials.
Metcalf, R.L., 1981. Insect control technology. In Kirk-Othmer Encyclopedia of
Chemical Technology. 3rd ed., Vo. 13. New York, N.Y.: Wiley.
Miller, G.T., 1985. Living in the environment. Belmont, Calif.: Watsworth Pub. Co.
Parsons, T.B., 1977. Industrial process profiles for environmental use. Chapter 8;
pesticide industry. Radian Corp. EPA-600-2-77-0234. Research Triangle Park, N.C.:
U.S. Environmental Protection Agency.
Pimental, D., 1981. Handbook of pest management in agriculture. CRC Press.
Plummer, J.R., 1981. Herbicides. In Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed., vol 12. New York, N.Y.: Wiley.
Smith, G., 1985. Insects help farmers to spray less, grow more. L.A. Times. August
23 issue, p.l, view section.
TRW, 1983. TRW Systems Group. Recommended methods of reduction, neutralization,
recovery, or disposal of hazardous waste. Vol. 5. EPA-670-2-73-053e. Washington,
DC: U.S. Environmental Protection Agency.
USC, 1985. U.S. Congress, Office of Technology Assessment. Pest management
strategies in crop protection. Washington, D.C.: U.S. Government Printing Office.
USDC, 1985. U.S. Department of Commerce, Bureau of the Census. Agricultural
Chemicals. In 1982 Census of Manufacturers. MC82-I-2886. Washington, D.C.: U.S.
Government Printing Office.
USEPA, 1976. U.S. Environmental Protection Agency, Office of Water and Waste
Management. Development document for final effluent limitation quidle lines for the
pesticide chemical manufacturing industry. EPA-440-l-73-060d. Washington, D.C.:
U.S. Environmental Protection Agency.
R2-19
-------
, 1979. U.S. Environmental Protection Agency, Office of Water and Waste
Management. Development document for proposed effluent guide lines, new source
performance standards, and pretreatment standard for the paint formulation point
source category. EPA-440-l-79-049b. Washington, D.C.: U.S. Environmental
Protection Agency.
13. INDUSTRY CONTACTS
Confidential source.
B2-20
-------
1. PROCESS: ELECTROPLATING
2. SIC CODES: 3471
3. INDUSTRY DESCRIPTION
Electroplating is a process by which a metal object is coated with one or more thin
layers of another metal by means of an electrical current, in order to alter the surface
characteristics of the object. The electroplating industry is generally divided into two
major segments: job shops which process materials owned by others on a contract basis
and captive shops owned by larger manufacturing facilities.
3.1 Company Size Distribution
There were an estimated 13,000 job and captive electroplating shops in the United
States (USEPA 1979). Roughly 69% of these are small captive shops which employ less
than 20 people. Table 3-1 summarizes the industry size distribution for job shops;
specific size distribution for captive shops is unknown.
Table 3-1 Company Size Distribution for U.S. Electroplating Job Shops in 1982
No. of employees per facility
Total
1-19 20-49 50-99 100+
No. of establishments 3,450 2,394 725 230 101
No. of employees^ 61,900 13,080 21,750 14,950 12,120
Source: Dunn's Marketing Services (1983), 1982 Census of Manufactures (USDC 1985)
(a) Breakdown of total was approximated from number of establishments and
corresponding company size.
R3-1
-------
3.2 Principal Producers
The industry is dominated by small captive shops with less than 20 employees each.
Thus, there are no major producers with a very large share of the entire market.
3.3 Geographical Distribution
The geographical distribution of electroplating shops is shown in Figure 3-1. As seen,
the highest concentration of shops are in California, the Midwest, and the East Coast.
The EPA regional breakdown is given below in Table 3-2.
Table 3-2 Geographical Distribution of U.S. Electroplating Industry by EPA Region
EPA Region
I
II
III
IV
V
VI
VII
VIII
IX
X
Total
Number of Establishments
343
415
137
208
1088
190
69
negligible
615
32
3103^)
Source: 1982 Census of Manufacturers ( USDC 1985).
(a' The discrepancy between Table 3-2 and Table 3-1 is caused by exclusion of states
with less than 150 employees.
4. PRODUCTS
Principal product areas in which electroplating processes are involved include:
automotive exterior/interior parts hydraulic cylinders
boat hardware shafts
plumbing fixtures bearings
cabinet hardware seals
R3-2
-------
VIII
(CO
oo
0-30
131-200
31-60
201-300
61-130
over 300
Roman numerals show EPA regions
Figure 3-1 Electroplating Plants in the U.S.
-------
Products (Continued)
kitchen appliances solar heated panels
utensils defense hardware
tools computer chassis
valve trim printed circuit boards
5. RAW MATERIALS
Raw materials used in the electroplating industry include cleaning agents, plating
solutions and metal anodes. Nationwide plating metal consumption rates are shown in
Table 5-1 for selected metals.
Table 5-1 1982 Consumption of Metals by U.S. Electroplating Industry
Metal
tin
nickel
silver
gold
cadmium
zinc
copper
Consumption
(metric tons/yr)
11,434
20,889
123
14.5
971 •
342,044^)
no data
Source: 1982 Minerals Yearbook (USDC 1982).
'a^ Includes hot dip galvanizing.
6. PROCESS DESCRIPTION
Electroplating is a process in which metal is coated with one or more other metals by
electrodeposition. Electroplating is used to increase the corrosion resistance of a
metal, to improve or alter the appearance of a metal, or to otherwise create a product
which serves some useful end.
Electrodeposition is achieved by passing an electric current through a solution
containing dissolved metal ions as well as the metal object to be plated. The metal
object becomes a cathode in an electrochemical cell, resulting in the deposition of the
P3-4
-------
dissolved metal ions onto the surface of the metal object. Ferrous or nonferrous
objects are commonly electroplated with aluminum, brass, bronze, cadmium,
chromium, copper, iron, lead, nickel, tin, and zinc. Precious metals such as gold,
platinum, and silver are also used widely.
The following processes are specifically excluded from this study since they are
covered in other studies of this report:
o Electronic Circuit Board Manufacturing
o Anodic Oxidation
o Electroless Plating
An electroplating process generally calls for moving the object to be coated
(workpiece) through a series of baths arranged in a sequence designed to produce the
desired end product. Typically, the sequence consists of cleaning, rinsing, and a
number of alternating electroplating and rinsing steps. The workpiece can be carried
on racks or in barrels. Large parts to be plated are mounted on racks which transfer
the workpiece from bath to bath. If many small parts are to be plated, they can be
contained in barrels which rotate in the plating bath. As an example, a flow chart of a
chromium plating operation is shown in Figure 6-1 (Tavlarides 1982). A number of
excellent detailed process descriptions are available (ASM 1964, Lowenheim 1979).
Electroplating produces a large variety of wastes, as discussed in Section 7. Most of
the wastes produced contain metals and other compounds used in the various plating
baths. Table 6-1 lists the constituents of the most commonly used plating baths.
7. WASTE DESCRIPTION
The ten primary electroplating process wastes and their respective points of origin and
composition are listed in Table 7-1. The wastes produced at a particular plant will be
similar to those listed, but their precise composition will depend upon the specific
process employed. Some or all of the ten waste types may be combined into a single
stream before treatment and disposal. It is common to combine the highly-
concentrated cyanide wastes from the plating and cleaning solutions with filter
sludges. These are kept separate from the acidic wastes and from the dilute cyanide
solutions.
B3-5
-------
CLEAN
WTER
FINISHED PRODUCT
i PROCESS IASTE CATESORIES:
0 SPENT ALKALINE CLEANIN* SOLUTIONS
0 SPENT ACID CLEANINS SOLUTIONS
0 SPENT CYANIDE CLEANINI SOLUTIONS
0 SPENT PLATINS SOLUTIONS
0 FILTER SLUDSES
0 HASTE RINSE HATER
(?) KASTENATER TREATNENT SLUOSE
Figure 6-1 Blocic Flo* Diagra for Chroiiui Plating Operation
R3-6
-------
Table 6-1 Common Electroplating Bath Compositions
Electroplating Bath
Composition
Brass & Bronze
Cadmium Cyanide
Cadmium Fluoroborate
Copper Cyanide
Copper Fluoroborate
Acid Copper Sulfate
Copper Pyrophosphate
Fluoride Modified
Copper Cyanide
Chromium
Chromium with
Fluoride Catalyst
Copper cyanide
Zinc cyanide
Sodium cyanide
Sodium carbonate
Ammonia
Rochelle salt
Cadmium cyanide
Cadmium oxide
Sodium cyanide
Sodium hydroxide
Cadmium fluoroborate
Fluoroboric acid
Boric acid
Ammonium fluoroborate
Licorice
Copper cyanide
Sodium cyanide
Sodium carbonate
Sodium hydroxide
Rochelle salt
Copper fluoroborate
Fluoroboric acid
Copper sulfate
Sulfuric acid
Copper pyrophosphate
Potassium hydroxide
Ammonia
Copper cyanide
Potassium cyanide
Potassium fluoride
Chromic acid
Sulfuric acid
Chromic acid
Sulfate
Fluoride
83-7
-------
Table 6-1 (continued)
Electroplating Bath
Composition
Gold Cyanide
Iron
Lead Fluoroborate
Lead-Tin
Nickel (Watts)
Nickel-Acid Fluoride
Black Nickel
Metallic gold
Potassium cyanide
Sodium phosphate
Ferrous sulfate
Ferrous chloride
Ferrous fluoroborate
Calcium chloride
Ammonium chloride
Sodium chloride
Boric acid
Lead fluoroborate
Fluoroboric acid
Boric acid
Gelatin or glue
Hydroquinone
Lead fluoroborate
Tin fluoroborate
Boric acid
Fluoroboric acid
Glue
Hydroquinone
Nickel sulfate
Nickel chloride
Nickel fluoroborate
Boric acid
Nickel sulfate
Nickel chloride
Nickel sulf amate
Boric acid
Phosphoric acid
Phosphorous acid
"Stress-reducing agents"
Hydrofluoric acid
Nickel carbonate
Citric acid
Sodium lauryl sulfate
(wetting agent)
Nickel ammonium sulfate
Nickel sulfate
Zinc sulfate
Ammonium sulfate
Sodium thiocyanate
-------
Table 6-1 (continued)
Electroplating Bath
Composition
Silver
Acid Tin
Stannate Tin
Tin-Copper Alloy
Tin-Nickel Alloy
Tin-Zinc Alloy
Acid Zinc
Zinc Cyanide
Silver cyanide
Potassium cyanide or Sodium cyanide
Potassium carbonate or Sodium carbonate
Potassium hydroxide
Potassium nitrate
Carbon disulfide
Tin fluoroborate
Fluoroboric acid
Boric acid
Stannous sulfate
Sulfuric acid
Cresol sulfonic acid
Beta naphthol
Gelatin
Sodium stannate
Sodium hydroxide
Sodium acetate
Hydrogen peroxide
Copper cyanide
Potassium stannate
Potassium cyanide
Potassium hydroxide
Rochelle salt
Stannous chloride
Nickel chloride
Ammonium bifluoride
Sodium fluoride
Hydrochloric acid
Potassium stannate
Zinc cyanide
Potassium cyanide
Potassium hydroxide
Zinc sulfate
Ammonium chloride
Aluminum sulfate or Sodium acetate
Glucose or Licorice
Zinc oxide
Sodium cyanide
Sodium hydroxide
Zinc cyanide
R3-9
-------
Table 7-1 Electroplating Process Wastes
No.
17"
Waste Description
Process Origin
Composition
RCRA
Codes
00
2.
3.
4.
5.
6.
7.
Spent alkaline cleaning
solution
Spent acid cleaning
solutions
Degreaser sludges
Solvent recycle
still bottoms
Spent plating solutions
Filter sludges
Waste rinse water
Aqueous cleaning
Acid pickling
Solvent cleaning
Solvent recycling
Electroplating
Electroplating
Drag-out, equipment
cleaning, spills
NAOH,Na2CO3,Na2SiO3, E009
(NaPO3)4, cyanide,
soils, EDTA+ Mg/Ca,
saponified and/or
emulsified grease
HC1,H2SO4,HNO3,
HBE4,Me+, oils,
soils
kerosene, naphtha,
toluene, ketones,
alcohols,ethers,
halogenated
hydrocarbons, oils,
soils, water
same as above solvents.
May contain HC1 from
solvent decomposition
see Table 6-1
Silica, silicides,
carbides, ash, plating
bath constituents
same as No. 1 and 2, but
in lower concentrations
E001, F002
F003, F005
F001, F002
F003, F005
F007
F008
8.
9.
10.
Wastewater treatment
sludge
Vent scrubber wastes
Ion exchange resin
reagents
Wastewater treatment
Vent scrubbing
Demineralization of
process water
Metal hydroxides,
sulfides, carbonates
similar to No. 7
brine, HC1, NaOH
F006
-------
Contaminated rinse water accounts for the large majority of waste produced on a
volume basis. Rinse water is used to remove the drag-out from a workpiece after it is
removed from a bath. Drag-out refers to the excess cleaning or plating solution that
adheres to the workpiece surface, and gets carried out of the solution bath upon
withdrawal of the workpiece from the bath. In general, the use of small part barrels in
the plating process (barrel plating) produces more drag-out than rack plating. This is
because a barrel carries with it more plating solution upon withdrawal from the bath
than a rack does, and because drainage of the drag-out back into the bath is more
difficult with barrels. If the drag-out from one bath is carried into the next bath in
sequence due to incomplete rinsing, it is referred to as "drag-in", and is considered a
contaminant in the later bath. Large amounts of rinse water are used on the
workpiece at several points in a typical process, as shown in Figure 6-1.
Spent cleaning and plating solutions are another waste source. Several types of
cleaning solutions are used to prepare a metal surface for electroplating. Stripping
wastes are a special type of cleaning waste. They result from the stripping off of the
old plated deposit prior to the deposition of a new metal plate. As shown in Table 7-1,
cleaning solutions may be acidic or basic, and may contain organics. Heavy metals are
usually not present, although some cleaning solutions contain cyanide. Spent plating
solutions contain high concentrations of metals. These solutions are not regularly
discarded like cleaning solutions, but may require purging if impurities build up.
Wastes produced from spills and leaks are usually present to some extent in an
electroplating process. Water is used to wash away floor spills, and the resulting
wastewater contains all of the contaminants present in the original solutions.
Wastewater is also produced from the wet scrubbing of ventilation exhaust air.
The wastewater produced in the electroplating process may contain a variety of heavy
metals and cyanide. The metals are removed by adding lime or other precipitating
agents, and precipitating under alkaline pH. The resulting metal hydroxide precipitate
forms a dilute sludge, which is thickened and then disposed of by landfilling. Figure
7-1 shows a typical electroplating wastewater treatment system.
B3-11
-------
ca
OJ
i
HEXAVALENT
CHROMIUM
I _
ACID >
S02 OH _^
NAHS04
CAUSTIC ->
CL2 OH f,
NAOCL
CHROMIUM
REDUCTION
r
CYANIDE
OXIDATION
I _J
1 ACID >
CN-
CAUSTIC >
ACID /ALKALI
WASTES
£^
"1
I I
FLOCCULANTS SOLID NASTE
DISPOSAL
n
f V
NEUTRALIZA
i ti IIRRF
"ON MIXER " ^UDGE
* blUHAbt UH
SYSTEM CLARIFIEH 1 J THICKENING
EFFLUENT
DISCHARGE
Figure 7-1 Mastenater Treatment System for Electroplating Hastes
-------
8. WASTE GENERATION RATES
Very little data were available on overall waste generation rates from the electro-
plating industry. In 1976, Battelle Columbus Laboratories reported waste generation
rates of metal hydroxides and degreaser sludges from the industry. Since these rates
are 10 years old and narrow in scope (they do not account for current industry
practices, nor do they cover all wastes generated), no annual volumetric rates could be
obtained. Fractional waste generation rates (the percentage each waste represents of
the total waste generated ) were estimated by project staff, based on the available
information and on industry comments. These values are shown in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The ten primary waste streams are presented in Table 7-1 along with their process
origin. From a source reduction standpoint, these waste streams can be grouped as
follows to reflect their common process origin:
Work cleaning wastes
Spent plating solutions and sludges
Waste rinsewater
Treatment wastes
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" have also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practices:
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
R3-33
-------
Material tracking and inventory control
Scheduling
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Work Cleaning Wastes (Nos. 1-4, Table 7-1)
Work cleaning wastes from electroplating processes are similar to the cleaning wastes
produced in many other manufacturing processes. A detailed discussion of source
reduction methods for cleaning wastes is provided in a separate practice study entitled
"Metal Parts Cleaning", Section 820.
9.1.2 Spent Plating Solutions and Sludges (Nos. 5 and 6, Table 7-1)
Plating solutions such as those listed in Table 6-1 contain high concentrations of heavy
metals, cyanides, and other toxic constituents. They are not discarded frequently, but
rather are used for long periods of time. Nevertheless, they do require periodic
replacement due to impurity build-up or the loss of solution contituents by drag-out.
When a plating solution is contaminated or exhausted, the resulting waste solution is
highly concentrated with toxic compounds and requires extensive treatment. The
source control methods available for reduction of spent plating waste include:
o Increasing the longevity of the plating solution.
The lifetime of a plating solution is limited by the accumulation of
impurities and/or by depletion of constituents due to drag-out. The
impurities come from five sources: racks, anodes, drag-in, water make-up,
and air. The impurities buildup can be limited by the following techniques:
Purer anodes.
During the plating process, metal from the anode is dissolved in the
plating solution and deposited on the cathode (workpiece). Some of
the impurities contained in the original anode matrix will stay behind
B3-14
-------
in the plating solution, eventually accumulating to prohibitive levels.
Thus, the use of purer metal for the anode will extend the plating
solution life.
Drag-in reduction by better rinsing.
Efficient rinsing of the workpiece between different plating baths
reduces the drag-in of plating solution into the next bath.
Use of deionized or distilled make-up water.
To compensate for evaporation, water is required for makeup of
plating solutions. Using deionized or distilled water is preferred over
tap water, since tap water may have a high mineral or solids content,
which can lead to impurity buildup.
Plating solution regeneration through impurity removal.
There are methods that have been successfully used to increase the
longevity of plating solutions through impurity removal. More
efficient filtering of a plating solution kept levels of impurities low
and extended solution life (McRae 1985). In cyanide baths, carbonate
tends to build up in solution over time due to CC>2 absorption from
the air. This leads to solution deterioration. Reducing the carbonate
concentration has been accomplished using a technique developed by
the U.S. Army (U.S. Pat., 4,365,481), which involves freezing the
carbonates out of solution. A metal box containing dry ice and
acetone is immersed in the plating bath. Carbonates are precipitated
directly onto the outside metal surface of the box, which is then
removed from the solution. The carbonates are scraped off the box
and discarded as solid waste, with a volume considerably smaller than
that of the sludge associated with the spent plating solution.
Proper rack design and maintenance.
Corrosion and salt buildup deposits on the rack elements will
contaminate plating solutions upon chipping and falling into the
solution. Proper design and maintenance (cleaning) will minimize this
form of contamination.
B3-15
-------
Replacement of cyanide plating solutions with cyanide-free solutions
A cyanide-zinc solution was replaced with a non-cyanide, non-chelated
alkaline zinc solution (Olsen 1973, Lowenheim 1979). This eliminated the
problem of handling cyanide-containing wastes. Other cyanide-free zinc
solutions*, along with cyanide-free pyrophosphate copper plating solutions,
have been used as replacements (Lowenheim 1979). Replacing cyanide
solutions with non-cyanide solutions, however, often requires upgrading of
the degreasing/cleaning techniques used. This is because the non-cyanide
replacements may require a much more thoroughly cleaned surface to
ensure high quality plating (USEPA 1981). The primary barrier to non-
cyanide bath use is that military contracts often specify the use of cyanide
solutions, thereby preventing electroplaters from using non-cyanide
replacements**.
Replacement of cadmium-based plating solutions with zinc solutions.
The use of cadmium has been replaced with zinc in many applications*.
See Section 10.1.1 for a detailed discussion of cadmium plating
alternatives.
Replacement of hexavalent chromium with trivalent chromium.
Trivalent chromium, which can easily be precipitated from wastewater, has
been used in place of toxic hexavalent chromium (Lowenheim 1979).
However, trivalent chromium produces a lower quality surface, and has not
seen widespread use.
Return of spent plating solution to manufacturer.
This option requires on-site segregation of solutions according to the metal
in the solution. Only a few suppliers (Harshaw, CP Chemical) reprocess
some spent bath solutions from their customers.
* Alexandria Metal Finishers 1985: Personal communication.
** National Association of Metal Finishers 1985: Personal communication.
B3-16
-------
9.1.3 Waste Rinse Water (No. 7, Table 7-1)
Waste rinse waters account for the largest fraction of waste volume produced in
electroplating processes. Any method of reducing the amount of rinse water used will
significantly reduce the total waste volume from a process. Large amounts of rinse
water are used to rinse off drag-out on a metal surface after the metal is removed
from a plating or cleaning bath. Rinse waters usually contain dilute solutions of
several toxic materials, such as cyanides and heavy metals. There are several
methods available to reduce the amount of toxicity of waste rinse water produced.
They can all be grouped into two major techniques: drag-out minimization and rinse
water minimization. It must be pointed out that reducing drag-out will result in a
decrease of the heavy metal content of the ultimate waste (treatment sludge), but the
decrease in water consumption will affect only the volume of the co-precipitated
calcium and magnesium hardness. Decreasing the amount of rinsewater without
reduction in drag-out may thus result in the smaller, but more highly toxic, volume of
treatment sludge.
9.1.3.1 Drag-out Minimization
By minimizing the amount of drag-out carried from a plating or cleaning bath to a
rinsing bath, a smaller amount of water is needed to rinse off the workpiece. Also,
less of the plating solution constituents leave the process, which ultimately produces
savings in raw materials and treatment/disposal costs. The amount of drag-out
depends on the following factors:
Surface tension of the plating solution.
A plating solution with a high surface tension tends to be retained in
the crevices and surface imperfections of the workpiece when it is
removed from the plating bath, thus increasing drag-out.
Viscosity of the plating solution.
Highly viscous solutions result in larger amounts of drag-out.
Physical shape and surface area of the workpiece.
The shape of the workpiece affects the amount of plating solution
that gets dragged out of the bath. With all other parameters
B3-17
-------
remaining the same, a larger workpiece surface area results in more
drag-out. It must be noted that barrel plating operations produce
more drag-out than rack plating.
Speed of workpiece withdrawal and drainage time.
The rate at which the workpiece is withdrawn, the time allowed for
drainage over the mother tank, as well as the orientation with which
the work is withdrawn from the bath, affect the amount of drag-out
produced.
Generally, drag-out minimization techniques include:
o Increasing plating solution temperature.
The increased temperature lowers both the viscosity and surface tension of
the solution, thus reducing drag-out. The resulting higher evaporation rate
may also inhibit the carbon dioxide absorption rate, slowing down the
carbonate formation in cyanide solutions. Unfortunately, this benefit may
be lost due to the formation of carbonate by the breakdown of cyanide at
elevated temperatures*. Additional disadvantages of this option would
include higher energy costs, higher chance for contamination due to
increased make-up requirement, and increased need for air pollution
control due to the higher evaporation rate.
o Lowering the concentration of plating bath constituents.
A decrease in the concentration of metal salts and other components of the
plating solution leads to lower solution viscosity. This results in less
dragout volume and lower metal losses. Additionally, lower concentration
will also reduce the rinsing requirement. For example, it has been found
that acceptable chromium plate can be obtained from baths containing only
25-50 g/1 CrC>3 compared to traditional concentration of 250 g/1 003
(USEPA 1981). The lower chromium concentration also results in a lower
solution viscosity, which reduces drag-out.
* National Association of Metal Finishers 1985: Personal communication
B3-J8
-------
Reducing the speed of withdrawal of workpiece from solution and allowing
ample drainage time.
The faster the workpiece is removed from the bath, the higher the drag-out
will be. The workpiece should be removed as slowly and as smoothly as
possible. Ample time should be allowed for draining the solution back to
the tank, especially for higher viscosity solutions. Usually, 30 seconds
allows most of the dragout to drain back to the tank. However, in
applications where quick drying is a problem, a 10 second draining still
permits good drag-out recovery*.
Use of surfactants.
Wetting agents have been used to lower the surface tension of plating
solutions. A solution with a high surface tension is retained in the crevices
and surface imperfections of the workpiece upon removal from the plating
bath. By reducing the surface tension, drag-out is greatly reduced.
Applied in only small amounts, wetting agents can lower a solution surface
tension enough to reduce drag-out by up to 50% (USEPA 1981). Only non-
ionic wetting agents, which will not be degraded by electrolysis in the
plating bath, should be used. The use of surfactants is sometimes limited
due to their adverse effect on the quality of the plate produced**.
Proper positioning of the workpiece on the plating rack.
When a workpiece is lifted out of a plating solution on a rack, some of the
excess solution on its surface (drag-out) will drop back into the bath.
Proper positioning of the workpiece on a rack will facilitate the dripping of
the drag-out back into the bath. The position of any object which will
minimize the carry-over of drag-out is best determined experimentally,
although the following guidelines were found to be effective (USEPA 1981):
Orient the surface as close to vertical as possible.
Rack with the longer dimension of the workpiece horizontal.
Rack with the lower edge tilted from the horizontal so that the
runoff is from a corner rather than an entire edge.
* Westinghouse Electric Corporation 1985: Personal communication.
**National Association of Metal Finishers.
83-19
-------
By redesigning racks for better drainage and reducing the metal concentra-
tion in the plating bath, drag-out was reduced at one plant by 55%. This, in
turn, greatly decreased rinse water requirements (Anonymous 1985a).
o Improved drag-out recovery.
A drain board positioned between a plating bath and rinse bath can capture
the dripping solution off of a workpiece and route it back to the plating
bath. This is a simple operation which has been incorporated into many
electroplating processes. The drain board can be made of either plastic or
metal. For acidic solutions, it has been suggested that drain boards be
made of vinyl chloride, polypropylene, polyethylene, or Teflon*-lined steel
(USEPA 1981). Another commonly practiced option is to incorporate a drip
tank between the plating bath and the rinsing bath. The drip tank is an
empty tank for collecting the dripping solution, which can be returned to
the plating bath.
9.1.3.2 Rinse Water Minimization
Another method of waste rinse water reduction is to rinse off the workpiece in the
most efficient manner, using the smallest volume of rinse water. Traditionally, a
workpiece would be immersed into a single rinsing bath following a plating bath, and
then moved on to the next step in the process. Several methods exist which use less
rinse water than the traditional rinsing method, while still adequately rinsing the
workpiece. These include:
o Multiple rinsing tanks.
The use of multiple rinsing tanks is one of the most commonly used source
reduction techniques. Virtually all new electroplating plants are designed
with multiple rinsing tanks, which can reduce rinse water requirements by
66%**' with possible theoretical reductions of over 90% reported (USEPA
1983, USEPA 1981, Olsen 1973). Multiple rinsing tanks are aligned either
in series or in parallel. In a three-tank rinsing system, a workpiece is
moved from a plating solution to the first rinse tank, then onto the second
rinse tank, then to the third, and finally onto the next step in the process.
* National Association of Metal Finishers 1985: Personal communication.
** Registered trademark of E. I. Du Pont.
B3-20
-------
In the parallel arrangement, each rinsing tank is individually fed with fresh
rinsing water. The more common series arrangement differs in that it is a
counterflow system. In a three tank counter-current system, the
workpiece enters the first rinse tank, which has the most contaminated
rinse water. It is then moved to the second tank, and then to the last,
where it contacts fresh rinse water. Fresh rinsewater enters only the last
(third) rinsing tank. The water from the third tank then flows into the
second tank, then into the first tank from which it can be routed into
treatment and/or to the plating tank as a make-up.
Fog nozzles and sprays.
Spraying water droplets directly onto a workpiece is much more efficient
than immersing a workpiece into a liquid water bath. The only limitation is
that spraying is not effective on oddly-shaped objects, since the spray
cannot make direct contact with the entire surface of the object. For
simple workpieces, such as sheets, it is highly effective.
A variation on the spray nozzle is the fog nozzle. A fog nozzle uses water
and air pressure to produce a fine mist. Much less water is used than with
a conventional spray nozzle. It is possible to use a fog nozzle directly over
a heated plating bath to rinse the workpiece. This allows for simultaneous
rinsing and replenishment of the evaporated losses from the tank. Spray or
fog rinsing is used on rack plating, which represents a third of all plating
operations. Nozzles are not applicable for barrel plating (two thirds of all
plating operations) because of the odd shape of the parts *.
Rinse water reuse.
By using the same rinse water stream at more than one step in the process,
the total amount of waste rinse water produced is drastically reduced.
After a rinse water is used once, it picks up contaminants from the
workpiece that was rinsed. If these contaminants do not interfere with the
quality of a subsequent plating step, the same water can be used again.
^or example, in a nickel plating process, the same rinse water stream
* Baxter and Wardman Eng. 1985: Personal communication.
33-2.1
-------
was used for the rinses following the alkaline cleaning, acid dip, and nickel
plating tanks. Instead of having three different rinse streams, only one
stream was used, greatly reducing the overall rinse water requirements
(USEPA 1983).
o Still rinsing.
Installing a still (or dead) rinsing tank immediately after a plating bath
allows for metal recovery and lowered rinse water requirements. In such a
system, the workpiece is immersed in a still rinse tank following the
plating bath. Since the still rinse has no inflow or outflow of water, the
concentrations of the plating bath constituents build up in it. When the
concentration becomes sufficiently high, the contents of the still bath are
used to replenish the upstream plating bath.
o Automatic flow controls.
The lowest possible rinse water flow rate which can efficiently rinse a
workpiece can be determined for all systems. This flow then can 'be
automatically controlled to avoid variations associated with water line
pressure changes and manual control by operator.
i
o Rinse bath agitation.
Agitating a rinsing bath mechanically or with air increases the rinsing
efficiency and cuts down on water demand.
9.1.4 Treatment Wastes (Nos. 8-10, Table 7-1)
In electroplating, toxic metal sludges result from the conventional treatment
processes used to remove metals from aqueous wastes. Metals are usually precipitated
as hydroxides or carbonates, resulting in a sludge which requires further treatment and
disposal. Lime is commonly used as the precipitating agent. The volume and toxicity
of the sludge produced can be lowered by reducing the metal content in the plating and
rinse waste waters, or by using different precipitating agents. Methods available to
accomplish this include:
o Use of different precipitating agents.
Normally, hexavalent chromium in waste rinse water or plating solutions is
treated by being reduced to trivalent chromium with a reducing agent,
B3-22
-------
followed by precipitation with lime. In one instance, sodium hydroxide was
used in place of lime (Chacey 1983). Lime precipitation produced 2.24 dry
solids/lb Cr (VI), while sodium hydroxide precipitation produced only 1.98 Ib
dry solids/lb Cr (VI).
Use of trivalent chromium instead of hexavalent chromium for plating.
One operation reported a 70% reduction in sludge production when
trivalent chromium was used for plating instead of hexavalent chromium
because of avoided necessity to precipitate gypsum associated with the
excess sulfate ions that would have been introduced during reduction step.
(Anonymous 1985a).
More efficient sludge dewatering.
The volume of sludge produced can be greatly reduced through the use of
new dewatering technologies which remove a greater percentage of water
than traditional dewatering techniques (Anonymous 1985b).
Better operating practices (stream segregation).
The type of waste requiring treatment can be controlled to some extent
through waste stream segregation. By isolating cyanide-containing waste
streams from waste streams containing iron or complexing agents, the
formation of cyanide-complexes is avoided, and treatment made much
easier (Dowd 1985). Segregation of wastewater streams containing
different metals also allows for metals recovery or reuse. For example, by
treating nickel-plating wastewater separately from other waste streams, a
nickel hydroxide sludge is produced which can be reused to produce fresh
nickel plating solutions. In one instance, the scrubber waste from a
chromium plating bath was segregated and could then be returned to the
bath. This resulted in less discarded waste and increased the longevity of
the plating solution.
Metal recovery techniques.
Decreasing the heavy metal content of wastewater can be accomplished
through drag-out minimization and also through recovery of these metals
prior to treatment. By recovering the metals from the rinse waters, the
toxicity and volume of the treatment sludge will be reduced, if not
B3-23
-------
eliminated. In addition, better raw material utilization will help to offset
higher operating costs. The metal recovery techniques examined include:
Evaporation
Reverse osmosis
Ion exchange
Electrolytic metal recovery
Electrodialysis
Use of separate treatments for individual solutions
Evaporation - Waste rinse water is evaporated by heating, leaving behind a
concentrated solution. The equipment used includes single or multiple
effect evaporators. Also, vapor recompression applications were reported
(Seaburg and Bacchetti 1982). The solution is concentrated until its metal
concentration is equal to that of the plating bath, then this solution is
reused. Using this method, 90-99% efficient metal recoveries can be
achieved (Clark 1984). Depending on the design, the evaporated water
vapor can either be condensed and re-used as rinse water, or it can be
vented off into the atmosphere (Campbell and Glenn 1982). Evaporation
has been used frequently for chromium recovery. One plant was able to
recover 8,000 Ibs of chromium per month, resulting in savings of
$100,000/yr, with a one-year return on investment (Campbell and Glenn
1982). It has also been proven effective for the recovery of gold, nickel,
copper, and cadmium (Kohl and Triplett 1984). Evaporation is best
established of all the metal recovery techniques used in electroplating.
Although it is the most energy intensive recovery technique, its simplicity
and reliability make it an attractive option for metal recovery. In order
for evaporation to be economical, multiple counter-current rinse tanks or
spray/fog rinsing should be used to minimize the amount of rinse water
being processed (MDEM 1984).
Apart from the energy cost, a distinct disadvantage is that the
concentrates may also contain the calcium and magnesium salts originally
present in the rinsewater. Adding them to the plating solution may result
in more rapid deterioration. This problem is alleviated in situations where
rinsewater is de-ionized or softened prior to use.
B3-24
-------
Reverse Osmosis - In reverse osmosis, a semi-permeable membrane is used
to concentrate a dilute waste rinse water stream. The dilute waste is
applied with high pressure to a membrane which allows the water to pass
through, but retains the metals and other solutes present. Thus, on one
side of the membrane, a concentrated metal solution is produced which can
be returned to the plating bath; on the other side of the membrane, pure
water is obtained which can be reused as rinse water.
Reverse osmosis has been used most successfully for nickel recovery,
although it is also used to recover copper, zinc, and cadmium (Campbell
and Glenn 1982, Kohl and Triplett 1984). It has not been extensively used
for chromium recovery (Campbell and Glenn 1982). Reverse osmosis is less
energy intensive than evaporation, but the characteristics of the
membranes available limits the type of waste streams that can be treated.
For example, only very dilute streams can be treated, and the solution
must be pre-filtered to extend membrane life (MDEM 1984).
Ion Exchange - Ion exchange involves exchanging one ion from a solution
with another ion in order to recover certain ions (such as metal ions) or to
purify the solution. This occurs by passing a solution over an ion exchange
resin which has the capacity of exchanging one of its own ions for an ion in
solution. Once a resin has reached its capacity in terms of ions exchanged,
it must be regenerated. Regeneration is usually accomplished using an acid
or a base, depending on the type of resin. The acid or base removes the
accumulated metal from the resin. Another step may be necessary to
remove the metal from the acid or base before the metal can be reused.
Ion exchange is often used for purification of rinse water, rather than
simple metal recovery. Ion exchange has been effective for recovery of
nickel, chromium, cyanide, gold, silver, and other metals (Kohl and Triplett
1984).
Ion exchange is a relatively complicated and delicate process which
demands a high level of process control and maintenance. For this reason,
as well as for other technical reasons such as regeneration problems, ion
exchange is not currently a popular method for metal recovery.
B3-25
-------
Electrolytic Metal Recovery - In electrolytic metal recovery, metal ions in
solution are reduced electrochemically onto a cathode surface within the
solution. When the cathode becomes fully coated with the metal, it is
removed from the solution and placed into a plating bath as an anode,
replenishing the bath with the metal. Up to 99% of the metal in waste
rinse water can be recovered using this method (Campbell and Glenn 1982).
One great advantage of the electrolytic method over other metal recovery
techniques is that it recovers only the plating metal, not the impurities,
from the waste rinse water. Electrolytic metal recovery is most efficient
on concentrated solutions. For solutions with less than 100 mg/1 of the
metal ion, low current efficiencies limit the process effectiveness.
Electrolytic recovery systems have been used to recover copper, tin, gold,
silver, cadmium, and other metals (Campbell and Glenn 1982, Kohl and
Triplett 1984). In one application, the metal was recovered directly from a
cyanide - destruct tank*.
Electrodialysis - In electrodialysis, an electric current and selective
membranes are employed to separate the positive and negative ions from a
solution into two streams. This is accomplished by feeding a solution
through a series of alternating cation and anion selective membranes,
through which a current is passed. Electrodialysis is used mainly to
concentrate dilute solutions of salts or metal ions. Electrodialysis has been
used to remove nickel, copper, cyanide, chromium, iron and zinc from
waste rinse water (MDEM 1984, Kohl and Triplett 1984). This technology
has not been used as widely in the electroplating industry as have other
metal recovery techniques (Campbell and Glenn 1982, Kohl and Triplett
1984).
Use of Separate Treatments - Use of separate treatments for individual
solutions results in a sludge that bears a single metal. The sludge (metal
hydroxide) can then be sold, e.g., to a chemical producer.
* National Association of Metal Finishers 1985: Personal communication.
B3-26
-------
9.2 Implementation Profile
The use of multiple rinse tanks, often with counterflow arrangements, is currently one
of the most effective and widely used methods of waste reduction available. Most
large plating operations, in fact, are built with multiple rinse tanks. Smaller shops,
however, often use only a single rinse following immersion in cleaning and plating
solutions.
The installation of a second and/or third rinse tank following a plating bath drastically
reduces rinse water consumption, as discussed earlier. It also involves substantial
process modifications. Since all of the baths and mechanical equipment in most
electroplating shops are built in a series type of arrangement, the inclusion of
additional tanks may require some changes in the shop lay-out. Besides the tanks
themselves, piping, pumps, flow controllers, and racks for transporting the workpieces
will need to be added. In some instances, the availability of space may be the
determining factor in the use of multiple rinse tanks.
In smaller electroplating shops, rinsewater tanks may be over-sized for their specific
applications. Instead of installing a second tank, it is often possible to simply divide
the existing tank in half by constructing a wall (Dowd 1985). This effectively converts
the single large tank into two smaller tanks. The additional piping needed is minimal.
A second waste reduction method with a high implementation potential is the reuse of
rinse water. Electroplating operations use rinse water at several stages in the process,
and it is often possible to use the same stream at more than one step.
Adding rinsing steps to an existing operation would necessitate modification of the
piping system. The main technical problem with this technique is that the quality of
the product must be monitored carefully. A rinse stream can be used a second time
only if the contaminants from the first rinse do not interfere with the quality of the
second rinse. Monitoring the rinse water composition and the quality of the workpiece
after each rinse might be required.
B3-27
-------
As hazardous waste treatment and disposal costs rise, there is an incentive to
implement waste reduction methods for the purpose of achieving cost reductions.
Some of the waste reduction techniques discussed here involve minimal capital
outlays. Examples of these would be good operating practices and rinse water reuse,
which often entail minor equipment changes and/or simple changes in procedure.
Smaller shops (roughly two-thirds of the industry) may find these methods practicable,
but may be capital-constrained in the implementation of more extensive
modifications. It has been shown that the use of multiple rinse tanks and other rinse
water conservation methods, which vary considerably from one to the other in
expense, can reduce waste treatment costs by $10,000 -$99,000 per year (David 1985).
9.3 Summary
The waste sources and their respective source control technologies are summarized in
Table 9-1. The ratings listed in this table are based on a scale of 0-4.0 and are used to
evaluate each technique for its waste reduction effectiveness, extent of current use,
and further application potential. The ratings shown are based on a review of the
available literature, as well as the opinions of industry personnel.
The most effective waste reduction methods include the use of multiple rinse tanks,
the installation of drain boards and drip tanks, the reuse of rinse water, and the
installation of metal recovery systems. Since waste rinse water accounts for the
majority of the volume of waste produced, the methods which decrease rinse water
consumption offer the greatest potential for waste reduction.
Based on the measures currently undertaken to reduce waste, a current reduction
index of 1.8 (45 percent) has already been achieved by the electroplating industry. By
implementing additional waste reduction methods or increasing the use of existing
measures, the current quantity of waste produced could be reduced to the level
corresponding to a future reduction index of 0.8 to 1.9 (20 to 48 percent). The above
values mean that currently employed measures have reduced waste generation by 45
percent from the levels that otherwise would exist, and that future measures could
reduce wastes by 20 to 48 percent from the current waste generation levels.
B3-28
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY FOR THE ELECTROPLATING INDUSTRY
m 1
OL>
10
Naste Stream
Cleaning
r ] j. lnnc * 4 \
Spent Plating
Solutions and
Sludges (•)
Waste Rinse Hater
Treatment
Hastes («)
All Sources
i
I
i
1
II-
12
13.
I*
i
n.
12
13.
|4
|5
16
IT
|8
1'
|10
111
|12
1
|1.
!2
13.
!'•
15.
1
1
| Found Documentation i
I Quantity |
See study on Metal Parts Cleaning | — '
Overall ! — |
Increase plating solution life | 1 \
Use non-cyanioe olat'ng solutions | 1 |
Replace cadmium plating with zinc | 1 I
Replace hexavalent chromium with tri | 1 |
Overall j 1 00 I
Increase solution temperature | 1 |
Use less concentrated plating soln | 1 1
Withdraw workpiece slowly from soln. | 2 1
Add wetting agents to plating soln. | 2 '
Position workpiece properly on rack | 3 |
Recover drag-out of plating solutions] 2 |
Install multiple rnse tanks | 3 |
Install fog nozzles and sprays ) 2 |
Reuse rinse water elsewhere in plant | t |
.Install still rinsing tanks | 1 |
.Install automatic flow controls | 2 |
.Use mechanical/ai" agitation of bath | 1 |
Overall | 1 75 |
Use efficient precipitating agents | 1 |
Use trivalent instead of hex chromium) 1 I
Install sludge dewatering systems | 1 |
Implement better operating practices | 1 |
Install metal recovery systems | 3 |
Overall | 1.10 |
All Methods
i
Quality |
—
— 1
1 1
2 1
2 1
2 1
t 75 |
2 1
2 i
2 1
1 1
3 1
1 1
3 |
2 1
2 1
1 I
1 1
1 i
1.75 |
2 1
2 1
2 1
2 1
3 1
2 20 |
haste | Extent of | Future | Fraction o' | Current
Reduction | Current Use | Application 1 Total Waste | Reduction
Effectiveness |
— 1
— 1
' 1
2 1
2 1
3 1
2.00 |
1 1
3 1
2 1
1 1
2 1
2 1
« 1
3 i
< 1
2 1
3 1
1 1
2.33 |
2 1
3 1
• 1
3 1
3 1
3 00 |
| Potential |
— ! — I
— 1 — 1
1 | 2 |
1 1 2|
1 ! 2|
' 1 1 1
1.00 | 1.75 |
2 | 1 |
1 | 2 |
'I 3 1
1 | 2 |
3 I 2|
2 | 3 |
2 1 2 |
21 31
1 1 3 |
2 ! 3|
2 1 2 |
2 | 1 !
1.75 | 2 25 |
1 2 |
1 ' 1
1 ' 1
1 2 |
1 3 |
1 00 | 1 80 |
| Index
I 2
0 10 I 2
! o
i °
1 o
1 o
0 IS | 0
1 0
0
0
1 o
1 1
1 1
1 2
1 1
1 '
1 1
| 1
I 0
0 65 i 2
0
1 O
1 .1
0
0
0 10 1
1 00 | 1
1
1-
1
0 1
o !
1 !
1 !
1 1
2 I
2 I
5 1
8 i
5 1
3 1
5 1
0 1
0 1
5 1
0 1
0 1
5 1
5 1
0 1
5 I
8 1
0 1
8 1
8 '
0 1
8 !
Future Reduction Index '
Probable
1
1
0
0
0
0
0
0
t
1
0
0
0
1
t
2
0
0
0
0
0
0
0
1
1
1
0
1
2 1
2 1
4 I
8 1
8 1
6 1
6 1
1 1
1 1
1 1
» 1
3 I
8 1
0 1
1 !
3 I
8 1
8 1
i i
8 1
* I
6 1
8 i
1 1
7 I
o !
8 1
1
Maximum |
1 9 '
1.9 !
,
0.8 |
0 8 1
1
0 8 |
1
I
1
1
1
1
1
2 3 |
1
1
1
2 3 !
i
1
1
i
1 7 1
1 1 '
1 9 1
(*) These waste streams include listed "F" and/or "K" RCRA wastes
-------
10. WASTE REDUCTION THROUGH PRODUCT SUBSTITUTION
This study has so far examined methods of reducing the amount of waste produced in
electroplating operations through process modifications or other changes. It is also
possible to reduce or eliminate the generation of certain wastes through product
substitution. A complete investigation of product substitutions has not been
attempted; rather, two possible product substitutions are briefly discussed below.
10.1 Cadmium Plating Alternatives
Cadmium is used in a wide variety of products for its excellent protective properties.
Cadmium-plated products are highly resistant to corrosion in land and marine
environments. For this reason, the U.S. military specifies cadmium plating for a large
variety of naval and aerospace equipment*. Roughly 35-40% of the total amount of
cadmium consumed in the country is used by the U.S. military**. Unfortunately,
cadmium is extremely toxic and there is concern over the introduction of soluble
cadmium salts into the environment.
It may be possible in some instances to replace cadmium plating with other materials
such as:
o Zinc using plating (zinc graphite has been considered an alternative to
cadmium).
o Titanium dioxide using vapor deposition.
o Aluminum using ion vapor deposition (Ivodizing).
None of the above-mentioned coatings have exactly the same properties as cadmium,
but nonetheless may prove to be satisfactory substitutes. Aluminum ion vapor
deposition is a very attractive process, but is considerably more expensive than
electroplating*.
* Cadmium Council 1985: Personal communication.
** National Association of Metal Finishers 1985: Personal communication.
B3-30
-------
The elimination of cadmium from industrial use has already been attempted. In 1979,
Sweden began the first stage of a ban on the use of cadmium in electroplating and in
various pigments. It was hoped that this would lower the concentration of cadmium
found in the environment over time. The ban has not achieved this end, and Sweden is
considering removing the current regulations on the use of cadmium*.
The elimination of the use of cadmium is not being endorsed by the electroplating
industry. The implementation of such a change would require major modifications to
many electroplating shops, and could result in the closing down of some shops
altogether.
10.1.2 Chromium Plating Alternatives
Chromium is used to plate a multitude of products ranging from automobile parts to
paper clips. In some instances, the chromium plate is required for its hardness and
durability. In other cases, it is used purely for decorative reasons.
Because there is a substantial amount of waste produced during a chromium plating
operation,'the elimination of any unnecessary use of chromium would be beneficial
from the environmental standpoint. For example, chromium-plated automobile
bumpers could be replaced with nickel-plated bumpers, although customer preference
for a shinier finish may play a major role. Of course, the substitute product must have
less waste associated with its production than does the original product. Many
automobile bumpers are currently being painted rather than plated during finishing
operations*.
As with cadmium, industry does not seem to be promoting the elimination of chromium
use on certain products. The economic reasons that could justify such a change today
are not present. Since the majority of electroplating operations are small and employ
less than 20 people, (Dun's 1982, USDC 1985) the capital needed to implement major
process changes is often not available.
General Motors Corp., 1985: Personal communication.
33-31
-------
11. CONCLUSIONS
To reduce the problems of waste treatment and disposal in the electroplating industry,
many methods of waste reduction through source control have been identified. It is
estimated that the total waste output from electroplating can be further reduced by
20 to 48% by implementing or expanding the use of the source reduction methods
discussed. The most efficient methods available for waste reduction are using
multiple rinse tanks, installing drain boards and drip tanks, and reusing rinse water.
The electroplating industry already appears to have implemented waste reduction
measures to a fairly high degree, as evidenced by the current waste reduction index
value of 1.8 (45 percent).
The technology associated with the waste reduction methods discussed is readily
available and well understood. Economic constraints have been the major barrier to
implementation.
12. REFERENCES
Anonymous, 1985a. Finishers face increasing environmental pressure. Plat. Met.
Finish. 72 (4): 36-9.
1985b. Plat. Surf. Finish. 72 (4): 20-4.
ASM, 1964. American Society for Metals. Metals Handbook. Metals Park, Ohio:
American Society for Metals.
BCL, 1976. Battele Columbus Lab. Assessment of industrial hazardous waste
practices; Electroplating and metal finishing industries job shops. EPA - 530 -SW -
136C. Washington, D.C.: U.S. Environmental Protection Agency.
Campbell, M.E, and Glenn, W.M., 1982. Proven profit from pollution prevention.
Toronto, Canada: The Pollution Probe Foundation.
CDHS, 1984. State of California, Department of Health Services. Second Biennial
Report. Alternative technologies for recycling and treatment of hazardous wastes.
Chacey, K., et. al. 1983. Chrome electroplating waste BAT. Poll. Enq. 15 (4): 20-3.
Clark, R., ed. 1984. Massachusetts hazardous waste source reduction. Conference
proceedings, October 17, 1984. Boston, Mass.: Massachusetts Department of
Environmental Management.
Dowd, P. 1985. Conserving water and segregating waste streams. Plat. Surf. Finish.
72 (5): 104-8.
Dunn's Marketing Services, 1983.
B3-32
-------
Hsu,G.F. 1984. Zinc-nickel alloy plating: an alternative to cadmium. Plat. Surf.
Finish. 71 (4): 52-5.
Kohl, J., and Triplett, B. 1984. Managing and minimizing hazardous waste metal
sludges. North Carolina State University.
Lowenheim, F.A. 1979. Electroplating in Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed. Vol. 8, pp. 826-69. New York, N.Y.: Wiley.
McRae, G.F. 1985. In-process waste reduction: part 1. Plat. Surf. Finish. 72 (6): 14.
MDEM 1984, Massachusetts Department of Environmental Management, Massachusetts
hazardous waste source reduction: metallic waste session. Conference proceedings
May 23, 1984. Boston, Mass.: Massachusetts Department of Environmental
Management.
Miller, G.T. 1985. Living in the environment; Wadsworth Publ. Co.
Olsen, A.E. 1973. Upgrading metal finishing facilities to reduce pollution. Oxy Metal
Finishing Corp., EPA-625-3-73-002, Washington, D.C. : U.S. Environmental
Protection Agency.
Radimsky, J., Piacentini, R., and Deibler, P. 1983. Recycling and/or treatment
capacity for hazardous waste containing cyanides. Staff report of the Department of
Health Services, State of California March, 1983.
Seaburg, J.L., and Bacchetti, J.A, 1982, Chemical Processing 45 (12): 30-31.
Tavlarides, L.L. 1982, Process modifications toward minimization of environmental
pollutants in the chemical processing industry. Chicago, 111.: Illinois Institute of
Technology.
USDC. 1982, U.S. Department of Commerce, 1982 Mineral Yearbook. Washington,
D.C.: U.S. Government Printing Office.
. 1985. U.S. Department of Commerce, Bureau of the Census
Electroplating. 1982 Census of manufactures. Washington, D.C.: U.S. Government
Printing Office.
USEPA. 1979. U.S. Environmental Protection Agency. Development document for
existing sound pretreatment standards for the electroplating point source category.
EPA-440-1-79-003. Washington, D.C.: U.S. Environmental Protection Agency.
. 1981. U.S. Environmental Protection Agency, Industrial Environmental
Research Lab. Inplant changes for metal finishers. Cincinnati, Ohio: U.S.
Environmental Protection Agency.
. 1983. U.S. Environmental Protection Agency, Office of Water Regulations
and Standards. Development document for effluent limitation guidelines and standards
for the metal finishing point source category. EPA-440-1-83-091. Washington, D.C. :
U.S. Environmental Protection Agency.
VERSAR, Inc. 1984. Technical assessment of treatment alternatives for wastes
containing metals and/or cyanides. Versar, Inc. Contract no. 68-03-3149, final draft
report for U.S. Environmental Protection Agency. Springfield, Va. : Versar, Inc.
83-33
-------
WSDE. 1985. Washington State, Department of Ecology. Progress report, priority
waste management study for Washington State hazardous waste. Washington State:
Department of Ecology.
U.S. Pat 4,365,481. To the U.S. Army.
13. INDUSTRY CONTACTS
Art Pierdon, Past president of the American Electroplater Society, Alexandria Metal
Finishers, Fairfax, VA.
Patrick Dowd, Baxter and Woodman Engineers, Crystal Lake, IL.
Technical Representative, Cadmium Council, New York, NY.
Jeff Jolicoeur, Chemical Consultants, City of Industry, CA.
Joseph P. Chu, Environmental Activities Staff, General Motors Technical Center,
Warren, MI.
David Anzures, National Association of Metal Finishers, San Fernando, CA.
Walter G. Vaux, Principal Engineer, Chemical and Process Engineering, Westinghouse
Electric Corp, Pittsburgh, PA.
B3-34
-------
1. PROCESS: EPICHLOROHYDRIN MANUFACTURE
2. SIC CODE: 2869
3. INDUSTRY DESCRIPTION
The epichlorohydrin manufacturing industry comprises firms engaged in both the
production and use of epichlorohydrin. The establishments are part of large
multiproduct corporations.
3.1 Company Size Distribution
In the U.S., crude and refined epichlorohydrin is currently produced at three locations
by only two companies. The total U.S. production capacity of the epichlorohydrin
industry in 1982 was 290,000 short tons. An epichlorohydrin facility employs
approximately 50-60 persons; the work force includes production workers (operators),
supervisory personnel, and maintenance personnel (Bales 1978).
3.2 Principal Producers
There are two major producers of epichlorohydrin in the United States:
Dow Chemical Company
Shell Chemical Company
3.3 Geographical Distribution
Of the three epichlorohydrin manufacturing facilities in the United States, two are in
Texas, and one is in Louisiana. All are located in EPA region VI. See Table 3-1 for a
listing of company locations.
B4-1
-------
Table 3-1 Epichlorohydrin Producers in the United States
Company
Production
tons/year
Capacity
tons/year
Percent
of Total
Dow Chemical USA
Freeport, Texas
Shell Chemical Co.
Deerpark, Texas
Norco, LA
123,000
33,500
33,500
210,000
80,000
incl.
72
28
incl.
Total
190,000
290,000
100.0
Source: Chemical Economics Handbook (SRI 1982) and industry comments.
4. PRODUCTS AND THEIR USE
Based on 1982 statistics, approximately 19% of the crude epichlorohydrin produced in
the U.S. was used to make synthetic glycerin. Glycerin is used in the making of alkyd
resins, explosives, polyether polyois, cosmetics and drugs, cellophane, food and
beverages, and in other miscellaneous applications. Approximately 62% of the
epichlorohydrin (after further refining) was used in the making of unmodified epoxy
resins. These resins are used for making protective coatings, reinforced plastics, and
adhesives. About 2.3% of the refined epichlorohydrin was used to make
epichlorohydrin elastomers. The remaining 14% was used for making a variety of
other miscellaneous products including glycidyl ethers, some types of modified epoxy
resins, wet strength resins for the paper industry, water treatment resins, surfactants,
and ion exchange resins. Table 4-1 presents a breakdown of epichlorohydrin uses.
Table 4-1 1982 Epichlorohydrin Products and Use Distribution
Epichlorohydrin Use
Breakdown
Percent
Glycerin
Unmodified epoxy resins
Epichlorohydrin elastomers
Miscellaneous
18.6
62.0
2.3
14.1
Total
97.0(a)
Source: Chemical Economics Handbook (SRI 1982).
a) Some epichlorohydrin is lost during the refining process.
B4-2
-------
5. RAW MATERIALS
The raw materials used to produce epichiorohydrin are shown in Table 5-1. Both Dow
and Shell Chemical make epichiorohydrin from an allyl chloride intermediate
synthesized from propylene and chlorine.
Table 5-1 Raw Materials Used by the Epichiorohydrin Manufacturing Industry
Raw Materials _ Usage _
Ib/lb product
Propylene Not Found
Chlorine Not Found
Allyl Chloride .997
Chlorine .9025
Slaked Lime 1.009
Source: Assessment of Hazardous Waste Practices (Gruber 1975).
6. PROCESS DESCRIPTION
Currently, both producers of epichiorohydrin use a process involving the use of allyl
chloride as an intermediate reactant. Simplified process flow diagrams are depicted
in Figures 6-1 and 6-2. The first step is the production of allyl chloride from
propylene and chlorine:
H2C = CHCH3 + Cl2 - — H2C = CHCH2C1 + HC1
The byproducts include 1,2-dichloropropane and isomers of 1,3-dichloropropene.
The second step is chlorohydroxylation of allyl chloride to dichlorohydrin (C3H60C1):
C1CH2 - CH = CH2 + HOC1 --
C1CH2 - CHC1 - CH2OH 1,2 - dichlorohydrin
The byproduct formed in this reaction is 1,2,3 - trichloropropane.
The third reaction step is the conversion of dichlorohydrin to epichiorohydrin:
2 C3 H6 O CL2 + Ca(OH)2 --72C1CH2 - CH - CH2 + CaCl2 + 2H2O
or \ / or
2NaOH O 2NaCl
B4-3
-------
CHLORINE
o
VAPORIZER
PROPYLENE
fi
DRYERS
PROPYLENE
PROPYLENE PREHEATER
RECYCLE
PROPYLENE
HCL
SEPARATOR
PHE-
FRACTIONATOR
V
LI6HT
ENDS
COLUNN
Y
X!
REACTOR
COOLER
X!
REACTOR
COOLER
ALLYL CHLORIDE
PURIFICATION
COLUNN
r
j PROCESS HASTE CATEGORIES!
(?) HCL BY-PRODUCT
(?) LIBHT ENDS
(T) HEAVY ENDS
Figure 6-1 Allyl Chloride Synthesis froi Propylene and Chlorine
B4-4
-------
CL
ALLYL CHLORIDE
PROCESS NASTE CATEGORIES!
©
©
®
OFF 6AS
SALT BRINE BOTTOMS
HEAVY ENDS
STEAM
STRIPPER
HATER AND
DICHLOROHYORIN
TO RECYCLE
EPICHLORO-
HYORIN
TRICHLORO-
1 PROPANE
SOLVENT
TO RECYCLE
FRACTIONATOR
Figure 6-2 Epichlopohydrin Production fro« Allyl Chloride
84-5
-------
In the first step, excess propylene and chlorine gas are continuously fed into an
adiabatic non-catalytic reactor at 930-950°F and 30 psia to produce allyl chloride and
HC1. The reactor effluent is cooled to about 120°F and sent to a pre-fractionator to
separate HC1 and propylene. The overhead stream gas stream is subsequently split at
cryogenic conditions; propylene is recycled to the propylene preheater and HC1 is
taken off for other uses in the integrated halogenated hydrocarbon facility.
After initial removal of HC1 and propylene, the organic chloride fraction is separated
in a two-step distillation. Low boiling constituents are taken overhead in the first
column. These include various saturated chlorides such as 2-dichloropropane. The
heavy-boiling fraction which is taken off as a bottom product in the second column, is
made up largely of unsaturated dichlorides and other allyl chloride degradation
products such as benzene, tars, and carbon (DeBenedictis 1979). This fraction, rich in
1,3 - dichloropropene, is used as a feedstock to the process of manufacturing
fumigants or nematicide formulations, such as D-D (Shell Chemical Co.) or Telone II
(Dow Chemical)
The allyl chloride overhead stream from the second column is fed continuously to a
stirred tank where it reacts in the liquid phase (at atmospheric pressure and 85-105°F)
with a solution of hypochlorous acid. The hypochlorous acid is produced in a packed
tower by dissolving chlorine in water. The reaction tank effluent is fed to a gravity
separator: the upper layer (aqueous phase) is recycled to the hypochlorous acid tower.
The underflow, chiefly dichlorohydrins, is fed to the second reactor, where virtually
complete conversion to epichlorohydrin by reaction with aqueous NaOH or lime takes
place. Trichloropropane, one of the reaction co-products is used as a solvent for the
epichlorohydrin. The effluent from the second reactor is steam stripped, removing
epichlorohydrin as the water azeotrope. The undercut (calcium chloride or sodium
chloride solution and excess lime in suspension) is sent to by-product recovery. The
distillates, water, and organic phases are separated, with the undercut fed to a
fractionating tower for recovery of epichlorohydrin and solvent. The epichlorohydrin
is purified and sent to storage. The recovered trichloropropane solvent is recycled to
the product reactor. Some process configurations, however, do not incorporate this
recycle stream.
B4-6
-------
The small amount of dichlorohydrins carried over from the steam stripper in the
epichlorohydrin-water azeotrope is mostly discharged in the water phase from the
second separator. This aqueous dichlorohydrin is recycled. The heavy ends, discharged
as still bottoms from the fractionator, contain some epichlorohydrin, 5 percent
dichlorohydrin, 7 percent chloroethers, 35 percent trichloropropane, and 50 percent
heavies or tars (Gruber 1975).
7. WASTE DESCRIPTION
The wastes evolved in the epichlorohydrin manufacturing process come primarily from
the off-streams or bottom streams of the various recovery steps. The specific RCRA
codes and waste stream compositions are given in Table 7-1. Most of the by-products
from the process have found application as feedstocks to other processes. For
example, light ends from the allyl chloride synthesis step are oxidized to recover HC1
and heat of combustion. Therefore, the label "waste stream" would not apply to those
facilities which reprocess the light ends in the above fashion. Sections below give
descriptions of treatment and disposal practices found in the literature or through
contacts with the industry*.
o HC1 Byproduct from Allyl Chloride Synthesis.
In an integrated chlorohydrocarbon facility, HC1 is recovered, e.g., by
routing it to the oxychlorination unit associated with the vinyl chloride
monomer production.
o Light Ends
The light ends from the allyl chloride synthesis step are oxidized to recover
HC1 and heat.
o Heavy Ends
The heavy ends from the allyl chloride purification column contain 1,3 -
dichloropropylene, an active ingredient of fumigant or nematicide
formulations. Hence, this stream is often used as a feedstock to a process
to manufacture such formulations. The remainder is either landfilled or
oxidized for HC1 or heat recovery. The heavy ends from the
* Dow Chemical Co. 1985: Personal communication
B4-7
-------
Table 7-1 Epichlorohydrin Process Effluent Streams
CO
I
oo
No.
1.
2.
3.
4.
5.
6.
7.
8.
Waste/Residual Process Origin
HC1 By-Product HC1 Separator
Light Ends Light Ends Column
Heavy Ends Allyl Chloride
Purification
Column
Epichlorohydrin
Fractionator
AC Reactor Tars Allyl Chloride Reactor
Off Gas Tail Gas Absorber
•
Epoxidation Reactor
Salt Brine Bottoms Salt Stripper
Equipment Cleaning Sludge Cleaning of Columns, Tanks
and Heat Exchangers
Spills and Leaks
Composition^3) RCRA
(Ib/lb epichlorohydrin) Codes
HC1 (1 Ib/lb allyl
chloride)
N/A
N/A
Heavy Ends
Chloroethers
Epichlorohydrin
Dichlorohydrin
Trichloropropane
N/A
Allyl Chloride
Chlorine
HC1
Allyl Chloride
Epichlorohydrin
Trichloropropane
Chlorine
HC1
CaCl2 or NaCl
N/A
N/A
—
—
—
.053 K017
.0074
.001
.0057
.037
F024
.002
Trace
Trace
.002
.0015
.0005
Trace
Trace
.600
F024
—
(a) Source: Assessment of Industrial Hazardous Waste Practices (Gruber 1975) for composition data.
-------
epichlorohydrin fractionator are either landfilled, burned to recover HC1 or
energy or used as feedstock to other processes. They contain high
concentrations of 1,2,3-trichloropropane, a byproduct of chlorohydroxyla-
tion reaction.
o Salt Brines
Salt brines from the epichlorohydrin stripper leave the process via bio-
oxidation treatment units to remove organics. The water solution of salts
(CaCl2 or NaCl) is not considered hazardous.
o Miscellaneous Process Equipment Cleaning Wastes
The cleanup sludges from heat exchanger tube cleaning and steamout oils
from columns or tanks are generated infrequently, (at 1 to 2 year
intervals). The main waste stream in this category is generated by
cleaning of the allyl chloride reactor, which occurs presently every
2 months*. The current disposal practice at all sites is not known,
although one facility reported that these wastes are recycled to the
process.
o Spills and leaks
Because of the toxicity of epichlorohydrin and its process intermediates,
the potential leak sources, such as valve packings, pump seals, and flange
gaskets are routinely inspected and maintained. Consequently, leakage is
minimal (Bales 1978).
Epichlorohydrin is shipped in drums, tank cars, or via pipeline. Epichlorohydrin is
pumped directly from storage tanks and introduced into the drums through a
retractable pipe with a cut off valve immediately above the drum. During the filling
operation, the drum opening is surrounded by a hinged ventilation hood. Exhaust is
discharged to the atmosphere following scrubbing (Bales 1978).
8. WASTE GENERATION RATES
The composition of selected waste streams from the production of epichlorohydrin
were given in terms of Ib. of waste/lb. of product (Gruber 1975), and are reported in
Dow Chemical Co. 1985: Personal communication.
B4-9
-------
Table 7-1. Overall waste generation rates for the whole industry, however, could not
be determined due to the lack of sufficient data. Fractional waste generation (the
percentage each waste represents of the total waste generated) was estimated based
on available information and industry comments. These values are shown in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
This section deals with the problem of how to reduce the volume and/or toxicity of the
epichlorohydrin production process waste streams. Off-gas and HC1 generation will
not be considered in this report. HC1 has various industrial uses and is a salable by-
product; the off-gases, containing relatively minute amounts of substances such as
allyl chloride, chlorine, and epichlorohydrin are flared.
9.1 Description of Techniques
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
procedural or institutional changes which result in a reduction of waste. The following
items highlight the scope of good operating practices:
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency preparedness
B4-10
-------
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified. A
summary of the source control methodology is given in Table 9-1. Sections below
describe the listed methods.
9.1.1 Heavy and Light Ends from Allyl Chloride Synthesis Step
The reduction of the volumes of these effluent streams will be considered as waste
minimization only to the extent that these streams are still land disposed. In the
facilities which currently process these streams, e.g., recover HC1 following thermal
oxidation, source reduction methods may be viewed as either not applicable or as
means to increase the allyl chloride yield.
Formation of heavy and light ends is governed by side reactions. In principle, the
desired substitution reaction to form allyl chloride (~ 570° F) is accompanied by: a
low temperature (<400°F) addition chlorination reaction, an unsaturated
monochloride isomer formation, an unsaturated dichloride formation by further
chlorination of allyl chloride, and by thermal degradation (>1100° F) to tars, carbon,
and benzene. The following source reduction methods were considered:
o Alternate reactor design.
Present reactor design encompasses the features of a strongly backmixed
plug flow reactor. A thorough review of the kinetic data for the reactions
may reveal that the current degree of mixing can be improved with an
attendant yield increase for allyl chloride. At one facility, reactor design
modifications to provide better mixing resulted in a drastic decrease in the
tars formation rate*. In evaluating different modification schemes,
consideration should be given to staged addition of chlorine. Also, the
concept of a fluidized bed reactor with inert solids (e.g., sand) appears
worthy of consideration because of excellent mixing and thermal stability
characteristics.
* Dow Chemical Co. 1985: Personal communication
B4-1]
-------
o Preheat of chlorine feed.
It is postulated that curtailment of the low temperature addition reaction
can be achieved through preheating of the chlorine feed (e.g., utilizing
reactor effluent waste heat) prior to mixing with hot propylene. The
preheat temperature of the propylene can be consequently decreased.
o Thermal oxidation with HC1 and heat recovery.
This efficient technique is extensively used in the integrated
chlorohydrocarbon facilities. Energy recovery results in a lowered steam
requirement, leading to lower boiler blowdown rates and associated
wasteloads. Recovered HC1 can be routed to oxychlorination units serving
other processes (such as VCM or PCE/TCE). This technique may be viewed
both as recycling or as source control, depending on the process boundary.
o Purified feed stock.
In allyl chloride manufacturing, the propylene feed impurities usually
encountered are water and propane (De Benedictis 1979). Water may react
with chlorinated hydrocarbons to form undesirable by-products and may
also provide a corrosive environment. Propane and other hydrocarbons are
chlorinated to undesirable saturated chlorides. The level of feedstock
purification required to offset the undesired byproduct formation can be
worth re-assessing in light of increasing waste generation costs.
9.1.2 Heavy Ends from Epichlorohydrin Production
Heavy ends come from the epichlorohydrin fractionator and originate in the
chlorohydroxylation reactor. A decrease in the volume of heavies production could be
achieved through implementation of the following:
o Thermal oxidation with HC1 and heat recovery.
Currently this technique is extensively used in the highly integrated
chlorinated hydrocarbon facilities.
B4-12
-------
o Alternate process routes.
The alternate method for producing epichlorohydrin starts with the
chlorination of acrolein to give 2,3-dichloroproprionaldehyde (Furman,
Fiach, and Hearn 1958). Reduction of the aldehyde with sec-butyl alcohol
(using aluminum sec-butoxide as a catalyst) yields glycerol and
dichlorohydrin. Dehydrochlorination with lime then forms epichlorohydrin
as is currently done. Epichlorohydrin can also be prepared by the
epoxidation of allyl chloride with peracids (Phillips and Starchier 1957,
Kellaetal 1974, Kleemann et al. 1971), perborates (Bruenie and Crenne
1965), or by epoxidation with tert-butyl hydroperoxide over vanadium
(Sakan, Sano, and Hattori 1970), tungsten (Sakan, Sano, and Hattori 1970),
or molybdenum compounds (Sakan, Sano, and Hattori 1970, Oshin,
Shakhovtseva, and Krasotkina 1975) or by oxidation with air over a cobalt
catalyst (Mokrousia, Oshin, and Tregar 1976). One route relying on
chlorination of allyl alcohol to dichlorohydrin followed by epoxidation was
patented in Japan. These alternate process routes may have proven to be
economically unfavorable in the past. However, advancing technology and
recent increases in disposal costs could warrant new investigations of
economic feasibility for some of these alternate routes for new grass roots
plants.
o Reevaluation of chlorohydroxylation reaction kinetics and reactor designs.
Clear understanding of how byproducts (such as 1,2,3 -trichloropropane) are
formed in the chlorohydroxylation reactor is necessary to examine low-
waste alternative designs of the reactor (i.e., the designs which minimize
byproduct formation). Such study may be warranted in light of increasing
disposal costs.
9.1.3 Allyl Chloride Reactor Tars
The formation of these tars can be reduced by alteration of the reactor design, as
discussed in Section 9.1.1. Provision for more mixing to eliminate stagnant areas in
84-13
-------
the reactor resulted in the decrease of tar cleanup frequency from 10 days to 2 months
at one facility*. The tars can be combusted for heat and HC1 recovery in certain
incinerator designs.
9.1.4 Miscellaneous Equipment Cleanup Wastes
Usually, the wasteloads associated with equipment cleaning are small and periodic in
nature (once every 1 or 2 years). Further reductions may be obtained through:
o Complete drainage of process piping or equipment prior to cleaning.
o Use of non-stick (electropolished or Teflon**) heat exchanger tubes to
reduce deposit clingage.
o Use of in-process heat exchanger tube cleaning devices (Anonymous 1985a).
o Lower process film temperatures and increased turbulence at the heat
exchanger surfaces to reduce fouling rates.
All the suggestions listed above will have only a minor impact on overall waste
generation, since equipment cleaning wastes are but a small fraction of the total
waste. The reader is further referred to the separate study of process equipment
cleaning in this appendix.
9.1.5 Spills and Leaks
As mentioned before, spills and leaks constitute a rather minor waste stream because
of the extensive level of preventive maintenance practiced. Further source reduction
is possible, in principle, through better operating practices (see separate process study
entitled "Good Operating Practices"). Additionally, some consideration should be
given to:
o Replacing single mechanical seals with double mechanical seals on pumps
or using canned seal-less pumps.
* Dow Chemical Co. 1985: Personal communication
** Registered trademark of E.I. Du Pont.
B4-14
-------
o Using leak detection systems and portable monitors.
o Enclosed sampling and analytical systems.
9.2 Implementation Profile
The identified source control options require considerable engineering and economic
analyses before implementation. The two U.S. producers of epichlorohydrin are both
large organizations with excellent technical capabilities. Therefore, analyses of
technical and economic feasibility are best performed by their resident technical
staff. No process-specific source control implementation avenues have been
identified.
9.3 Summary
Table 9-1 represents a summary of proposed source control methodologies for the
epichlorohydrin manufacturing industry. Each method was rated for its effectiveness,
extent of current use and future application potential. Based on these ratings, a
current reduction index of 3.1 on a scale of 0-4 (78 percent) was derived, which is
indicative of the high level of waste reductions already achieved by the
epichlorohydrin industry. (The current reduction index represents the amount of waste
that was reduced compared to the waste volume that would currently be generated
without all the measures listed and their current level of application.) It appears that
by implementing additional waste reduction measures, the amount of waste currently
being generated can be further reduced to the level characterized by a future
reduction index of 0.7 to 0.9 (18 to 22 percent reduction from current waste
generation levels). Potentially, the most effective measures for achieving reductions
in waste are those that are characterized by high value of the future reduction index.
As shown in Table 9-1, these include: use of an alternate allyl chloride reactor design,
further application of burning the chlorinated waste for HC1 and heat recovery,
additional purification of propylene feedstock, reevaluation of chlorohydroxylation
reactor design and use of non-stick heat exchanger tubes.
B4-15
-------
TABLE 1-1 SINURY OF SOURCE CONTROL NETHOOOL06Y FOR THE EPICHLOROHYORIN MNUFACTURIN6 INDUSTRY
03
f
I—"
ON
Hast* Stream
Heavy/Light Ends
Ally! Chloride
Production (')
Htavy Ends
Eplchlorohydrln
Production (')
Tirs
Ally! Chloride
Reactor (')
Equ1p«mt Cleaning
Hastes (»)
Spllli and Leaks
All Sources
(
1
1
1
I'-
|2-
|3-
!<•
1
|1.
12-
13.
1
1'.
12.
1
I'-
ll-
|3.
l«-
1
!'•
|2.
13.
l<.
1
1
• + --
1
«**«_*.__! M.*-k.J« 1 «_,, 1
Found Documentation j
| Quantity | Quality | Effe
Alternate reactor design |
Preheat the chlorine feed |
Recover the HCl/heatlng value |
Purify feedstock |
Overall |
Recover the HCl/heatlng value |
Use alternate process route |
Re-evaluate chlorohydroxylatlon |
Overall |
Alternate reactor design |
Recover the HCI/heattng value |
Overall |
Increase equipment drainage tlie |
Lower heat exchanger file twperature|
Non-stick heat exchanger tubes |
Use In-process H.X. cleaning devices |
Overall |
Use double Mchanlcal seals on puips |
Use leak detectors |
Enclosed saipllng and analy. systees |
Use vapor recovery systeis |
Overall |
All Methods
*
2 1
0 1
2 1
1 1
1.25 |
2 I
3 I
0 I
1.67 |
2 I
2 I
2.00 |
1 1
1 1
2 1
2 1
1.SO |
1 I
3 I
3 I
2 I
2.25 |
A
2 I
0 I
2 I
1 1
1.25 |
2 I
2 I
0 I
1.33 |
2 1
2 1
2.00 |
1 1
1 1
1 1
1 1
1.00 |
1 1
3 1
2 1
2 1
2.00 |
1
Waste | Extent of | Future | Fraction of | C
duct Ion | Current Use | Application | Total Haste | Re
ictlveness | | Potential | |
3 1
1 1
4 1
2 1
2.50 |
4 1
1 1
2 1
2.33 |
3 1
< 1
3.50 |
3 1
2 1
3 1
2 1
2.50 |
3 1
2 1
2 1
4 1
2.75 |
2 1
0 1
3 1
1 I
I 50 |
3 I
0 I
0 I
1.00 |
2 1
3 1
2.50 |
3 1
1 1
0 1
1 I
1.25 |
4 1
3 1
3 1
< 1
3.50 |
2 1 1
1 1 1
3 1 1
2 1 1
2.00 | 0.35 |
3 1 1
' 1 1
2 1 1
2.00 | 0.35 |
2 1 1
3 1 1
2.50 | 0.24 |
2 1 1
1 1 1
2 1 1
2 1 1
1.75 | 0.05 |
2 I I
1 1 1
' 1 1
1 1 1
1.25 | 0.01 |
1 LOO |
A A
jrrent | Future Reduction
Index |
Index | Probable | Maxlftue |
1.5 |
0.0 |
3.0 |
0.5 |
3.0 |
3.0 |
0.0 |
0.0 |
3.0 |
1.5 |
3.0 |
3.0 |
2.3 |
0.5 |
0.0 |
0.5 |
2.3 |
3.0 |
1.5 |
1.5 |
3.9 |
3.9 |
3.1 |
A
0.8 |
0.3 |
0.8 |
0.8 |
0.6 |
0.8 |
0.3 |
1.0 |
0.7 |
0.6 |
0.8 |
0.8 |
0.4 |
0.4 |
1.5 |
o.e |
0.8 |
0.0 |
0.1 |
0.1 |
0.0 |
0.1 |
0.7 i
0.8 |
1
0.8 |
0.8 |
0.8 |
1
1
1.0 |
1.0 |
0.8 |
0.8 |
0.8 |
1
1
1.5 |
1
1.5 |
1
0.1 |
0.1 |
1
0.1 |
O.S |
(*) These waste streais Include listed T and/or 'K1 RCRA wastes.
-------
10. PRODUCT SUBSTITUTION ALTERNATIVES
There does not appear to be any viable alternative for epoxy resins, a principal
consumer of epichlorohydrin. Epoxies have uniquely desirable physical properties, such
adhesiveness combined with toughness and resistance to chemical attack. It would be
difficult to find an epoxy substitute of comparable quality.
Approximately 19% of the crude epichlorohydrin produced is used to make synthetic
glycerol. Glycerol has wide applications, as listed in Section 4. The alternative to
synthetic glycerol is the glycerol obtained as a by-product during the production of
soaps from animal and vegetable fats and oils (a principal source of glycerol prior to
1948). Alternatively, sorbitol offers a viable substitute for glycerol in pharmaceutical
and cosmetic applications.
11. CONCLUSIONS
As is the case with other organic chemical processes, -the epichlorohydrin industry
appears to have reduced their wastes considerably by implementing the source control
techniques noted. However, it is also apparent that moderate further reductions are
possible. Based on the available informatipn, it appears that the following methods
deserve further consideration:
o Alternate allyl chloride reactor design to provide more mixing.
o Combustion of waste with attendant HC1 and heat recovery.
o Use of purer feedstock.
o Reevaluation of kinetics and design for chlorohydroxylation reactor.
o Use of non-stick heat exchanger tubes.
The identified product substitution alternatives were natural glycerol from vegetable
fats and oils instead of from epichlorohydrin, and use of sorbitol in place of glycerol.
B4-17
-------
12. REFERENCES
Anonymous. 1985a Chemical Engineering Progress, 81(7):7.
. 1985b Chemical Engineering Progress, 81(7):104-5.
Bates, R.E. 1978. Epichlorohydrin manufacture and use. Industrial Hygiene Survey.
NIOSH.
Buienie, J.C., and Crenne, N. 1965. Fr. Pat. 1,447,267 (June 1, 1965) to Societe
Chimique des Usines Rone-Poulenc.
Brzezicki, A., et al. 1984. A mathematical model of epichlorohydrin synthesis.
Inzynieria Chemiczna Procesowa. 5(2): 201-14.
DeBenedictis, A. 1979. Allyl chloride. In Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed. vol. 5, pp. 763-769. New York, N.Y.: Wiley.
Faith, W.L., Keyes, D.B., and Clark, R.L. 1975. Industrial Chemistry, 4th ed. New
York, N.Y.: Wiley.
Furman, K.E., Fiach, H., and Hearne, G.W. 1958. U.S. Pat. 2,860,140 (Nov. 11, 1958).
to Shell Development Co.
Gruber, G.I. 1975. Assessment of industrial hazardous waste practices, organic
chemicals, pesticides and explosives industries. TRW Systems Group. EPA-530-SW-
118C. Washington, D.C.: U.S. Environmental Protection Agency.
Henderson, J.B., and McKay, N.H. 1952. Paper presented at an annual meeting of the
American Chemical Society, Southwest Region, Little Rock, Ark.
Keller, R., et al. 1974. U.S. Pat. 3,799,949 (March 26, 1974) to Degussa.
Kleemann, A., et al. 1971. OLS (Ger. Pat. Disci.) 1,942,557 (March 18, 1971). To
Degussa.
Mokrousua, I.Y., Oshin, L.A., and Tregard, Y.A. 1976. Kinet. Katal. 17(2):515.
Oshin, L.A., Shakhovtseva, G.A., and Krasotkina, B.E. 1975. Neftekhimija. 15:281.
Phillips, B., and Starchier, P.S. 1957. Brit. Pat. 784,620. (Oct. 9, 1957). To Union
Carbide Corp.
Sakan, S., Sano, M., and Hattori, K. 1970. Jpn. Pat. 7,017,645 (June 18,1970). To
Japanese Chemical Industries.
SRI. 1982. Stanford Research Institute. Chemical Economics Handbook, 1982. Menlo
Park, Calif.: Stanford Research Institute.
Steen, D.E. 1960. U.S. Pat. 2,966,525 (Dec. 27 1960). To Monsanto Chemical Co.
B4-18
-------
USDC. 1985. U.S. Department of Commerce, Bureau of the Census. Industrial
organic chemicals. In 1982 Census of manufacturers. MC82-I-28F. Washington D.C.:
U.S. Government Printing Office.
13. INDUSTRY CONTACTS
S.L. Arnold, Manager, Environmental Information Clearinghouse, Dow Chemical Co.
Midland, MI.
B4-19
-------
-------
1. PROCESS: INORGANIC PIGMENTS MANUFACTURE
2. SIC CODE: 2816
3. INDUSTRY DESCRIPTION
As defined by Standard Industrial Classification (SIC) 2816, the inorganic pigment
industry includes establishments engaged in the manufacture of inorganic black
pigments (except furnace and channel carbon black), white pigments, and colored
pigments.
3.1 Company Size Distribution
In 1982, the industry included 106 establishments nationwide, and employed 11,200
people. Roughly 63% of the total work force was employed at 13 large establishments.
Table 3-1 summarizes company size distribution.
Table 3-1 1982 Company Size Distribution
No. of employees per facility
No.
No.
of establishments
of employees
Total
t
106
11,200
1-19
43
400
20-99
39
1,800
100-499
17
4,100
500+
7
4,900
Source: 1982 Census of Manufacturers (USDC 1985).
3.2 Principal Producers
The principal producers of inorganic pigments in the U.S. include the companies listed
below (Versar 1980, Williams et al. 1976):
American Cyanamid Co. Kerr-McGee Corp.
CIBA-GEIGY Corp. NL Industries, Inc.
E.I. du Pont de Nemours & Co., Inc. SCM Corp.
Hercules
B5-1
-------
3.3 Geographical Distribution
The geographical distribution of inorganic pigment plants is shown in Figure 3-1 and
Table 3-2 below. Half of the total number of plants are located in just five states, and
the majority are clustered in the industrialized regions stretching between Illinois and
New Jersey.
Table 3-2 Geographical Distribution of Inorganic Pigment Plants by EPA Region
EPA Region Number of Establishments
I
II 24
III 16
IV 11
V 22
VI
VII 3
VIII
IX " 10
X
National 106 ({
Source: 1982 Census of Manufacturers (USDC 1985).
(a' The discrepancy between the national total and the sum of the ten EPA regions
listed above is due to the exclusion of establishments in states with less than 150
employees.
4. PRODUCTS AND THEIR USES
Inorganic pigments include black, white, colored, colorless, and metallic pigments and
are used for a variety of decorative, protective, and functional purposes. They are
used for automotive finishes, industrial coatings, oil and latex paints, and many paper,
plastic, rubber, glass, cement, and porcelain products (Scheik 1982). A list of the
various inorganic pigments produced in the U.S. is given in Table 4-1.
B5-2
-------
VIII
CO
tn
i
CHJ 0-2
/tt 6-10
JZtA over 10
Roman numerals show EPA regions
Figure 3-1 Inorganic Pigment Plants in the U.S.
-------
Table 4-1 Inorganic Pigments Produced in the U.S.
Quantity Produced
Product (thousands of tons)
Titanium pigments 886.0
Other white opaque pigments, including basic 6.6
carbonate and sulfate pigments, white lead, excluding
white lead in oil
Zinc oxide pigments 128.6
Chrome colors
Chrome oxide green 4.0
Chrome yellow and orange 21.0
Molybdate chrome orange 6.2
Zinc chromate
other chrome colors
White extender pigments, including barytes, blanc 51.1
fixe, whiting colors and lakes and toners
Color pigment other than chrome colors, lakes, and toners
Iron oxide pigments
Colored lead pigments
Carbon black (excluding furnace and channel carbon 5.0
black and charcoal)
Cadmium sulfide pigments 2.9
Ceramic colors and all others
Source: 1982 Census of Manufacturers (USDC 1985) and industry comments.
5. RAW MATERIALS
The raw materials used in the production of inorganic pigments consist of a broad
range of chemicals including basic organic and inorganic chemicals as well as a variety
of ores and minerals. These materials, along wih their respective consumption rates,
are listed in Table 5-1.
B5-4
-------
Table 5-1 Raw Materials Used in the Production of Inorganic Pigments for 1982
Consumption Rate
Raw Material (thousands of tons)
Organic chemicals
Alcohol, ethyl *
Other alcohols *
Plastic resins 8-3
Other organic chemicals *
Inorganic chemicals
Acids, except spent acids
Hydrochloric acid 13.6
Hydrofluoric acid *
Nitric acid 5.7
Phosphoric acid 1.0
Sulfuric acid 254.8
Ammonia 5.1
Chlorine 304.1
Phosphorus *
Sodium carbonate *
Sodium hydroxide 62.1
Salt in brine *
Acetylene and other industrial gases compressed and *
liquified, including argon, carbon dioxide, nitrogen,
nitrous oxide, etc.
Crude materials
Bauxite *
Phosphate rock *
Sulfur *
Sulfuric acid sludge *
Zinc and zinc based alloy refinery shapes *
Iron and ferrous alloy ores *
Nonferrous metal ores *
Crude non-metallic minerals *
* No data
Source: 1982 Census of Manufacturers (USDC 1985).
6. PROCESS DESCRIPTION
There are a variety of processes used to produce the pigments listed in Table 4-1.
Since titanium dioxide (TiO2) is the most widely used inorganic pigment, the
B5-5
-------
remainder of this study will focus on the production of TiO2 and associated 'Source
reduction practices.
Titanium dioxide is produced by two major processes - the sulfate process and the
chloride process. While both processes are widely used at present, the chloride process
is likely to displace the sulfate process in the future. The sulfate process has been
hurt by a combination of economic, environmental and market forces. Environmental
problems and waste disposal costs associated with the large amounts of spent acid and
iron salt wastes which are produced during the sulfate process have resulted in the
closure of certain U.S.-based plants (Ryser 1985). In addition, TiO£ pigment from the
chloride process is preferred for use in paint products, which accounts for roughly half
of TiO2 use. For these reasons, further discussion will be limited to the production of
TiO2 by the chloride process.
In the chloride process, two types of raw materials may be used. Ilmenite ore, which
contains 40-70% TiO2, is one raw material. The other is either rutile ore or upgraded
ilmenite ore, both of which contain more than 90% TiO2 (Versar 1980, USEPA 1980a,
Ryser 1985). The ore is mixed with coke and reacted with chlorine at 800 to 1000°C in
a fluidized bed reactor. The reaction produces titanium tetrachloride, carbon dioxide,
carbon monoxide, iron chlorides, and small amounts of other metal chlorides. The
gases are cooled stepwise to remove impurities. The first cooling step lowers the
temperature to 250°C to remove iron and other heavy metal chloride. These are
removed as solids by filtration. Further cooling of the remaining gases to ambient
temperatures condenses the titanium tetrachloride and some dissolved chlorine. The
uncondensed gases remaining include carbon monoxide, carbon dioxide, unreacted
chlorine, hydrochloric acid, some titanium tetrachloride, and other trace gases. The
uncondensed gas stream is scrubbed before being discharged to the atmosphere or
incinerated for its CO (heating) content. The condensed titanium tetrachloride stream
contains impurities such as aluminum chloride and silicon tetrachloride which are
removed through distillation. Copper and/or proprietory organic complexing agents
are added during the distillation step to aid in the decomposition of trace phosgene
(present as an impurity) and to help separate titanium tetrachloride from other
chlorides.
The purified titanuim tetrachloride product is then oxidized with air or oxygen at
150QOC to produce titanium dioxide and chlorine gas. The chlorine gas is often
B5-6
-------
CHLORINE
COPPER —
UPSRADED ILNCNITE
OR RUTILE 0«C
REACTION
1
COOLINI
V
1
FILTRATION
1
COOLINI AND
CONDENSATION
-» 4 rl
PURIFICATION
AND
DISTILLATION
©
1 r-
OXIDATION
TIOj SOLIDS
SEPARATION
1
^ f~
•ASHIM AND
FILTRATION
©
1
•1 COKE
LituiD n
©©
IIUIO
TICL.
AIR/ OXYOEN
LINE OR CAUSTIC -i
CHLORINE
j J ABSORPTION ' ^ tgmm
i LIRUIFICATIO* 0
1
t 1
' HTPOCHLORITC
DECONPOIITION
^ '
!
| HILLINI
no, PIIHCNT
PROCESS «ASTE CATEOORIEi:
I 0 IASTENATER
I 0 FILTER ILUDKI
I 0 DIITILULTION IOTTONI
| 0 VENT IAIEI
Figure 6- 1 Titaniui Dioxide Prediction by the Chloride Process
B5-7
-------
recycled to the chlorination reactor. The solid titanium dioxide is separated from the
gas phase using filters, cyclones, or other proprietary methods. The gases from the
oxidation reaction are scrubbed with lime or caustic soda before being discharged to
the atmosphere. The solid titanium dioxide is then washed with alkali to remove
hydrochloric acid and residual chlorine. It is then dried, milled, coated with an
inorganic oxide such as alumina to increase its weathering resistance, and packaged
(Versar 1980, USEPA 1980a).
7. WASTE DESCRIPTION
The major waste streams along with their process origins and compositions are listed
in Table 7-1. Air pollutants are generated during oxidation, distillation, and drying.
The major air pollutants produced include titanium tetrachloride, hydrogen chloride,
and carbon monoxide from the chlorination and oxidation reactions, as well as chlorine
and particulates from drying and milling. The gases from the chlorination and
oxidation reactions are normally treated in wet scrubbers producing a wastewater
stream which must undergo further treatment.
Wastewater is also generated from washing, filtering, drying, milling, and equipment
cleaning, and from processing of distillation bottoms. Wastewater is typically treated
by in-plant systems using such conventional treatment processes as pH adjustment,
precipitation, and clarification. At plants where ilmenite ore with a high iron content
is used as a raw material, large amounts of acidic iron chloride wastes are produced.
This waste stream has been disposed of by deep-well injection or ocean dumping at
some locations. Solid wastes produced during the chlorination reaction include
unreacted ore, coke, and metal salts. These are either recycled or disposed of in
landfills. For those plants using ilmenite with a high iron content, ferric chloride
waste poses a major disposal problem. However, markets exist for the use of ferric
chloride as a flocculant or a wastewater treatment chemical. In these application,
ferric chloride can be used to clarify drinking water, to treat sewage, and to remove
phosphorus from wastewater (Du Pont 1985a). Du Pont presently recovers and sells
roughly 100,000 tons per year of ferric chloride from one of its titanium dioxide
plants. While the supply of ferric chloride appears to be greater than its demand as a
wastewater treatment chemical, this market does absorb large quantities of the waste.
Ferric chloride produced by titanium dioxide plants competes with the ferric chloride
B5-8
-------
Table 7-1 Manufacturing Wastes From TiO2 Production (Chloride Process)
No. Waste Description
Process Origin
Composition
Concentration (lb/1000 Ib product) RCRA
95% TiO2 ore 65% TiO2 ore Codes
i
\O
Chlorinator wastes Chlorination reactor
Chlormator scrubber Chlorination gas
wastes scrubber
Distillation
Oxidation scrubber
wastes
Finishing
operations
wastes
Distillation
Oxidation tail gas
scrubber
Washing, filtering,
drying, milling
unreacted ore 16
coke 37
iron chlorides 2
other metal chlorides
titanium tetrachloride 25
HC1 130
titanium tetrachloride 25
HC1 50
chlorine
phosgene
metal oxides and chlorides
organic complexing agents
hypochlorite
titanium dioxide 10
NaCl + NaoSO^ 16
25
68
380
-------
produced by the steel industry during steel pickling operations for the wastewater
treatment market.
8. WASTE GENERATION
The 1980 waste generation rates from the production of titanium dioxide by the
chloride process are given in Table 7-1. The exact types and quantities of wastes
produced are dependent upon the raw ore used in the process. Waste rates for
processes using high quality ore (95% TiC^) and lower quality ore (65% TiC^) are both
given. The major difference in waste output between the two is the amount of ferric
chloride produced. Plants using lower grade ilmenite ores produce much greater
amounts of iron chloride waste than do plants using high grade ore.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The wastes which are produced during the manufacturing of titanium dioxide by the
chloride process are listed in Table 7-1. Various source reduction methods available
for reducing these waste streams are discussed below.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" have also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel Practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
B5-10
-------
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Ferric Chloride Waste
As mentioned previously, for plants using ores with high iron content, the ferric
chloride produced during the chlorination reaction poses a major disposal problem. In
addition, chlorination of the unwanted iron represents an economic disadvantage since
the chlorine must then be disposed of as waste. Source reduction measures to reduce
the amount of ferric chloride produced include:
o Use of high-purity ores.
The use of a high purity ore will minimize the amount of impurities
entering the process, greatly reducing waste generation rates. As
discussed earlier, two types of ore, ilmenite and rutile, have traditionally
been used to produce titanium dioxide by the chloride process.
Plants utilizing ilmenite ore generate large amounts of iron chloride
wastes. Ilmenite ore contains 40-70% titanium dioxide and large amounts
of iron. During the chlorination reaction this iron is converted to iron
chlorides which are either sold as a by-product (see below) or disposed of as
a waste, often by deep well injection.
Rutile ore, containing greater than 90% titanium dioxide and only small
amounts of iron, results in substantially less waste generation. Rutile ore
however, is much less abundant than ilmenite ore, and is more expensive.
To achieve the benefits of a higher purity ore, ilmenite can be blended with
ferro-titanium slag to produce a raw material of roughly 75% titanium
dioxide (McNaulty 1986).
B5-11
-------
A third type of ore called anatase (25% titanium-dioxide) is being tested
for use at a few titanium dioxide facilities (Ryser 1985). The main
advantage of using anatase is economic since it may only cost 10% as much
as rutile ore. The process, designed by Du Pont, is a modification of an old
chloride route developed several decades ago. No other technical
information is available and the comparative waste generation character-
istics are unknown.
Recovery of valuable products.
Chlorine can be recovered from ferric chloride by oxidizing the material
with oxygen or air in a fluidized bed reactor. Based on pilot plant tests,
99% recovery of chlorine was achieved (Ogawa 1980). An important part
of the process was that the ferrous chloride was mixed with iron oxide
particles (also produced in the reaction) before being introduced into the
reactor. The oxide acted as a catalyst and increased the reaction rate.
The major disadvantage of the process, however, was low conversion of
oxygen (only 75 percent). Oxygen, if present with the recycled chlorine,
would increase the amount of coke required in the chlorination reactor and
the amount of off-gas generated. Going to a larger oxidation reactor was
expected to increase the conversion of oxygen to 85 percent. It was
reported that a Canadian firm attempted to recover chlorine from ferric
chloride in the early 1970's but that no current work is being performed*.
Currently, Du Pont continues to investigate this option but their efforts
have not yet overcome the difficulties encountered in previous attempts.
Another process for converting ferric chloride waste into valuable products
is to first slurry the waste with water and then spray the slurry into a
roasting furnace operating at 1000°C (Setoguchi 1980). The reaction
produces iron oxide and HC1 containing gases which are scrubbed to
produce 18 weight percent hydrochloric acid. This acid can be used
elsewhere in the facility or further processed to produce chlorine and
hydrogen (for example, using the Kel-Chlor or electrolysis process). Fuel
for operating the roasting furnace can be taken from the chlorination
reactor off-gas which contains a high level of CO.
Confidential source 1985: Personal communication.
85-12
-------
o Ore pretreatment.
Ilmenite is the major type of titanium ore available domestically. As
noted above, ilmenite often has a high iron content which leads to the
generation of iron salts as a waste. One possible way to reduce this waste
stream is to pretreat the ore prior to digestion to remove the iron.
Ilmenite can be converted to an alkaline earth titanite, such as sodium
titanate. When sodium titanate is used in the chloride process, the waste
produced is sodium chloride or sodium sulfate, which leaves the process as
inert dissolved solids (Versar 1980). While this process is attractive
environmentally, it may not be practical from an economic standpoint*.
Other methods of upgrading ilmenite ore for TiO2 production are the
Benilite process (which has been commercialized in India, Japan, and
Taiwan) and the Murso process. In both processes, iron is leached out of
the ore using hydrochloric acid thereby producing a feed material of
greater than 90% titanium dioxide. After the iron is leached out as ferrous
chloride, it is converted to iron oxides and hydrochloric acid. The
hydrochloric acid can be reused within the process (Versar 1980).
Pretreatment of the ore (oxidation and reduction) and addition of ferrous
chloride to the hydrochloric acid leach solution have been reported to
increase or improve the teachability of the ore and reduce fines generation
(Sinha 1980).
9.1.2 Scrubber Wastes
Air emissions are commonly treated in a wet scrubber, producing substantial amounts
of wastewater. In the production of titanium dioxide by the chloride process, scrubber
water accounts for as much as 20 percent of the total wastewater stream (USEPA
1980a). Unreacted chlorine gas from the chlorination reactor is removed from the tail
gas by scrubbing with either water or caustic. Reduction of chlorine content in the
tail gas will result also in reduction of scrubber wastewater, treatment chemical usage
and chlorine loss. The discharge of unreacted chlorine in the tail gas can be reduced
or eliminated by chlorine liquefaction achieved by pressurization and subsequent
refrigeration to condense chlorine from the tail gas (USEPA 1980a).
*E.I. du Pont de Nemours & Co. 1986: Personal communication.
B5-13
-------
9.1.3 Filtering and Washing Wastes
The ore impurities will at some point have to be separated from the desired product,
resulting in a waste stream which must be treated. A substantial portion of
wastewater generated in TiO2 plant is from the rinsing and washing of the pigment.
Rinsing and washing serves the purpose of removing residual chlorine and
hydrogen chloride, soluble salts, and other impurities from the final product. The
amount of impurities is largely controlled by ore composition and the efficiency of the
upstream distillation process to purify titanium tetrachloride. However, the use of
water can be reduced using countercurrent rinsing and the amount of pigment carry-
over can be reduced by using more efficient filtration.
9.2 Implementation Profile
The major source of waste from titanium dioxide pigment manufacturing is simply the
impurities separated out from the ore. Thus, the most important source control
methods are concerned with the quality of ore used. The use of high-purity ores
reduces the amount of impurities introduced into the process. However, a manu-
facturer may have little control over the quality of the ore which is used if the
availability of high-puriy ore is limited. In this case, ore pre-treatment and the
recovery of chlorine and iron oxide gain significance. The use of efficient rinsing and
washing techniques for product purification along with the reduction of scrubber
wastes through chlorine recovery from the tail gas are, in essence, water conservation
measures and, as such, may not be prime waste minimization issues of current
environmental concern.
9.3 Summary
The waste sources and their respective source control techniques are summarized in
Table 9-1. The ratings listed in this table are based on a scale of zero to four and are
used to evaluate each technique for its waste reduction effectiveness, extent of
current use, and future application potential. The ratings were derived by the project
staff based on the available information.
It appears that the current level of waste minimization in titanium dioxide manu-
facturing based on the chloride process is significant. This is evidenced by the current
B5-14
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY FOR THE TITANIUM DIOXIDE MANUFACTURING INDUSTRY
1
1
| Ferric Chloride
Haste
Scrubber Wastes
Filtration and
Washing Wastes
All Sources
Control Methodology
J
1. Use high-purity ore
2. Recover products from «aste
3. Pretreat ore to remove iron
Overall
t Chlorine Hquifaction for reuse
Overall
|1. Use more efficient filters
2 Use efficient washing/rinsing methods
3. Process control/automation
Overall
All Methods
round Documentation |
- — i
Quantity | Quality f
31 3|
21 2|
3 1 3 |
2.67 | 2.67 |
1 1 2 |
1 00 | 2.00 !
1 | 2 I
2| 2 |
1 1 11
1.33 I 1.67 |
Haste 1 Extent of I Future | Fraction of
Effectiveness | | Potential |
« 1 2 1 1 1
31 0| 1 |
2 I 2| 21
3.00 | 1.33 | 1.33 | 0.05
2 1 2 | 1 |
2.00 I 2.00 | 1.00 | 0.20
1 1 2 1 2 1
3| 3| 1 |
21 2 ! 2 |
2.00 | 2.33 I 1.67 | 0.75
| 1.00
Current | future Reduction Index |
Index j Probable | Maximum |
2.0 | 0.5 | I
0 0 | 0.8 | 0.8 |
1.0 | 0.5 | |
2.0 I 0.6 | 0.8 I
1.0 | 0.3 | 0.3 |
1.0 I 0.3 | 0.3 I
0.5 | 0.3 | |
2 3 | 0.2 | 1
1.0 | 0.5 | 0.5 I
2.3 | 0.3 | 051
2.1 | 0.3 | 0.5 |
(*) These streams include listed "F1 and/or "K" RCRA wastes.
-------
reduction index (CRI) of 2.1 (53 percent), which is a measure of reduction of waste
that would have been generated if none of the listed methods were applied at their
current level of use.
Future reductions appear to be rather modest as evidenced by the future reduction
index (FRI) of 0.3 and 0.5 (8 to 12 percent). FRI is the measure of waste reduction
achievable through the implementation of the listed techniques to their full rated
potential. From the standpoint of future application, the most promising source
control techniques found were product recovery from waste (Cl2 and Fe2O3), ore
pretreatment to remove iron, and use of high purity ore.
10. PRODUCT SUBSTITUTION ALTERNATIVES
10.1 Paint Usage
Paint and paint-related products, such as automotive finishes, account for a large
portion of total titanium dioxide pigment usage. No substitute for titanium dioxide
was identified in paint pigment application. However, any reductions in the use of
paint will have a direct impact on the demand for titanium dioxide. For additional
discussion of this issue the reader is referred to the study of paint manufacturing
included in this appendix.
10.2 Paper Products
Paper products account for roughly 18% of the total use of titanium dioxide, which is
the most widely used inorganic pigment (USEPA 1980a). In paper manufacturing,
titanium dioxide (TiO2) is used as an opacifier. In place of TiO2, alumina or silica
clays have been tried, however the main advantage of TiC>2 over the other substances
is that it has a superior reflectance.
11. CONCLUSIONS
The facilities using the chloride process for production of titanium dioxide appear to
have minimized their waste significantly by using high-purity ore. Water use was
reduced by the application of efficient washing and rinsing techniques. Any future
reductions appear to be modest, in the range of 8 to 12 percent, primarily resulting
B5-16
-------
from wider adoption of ore pretreatment, use of high purity ore and, possibly, recovery
of C\2 and Fe£O3 from FeCl3 waste stream (potentially, the most effective
technology which has not yet been commercialized).
12.REFERENCES
Campbell, M.E., and Glenn, W.M., 1982. Profit from pollution prevention. Toronto,
Canada: The Pollution Probe Foundation.
E.I. Du Pont de Nemours and Co. 1985a. Statement to Tennessee water quality
control board concerning consideration of adoption of rules of the water guality
control board. Chapter 1200-4-6. Underground Injection Control. Feb. 7, 1985.
1965b Ocean dumping permit. Certified report on implementation
plan (special condition - 7(a)(2)). Submitted to the Marine and Wetlands Protection
Program. U.S. Environmental Protection Agency, Region II, New York, NY. June 13,
1985.
FTA, 1983. Frontier Technical Associates, Inc. Development document for effluent
limitations guidelines and standards for the inorganic chemicals point source category.
Contract no. 68-01-6701, revised working draft for Office of Water and Waste
Management. Washington, D.C.: U.S. Environmental Protection Agency.
Ogawa, Minour et. al. 1980. A study of titanium resources and its chlorination
process. In Titanium' 80. Science and Technology. Proceedings of the Fourth
International Conference on Titanium. 3rd vol. Kyoto, Japan. May 19-22, 1980.
Ryser, J. 1985. New feed, new technique enliven the TiO9 scenario. Chem. Enqr.
Nov. 25, 1985: 18-20.
Scheik, R.C., 1982. Inorganic pigments. In Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed., vol. 17, pp. 788-838. New York, N.Y.: Wiley.
Sinka, H.N. 1980. Effects of oxidation and reduction temperatures, and the addition
of ferrous chloride to hydrochloric acid, on the leading of Ilmenite. In Titanium' 80.
Science and technology. Proceedings of the Fourth International Conference on
Titanium. 3rd vol. Kyoto, Japan May 19-22, 1980.
Setoguichi, Masahiko 1980. Pollution prevention for titanium tetrachloride plant. In
Titanium' 80. Science and technology. Proceedings of the Fourth International
Conference on Titanium. 3rd vol. Kyoto, Japan May 19-22, 1980.
UNESC, 1979. Urated Nations Economic and Social Council. Production of titanium
white from ilmenite by the sulfate method with reprocessing of the quantitatively
most import waste products. ENV/WP.2/5/Add. 21.
USDC, 1985. U.S. Department of Commerce, Bureau of the Census. Industrial
inorganic chemicals. In 1982 Census of manufacturers. Washington, D.C.: Govern-
ment Printing Office.
USEPA, 1979. U.S. Environmental Protection Agency, Office of Water and Waste
Management. Development documents for proposed effluent guidelines, new source
B5-17
-------
performance standards, and pretreatment standards for the paint formulating point
source category. EPA-440-1-790-0406. Washington, D.C.: U.S. Environmental
Protection Agency.
, 1980a. U.S. Environmental Protection Agency, Office of Water and Waste
Management. Development documents for effluent limitation guidelines and standards
for the inorganic chemicals point source category. EPA-440-1-79-007. Washington,
D.C.: U.S. Environmental Protection Agency.
, 1980b. U.S. Environmental Protection Agency, Office of Research and
Development. Treatability Manual; Vol. 2. Industrial descriptions. EPA-600-8-80-
042b. Washington, D.C.: U.S. Environmental Protection Agency.
Versar, 1980. Versar, Inc. Multimedia assessment of the inorganic chemicals industry.
Cincinnati, Ohio: U.S. Environmental Protection Agency.
Williams, R., et ah, 1976. Economic assessment of potential hazardous waste control
guidelines for the inorganic chemical industry. Arthur D. Little, Inc. EPA-530-SW-
134C. Washington, D.C.: U.S. Environmental Protection Agency.
13. INDUSTRY CONTACTS
Confidential Source
C.R. Steward, E.I. Du Pont de Nemours & Co. Wilmington, DE..
B5-18
-------
1. PROCESS: METAL SURFACE FINISHING
2. SIC CODE: 3471
3. INDUSTRY DESCRIPTION
Metal finishing is a part of the plating and polishing industry classified under Standard
Industrial Classification (SIC) 3471. This industry is comprised of establishments
primarily engaged in all types of electroplating, plating, anodizing, coloring, and
finishing of metal and formed products for the trade. Most of the work performed by
this industry is done on materials owned by others. Though metal surface finishing is
classified under SIC 3471, metal finishing operations are performed by many other
industries included in SIC groups 34 through 39 (USEPA 1980). The metal finishing
process includes some 44 unit operations and only the metal surface finishing
operations that use chemical means are discussed in this report.
3.1 Company Size Distribution
Since metal surface finishing operations are performed by various industries classified
under many SIC codes, company size distribution data for metal surface finishing as an
industry was not separately available. In 1980, there were approximately 160,000
manufacturing facilities in the U.S. which were covered by the metal finishing
category (USEPA 1980). These facilities varied greatly in size, age, and number of
employees. They ranged from very small independent job shops with less than ten
employees, to small shops within large corporations, to large facilities employing large
work forces.
3.2 Principal Producers
The metal surface finishing industry is dominated by small job shops employing less
than 20 employees each. There are no major producers who control a large share of
the market.
3.3 Geographical Distribution
The geographical distribution of the metal surface finishing industry was not available
due to the reasons stated above.
B6-1
-------
4.
PRODUCTS AND THEIR USE
The metal surface finishing industry deals mostly with the treatment of metallic or
non-metallic products manufactured by others. Each product requires a specific
process sequence to obtain the desired physical, chemical, or aesthetic properties
desired by the user. The principal products of the metal surface finishing industry
include:
Printed circuit boards
Coil coating
Automotive parts
Kitchen utensils
Jewelry
Mechanical (non-automotive) parts
5. RAW MATERIALS
Reagents phosphoric acid, secondary or tertiary metal phosphates, sodium
dichromate, sodium nitrate, sodium cyanide, barium chloride, sodium
chloride, sodium carbonate, sodium cyanate, ammonia, silicon tetra-
chloride, zinc oxide, chromic acid
Accelerators quinoline, toluidine, nitrophenols, various oxidizing agents such as
peroxides, and sulfites
Metals zinc, aluminum, chromium, cadmium, magnesium, iron, nickel,
copper, silver, molybdenum, vanadium, tungsten
Alloys tin, lead-tin alloys, bronze, brass
6. PROCESS DESCRIPTION
Metal surface treatment consists of various processes such as electroplating,
electroless plating, anodizing, chemical conversion coating, cleaning, etc. (USEPA
1980, BCL 1976, Schneberger 1981, Durney 1984). Since electroplating and metal
surface cleaning are discussed in separate studies in this appendix, this study
considers only chemical surface treatments such as electroless plating, chemical
B6-2
-------
conversion coating, etching, chemical milling, and some forms of case hardening
treatment. Metal surface treatment is performed mainly to modify the metal surface
to be less reactive and more corrosion-resistant and is dependant on the type of
required surface modification. All the metal surface treatment processes have three
basic steps: surface cleaning or preparation, surface treatment, and rinsing or post-
finishing operations. The cleaning and post-finishing operations are specific to the
surface treatment method used. The following sections describe various surface
treatment methods with the exception of coil coating. All chemical surface treatment
operations are essentially batch operations, where the metal object is dipped in a bath
containing various reagents to achieve the required surface modification.
6.1 Electroless Plating
Electroless plating allows for the deposition of metal on an object's surface without
the use of external electrical energy. This is achieved by a chemical reduction process
which depends upon the catalytic reduction of a metallic ion in an aqueous solution.
This process has found widespread use due to several advantages over conventional
electroplating which include ability to produce a uniform coat on all areas of the part
regardless of its geometry without the need to supply external electrical energy.
Copper and nickel electroless plating are the most common.
The basic ingredients of electroless plating solutions are a source of metal ions
(generally copper or nickel), a complexing agent to maintain ions in solution at the
operating pH value, a compatible reducing agent, a material to adjust the pH of the
bath, and stabilizers, wetters, stress relievers, etc. Table 6-1 lists the bath
constituents for copper and nickel electroless plating operations.
The electroless plating operation consists of cleaning the object surface, immersing it
in a bath containing the previously described constituents for a specific period of time,
and then rinsing the object (after removal from the bath) to remove process solutions
adhering to the surface.
Plating bath solutions last only a few hours because catalytic particles precipitate in
the bath. The bath life is usually increased by periodic filtration, and by the addition
R6-3
-------
Table 6-1 Bath Constituents for Copper and Nickel Eiectroless Plating
Function
Copper Plating Operation
Nickel Plating Operation
Metal ion source
Complexinq agent
oReducing agent
ON
i
pH adjuster
Stabilizers
Other additives
Cupric sulfate (3-15 g/1)
Rochelle salt (sodium potassium tartarate,
ethylenediaminetetracetic acid (EDTA), sodium
salts of EDTA, nitrilotriacetic acid, gluconic
acid, gluconates, triethanol amine, n-hydroxy
ethylene diamine tetracetate (20-50 g/1).
Formaldehyde, paraformaldehyde, trioxane,
dimethylhydantoin, sodium and potassium
borohydride.
Sodium or potassium hydroxide
2-mercaptobenzothiazole, thiourea, methanol
Water soluble metal cyanides, polysiloxanes,
methyl dichloro si lane
Nickel chloride, nickel sulfate, nickel
sulfamate, nickel hydrophosphite
Lactic acid, dicarboxylates
Sodium hyposphosphite
Sodium or potassium hydroxide
Molybdic acid anhydride, arsenious acid,
hydroxyl amino sulfate, hydrazine.
Thiourea, soluble fluorides, alcohol
sulfonates, ethylene oxide derivatives,
sodium sulfate.
Source: Development Document for Effluent Limitation Guidelines (USEPA 1980).
-------
of stabilizers. The bath is eventually dumped thereby generating a waste stream.
Other waste streams are generated by rinsing operations and periodic cleaning of the
process equipment. Other electroless plating operations, such as vapor deposition, are
purely physical operations and are not considered in this report.
6.2 Chemical Conversion Coating
This operation includes phosphating, chromating, metal coloring, and passivating. The
coating deposited on metal objects is for decorative or corrosion protection purposes,
and in some instances to prepare the surface for painting. The mode of operation and
waste generation are similar to the electroless plating operation described in Section
6.1. The following sections discuss the four different chemical conversion coating
methods listed above.
6.2.1 Phosphating
Phosphate coatings are formed on the surfaces of iron, steel, galvanized steel,
aluminum, and electrodeposited zinc and cadmium to promote adhesion of organic
coatings, to retard interfacial corrosion, to retain and enhance the performance of
corrosion resistant oils, and to assist in cold deformation processes. Small parts are
coated in barrels immersed in the phosphating solution and large parts are spray
coated or continuously passed through the phosphating solution. The object to be
coated may be dipped successively in a series of processing tanks.
The phosphating solution consists of a phosphoric acid solution of metal dihydrogen
phosphate. The coating time and temperature depends on the type of metal to be
coated and whether a spray or immersion coating scheme is used. Sometimes
accelerators (to improve quality), stabilizers (to prolong bath life) and oxidizing agents
(to control the coating rate) are added to the phosphating solution.
There are certain parameters, such 'as the ratio of free to combined phosphoric acid,
total acid, metal-ion concentration, accelerator concentration, and the process
temperature that must be controlled to achieve a suitable coating and maintain the
R6-5
-------
integrity of the bath. The coated object, after phosphating, may be given a
conditioning rinse with dilute chromic and/or chromic-phosphoric acid for surfaces to
be protected against corrosion, or an alkaline rust-inhibiting treatment for objects to
be cold-deformed. Concerns about the environmental effects of chrome have led to
the development of chrome-free rinses. Though these are less effective than their
chrome-containing counterparts, they are suitable for applications where corrosion is
not a major concern.
6.2.2 Chromating
Chromate coatings are most frequently applied on zinc, cadmium, aluminum,
magnesium, copper, brass, bronze, and silver to minimize rust formation and to
guarantee paint adhesion. Chromate-type conversion coatings are produced primarily
by a simple immersion process although a spray or brush treatment can be used.
The chromating solution consists of chromic acid, one or two mineral acids such as
sulfuric or nitric, and often some activating compounds. Chromate conversion
•
coatings are formed because the metal surface dissolves to a small extent, causing a
pH rise at the surface-liquid interface. This results in the precipitation of a thin
complex chromium metal gel on the surface, composed of hexavalent and trivalent
chromium and the coated metal itself. This gel is normally so'ft when formed and must
be handled carefully. After drying, the coating becomes hard and relatively abrasion-
resistant.
The thickness and color of the chromate coating depends on the solution composition,
temperature, pH, and the length of treatment. The coated object is usually rinsed
with cold water containing sodium hydroxide or sodium carbonate to provide a clear
noniridescent coating. This is followed by a warm water rinse to prevent the removal
of the coating.
6.2.3 Metal Coloring
While coloring of steel, copper, aluminum, and their alloys is done primarily for
aesthetic purposes, this surface treatment often imparts other favorable properties
such as improved corrosion resistance and better abrasion and wear characteristics.
The coating operation is primarily of the immersion type.
B6-6
-------
Many formulations are available for the coloring of metals and most of them are
proprietary. The major coloring process for steel uses a treatment solution of sodium
hydroxide and sodium nitrate in water. The processing temperature may vary from
275-320°F, and the immersion time may vary from 5 to 30 minutes. The coating color
and characteristics are largely a function of the alloy being treated, surface
characteristics, concentration of the bath, temperature, and immersion time.
6.2.4 Passivation
Passivation refers to forming a protective film on metal, particularly stainless steel
and copper, by immersion in an acid solution. Stainless steel is passivated to dissolve
embedded iron particles and to form a thin oxide film on its surface. A typical
treatment solution for stainless steel is nitric acid or nitric acid with sodium
dichromate. Copper is passivated using a solution of ammonium sulfate and copper
sulfate.
6.3 Chemical Etching
Chemical etching is used to produce specific design configurations and tolerances on
metallic or metal-clad plastic (printed circuit boards) by controlled dissolution of the
metal with chemical etchants. Typical etching solutions are ferric chloride, nitric
acid, ammonium persulfate, chromic acid, cupric chloride, hydrochloric acid, etc.
"Bright dipping" is a special form of chemical etching used to remove oxide layers
from ferrous and non-ferrous materials.
6.4 Cyanidinq
Cyaniding is a type of case hardening that produces a hard surface on a metal whose
core remains relatively soft. The product is a hard, wear-resistant surface backed by
a strong, ductile, and tough core. Carbon and alloy steels are usually immersed in the
cyaniding bath for a specific period of time to achieve the required degree of surface
hardening.
The most common cyaniding solution consists of 30 percent sodium cyanide, 40 percent
sodium carbonate, and 30 percent sodium chloride. Baths containing 97, 75, and 45
percent sodium cyanide are also used. Oxygen from the air oxidizes the sodium
B6-7
-------
cyanide to sodium cyanate which, at high temperatures, decomposes to form nascent
carbon and nitrogen. The carbon and nitrogen are absorbed by the steel, which
increases surface hardness. The processing temperatures may vary from 1200-1350°F
for low penetration to 1650-1725°F for high penetration. A combination treatment
using high temperature immersion followed by low temperature immersion is also used.
The depth of surface that is hardened is controlled by the temperature and the cyanide
content. As drag-out and carbon depletion occur, special salt compositions are added
to replenish and regenerate the bath. At the end of the treatment, the objects are
immersed in a water or oil bath, where quenching and rinsing is accomplished
simultaneously. The quench water is potentially hazardous and is treated for cyanide
destruction followed by clarification prior to discharge. The quench oil is also
potentially hazardous and is disposed of by incineration.
7. WASTE DESCRIPTION
The primary wastes associated with metal surface finishing, along with their process
sources, are listed in Table 7-1. The wastes produced in metal surface finishing
operations come mainly from two sources: dumping of process tanks, and rinse waters
used to wash off process solutions adhering to the product surface or entrapped in the
crevices due to the shape of the product piece (Durney 1984, AESI 1981, CP Staff
1984). Additional waste is generated as a result of process solution filtering. The
process solutions are periodically filtered to remove precipitated metals and are
reused. These filtered solids are mixed with solids removed from the rinse waters and
are either landfilled or sold for metal reclamation.
Other wastestreams include spills and leaks plus stripping wastes.
Spent Bath Solution
The activity of the plating solution decreases with time due to the precipitation of
salts and depletion of constituents. After 3-6 regeneration cycles, the bath is
eventually discharged*. This waste stream usually contains cyanides or metallic
*National Association of Metal Finishers 1985: Personal communication.
B6-8
-------
03
ON
Table 7-1 Metal Surface Finishing Wastes
No.
Waste
Description
Process Origin
Composition
RCRA
Code
1. Spent bath solution
Dumping of the process
solutions after depletion
OT loss of activity.
cyanides, cyanide com-
plexes, hexavalent
chrome, copper, nickel,
zinc, cadmium, and other
metals and their salts
in water
F011
Waste rinse water
Rinsing treated objects,
equipment cleaning,
quenching of case
hardened steel.
same as spent bath
solution
Filter waste
Filtration of process
solution, spent baths,
and treated waste rinse
water.
Complexes of various F010
metals, cyanides, etc. F012
Spills and leaks
Overflows and leaks
from various process
equipment
same as
(1)
Stripping waste
Removal of coatings
from improperly treated
objects.
not available
-------
complexes. In the recent past, this stream was treated together with the rinse water
stream. However, the need for obtaining the necessary permits and also the perceived
regulatory compliance difficulties have discouraged treatment of spent baths*. This
stream is often sent off-site for disposal.
Waste Rinse Water
Rinse water is used to wash off process solutions adhering to the product surface or
entrapped in its crevices due to its geometrical shape. The rinse water stream is
usually discharged to municipal treatment facilities with or without any treatment,
depending on its composition. The treatment procedure includes oxidative destruction
of cyanides reduction of chromates, neutralization, and solids removal.
Filter Waste.
The filtration step in the regeneration of plating solution and the solids removal in
rinse water treatment each generate a solid waste. These solids contain oxides or
complexes of metals and are either landfilled or sent off-site for metal reclamation.
Spills and Leaks
The overflow and leaks from various process equipment are usually mixed with the
rinse water stream and disposed of as explained above.
Stripping Waste
Before coating an object, the previous coatings on it are removed by a striping
operation. This is also done to remove coatings from an improperly coated object. In
small job shops, the same stripping solution could be used for removing different types
of coating. The disposal of the spent bath from such operation is similar to that of the
spent baths discussed earlier. The rinse waters are usually discharged to municipal
treatment facilities with or without treatment. The untreated streams may contain
various cyanides and cyanide complexes, hexavalent chrome, copper, nickel, zinc,
*Westinghouse Electric Corporation, 1985: Personal communication
B6-10
-------
cadmium and other metals. The treatment includes oxidative destruction of cyanides,
reduction of chromates, neutralization, and solids removal. The treatment sludge is
expected to be landfilled. Spills and overflows that occur can be mixed and treated
with other liquid wastes and treated as discussed.
8. WASTE GENERATION RATES
Since metal finishing operations are often performed along with electroplating and
other operations, the waste generation rates specifically attributable to metal
finishing are difficult to determine. No waste generation data were in evidence at the
time of the final document preparation. While no specific waste generation rates were
reported, fractional rates were estimated by project staff based on the available
information and engineering judgements. These values are shown in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The list of individual waste streams and sources and their corresponding source
reduction methods is presented in Table 9-1. The following sections discuss the
various waste reduction methods based on a literature survey and industry contacts.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
B6-1J
-------
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Spent Bath Solutions
Most of the metal surface finishing operations are performed by immersing an object
into tanks containing specific reagents. Due to the precipitation of salts and depletion
of constituents, the bath becomes ineffective and must be regenerated or discarded
when necessary. This waste, containing cyanide, cyanide complexes of metal, and
other metallic complexes, is often sent to a treater for disposal. The following
methods could reduce this waste stream:
o Extending bath life.
During the surface treatment operation, many insoluble salts (such as
ferric phosphate in the case of ferrous metal phosphating) precipitate out
of solution and thereby decrease its effectiveness. In addition, the
depletion of metal in the solution causes the activity of the bath to
decrease. The bath life, if prolonged, can contribute toward waste
reduction since the frequency of process solution dumping decreases. The
bath life can be increased by periodic or continuous filtering of the bath,
regeneration of the spent bath solution, and preventive measures against
bath contamination. As the insoluble metallic salts precipitate onto the
cooling/heating coils and the object, the effectiveness of the bath goes
down. By periodically filtering the process solution, its activity can be
maintained (Durney 1984, Saubestre 1957). This is already practiced at
many metal finishing operations, especially those with chemical conversion
coatings.
Removal of the soluble salts formed during the treatment process can also
extend bath life. The soluble metallic salts that form during the treatment
process accumulate in the bath and reduce its effectiveness. These salts
86-12
-------
can sometimes be removed by temporarily lowering the bath temperature
so as to form solid crystals. In the case of electroless nickel plating, the
sodium sulfate that forms can be crystallized by lowering the bath
temperature to 41-50°F (Durney 1984). The crystals can then be removed
by filtration.
Another process, patented by the U.S. Army, involves the removal of
sodium carbonate from sodium cyanide-based plating baths by cooling. Dry
ice is used to cool the plating bath, thereby causing the precipitation of the
carbonates. The plating solution, free of carbonates, can then be reused.
So far, this method has not found widespread use (Arienti 1985, Versar
1985).
Use of an electrolytic diaphragm cell for regenerating spent chromic acid
used in etching operations has been reported (AESI 1981). The process uses
electrolytic diaphragm cell to oxidize trivalent chromium to hexavalent
chromium and remove contaminants. The quality of the regenerated
etchant was reported to be equal to or better than fresh etchant. This
process, which was still in the developmental stage, would have a great
potential for reducing spent chromic acid waste. In one such application,
extensively tested at the U.S. Bureau of Mines in Rolla, Mo., copper
etching solution was regenerated and metallic copper recovered at the
same time. Recovery was accomplished by depositing the copper onto the
cathode of the electrolytic diaphragm cell (Basta 1983).
Other measures that can extend bath life are to use automatic control
systems, maintain all rack and barrel systems, and practice good house-
keeping at all times. Automatic control devices can be used to maintain
the concentration by conductivity measurement. A significant change in
conductivity would initiate pumping of fresh concentrate into the tank.
This type of automatic control is used by large facilities especially for
chromate conversion coating (Durney 1984). If the racks or barrels (used
for transferring objects to be coated) do not have proper protective
coatings, the bath could become contaminated. The process solution can
attack the weak spots in the racks or barrels, causing the formation of
metallic salts in the bath and thereby lower its activity. Fluorocarbon
B6-13
-------
coatings applied to the racks or barrels have been found to be effective in
minimizing such contamination (Lane 1985). Such a coating will also be
helpful in lowering dragout since less bath solution that remains in the
corroded crevices on the racks or barrels. Special measures that constitute
good operating practices are discussed at the end of this section.
Metal/acid recovery from spent bath solutions.
By using electrolysis on the spent bath solutions, useful metals can be
recovered and at the same time the hazardous nature of the spent bath can
be reduced (Lewis 1980, Campbell and Glenn 1982). This method was
implemented in one facility by insertion of electrodes directly into the
cyanide destruction tank*. However, electrolysis is used only to a limited
extent by the industry (USEPA 1982).
In addition to recovering metals from the spent bath, spent acid can also be
recovered and recycled by means of ion-exchange (Basta 1983). Eco-Tec
Ltd., in Ontario, Canada, markets an acid purification system that uses a
propcietary resin that recovers mineral acids. The metals are recovered in
a concentrated (but still dissolved) form. The concentrated metals can
than be recovered by electrolytic means. This process is used by Modine
Manufacturing, in Trenton, Mo., to treat copper-contaminated sulfuric
acid/hydrogen peroxide solution which was used to brighten brass (Basta
1983). Sodium phosphate salts, formed in nickel/copper electroless plating,
can be converted into useful hypophosphite salts by using ion exchange
resins activated with hypophosphorous acid. The use of ion exchange resins
for regeneration, however, suffers from the disadvantage of generating
additional wastes such as spent resins and resin regeneration solutions.
Another noneiectrolytic means of metal reclamation, still in developmental
stages, is the use of bacteria (Basta 1983). Here, the microbes form
complexes with the metals in solution, creating a biomass. The biomass
can then be burned to recover elemental metal. Laboratory studies in this
area are in progress at Polbac Corp. (Allentown, Pa.) and the O'Kelly
Companies (Tulsa, Ok.).
* National Association of Metal Finishers 1985: Personal communication
B6-14
-------
Another new trend is the use of so-called liquid membranes. These
membranes are composed of polymeric materials loaded with an ion-
carrying solution (Basta 1983). Liquid membranes were used at Bend
Research Inc., in Bend, Ore., to remove chromium from rinse waters and
spent baths. Here, chromium in the form of dichromate is drawn across
the membrane, forming a tertiary amine metal complex. This complex is
then broken down on the other side of the membrane with sodium hydroxide
solution. Similar experiments are in progress at the Warren Springs
Laboratory in Stevenage, England, for treating wastes from printed circuit
board manufacturing.
Spray/brush methods instead of immersion methods.
The use of spray/brush methods, which use reagents in a more efficient
manner, will reduce the total amount of spent reagents generated. How-
ever, spray methods are useful only in continuous applications and when the
objects are flat or geometrically simple. Brush application methods are
also useful only where the geometry of the object and/or the nature of the
operation permits it.
Use of thinner foils in printed circuit boards manufacture.
Printed circuit boards are made by the controlled dissolution of the metal
foil present on a plastic substrate in an etching solution. By using a thinner
metal foil on the plastic board, lower amounts of etching solution are
needed and more dilute solutions can be used. This will result in less spent
bath wastes. Other techniques allowing for reduction in etching solution
use are described in the study of printed circuit boards manufacture (Bll).
Alternatives to conventional metal surface treatment techniques.
In cyaniding, liquid cyanide salts are used to generate nascent carbon and
nitrogen that diffuse into the metal surface, causing it to harden. The
same result can be achieved when carbon and steel are exposed to ammonia
gas. Ammonia decomposes at about 1550°F to produce nascent nitrogen.
Combined with carbon, the gas will diffuse into the metal surface. The use
of gas phase carbonitriding alleviates the need for a cyanide solution bath.
In addition, the quench-rinsing sequence present in the liquid phase
carbonitriding (cyaniding) process would also be avoided. Thus, the waste
B6-15
-------
generated by spent bath solutions can be eliminated totally. Cyaniding,
however, does have several advantages over gas carbonitriding
(Schneberger 1981). These include more flexible operation illustrated by
the ability to simultaneously treat many small batches which require
different cycle times, and a higher heating rate. This is important for
cases where the time needed for the object to reach the conditioning
temperature may be a large fraction of the total cycle time.
Vacuum evaporation methods for coating nickel, aluminum, and other
metals have been developed. Here, metals are evaporated at low pressure
using an electron beam and the vapor condenses, as a coating, onto the
substrate. This method could be a viable alternative to electroless nickel
plating, which generates spent bath wastes. However, the vacuum
evaporation method has several disadvantages, e.g., it requires relatively
expensive equipment which becomes effective only when a large number of
substrates are to be coated, and the uniformity of coating thickness is
generally very difficult to control.
Chromium and cadmium can be deposited on steel using ion plating
methods instead of electrodeposition. Ion plating uses high-energy ions
that bombard the depositing metal which evaporates and then condenses on
the substrate. In the U.S., this method is used only when other techniques
are found inadequate. However, ion plating is in wide use in Japan where it
successfully competes with electrodeposition (Durney 1984).
Chemical vapor deposition (where a chemical reaction decomposes the
reactant gases to produce the desired coating material which then
condenses on the substrate) can be used for almost any coating operation
(Durney 1984). However, its use has been limited to the semiconductor
industry, and commercial systems for other applications are not available
at the present time.
Ion beam processing techniques, still in the development stage, will be an
excellent alternative to case-hardening treatments (Anon 1984). Here, a
high energy ion beam is used to harden the surface by implanting the ions
in the material.
B6-16
-------
Use of less toxic solutions.
Whenever possible, the use of less toxic solutions for metal finishing will
lower the hazard of spent solutions. In the case of electroless copper
plating, water soluble cyanide compounds of many metals are added to
eliminate or minimize the internal stress of the deposit. It was found that
polysiloxanes, such as General Electric silicon fluid SF-96, are also
effective stress relievers (Durney 1984). By substituting cyanides with
polysiloxanes, the hazardous nature of the spent bath solution can be
reduced. Use of trivalent chrome instead of hexavalent chrome in
chromate conversion coating can eliminate the toxicity of the spent
electroplating baths. Though some manufacturers use trivalent chrome*,
its use is not growing rapidly because the quality of trivalent chromium
coatings is not as good as that of hexavalent coatings in many
applications.* Currently, there were at least five companies that offer
trivalent systems (Chementator 1982). Trivalent chromium baths can also
use lower metal concentrations. One such solution, developed by
W. Canning Materials Ltd. in Birmingham, England, uses only 3.5 gms/liter
of total chromium compared to the level of 130 gms/ liters used in
traditional hexavalent baths.
More dilute process solutions.
The use of dilute bath solutions, whenever possible, would also reduce the
hazardous nature of the dumped bath. In the case of cyaniding, a typical
bath solution composition is 30 percent sodium cyanide, though some
facilities use 45, 75, and 92 percent solutions. By using a 30 percent
solution instead of a solution of higher concentration, substantial reduction
in the cyanide content of the spent bath solution can be achieved. In
electroless copper plating for printed circuit board manufacture, dilute
solutions have been tried successfully by many manufacturers (USEPA
1981). The use of dilute bath solution could also lower subsequent rinse
water requirements and metal dragout into rinsewater.
* National Association of Metal Finishers 1985: Personal communication.
B6-17
-------
o Better operating practices.
By frequent monitoring of the bath activity and regular replenishment of
reagents or stabilizers, bath life can be prolonged (Durney 1984). These
reagents or stabilizers differ from process to process. Stabilizers such as
2-mercaptobenzothiozole and methanol are found effective in electroless
copper plating used for manufacturing printed circuit boards. The addition
of stabilizers can sometimes decrease the deposition rate, but can still be
economical in the long run.
Good control of the bath temperature is important from the viewpoint of
performance predictability and is another method of prolonging bath life.
Many of the surface treatment operations use tanks with immersed
cooling/heating coils. As the salts precipitate and form scales on the coils,
the heat transfer is impeded and temperature control becomes increasingly
difficult. The heat transfer efficiency can be maintained by periodic
cleaning of the coils or by using jacketed tanks instead of coils.
Proper storage of the process solutions can also reduce waste generation.
Usually, the process solutions are stored as a two-part solution and are
mixed when a batch is needed. Prolonged storage of these solutions may
allow some chemical reactions to occur that could generate contaminants
that reduce bath life. In electroless copper plating, if formaldehyde (using
as a reducing agent) is stored with a hydroxide, the hydroxide can cause the
formaldehyde to breakdown into formic acid and methyl alcohol. Thus, it
is better to only store non-reactive mixtures of materials or to store each
item separately.
9.1.2 Waste Rinse Water
At the end of surface treatment, objects are rinsed with water, alkali, or acid to
remove reagents adhering to the surface or trapped in the crevices of the object. This
rinsing may be repeated several times, and the rinse water represents about 90 percent
B6-18
-------
of the total waste volume gererated (AESI 1981). This stream, containing cyanides and
cyanide complexes along with other metal complexes, is treated for cyanide
destruction and discharged to the sewer. The following source control methods are
suggested:
o Reduction of drag-out.
In an immersion-type treatment process, small objects are placed inside
barrels and bigger objects are supported on racks for immersion into the
bath. When the object is removed from the bath, the rack or barrel (and
object) carry some reagents with it, called "drag-out". Methods that can
be used to reduce drag-out and subsequently lower rinse water
requirements are: proper racking of the parts; keeping the racks free of
metal buildup and corrosion; increasing drainage time above the process
tank; using stationary recovery rinses by installing save rinse or drip rinse
tanks; using air blowoff or tumbling to ensure drainage; and using drainage
agents (Cheremisinoff, Peina, and Ciancia 1976, AESI 1981, and Cook et.
al. 1984).
o Effective rinsing methods.
By using an efficent rinsing sequence, the quantity of rinse water required
can be reduced substantially. Rinsing efficiency can be improved by using
properly designed rinse tanks, using air agitation in the rinse tanks, using
fog sprays, using automatic valves that control flow rates based on the
movement of parts through the processing line, and by using counter-
current rinsing (Cheremisinoff, Peina, and Ciancia 1976, AESI 1981, and
USC 1983).
An estimated 90 percent reduction in rinse water can be achieved by using
a countercurrent rinse instead of a single running rinse (AESI 1981).
Converting to a countercurrent rinse requires only the addition of one or
more tanks, appropriate plumbing, and an air agitation system. Because of
space limitations or the use of preprogrammed hoist lines, installation of
an additional rinse tank may not be possible at many job shops. However,
many facilities have reported substantial savings by converting to
countercurrent rinsing.
86-19
-------
The use of flow control valves can reduce rinse water use by 50 percent
with minimal capital costs (AESI 1981). In many instances, excess water is
used indiscriminately to ensure total rinse to protect product quality.
Though many facilities have installed flow control valves, concerns about
reduced product quality have contributed to opposition to such measures
(AESI 1981).
Fog sprays, though efficient, are not suitable for all applications. In
instances where the coating has little strength initially (as in the case of
chromate coatings), fog sprays are generally not used. Use of air agitation
in the rinse tanks promotes turbulence in the tank which increases the
rinsing efficiency. The use of air agitation in tanks is fairly widespread
(AESI 1981).
o Use of immiscible rinses.
The use of non-aqueous immiscible solvent for rinsing would allow the
rinsed solution to either sink or rise during decantation, and the solution
could then be returned to the surface treatment bath for reuse without any
pretreatment. The rinsing solvent could also be recycled. This process, if
feasible, could reduce or eliminate rinse water wastes. Tests with five
solvents for use in the chromating process were conducted at the United
Technologies Research Center (AESI 1981). A major disadvantage of this
method would be the potential for increased air emissions and the need to
dispose of spent solvent.
o The use of no-rinse coating processes.
As the name suggests, no-rinse coatings do not require rinsing after a
coating is formed and dried, as there are no residuals left to interfere with
the subsequent treatment. Recent developments in chromate conversion
coating for the coil coating industry have resulted in a solution that can be
applied to steel, galvanized steel, or aluminum, without the need for any
subsequent rinsing (USEPA 1982). After the coating is formed, it is dried
in air at about 150°F. This no-rinse process, though used by only a few coil
coating facilities, can be used for other coating applications (USEPA 1982).
B6-20
-------
The no-rinse process does have some disadvantages. These are: the
inherent high speed of operation requires very efficient control; existing
facilities are difficult and expensive to adapt to no-rinse coating
operations; and the no-rinse conversion coatings are not FDA approved for
food grade coatings.
Reuse/recycle of rinse water.
A literature review on the reuse and recycle of process water in the metal
finishing industry was conducted (Mathews 1980). The following paragraphs
deal with some of the reuse/recycle options that are already in use or were
suggested for use.
In the chromate process, the first rinse (high in chromic acid), can be
recycled to the chromating tank (UNECE 1982). The last rinse can be
regenerated using ion-exchange resins and can then be recycled to the
rinsing process. Or, by sending the rinse water to an evaporator, the
concentrate from the evaporator can be recycled to the coating bath, and
the overhead from the evaporator can be condensed and used for rinsing
(AESI 1981, Basta 1984). This scheme (illustrated in Figure 9-1) can reduce '
rinse water requirements drastically. The method requires sufficient water
to be evaporated and recycled to satisfy rinsing requirements. The
economics of such systems has been discussed in detail elsewhere (AESI
1981). Case studies of facilities using evaporators have also been reported
(Lewis 1980). The use of atmospheric evaporators, while decreasing energy
requirements, requires significantly more floor space than conventional
evaporators. Such systems are in use at some facilities (Campbell and
Glenn 1982).
Concentration of rinse water for recycle to plating baths or for reuse can
be achieved using ion exchange resins. A reciprocating flow ion exchanger
for drag-out recovery from chromium, copper, and mixed coating processes
is in current usage. The use of similar ion exchange columns for rinse
water treatment and reuse is expected to have great potential in the
future. The use of ion exchange columns does have some disadvantages.
These are: ion exchange is more expensive than conventional chemical
B6-21
-------
CD
ho
N3
HAKE UP _
SOLUTIONS
PRODUCT AND DRAG-OUT
. -
SURFACE
TREATMENT
BATH
CONCENTRATED
TREATING
SOLUTION
RINSE
TANK
r
RINSE
TANK
DISTILLATE
EVAPORATOR
FINISHED
PRODUCT
HAKE UP
RINSE HATER
Figure 9-1 Closed Loop Evaporation System for Metal Surface Treatment Haste Reduction
-------
treatment; regeneration steps generate their own wastes; and the process
is only favorable for streams with a low concentration of contaminants (up
to approximately 250 ppm).
Reverse osmosis for concentrating rinse waters has found limited use.
Here, water is preferentially forced through the pores of a semi-permeable
membrane and two streams are generated: a concentrated salt solution
that can be sent to the bath, and a pure water stream that can be used for
rinsing. Some of the limitations confronting the process are that the
membranes operate in the high pH range only (pretreatment to assure high
pH is necessary); the reliability or durability of the membranes under
variable conditions characteristic of mixed wastewater is questionable;
supplemental evaporation (to boost concentration) may be necessary before
the salt solution can be recycled to the coating process; and the process is
economical only for processes with high drag-out rates (something that can
be controlled by good operating practices). Cellulose acetate membranes
are used by some nickel coating facilities. Other potential membrane
applications include copper, zinc coating, and chromating operations.
The electrodialysis process uses alternately placed anion and cation
permeable membranes and separates the rinse waters into a dilute stream
that can be used for rinsing and into a concentrated stream that can be
recycled to the coating operation. The process is continuous and the major
operating cost is for electricity. Currently, some nickel, zinc, and
chromating facilities use this process.
Replacing hexavalent chromium with trivalent chromium.
In chromate conversion coatings, hexavalent chromium can be replaced by
trivalent chromium, which reduces the hazardous nature of rinse waters.
This method is discussed in more detail in Section 9.1.1.
Replace cyaniding by gas carbonitriding.
As explained in Section 9.1.1, cyaniding requires a quench-rinse operation
to wash cyanides present on the objects. The use of gas carbonitriding
eliminates the need for this rinsing operation that generates a hazardous
waste stream.
B6-23
-------
Use of still rinsing.
The work piece can be immersed in a still rinse tank following the metal
finishing bath. The still rinse tank has no inflow or outflow of rinse water,
and the finishing bath constituents build up in it. When the rinse water in
the still rinse tank becomes concentrated enough, it can be used to
replenish the finishing bath solution. The use of still rinsing can cut down
the quantity of rinse water required. This process, when used for initial
rinsing, can facilitate metal recovery from the rinse water. Still rinsing is
usually followed by a spray or countercurrent rinse to ensure the complete
removal of contaminants.
Metal reclamation from rinse water wastes.
By using electrolytic or non-electrolytic methods, as explained in Section
9.1.1, metals present in the rinse waters can be recovered. This reduces
the hazardous nature of this stream.
Changing rinse composition.
i
Changing the rinse composition to reduce its hazardous nature should be
attempted whenever possible. In a zinc-based phosphating bath, the final
rinsing of the object is usually done with dilute chromic acid. Environ-
mental concerns about chrome caused several facilities to develop and use
chrome-free rinses, even though these rinses were not as effective as the
chrome-containing rinses (Schneberger 1981).
Minimizing process water use.
By reusing rinse water effluents from certain operations in other
operations with lower rinse water quality requirements, overall aqueous
wastes can be minimized. The use of water from fume scrubbers has been
shown to be practical for rinsing in certain cases (Cheremisinoff, Peina,
and Ciancia 1976). Used cooling water or steam condensate might also be
used for rinsing if technically permissible and economically justified.
Better operating practices.
A typical metal finishing facility has various coating operations being
performed at the same time, and mixing of various rinse streams is not
B6-24
-------
uncommon. By segregating various rinses, their reuse or recycle can be
promoted. Metal reclamation by electrolysis from these various streams is
easier if they are not mixed together. In the recent past, rinse waters and
spent baths were mixed and treated together. Segregating spent bath
wastes from rinse water wastes is quite common at present.
9.1.3 Solid Wastes
Solid wastes are generated during the filtration of the bath to remove precipitated
metallic salts, and by the clarification, dewatering, and filtering operations performed
during waste treatment. These solid wastes are usually disposed of by landfilling or
are sent to an outside reclaimer. The following source reduction techniques are
considered:
o Metal reclamation from the solid waste.
It was reported by Lisanti and Helwick (AESI 1981) that a certain facility
hauled solids to another plant within its corporation for metal reclamation.
Depending on waste volume, such recovery can be an uneconomical
proposition in terms of capital costs. Individually, the quantities generated
might not be large enough to justify a dedicated recovery system. The
alternative to individual treatment systems for metal recovery would be a
centralized waste treatment facility, where the wastes could be
collectively treated. In such a facility, the recovery options would be more
economically attractive. Such centralized treatment facilities are in
operation in West Germany.
o Effective dewatering of the solids.
Solids are dewatered by some facilities using a filter press. By effectively
dewatering the solids, the quantity of waste sent for landfilling can be
decreased by about 50 percent (AESI 1981). More effective dewatering can
be achieved using disc, scroll, or basket type centrifuges, vacuum drums or
belt filters.
B6-25
-------
9.1.4 Spills and Leaks
Spills are the result of overflow from various process tanks and other inadvertent
discharges, such as valve closure failure, leaky gaskets, and drips. The following
source reduction method is postulated:
o Better operating practices.
By using splash guards and drip boards, spilled liquids can be recycled and
spilled cleaning wastes can be avoided. Use of float valves or alarm
systems are inexpensive options to prevent overflows. Other possibilities
include liquid level controllers. Good housekeeping measures such as
periodic inspection of process equipment and piping, periodic relining of
the tanks, and training and educating personnel to be cognizant of the
importance of controlling waste generation can reduce the number and size
of spills and leaks occuring in a facility.
9.1.5 Stripping Wastes
Stripping wastes, which are generated most commonly in small job shops, are the
result of removing old or bad coatings on objects prior to the required surface finishing
operation. The operation is essentially an etching type treatment, whereby the
coating is dissolved in an acid. Since the same bath may be used to strip various metal
objects, this waste is difficult to deal with since recycling or regeneration is not
generally possible. The following source reduction techniques are suggested:
o Use of non-chrome etchants.
Whenever possible, ferric chloride, or ammonium persulfate solutions
should be used instead of chromic-sulfuric acid etchants. However,
compatibility with the planned surface treatment must be carefully
examined. The use of such etchants will reduce the toxicity of the
stripping solution that ends up as a waste.
o Decrease the generation of off-spec coatings that require stripping.
B6-26
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY FOR THE METAL SURFACE TREATMENT INDUSTRY
1
Haste Stream |
1
Spent Bath |l
Solutions («) |2
|5
IS
8
1
Waste Rinse Hater |t.
12
|3
IS
is.
li-
lt-
|10
1
Filter Wastes (*} |1.
i
Spills and Leaks |1.
1
Stripping Haste |1.
|2.
I
| All Sources |
I
Control Methodology |~
I
Extend bath life I
Metal/acid recovery from spent baths |
Spray/brush Heirs instead of immerse )
Use thinner foil for P.C.B.'s |
Alternative treatment techniques I
Use of less toxic solutions |
More dilute process solutions |
Better operating practices |
Overall 1
Reduce dragout of solution from tank |
Employ effective rinsing methods |
Use of immiscible rinses |
Use of no-rinse coatings |
Reuse and recycle spent rinse water |
Replace hex chrome with trlvalent j
Replace cyanidlng with gas carbon. |
Use still rinsing technique |
Reclaim metal from rinse water wastes |
Change rinse composition |
Minimize process water use i
Better operating practices i
Overall |
Metal reclamation from solid waste |
Effective dewatering of the solids |
Overall |
Better operating practices |
Overall |
Use of non-chrome etchants |
Reduce generation of off-spec coat ing |
Overall |
All Methods
Found Documentation |
Quantity |
2 !
1
1
1
1
1
1
1.25 |
2 1
2 !
1 1
1 1
1 1
1
1 i
1 1
3 1
1 1
i !
' i
1.33 |
1 1
' 1
1.00 |
2 1
2 00 |
1 1
1 1
t 00 |
Quality 1
2 1
2 1
1 1
1 1
2 :
2 !
2 i
2 1
1.75 |
2 1
2 1
2 !
2 !
2 !
2 I
1 1
1 |
3 1
2 I
2 1
2 1
1 92 |
1 1
2 !
1.50 |
t !
1 00 1
1 I
1 1
i oo !
Haste | Extent of |
Reduction | Current Use |
Effectiveness | 1
t 1
2 1
1 1
2 1
2 I
2 i
2 1
2 1
K75|
2 1
3 I
3 1
2 1
2 1
3 1
2 1
2 I
2 1
2 1
2 i
2 17 !
2 1
2 |
2.00 |
2 1
2 00 |
2 1
2 1
2001
1 1
t 1
2 1
3 !
1 1
2 1
1 1
2 1
1.63 |
1 1
1 1
0 1
1 1
J 1
1 1
1 1
1 1
2 1
1 I
1 1
1 1
1 08 |
1 1
t 1
1.00 |
2 1
2.00 |
1 I
1 1
1 00 !
Future | Fraction of | Current
Application | Total Haste | Reduction
Potential | | Index
2 1
2 !
l 1
' 1
' 1
2 1
2 !
3 I
1.75 |
3 I
3 1
1 1
2 1
3 1
' 1
1 1
2 1
2 1
1 1
2 1
2 1
1.92 |
1 1
2 1
1.50 |
2 1
2.00 |
2 1
2 1
2.00 i
1
1 0
1 0
1 '
1 1
! 0
1 1
0.15 | 1
1 o
t o
1 o
1 '
1 ®
1 o
1 1
1 0
1 o
1 o
0.65 | 1
1 o
1 o
0.05 ! 0
1 1
0 OS | 1
1 o
i o
0 10 0
i oo i i
| Future Reduction Index |
i 1
1
1
3 1
5 1
5 1
5 1
5 I
0 1
5 1
0 1
5 1
5 1
• 1
0 1
> 1
o 1
1
1
1
1
1
1
5 1
0 1
5 I
5 1
5 1
0 1
0 1
5 1
5 I
5 !
0 1
Probable
0
0
0
0
0
0
0
0
0
1
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
| Maximum |
— * 4
4
8
1
1
4
5
8
8
5
t
7
3
1
8
4
6
8
5
4
8
8
8
4
8
1
0.8 |
1
1
1
1
O.I |
08 |
4
0.8 |
1
1.7 |
1
1
1
1
1
1
1
1
i
1
1.7 |
1
08 |
6 I 0 8 I
5| 0 5 |
5 I 05|
8| 0 8 |
8 | 08!
8 0 8 |
T I 1 3 !
CD
cr>
i
ro
(') Ihese streams Included listed 'F' and/or 'K' RCRA wastes
-------
9.2 Implementation Profile
The source control methods discussed for use by the metal surface finishing industry
have different potentials for application depending on whether the facility is a captive
shop or a job shop. Job shops are usually required by their clientele to follow certain
operating procedures, and most job shops buy their process solutions from
manufacturers. Thus, modifications in certain operating procedures and changes in
process solutions are not feasible in many job shops without the guidance and approval
of the customer. Also, because smaller operations usually have space limitations,
installation of additional tanks for countercurrent rinsing may not be practical. For
facilities which have preprogrammed hoist lines in operation, changes in operations
such as rinsing may be difficult. However, measures such as the installation of flow
control valves, splash guards, etc., which do not require much capital or space can be
used by all facilities. Efficient use of water can be made by reusing spent process
water at places that require low quality water. Improvements in operating practices,
such as spill and leak prevention, are additional measures which require little capital
to implement.
Reclamation of metals from wastes is an effective way to reduce or eliminate the
hazardous constituents of these wastes. In many job shops, the volume of wastes may
be too low for economical recovery of metals. In such cases, the use of centralized
waste treatment facilities to collect and treat wastes from various plants has proven
in some to be a long-term economical solution.
9.3 Summary
The summary of all noted source control techniques is given in Table 9-1. Each
technique was rated for its effectiveness, extent of current use and future application
potential on scale of 0 to 4. The ratings were derived by project staff based on review
of the available data and in consultation with the industry. The estimates of current
level of waste reduction achieved (current reduction index) and possible future
reduction (future reduction index) were obtained from the ratings in accordance with
the methodology presented in the introduction to this appendix.
B6-28
-------
The current reduction index (CRI) is a measure of reduction of waste that would be
generated if none of the methods listed were implemented to their current level of
application. For the entire metal surface finishing industry, CRI is 1.0 (25 percent)
which is indicative of the low to moderate level of waste minimization that already
has taken place. Current measures that have proven effective as a whole have been
the use of less toxic bath solutions, the reuse and recycling of spent rinse water, the
reclaiming of metal from waste, and the implementation of better operating practices.
The future reduction index (FRI) is an indication of the level to which the currently
generated waste can be reduced if all of the techniques noted were implemented
according to their rated potential. The FRI value of 0.7 to 1.3 (18 to 33 percent) is
indicative of the moderate extent of future waste reductions. Among the techniques
that were found most effective and applicable (as evidenced by high FRI value),
employment of effective rinsing methods, metal/acid recovery from spent baths, use
of more dilute process solutions, and further implementation of better operating
practices appear to promise the greatest level of reduction for the industry as a whole.
10. PRODUCT SUBSTITUTION ALTERNATIVES
Since many metal objects can be coated with several alternative metals to achieve the
same quality of treated surface, certain substitutions could contribute towards
hazardous waste reduction. The replacement of nickel by zinc is possible in some
instances. In other applications, zinc can be substituted by manganese or iron.
Electroless nickel plating, used by some printed circuit boards manufacturers, can be
substituted with electroless copper plating. Cadmium and silver can generally be
substituted with other metals.
11. CONCLUSIONS
The estimates indicate that moderate reductions of waste generated by the metal
surface finishing industry are possible, probably in the 18 to 33 percent range. Several
effective source control measures include use of efficient rinsing techniques,
recovering metal/acid from spent baths, use of more dilute process solutions and
further implementation of better operating practices.
B6-29
-------
12. REFERENCES
AESI. 1981. American Electroplater's Society, Inc. Conference on advanced pollution
control for the metal finishing industry. EPA-6QO-2-81-028. Cincinnati, Ohio: U.S.
Environmental Protection Agency.
Alban, L.E. 1981. Metal surface treatments: case hardening. In Kirk-Qthmer
Encyclopedia of Chemical Technology. 3rd ed. Vol. 15, pp. 313-24. New York, N.Y.:
Wiley.
Arienti, M. 1985. GCA Corporation. Waste category assessment report, Solvent
waste; draft final report. EPA Contract No. 68-03-3243. Cincinnati, Ohio: U.S.
Environmental Protection Agency.
Basta, N. 1983. Total metal recycle is metal finishers' goal. Chemical Engineering.
90(10)816-19.
Basta, N. 1984. Ion-beam processing: new surface treatment method. Chem.
Process. 91(16): 18-21.
BCL. 1976. Batelle Columbus Lab. Assessment of industrial hazardous waste
practices: electroplating and metal finishing industrial job shops. EPA-530-SW-136C.
Washington, D.C.: U.S. Environmental Protection Agency.
Campbell, M.E., and Glenn, W.M. 1982. Profit from pollution prevention. Ontario,
Canada: Pollution Probe Foundation.
Chementator. 1982. Chemical Engineering. 89(14):9-10.
Cheremisinoff, P.N., Peina, A.J., and Ciancia, J. 1976. Ind. Wastes. 22(6):31-4.
Cook, T.M., Cubbage, M.L., and Fister, L.J., 1984. Draining process solutions from
sheets, baskets, pipes, threads and fins, Metal Finishing, (7):33.
C.P. Staff. 1984. Hexavalent chromium waste detoxified by chemical system, dewater
through filter press. Chem. Process. 47(10):38.
Durney, L.J., ed. 1984. Electroplating engineering handbook. 4th ed. New York,
N.Y.: Van Nostraud Reichold Co.
Lane, C., 1985. Fluorocarbon coating eliminates corrosion of acid bath racks. Chem.
Process. 48(10): 72.
Lewis, T.A. 1980. Ind. Finish. April 1980.
Mathews, J.E. 1980. Industrial reuse and recycle of wastewater; literature review.
Robert S. Kerr Environmental Research Lab. EPA-600-2-80-183. Ada, Okla: U.S.
Environmental Protection Agency.
Olsen, A.E. 1973. Upgrading metal finishing facilities to reduce pollution. Oxy Metal
Finishing Operation. EPA-625-3-73-002 (USEPA Technology Transfer). Washington,
D.C.: U.S. Environmental Protection Agency.
B6-30
-------
Saubestre, E.B., 1957. Proc. Am. Electroplating Soc. 46:264.
Schneberger, G.L. 1981. Metal surface treatments: chemical and electroplating
conversion treatment. In Kirk-Othmer encyclopedia of chemical technology. 3rd ed.
Vol. 15, pp. 304-12. New York, N.Y.: Wiley.
UNECE. 1982. United Nations Economic Commision for Europe. Compendium on low
and non-waste technology. Vol. 4. Monograph 73. Geneva: United Nations Economic
Commission for Europe.
USC. 1983. U.S. Congress, Office of Technology Assessment. Technologies and
management strategies for hazardous waste control. Washington, D.C.: U.S.
Government Printing Office.
USEPA. 1980. U.S. Environmental Protection Agency, Office of Water Regulations
and Standards. Development document for effluent limitation; guidelines and
standards for the metal finishing industry. EPA-440-1-80-091A. Washington, D.C.:
U.S. Environmental Protection Agency.
. 1981. U.S. Environmental Protection Agency, Industrial
Environmental Research Lab. Implement changes for metal finishers. Cincinnati,
Ohio: U.S. Environmental Protection Agency.
: 1982. U.S. Environmental Protection Agency, Effluent
Guidelines Division. Development document for effluent limitations; guidelines and
standards for the coil coating industry. EPA-440-1-82-071. Washington, D.C.: U.S.
Environmental Protection Agency.
Versar, Inc. 1985. Versar. National Profiles Report for Recycling/a preliminary
assessment, Draft report for waste treatment branch. EPA Contract No. 68-01-7053.
U.S. Environmental Protection Agency.
13. INDUSTRY CONTACTS
Dr. J. Chu, Environmental Activities Staff, General Motors Corp., Warren, MI.
D. Anzures, National Association of Metal Finishers, San Fernando, CA.
W.G. Vaux, P.E., Chemical and Process Engineering, Westinghouse Electric Corp.,
Pittsburgh, PA.
B6-31
-------
-------
1. PROCESS: ORGANIC DYES AND PIGMENTS MANUFACTURE
2. SIC CODE: 28652, 28653
3. INDUSTRY DESCRIPTION
Manufacturers of organic dyes and pigments are included as part of the cyclic crudes
and intermediates manufacturing industry. The companies are engaged in the
conversion of cyclic organic chemicals into more complex intermediates, and/or in the
subsequent formulations of these products into dyes and pigments,
3.1 Company Size Distribution
The 1982 Census of Manufacturers (USDC 1985) reported a total of 143 companies
involved in the manufacturing of cyclic crudes and intermediates, with 189
manufacturing plants located throughout the U.S. As of 1983, 22 companies produced
only dyes, 21 produced only pigments, and 12 produced both dyes and pigments. Since
no specific size data were found for the organic dyes and pigments sector, Table 3-1
gives the company size distribution in terms of the number of employees for the entire
cyclic crudes and intermediates manufacturing industry.
Table 3-1 Company Size Distribution
No. of
establish-
ments
Total
189
No. of Employees Per Facility
1-49 50-99 100-249 250-499 500-999
93 25 41 18 8
1000-250C
4
No. of
employees 27,075 2,325 1,875 7,175 2,700 6,000 7,000
Source: 1982 Census of Manufacturers (USDC 1985).
B7-1
-------
3.2 Principal Producers
The principal producers of organic dyes and pigments are listed below:
Mobay Chemical Corporation
Sandoz Chemicals Corporation
Toms River Chemical Corporation
ICI United States, Inc.
Crompton & Knowles Corporation
Atlantic Industries, Inc.
3.3 Geographical Distribution
BASF Wyandotte Pigments
Sun Chemical Corporation
American Hoechst Corporation
Harshaw-Filtrol Partnership
Galaxie Chemical Corporation
Ciba-Geigy Corporation
According to data from 1977, there were 55 companies producing organic dyes and
pigments at 87 manufacturing sites. Among these, 89 percent of the plants were
located in 11 eastern states. The remaining sites were distributed throughout the
country as shown in Figure 3-1 and Table 3-2 below:
Table 3-2 Location of Organic Dyes and Pigments
Manufacturing Facilities in th U.S.
EPA Region
No. of Establishments
I
II
III
IV
V
VI
VII
VIII
IX
X
8
38
11
13
12
0
1
0
3
0
National
86
Source: Industrial Process Profiles (Radian Corp., 1977).
(a) The location of one plant was not reported.
B7-2
-------
VIII
CD
I
OJ
0
2-5
r^d 6-10
^;^j over 10
Roman numerals show EPA regions
Figure 3-1 Organic Dj^/Pigment Plants in the U.S.
-------
4. PRODUCTS AND THEIR USE
There are over 1000 different dyes and pigments being produced in the U.S., with
pigments being fewer in number (approximately 10 percent). These products are sold
as pastes, powders, lumps, dispersions in organics, or aqueous solutions with
concentrations between 6% and 100%. Of the total quantity of dyes produced, two-
thirds is used as colorants in the textile industry, one-sixth is used in coloring paper,
and the rest is used in producing organic pigments and in the dyeing of leathers and
plastics. For pigments, one-half of the quantity produced is used in printing, while the
rest is used as colorant for paints, plastics, textiles, and paper.
According to their applications, dye products are classified into 10 principal classes.
These are: (1) acid dyes, (2) basic dyes, (3) direct dyes, (4) disperse dyes, (5) reactive
dyes, (6) fluorescent brightening agents, (7) food, drug and cosmetic colorants, (8)
mordant dyes, (9) solvent dyes, and (10) vat dyes. U.S. annual production of dyes in
1983 was 122 thousand short tons per year (TPY) (US1TC 1983). The production of
direct and solvent dyes decreased as compared to the previous years, while the
production of the remaining 8 classes increased. Fluorescent brightening agents
accounted for the largest portion of dyes produced in 1983 (25%). For the remaining
classes, vat dyes, disperse dyes, direct dyes, and acid dyes accounted for 21%, 15%,
12%, and 11% of the annual total production, respectively.
Pigments are classified either as toners or lakes. Toners are full-strength organic or
metallo-organic colorants that do not contain any inorganic pigments or carriers.
Lakes, on the other hand, are combinations of a dye with an inorganic carrier, usually
alumina hydrate. The production rate of pigments in 1983 was 39 thousand tons per
year, with 99% of the amount produced as toners (USITC 1983).
5. RAW MATERIALS
A wide variety of organic and inorganic materials are used in the production of organic
dyes and pigments. Large manufacturing plants produce their own dye intermediates
from basic units such as benzene or naphthalene. Small manufacturers, however, will
bring in intermediates from other sources, often foreign. Most of the organic
intermediates are hazardous in nature, with some being identified as carcinogenic.
B7-4
-------
The common types of raw materials used in this industry are listed below (Catino and
Farris 1979, Bannister, Olin, and Stingle 1979, Fytelson 1982, Radian 1977, SRI 1984):
Organics:
Inorganics:
Acids:
derivatives of benzene, toluene, naphthalene, anthraquinone, and
other aryl systems including heterocyclics, which contain functional
groups such as sulfonic acid substituents, hydroxyl, amino,
alkylamino, chloro, nitro, carboxyl, alkoxyl and others; reagents such
as lower alcohols and amines, formaldehyde, phosgene,
dimethylsulfate, acetic acid, and glycerine.
alumina hydrate, aluminum salts, ammonium salts, calcium chloride,
sulfur, chlorine, bromine, chromium salts, copper and copper salts,
iron and iron salts, iodine, nickel salts, sodium salts, potassium salts,
lead peroxide, phosphorous chlorides.
sulfuric acid, oleum, nitric acid, hydrochloric acid, chlorosulfonic
acid, acetic acid, chloroacetic acid.
Bases:
ammonia, calcium carbonate, sodium hydroxide, calcium oxide.
Catalysts: zinc chloride, aluminum chloride, ferric chloride, iron, Raney nickel
catalyst, stannous chloride, magnesium oxide, manganese dioxide,
lead peroxide.
6. PROCESS DESCRIPTION
Organic dyes and pigments are produced from a great variety of cyclic intermediates
using many processes. Manufacturers often purchase organic chemicals, usually in the
form of cyclic benzenoids, and convert these into more complex intermediates, and
ultimately into dyes and pigments. The distinction between dyes and dye
intermediates is somewhat arbitrary, since some dye intermediates may be used as
dyes, or they can be further processed to form other dyes or pigments. Similarly,
pigments are differentiated from dyes based on application methods rather than
chemical constitution. Pigments are much less soluble and they retain their granular
form throughout the application process.
B7-5
-------
Although the number of different types of organic dyes and pigments produced is
large, the majority of these products can be classified into eight basic classes which
are based on the chemical structure of the common color-bearing units (chromophores)
in the molecules. These include:
o azo dyes and pigments,
o anthraquinone dyes and pigments,
o stilbene dyes and fluorescent brightening agents,
o diphenyl methane and triarylmethane dyes and pigments,
o methine and polymethine dyes and pigments,
o xanthene dyes and pigments,
o phthalocyanine dyes and pigments,
o sulfur dyes.
Of the eight classes, azo dyes and pigments represent the most important class
commercially produced in the U.S., accounting for 35 percent of all U.S. dye and
pigment production in 1978. Since data on U.S. production by structural class have not
been reported since 1978, the information on present relative and total production of
these dyes and pigments, by class, remains uncertain. Nevertheless, based on the
information provided in a recent EPA report on the waste generation from the organic
dyes and pigment industry (SRI 1984), it is believed that azo dyes and pigments still
represent the largest volume of dyes and pigments produced. Therefore, this report
will focus on the production of azo dyes and pigments. Their manufacturing process
will be examined in detail to exemplify the degree of complexity which can be
encountered in the production of organic dyes or pigments.
Figure 6-1 presents the flow chart for the manufacturing of azo dyes and pigments.
There are basically three general stages in the production of any dye or pigment.
These are:
Synthesis stage.
Operations performed in this stage vary widely among the dyes and
pigments due to the differences in the sequence of reactions required
to obtain the desired structure of a specific dye or dye intermediate.
B7-6
-------
ARYLANINE
•ATER —
OR
M
©
(RESIN 1 I
1 1
\ -I 1_
®i NANOg 1 !
\ ICE f I DIAZOTIZATION
COUPLER I SALT 1
1 * ~i
I";™ — > WUPLINS j
•ABTEIATER
SOLUBLE [7rY«H! i ^
DYEB r~lT10N©| _,
/J\ x^-v^ INSOLUBLE
DTE ( " PISNENTS
PRECIPITANT \N_J,/r ^
1 ! | FILTRATION
J PRECIPITATION | 1 .... SsT J
~*1 FILTRATION QJ
1 1
A DYE OR PISNENT
I PRESSCAKE OR SLURRY
© "^
BANK __ {
:oiu« P |
i i , 1 1
FLUBHINB^INORBANIc' "jjJJ^jjj"' f «ATE» 1 J NET BRINOINB j
, ®! "LT i © 1""F"-' i , ®i
1 ' r ' 1
* 1 *
"'*«1 ^2^. DYE OR
DISPERSION f ^ .i««eyT
„„ BTe / SROUNO PIBNENT PI8NENT
SOLUTION V °* DTE pA|TE
LNABHINB AND
DRYINt
~r^ ^
FINISHED DYES OR PISNENTS
PROCESS IABTE CATEBODIES'
© DISCARDED CONTAINERS ,
© SYNTHESIS NABTEB |
© PRODUCT FINISHINB HASTES <
© ERUIPNENT CLEANINB IASTES
1
© SPILLS AND OFF -SPEC PRODUCTS i
SYNTHESIS OF
AZO DYES
DYE PRECIPITATION AND
PRODUCTION OF
PISNENTS
FRO*
SELECTED DYEB
FOR THE
PRODUCTION OF
DYEB OR PIBNENTS
Figure 6- i Block Flo* Diagrai for the Production of Azo Dyes and Pigients
87-7
-------
Precipitation stage.
Dyes which are still soluble after the last of the reaction sequence
are often precipitated out by the addition of salts. Similarly,
pigments can be obtained directly or from the precipitation of dye
intermediates on resin, or by the addition of metal salts.
Finishing stage.
This stage is common to the production of all dyes and pigments. It
usually involves the processing of dyes and pigments into their final
forms through operations such as drying, grinding and standardizing
as required by the dye users.
The following sections describe in detail the steps involved in each of the stages for
the production of azo dyes and pigments.
6.1 Synthesis of Azo Dyes and Pigments
Azo dyes are compounds containing one or more azo groups (-N=N-) in their
molecules. Azo dye products consist of direct dyes, acid dyes, pigments, and a small
amount of azoic compounds which are incomplete dyes that require coupling during the
application process. Azo dyes and pigments are produced via a two-step process
consisting of the diazotization and coupling of aromatic amines.
The manufacturing process begins with a diazotization reaction. In this first stage,
arylamines such as anilene are reacted with nitrous acid to form diazonium chloride
salt according to the reaction:
ArNH2 + NaNO2 + 2HC1 - - , ArN = N+C1" + NaCl + 2H2O
where Ar is an aromatic group. Since nitrous acid is very unstable, it is produced as
needed from the reaction of an aqueous solution of sodium nitrite and hydrochloric
acid. The batch reaction is carried out at 32°F - 40°F in a period of 1-3 hours using
arylamine, NaNO2, and HC1 in a molar ratio of 1:2:3.
The product is an aqueous solution containing 15% diazonium chloride and 7% NaCl. In
the production of azo dyes, as well as of any other dyes and pigments, the reactions
B7-8
-------
are almost always carried out in batch reactors because the small total demand does
not justify the use of a continuous process. The reactions are carried out batchwise in
one or several reactor trains. Each train includes at least one diazotizing reactor, one
coupling reactor, and one filter press. The kettles are made from cast iron, stainless
steel, glass-lined steel, wood, or brick, and have a capacity of 500 to 10,000 gallons.
The reactors are equipped with mechanical agitators, temperature recorders, and pH-
probes. Jackets or coils are used for heating and cooling. Unjacketed reactors are
also used, and the temperature is controlled by direct introduction of steam or ice.
The yield from a diazotization reaction is fairly high. The use of HC1, however, may
result in the formation of byproducts. For example, the diazonium salt may
decompose into compounds which may be further chlorinated into polychlorinated
organic byproducts. Furthermore, the decomposition products may react with the
original arylamine to form secondary amine compounds. These secondary amine
compounds can then react with nitric acid and form insoluble compounds.
A clarification operation is often performed to remove these solid by-products before
the intermediate is transferred to the coupling reactor. The resulting solid waste is
often sent to land disposal. In addition to these byproducts, other major contaminants
include undecomposed diazotic acids. Once clarified, the intermediates from the
diazotization reactor are then transferred to the coupling reactor.
In the second stage where coupling occurs, the diazonium salt is reacted with another
aromatic compound, called a "coupling component," which has a strong electron-
donating substituent such as a phenolic, amino, or substituted amino functional group.
The reaction forms azo compounds in the following manner:
Ar-N=N+Cl-+ Ar'-H —. Ar-N=N-Ar' + HC1.
Coupling is generally conducted in a mildly alkaline condition by the addition of NaOH.
The reactor is operated at 32°F-40°F for 4-24 hours. The ratio of diazo salt to the
coupling component varies depending on the degree of coupling required. High pH is
often avoided so as to minimize non-coupling reactions leading to the formation of
undesirable byproducts, e.g. triazene compounds:
NaOH H
ArN2 + Cl- + Ar'NH2 v Ar-N=N-N-Ar' + H2O + NaCl
B7-9
-------
Furthermore, coupling reactions must compete with the decomposition of the
diazonium salts to form phenols as byproducts from the reaction:
x NaOH
ArN=N+Cl--- N2 + (Ar+Cl-) - ArOH+NaCl
In addition, undecomposed diazotic acids carried over from the diazotization reactor
may react in the basic medium to form diazotate salts. These salts are insoluble and
often require filtration of the coupling reactor effluent to remove them from the
soluble dye solution. The resulting filter press cake is sent to land disposal.
6.2 Soluble Dye Precipitation and/or Pigment Production from Selected Dyes
Dyes formed from the coupling reaction, or formed at the end of the synthesis
sequence in general, can be either soluble or insoluble. Insoluble dyes (or pigments)
are often filtered, and the products are collected in the form of filter press cakes.
These are then sent to dye/pigment finishing operations before being sold as final
products. Soluble dyes are precipitated by the addition of salts. The resulting slurry is
filtered, and the press cake is air blown to remove the mother liquor. The dyes are
then transferred to finishing steps. The mother liquor which contains byproducts and
residual dyes and pigments is treated on-site with other process wastewaters before
being discharged. Similarly, dyes that are too easily soluble in water and/or organic
solvents can be converted into pigments with the requisite low solubility
characteristics. These pigments (lakes) can be formed by precipitation of water
soluble dyes on an adsorptive surface such as resin or alumina hydrate, or from the
precipitation of an acid or basic dye using inorganic salts. Lakes formed from acid
dyes are precipitated by the addition of soluble salts of alkaline-earth metals, while
lakes formed from basic dyes are made from precipitation with regular metal salts.
The precipitation process is carried out by stirring dyes and dye precipitants in a steel
vat. The reacting solutions are heated to 120°F-170°F, and fed in at a prescribed
rate. The product slurry is then cooled to obtain the desired crystalline form of the
pigment products. The product slurry is filtered, and the resulting filter press cakes
are sent to finishing operations. The mother liquor containing residual dyes, pigments
and metal salts is not recovered and is sent to wastewater treatment.
B7-10
-------
6.3 Finishing Operations
After the dyes or pigments are collected in cake form or as slurries, they are sent
through a series of finishing steps before being sold as final products. The physical
form of the product plays an important part in the marketing of the material.
Pigment attributes such as crystal size, specific, surface, and particle shape have
enormous influence on the quality and performance of a pigment.
One important finishing operation is called flushing. This is used to transfer pigments
from a water-borne phase to an organic or oil-borne phase (67 percent of pigments are
sold in this organic form). The flushing operation is usually done in more than one
charge in steel tanks equipped with high speed stirrers or blade agitators. The residual
water from these charges is then removed by vacuum stripping. The finished products
are pigment dispersions or dye solutions. Resins and surfactants are usually added to
stabilize the dispersion. Wastewaters from the flushing operations containing organic
chemicals are sent to treatment before being discharged.
Conventional final finishing is accomplished through dry grinding of the product. The
crude product press cakes or slurries are dried and then ground. Typically, ball or
hammer mills are used. Drying is done in hot air or vacuum using ovens, rotary dryers,
drum dryers, or most frequently, spray dryers. The dry grinding process is losing its
popularity, however, due to dusting which causes air pollution problem. This operation
can be replaced by wet grinding which is done in ball or sand mills. Surfactants are
often added to facilitate the grinding process. The resulting product is a ground
aqueous paste which can be subsequently dried by spray drying.
The last important finishing step is a standardization step in which inert salts are
often added to meet product specifications, which can vary widely. For example, of
the total amount of organic pigments sold, 45 percent is sold as dry powder without
dilution, 40 percent is sold as colorants in oil with up to 65 percent dilution by inerts,
and the remaining 15 percent is sold as wet filter cake without further processing.
B7-11
-------
7. WASTE DESCRIPTION
The primary specific wastes associated with the organic dyes and pigments
manufacturing industry are listed in Table 7-1. Wastes or process residuals from this
industry were classified as follows:
Discarded Shipping Containers: Empty containers, paper bags, or fiber drum liners
that contain residual amounts of organic intermediates or other raw materials.
Synthesis Wastes; Generally, these include the byproducts of the various reaction
steps prior to finishing. The byproducts are formed as a result of undesirable side
reactions that occur in series or in parallel to the main reaction leading to the desired
product. Synthesis wastes also contain unconverted reactants and coproducts (e.g.
salts), all dissolved in the reaction medium such as water or solvent.
As to the form, composition, and the point of exit from the process, synthesis wastes
vary. In the case of soluble azo dyes, these wastes exit as a solid cake from filtration
following coupling and in liquor following filtration of the precipitated dye. In case of
anthraquinone dyes, the synthesis wastes would be present in the distillation column
bottoms, which also contain organic solvent.
Considerable variability in the treatment and ultimate disposal of synthesis wastes is
expected given the large number of manufacturing facilities. Some plants treat the
filter cakes with hypochlorite solution to reduce the amount of organic material prior
to land disposal of the left-over solids. The aqueous waste containing residual organic
and inorganic compounds can be also treated by oxidation to reduce oxygen demand,
color and turbidity to acceptable levels prior to discharge. The alternatives for
solvent waste include solvent recovery and/or disposal of residuals via landfilling or
incineration.
Product Finishing Wastes; Since equipment cleaning and spill/off-spec product wastes
are considered separately, the wastes from the finishing steps include dusts generated
during drying and grinding which are collected in the baghouse. Additionally, these
wastes would include residual water or organics from the flushing step. Usually, these
wastes are minor in comparison to other waste categories noted.
B7-12
-------
Table 7-1 Organic Dye/Pigment Manufacturing Process Wastes
No.
Waste
Description
Process Origin
Composition
RCRA
Code
Used containers
Unloading of raw materials
paper bags and fiber drums
containing residual amounts
intermediates
03
^l
i
Synthesis wastes
Product finishing
wastes
Equipment cleaning wastes
Spills and off-spec
products
Reaction steps
Product drying, flushing,
grinding, standardizing,
packaging
Cleaning of equipment
between batches
Cleaning of spills, off-
spec product generation
water, organic solvents,
metal salts, inter-
mediates, side reaction
byproducts
dyes, pigments, inter-
mediates, inert salts,
auxiliary oils
dilute alkaline or acidic
aqueous solutions or sol-
vents with small amounts
of residual organics
dyes, pigments, organic
intermediates.
F002
F003
F004
F005
-------
Equipment Cleaning Wastes; These include aqueous and solvent streams associated
with the cleaning of reactors, transfer lines and equipment. Cleaning is done between
consecutive dissimilar batches in order to prevent contamination. It is expected that
in an average facility employing a number of reactors, tanks and filter presses,
cleaning is performed using a clean-in-place system employing either flow-over or
fill-and-empty techniques. The ball mills and other mechanical equipment used in the
finishing operations is expected to be cleaned using fill-and-empty approach.
The cleaning medium can be alkaline aqueous solution or solvent. At times, an
oxidizing agent, such as hydrogen peroxide, may be added. The sludge separated from
the cleaning solution can be co-mingled with other solid process residuals prior to
possible treatment and disposal. The aqueous effluent is neutralized prior to
discharge, but can also be mixed with liquid streams from the synthesis step prior to
neutralization, treatment and discharge.
Spills and Off-Spec Products: These wastes are generated as a result of inadvertent
operation, usually attributable to human error. Again, diversity of chemical species,
availability of treatment, and plant operating procedures prevent one from compiling
precise information on handling and disposal of these wastes. However, it is expected
that a high degree of effort is expended in attempts to recover valuable chemicals
from spills or spoiled batches before spill wastes are disposed of as residuals, probably
through landfilling.
8. WASTE GENERATION RATES
Recent effort has been made to characterize the waste streams generated from the
manufacturing of organic dyes and pigments (SRI 1984). Information about the
quantity of these wastes, however, was not in evidence at the time of the final
document preparation. The relative proportions of each waste stream were estimated
by project staff based on the available information and are given in Table 9-1.
87-14
-------
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
A summary of the waste sources and the corresponding source reduction methods is
given in Table 9-1. This section deals with the description of the listed methods,
including known application cases.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Containers
Raw materials used in the production of organic dyes and pigments are usually shipped
in paper bags or fiber drums. Since some of these materials are toxic in nature, the
discarded containers containing residual amounts of these substances can be
considered as hazardous. The following source reduction techniques were noted:
yB7-15
-------
o Use rinsable/recycleable drums and plastic liners.
The use of polyethylene bags to line steel drums is already well established
in the industry along with the use of reusable tote bins. The residual use of
paper bags or liners should be eliminated.
o Maximize size of container.
Here, the underlying principle is that the same volume of material has a
smaller external surface if supplied in a single container as opposed to a
number of smaller containers. As the residual amount depends on the size
of the "wetted" surface, less waste residue will result from the use of
larger containers. Switching from drums or bags to mega-drums, tote bins
and, where possible, to bulk handling has already been accomplished to a
significant degree, but a moderate improvement margin probably exists.
o Container segregation.
Since only the organic intermediates and a few other reagents are
hazardous while a large amount of other raw materials, such as inert salts
used for dye and pigment standardization operations, are not, separation of
the different types of empty disposable containers can greatly reduce the
amount of potentially hazardous waste generated from this source.
9.1.2 Synthesis Wastes
In the azo dyes example, synthesis steps include diazotization, coupling and precipita-
tion. In a more general context, the synthesis of all dyes and pigments from raw
materials can result in the generation of hazardous waste due to (1) the formation of
by-products, (2) the use of toxic catalysts, or (3) the use of toxic raw materials.
Modification of the synthesis process can lead to substantial waste reduction. The
following source reduction techniques were noted:
o Eliminate the use of toxic catalysts.
Catalysts containing toxic metals are often used to manufacture dyes and
pigments. Mercury catalyst used to produce anthraquinone dyes for the
coloring of cottons was replaced with a less toxic substance (NJDEP 1985).
B7-16
-------
Reduce the use of toxic raw materials.
Reduction of the toxicity of hazardous waste can be achieved through the
elimination or reduction of the use of highly toxic reactants and
intermediates. For example, since benzidine has been identified as
carcinogenic, its use has been eliminated by most of the U.S. dye and
pigment manufacturers (Radian 1977). In addition, some manufacturers
who used to make intermediates for the production of anthraquinone dyes
(which requires the use of toxic mercury catalysts) have stopped their
production and import these intermediates instead (Catino and Ferris 1979,
Bannister, Olin, and Stingle 1979, Fytelson 1982). While this measure will
reduce the overall amount of hazardous waste generated in the U.S., it only
shifts the problem to the foreign suppliers. In general, the use of less toxic
substitutes should be explored most intensively in the area of auxiliary raw
materials, i.e. those that do not undergo conversion to the final product.
For example, the use of xylene as a solvent can be re-examined from the
standpoint of substitution with less toxic solvent, e.g. acetone. Also, use
of non-chromate treatment for cooling water will lower the toxicity of the
blowdown sludge (which is not a part of "synthesis waste"—its inclusion is
intended to illustrate the approach).
Minimize byproduct formation.
Modifications of the synthesis process to increase yields and reduce by-
product formation are extremely specific to the manufacturing of each
different dye or pigment. Each process should be investigated separately
to identify potential waste reduction techniques. To exemplify the
approach, the following process modifications applicable to azo dyes were
identified as candidates for future evaluation:
Use an optimum pH range for a particular combination of diazo
salt/coupling components. Since higher pH increases both the rate of
the coupling reaction and the rate of formation of undesirable
byproducts, it is likely that there is an optimum pH range which
maximizes the yield of the desired product.
Shorten the residence time before coupling to minimize the
decomposition of diazonium salts. Since diazonium salts can readily
B7-17
-------
decompose in the coupling medium to produce phenolic compounds as
byproducts, the time it takes to transfer these salts to the coupling
reactor should be minimized. This can be accomplished by combining
clarification operation with a fast transfer to the coupling reactor,
e.g. by using a pump with the associated filter. The piping and filter
casing can be emptied using pressurized nitrogen following the
transfer. Pre-cooling of the coupling reactor would help to limit
detrimental temperature rise.
In general, the avenues which may, upon exploration, lead to further increases in
reaction yields and thusly to the decrease of byproduct waste generation include:
search for more selective catalysts.
modification of reaction temperature trajectory.
modification of reaction addition strategy (rate of sequence).
modifications to provide higher degree of mixing.
o Modification of filtration process.
By eliminating the need for filter aids, the amount of solid waste generated
can be reduced. For cases in which a precoat-type pressure filter is being
used for clarification, switching to a bag- or a leaf-type filter (LWVM
1985) would eliminate the need for filter aids, which does result in a
reduction of the volume of discarded cake.
Generally, improvement in filtration efficiency will result in better
product recovery and, subsequently, in less waste. Such improvement can
be obtained through the use of multistage versus single stage filtration, the
recycle of mother liquor during the initial cake-building period, the use of
a cake-gasketed recessed plate instead of the conventional plate-and-
frame (or the use of recessed plate filters to reduce leaking and wearing of
the filter medium), or the use of low-temperature air to blow dry the filter
cake to prevent the filter cloth's deterioration, which can result in leaking,
and other techniques.
B7-18
-------
9.1.3 Product Finishing Wastes
Drying of dyes and pigments is done with rotary dryers, spray dryers, or drum dryers in
hot air or in vacuum ovens. The following source control technique is proposed:
o Replace steam jets with vacuum pumps.
In cases where steam jets are used to generate a vacuum, wastewater can
be generated through contamination by process vapors or dusts of the
cooling water used in the barometric condenser. This may represent a
significant source of waste from the production of dyes which require
extensive use of organic solvents (e.g. anthraquinone dyes). The waste-
water generated can be avoided through the use of vacuum pumps with
surface condensers. The use of a surface condenser will increase the
potential recyclability of condensate.
Solid waste is generated from the collection of dust in baghouses during material
handling, grinding, blending, and standardizing operations. The following source
control techniques are noted:
o Use wet instead of dry grinding.
Dye and pigment products can be ground wet using ball or sand mills and
subsequently spray-dried. This operation will reduce the amount of dust
emitted in comparison to the use of the conventional dry grinding step.
This method is used only for water-insoluble dyes and pigments.
o Increase the use of dust suppression techniques.
Dust emission can be suppressed through the use of atomized water sprays,
enclosed weigh-transfer hoppers, or better care in manual material
handling.
o Recycle baghouse fines.
Baghouse hopper emptying should be scheduled for possible recycling of the
dye or pigment fines. Baghouse hoppers should be cleaned before and after
the manufacturing of a large quantity of a particular dye or pigment to
recover these products. This practice is feasible only for plants producing
a very limited number of dyes or pigments.
B7-19
-------
9.1.4 Equipment Cleaning Wastes
These constitute a significant fraction of the total waste generated during organic dye
and pigment manufacture, which is typical for any batch process. The subject of how
to minimize equipment cleaning waste is discussed in a separate study (#322) found in
this appendix, and is of universal importance to the entire chemical process industry.
The reader is referred to that study for a general characterization of source control
approaches.
Based on the information available, it appears that a two-step rinse with a maximum
utilization of cleaning solution as a make-up in the next compatible batch is an
approach utilized to a significant degree by the organic dye and pigment industry.
9.1.5 Spills and Off-Specification Product Waste
As previously mentioned, spills and off-specification products result from inadvertent
operations usually attributable to human error. At times, such losses can also result
from inadequate provisions for contingencies. For example, a diazotization reaction
requires that solution is maintained at 32-40°F, which necessitates the use of coolant
maintained by a refrigeration system. If a refrigeration system malfunctions and
coolant flow is stopped, the batch can be spoiled. Provision for an adequate surge
capacity of chilled coolant may extend the time in which the refrigeration plant can
be brought back on stream without spoiling the batch. In general, avoidance of spills
and generation of off-spec products could be promoted by the following source control
techniques:
o Increased use of automation.
This includes automated batching systems, monitoring of the adequate
number of process function, and automatic control of fail-safe and "fail-
clean" shutdown features. Use of supervisory computer control is
recommended for exploration.
o More intensive in-process quality control.
B7-20
-------
o More thorough cleaning of process equipment and piping.
This measure would reduce contamination that may cause an off-spec
batch.
o Better operating practice.
This includes personnel training, better supervision, spill prevention
programs, and other measures that fall under the category of
procedural/institutional modifications. For a detailed discussion, the
reader is referred to the separate study in this appendix.
9.2 Implementation Profile
Reduction of wastes from the manufacture of organic dyes and pigments can be
achieved either through process modifications in the synthesis of the dye, or through
improvements in operating practices in the area of product finishing, equipment
cleaning and materials handling. A detailed characterization of optimal waste-
reducing process modifications which lower the generation of synthesis waste is
difficult to accomplish due to the large number of processes used to produce the wide
variety of dyes and pigments currently on the market. Process modifications may
require significant research and development expenditures and lengthy periods of time
to become established. Given the shift from domestic production to imports, some
companies may lack sufficient incentives to pursue additional development work.
Improvements in operating practices, on the other hand, are applicable to all
processes. These are easily implemented as they typically require only minor changes
in operational procedure or the addition of simple equipment or controls. However,
they are not as effective as process modifications in reducing the generation of
hazardous waste.
9.3 Summary
The sources of waste from the manufacture of organic dye and pigment and their
respective source control techniques are summarized in Table 9-1. The ratings listed
in the table are based on a scale of 0 to 4 and are used to evaluate each technique for
B7-21
-------
TABLE 9-1 SUMURY OF SOURCE CONTROL METHODOLOGY FOR THE ORGANIC DYE / PIGMENT MANUFACTURING INDUSTRY
4
Haste Source
Used Containers
Synthesis Haste (*
Product Finishing
Hastes
Equipment Cleanlni
Hastes
Spills and Off-
Specification
Product Hastes
| All Sources
1
1
1
li-
is.
1
)M
|3.
!«•
1
M-
12.
|3.
l«.
1
|1-
1
I'-
|2.
13
I'-
1
1
1
Control Methodology |-
1
Use liners and recyclable drums |
Max1n1ze container size |
Container segregation |
Overall |
Eliminate use of toxic catalyst |
Eliminate use of toxic raw materials |
Minimize byproduct formation !
Modify filtration process I
Overall |
Replace steam Jets with vacuum pumps |
Use wet grinding |
Dust suppression |
Recycle baghouse fines |
Overall 1
See study on equipment cleaning (B22)|
Overall |
Increased automation |
Intensify quality control |
Thorough equipment cleaning |
Better operating practices |
Overall 1
All Methods
Found Documentation | Waste | Cxtent of | Future | Fraction of | Current | Future Reduction index I
Quantity | Quality I Effectiveness | | Potential | | Index | Probable | Maximum |
1 | 1 | 3 | 3 | 2 | 2 3 | 0.4 | 04)
t | 1 I 2 | 3 | 1 | | 1.5 | 0.1 | |
1 | 1 1 3 | 3 | 1 | 2.3 | 02| i
1.00 | 1.00 | 2.67 | 3.00 | 1 33 | 0.02 | 2.3 | 0 2 | 0 4 |
t 1 1 I 3 | 1 | t I | 0 8 | 0 6 | |
1 | 1 | 3 | 1 | 2 | | 0 8 | 11| 1.1 |
t | t | 3 | 3| 3 | | 2.3 | 0.6 | |
0 | 0 | 1 | 3 | 2 | I 0 8 | 0 1 |
0.75 | 0.75 | 2.50 | 2.00 | 2 00 | 0.60 | 2.3 | 0.6 | 1.1
1 | 1 | 2 | 2 | 3 | | 1.0 | 0.8 | 08
0 | 0| 3 | 3 | 1 | I 23| 0.2 |
0 | 0 | 1 | 1 | 1 | | 0 3 | 02|
0 | 0 | 2 | 0 | 1 | | 0 0 | 0.5 | |
0.25 | 0.25 | 2 00 | 1.50 | 1.50 | 0 05 | 2.3 | 0.4 | 03
— | — | — | — | — | | 2 6 | 0 7 | 1.4
— | — | — | — | — | 0.32 | 2.6 | 0.7 | 1.4
1 | 1 | 3 | 2 | 3 | | 1 5 | 1 1 | 11
l| 1 | 2 | 3 | 2 | | 1.5 | 0.3 |
1 | 1 | 2 | 3 | 2 | | 1.5 | 0.3 | |
1 | 1 | 3 | 3 | 3 | 2.3 | 0.6 |
1.00 | 1 00 | 2.50 | 2.75 | 2 50 | 0.01 | 2.3 | 0.5 | It
| 1.00 | 2.4 | 0.6 | 1.2
00
(•) These waste streams contain listed "F" and/or 'K' RCRA wastes.
-------
its waste reduction effectiveness, extent of current use and future application
potential. The ratings were derived by project staff from the available information
and from industry comments.
It appears that the current level of waste minimization in the manufacture of organic
dyes and pigments is high. This is evidenced by the current reduction index (CRI) of
2.4 (60 percent) which measures the extent of reduction in the waste that otherwise
would be generated if none of the listed methods were applied as they are currently.
The potential for future reductions appears modest, as evidenced by the future
reduction index (FRI) of 0.6 to 1.2 (15 to 30 percent). The future reduction index is
the measure of waste reduction achievable through implementation of the listed
techniques to their rated extents. The greatest potential for future reductions appears
to reside in eliminating the use of toxic raw materials and increasing the degree of
automation within the production process.
10. PRODUCT SUBSTITUTION ALTERNATIVES
The production of organic dyes and pigments is inherently dynamic in nature since
changes in the demands for these products are easily induced by several factors. Such
factors include the possibility of creating new dyes with better qualities, the invention
of new products that require completely different types of dyes for coloring purposes,
or the improvement of existing dye technology which will affect the selection as well
as the quantity of the dyes being used. For example, disperse dyes have attained wide
usage due to the growth of synthetic fabrics. Furthermore, improvements in dyeing
operations, such as increased use of automation, or the introduction of new dyeing
techniques, such as low-liquor jet dyeing, will enhance the performance of one dye
over another, and thus will make its selection more favorable. The growth of inactive
dyes serves as a good example for this effect. Reactive dyes are used to color
cellulose-based materials, and are preferred over other dyes due to their wide range of
shades, excellent wet fastness, and simple application method. However, problems are
encountered in the use of these dyes due to the low degree of exhaustion and fixation
observed. This defect creates problems in the form of difficult pollution control and
excessive energy usage which tend to discourage the use of these dyes. Research,
however, is being done to counteract these problems through the use of fixation
B7-23
-------
accelerators or the development of short-liquor dyeing and low temperature dyeing
techniques (Abeta, Yoshida, and Imada 1984). Once these barriers are overcome, the
production of reactive dyes is expected to increase significantly.
In summary, changes in the dyeing industry, as well as increasing focus on the
selection and manufacture of less toxic dyes and pigments, will significantly affect the
future production of organic dyes and pigments.
11. CONCLUSIONS
Although significant waste reduction efforts have been made by the manufacturers of
organic dyes and pigments, opportunities for potential future reduction remain
numerous. Based on available information, an additional 15 to 30 percent of waste can
probably be reduced if the proposed source reduction techniques are implemented.
While source reduction measures associated with the dye synthesis stage of production
are individually quite effective, implementation of these techniques can be difficult
due to the diversity of the processes used to produce dyes and pigments in a single
plant. Furthermore, the extent of waste reduction in one plant may differ greatly
from another one. The most promising universal approach to waste reduction for
synthesis stage, however, would be to eliminate the use of toxic raw materials, or to
modify the existing process to achieve better product yield.
In addition to source reduction methods, changes in dyeing technology and/or the
development of new materials by the textile industry can also significantly affect the
production demand for organic dyes and pigments. This in turn can alter the waste
load generated from the manufacturing process.
12. REFERENCES
Abeta, S., Yoshida, T., and Imada, K. 1984. Problems and progress in reactive dyes.
American Dyestuff Reporter. 73(7): 26-31.
Anonymous, 1984. New products review. American Dyestuff Reporter. 73(12).
Bannister, D. W., Olin, A. D., and Stingle, H. A. 1979. Dyes and dye intermediates. In
Kirk-Othmer Encyclopedia of Chemical Technology. 3rd ed. Vol. 8, pp. 159-212. New
York, N.Y.: Wiley.
B7-24
-------
Catino, 5. C., and Farris, R. E. 1979. Azo dyes. In Kirk-Othmer Encyclopedia of
Chemical Technology. 3rd ed. Vol. 3, pp. 387-432. New York, N.Y.: Wiley.
Fytelson, M. 1982. Pigments: organic. In Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed. Vol. 17, pp 838-71. New York, N.Y.: Wiley.
Griffiths, J. 1984. Developments in the chemistry and technology of organic dyes.
Oxford: Blackwell Scientific Publications.
LWVM, 1985. The League of Women Voters in Massachusetts. Waste reduction; the
untold story. Seminar Proceeding at the National Academy of Science, Conference
Center, June 18-21. Woods Hole, Mass.: The League of Women Voters in
Massachusetts.
MRI 1981. Midwest Research Institute. Material balance for dyes and pigments from
benzidine and the benzidine derivatives. EPA-560-2-81-001. Washington, D.C.: U.S.
Environmental Protection Agency.
NJDEP, 1985. New Jersey Department of Environmental Protection, Division of
Waste Management. Source reduction of hazardous waste. Seminar Proceeding at
Douglas College, Rutgers University, August 22, 1985. New Jersey: N.J. Department
of Environmental Protection.
Radian Corp. 1977. Industrial process profiles for environmental use: Chapter 7;
organic dyes and pigments industry. EPA-600-2-77-023g. Cincinnati, Ohio: U. S.
Environmental Protection Agency.
SRI 1984. Stanford Research Institute. Wastes from manufacture of dyes and
pigments; Vols 1-9. PB 84 - 200 864. Washington, D.C.: U. S. Environmental
Protection Agency.
USDC 1985. U.S. Department of Commerce, Bureau of the Census. Industrial organic
chemicals in 1982 Census of Manufacturers. MC82-1-28F. Washington, D.C.: U.S.
Government Printing Office.
USITC 1983. U. S. International Trade Commission. Synthetic organic chemicals, U.S.
production and sales. USITC Publication No. 1588. pp. 57-89.Washington, D.C.: U. S.
Government Printing Office.
13. INDUSTRY CONTACTS
Newby, W.E., E. I. duPont de Nemours & Co., Inc. (retired), Wilmington, DE.
Bowers, D.P., Senior, Director of Environmental Control, Merck Chemical Manu-
facturing Division, Merck &. Co., Inc., Rahway, N.J.
B7-25
-------
1. PROCESS: PAINT MANUFACTURE
2. SIC CODE: 2851
3. INDUSTRY DESCRIPTION
As defined by Standard Industrial Classification (SIC) 2851, the paints and allied
products industry "comprises establishments primarily engaged in the manufacture of
paints (in paste and ready mixed form), varnishes, lacquers, enamels and shellacs,
putties, wood fillers and sealers, paint and varnish removers, paint brush cleaners, and
allied paint products". Establishments engaged in the manufacture of pigments
(organic or inorganic), resins, printing inks, adhesives and sealants, or artist materials
are not included.
3.1 Company Size Distribution
Of the 1,441 paint manufacturing facilities located in the U.S. in 1982, more than 55
percent employed less than twenty people each. Overall, the Bureau of the Census
(USDC 1985) estimated that 54,100 people were employed by the paint manufacturing
industry. Table 3-1 lists company size distribution as a function of total number of
employees at a given site.
Table 3-1 Company Size Distribution
No.
No.
of establishments
of employees
Total
1,441
54,100
No. of Employees per Facility
1-19 20-49 50-99
819 331 171
6,400 10,500 11,800
100+
120
25,400
Source: 1982 Census of Manufacturers (USDC 1985).
3.2 Principal Producers
The paint manufacturing industry is composed of a small number of multi-plant,
multi-product companies and a large number of single-plant companies. The eight
major paint producing companies in the U.S. are listed below:
B8-1
-------
Sherwin-Williams Clorox
PPG Industries Valspar
SCM Corp (Glidden) De Soto Paints
Inmont Insilco Corp.
3.3 Geographical Distribution
Due to the expense of transporting products over long distances, paint plants
tend to be clustered around population centers. Approximately 44 percent of all
paint sites are located in five states (California, New Jersey, New York, Illinois,
and Ohio), with 67 percent being located in ten states. Distribution by EPA
regions are shown in Figure 3-1 and Table 3-2 below.
Table 3-2 Location of Facilities by EPA Region
EPA Region
I
II
III
IV
V
VI
VII
VIII
IX
X
No.
1-19
55
133
54
110
191
58
46
12
134
26
of Employees
20+
31
87
51
99
168
50
31
5
81
19
per Facility
Total
86
220
105
209
359
108
77
17
215
45
National 819 622 1441
Source: Assessment of Industrial Hazardous Waste Practices (WAPORA 1975)
adjusted to reflect number of establishments listed in the 1982
Census of Manufacturers (USDC 1985).
B8-2
-------
00
03
-------
4. PRODUCTS AND THEIR USE
Most small plants produce paint in 10 to 500 gallon batches. Plants with more than 20
employees tend to produce paint in 1,000 to 3,000 gallon batches*. Overall, the paint
industry sold 8.6 billion dollars worth of product in 1983 ($3.9 billion for architectural
coatings, $3.0 billion for product coatings, and $1.7 billion for special purpose
coatings) (Webber 1984). The amounts and distribution of products manufactured by
the paint industry in 1983 are shown below.
Table 4-1 1983 Paint Products and Use Distribution
Architectural Coatings 463 million gallons
Product Coatings 331 million gallons
Metal containers 19%
Automotive 16%
Machinery 6%
Sheet, strip and coil 6%
Metal furniture 5%
Other 48%
Special Purpose Coatings 130 million gallons
High performance maintenance 31%
Automotive and machinery refinishing 29%
Traffic paint 14%
Other 26%
Source: Chemical and Engineering News (Webber 1984).
For an average paint plant located in the U.S., 60 percent of its total annual
production would be solvent-based paint, 35 percent would be water-based paint, and 5
percent would be allied products. While a large percentage of paint used for
architectural coating is water-based (more than 70 percent), solvent-based paint is
still predominately used for product and special purpose coatings.
5. RAW MATERIALS
Annual consumption rates of raw materials used by the paint manufacturing industry
are shown in Table 5-1.
•"•Confidential source 1985: Personal communication.
B8-4
-------
Table 5-1 Raw Materials Used by the Paint Manufacturing Industry in 1982
Materials
Usage
Resins
Solvents
Pigments
Extenders
Alkyd
Acrylic
Vinyl
Other
Aromatic
Aliphatic
Ketones
Alcohols
Other
Titanium dioxide
Inorganic^3'
Organic
Calcium carbonate
Talc
Clay
Other
Miscellaneous
Drying Oils
Plasticizers
Other
1844 million Ibs/yr.
33%
19%
19%
29%
3774 million Ibs/yr.
30%
27%
17%
12%
14%
1062 million Ibs/yr.
65%
33%
2%
1162 million Ibs/yr.
31%
25%
23%
21%
220 million Ibs/yr.
41%
18%
41%
Source: Chemical Economics Handbook (SRI 1981) data for 1977 adjusted for 1982
production rates.
(a) Approximately 60 percent of the inorganic pigments used consisted of iron
oxide, zinc oxide, zinc dust, and aluminum paste; 27 percent consisted of
lead and chrome compounds; and 13 percent consisted of other.
58-5
-------
The major raw materials used to manufacture paint are resins, solvents, drying oils,
pigments, and extenders. Based on the wide variety of paints produced, no one type of
material dominates the market.
6. PROCESS DESCRIPTION
Detailed process flow diagrams of paint manufacturing have been presented in the
open literature (Haines 1954, Payne 1961). The following description briefly highlights
the production of the industry's two main products: solvent-based paint and water-
based paint. At a typical plant, both types of paint are produced. A block flow
diagram of the steps involved in manufacturing paint is presented in Figure 6-1.
The production of solvent-based paint begins by mixing resins, ground pigment, and
pigment extenders together in a high speed mixer. During this operation, solvents and
plasticizers are also added. Following the mixing operation, the batch is transferred
to a mill for additional grinding and mixing. The type of mill is dependent on the types
of pigments being handled, so that no one style is universal. Next, the paint is
transferred to an agitated tank where tints and thinner (usually a volatile naphtha or
blend of solvents) are added. Upon reaching the proper consistency, the paint is
filtered to remove any non-dispersed pigment and transferred to a loading hopper.
From the hopper, the paint is poured into cans, labeled, packed, and moved to storage.
The typical batch size for a small plant will range from 10 to 500 gallons while the
average batch size for a large plant will be 1,000 to 3,000 gallons.
The typical paint plant produces about 60 percent solvent-based paint, 35 percent
water-based paint, and 5 percent allied product. The water-based paint process is very
similar to the solvent-based process. The major difference is the substitution of water
for solvent and the sequencing of material additions. Preparation of water-based paint
begins by mixing together water, ammonia, and a dispersant in a mixer. To this
mixture, ground pigment and pigment extenders are added. After mixing, the material
is ground in a mill and then transferred to an agitated mix tank. Four additions of
materials occur in this tank. First, resin and plasticizers are added to the mixture;
second, a preservative and an antifoaming agent are added; third, a polyvinyl acetate
emulsion is added; and fourth, water is added as a thinner. Following this mixing
operation, the handling of the paint is similar to that for solvent-based paints. At
B8-6
-------
RESINS ,_
PIGMENTS <
EXTENDERS
SOLVENTS
PLASTICIZERS
I
GRINDING AND 1
NIXING-..- _|
©0©|
(
MATER
"a AMMONIA
DISPERSANT
PI6NENT
EXTENDERS
TINTS
THINNER
V
1
1
GRINDING |
\
NIX
I
r«e L 5
r^
RESIN
PRESERVATIVE
ANTIFOAM
PVA EMULSION
MATER
FILTERING
PACKAGING
FINAL PRODUCT
PROCESS NASTE CATEGORIES:
(7) DISCARDED RAM MATERIAL CONTAINERS
(?) BA6HOUSE PISMENT DUSTS
(D OFF-SPECIFICATION PAINT
(7) FILTER CARTHIDBES
(?) EQUIPMENT CLEANING HASTES
Figure 6- 1 Block Flow Diagram for Paint Manufacture
B8-7
-------
many facilities the grinding and the mixing and grinding operation may be bypassed
with all the dispersion operations occurring in a single high-speed mixer.
7. WASTE DESCRIPTION
Typically, paint facilities segregate and store waste only to the degree required by the
waste disposal contractor. Since the degree of segregation can affect the amount of
material having to be classified as hazardous, and the cost of disposing of hazardous
material is increasing, paint facilities are taking a more active role in waste
management. The major wastes that the paint industry must manage are empty raw
material packages, dust from air pollution control equipment, off-specification paint,
spills, and equipment cleaning wastes. Equipment cleaning wastes are a dominant
waste stream.
The primary specific wastes associated with paint manufacturing are listed in Table
7-1. Waste generated by the industry is usually managed in one of four ways: on-site
reuse, on-site recycling, off-site recycling, and off-site treatment/disposal (Ryan
1984). On-site reuse involves the reuse of waste (without treatment) as a feed or wash
material for producing other batches of paint. Also included is the sale or in-house use
of off-specification paint as utility paint. On-site recycling involves the reclaiming of
solvent by distillation or recovery of heating values by incineration. Usually, on-site
recycling is performed by large companies (those that produce more than 35,000
gallons of solvent waste each year) while small companies (those that produce 20,000
gallons or less per year) send the waste to an off-site recycler. The fourth option, off-
site treatment/disposal involves incineration or land disposal. The extent of waste
currently being land disposed is unknown.
8. WASTE GENERATION RATES
The most recent published estimates of the nationwide waste generation rates by paint
industry date back to 1974 and are presented in Table 8-1 below. While the total
amount of waste due to bags and packages was reported to be 302,000 metric tons per
year (WAPORA 1975), the actual amount of hazardous waste (bags containing
hazardous pigments) was only 2000 TRY. Since the estimate of 302,000 TRY included
recycled drums and pallets, and because the current industry practice is to segregate
hazardous from non-hazardous waste, only the hazardous amount is presented so as not
B8-8
-------
Table 7-1 Paint Manufacturing Process Wastes
No. Waste Description
Process Origin
Composition
RCRA Codes
1. Leftover raw materials in original
containers
2. Pigment dusts from air pollution
equipment
3. Off-specification paint
Unloading of materials
into mixing tanks
Unloading of pigment
into mixing tanks
Color matching (small
scale) production
Paper bags with a
few ounces of left-
over pigments
Pigments
Paint
CD
co
i
\o
4. Spills
Accidental discharge Paint
5. Waste rinsewater
6. Waste solvent
7. Paint sludge
Equipment cleaning
using water and/or
caustic solutior\s
Equipment cleaning
using solvent
Equipment cleaning
sludges removed from
cleaning solution
Paint, water, caustic
Paint, solvent
E002
FOD3
F005
Paint, water, caustic,
solvent
8. Filter cartridges
Undispersed pigment Paint
-------
to distort the hazardous waste generation picture. In addition, the amount of solvent
discarded only represents the amount of solvent considered by WAPORA to be toxic.
Current waste generation rates were not in evidence at the time of final document
preparation. According to one source, these rates have changed dramatically in the
last six years due to RCRA*.
Table 8-1 1974 Nationwide Generation Rates of Specific Wastes from the
Paint Manufacturing Industry in metric tons per year
Wastestream
Empty bags and packages
Dust from air pollution equipment
Off-specification product
Spills
Cleaning of Equipment
Total
Total
TPY
2,000
1,600
4,900
5,400
82,000
95,900
Toxic
Solvents
TPY
--
--
580
85
13,600
14,265
Toxic
Metals
TPY
128
80
41
5
590
844
Source: Assessment of Industrial Hazardous Waste Practices (WAPORA 1975).
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The list of individual primary waste streams and their sources along with a list of
source reduction methods is presented in Table 9-1. The following sections discuss
recommended waste reduction methods and identified procedures. The basis for
identification came from published accounts in the open literature and through
industry contacts. It should be noted that not many published accounts were available.
The reason for the lack of published data is probably that many of the methods used by
the paint industry are common sense measures (i.e. good housekeeping) and are not
viewed as being special waste reduction techniques worth extensive documentation,
analysis, or presentation.
National Paints and Coatings Association 1986: Personal Communication.
B8-10
-------
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Bags and Packages
Inorganic pigments, which contain heavy metals and may therefore be classified as
hazardous are usually shipped in 50 pound bags. After emptying the bag, an ounce or
two of pigment usually remains inside. Empty containers of liquid raw materials that
constitute hazardous waste (e.g. solvents and resins) are typically cleaned or recycled
to the original raw material manufacturer or to a local drum recycler. Empty liquid
containers are excluded from the following discussion. The following waste reduction
techniques for bags and packages were noted:
o Use of water soluble bags for toxic pigments and compounds used in water-
based paints.
When empty, the bags could be dissolved or mixed in with the paint. Such a
method is commonly used for handling mercury compounds and other paint
BB-11
-------
fungicides*. This method could not be used, however, when producing high
quality, smooth finish paint since the presence of this material could affect
the paint's film forming property**.
o Use of rinseable/recyclable drums with plastic liners instead of paper bags.
o Better operating practices (Identified).
Through industry contacts, it was established that the most effective way
of reducing hazardous waste associated with bags and packages (or any
other waste stream) was to segregate the hazardous materials from the
non-hazardous materials As an example, empty packages that contained
hazardous materials should be placed into plastic bags (so as to reduce or
eliminate dusting which can lead to contamination of non-hazardous
material and should be stored in a special container to await collection.
9.1.2 Pigment Dust from Air Pollution Equipment
Some of the dusts generated during the handling, grinding, and mixing of pigments can
be hazardous. Therefore, dust collection equipment- (hoods, exhaust fans, and
baghouses) are provided to minimize a worker's exposure to localized dusting and to
filter ventilation air exhaust. As of 1974, 15 percent of the pigment dust collected in
baghouses was reused to make low-grade paint with the remaining 85 percent drummed
for disposal (WAPORA 1975). The waste reduction methods considered consist of:
o Dedicated baghouse system for pigment loading area (Suggested).
At Daly-Herring Co., in Kinston, N.C., (while Daly-Herring is engaged in
formulation of pesticides and not paints, there are many material handling
problems common to both industries) dust streams from several different
production areas were handled by a single baghouse. Since all of the
streams were mixed, none of the waste could be recycled to the process
that generated them. By installing separate dedicated baghouses for each
production line, all of the collected pesticide dust recycled (Huisigh et al.
* Confidental source 1985: Personal communication.
** E.I. Du Pont de Nemours & Co. 1985: Personal communication.
B8-12
-------
1985). While this example is not intended to imply that most of the dust
generated by the paint industry could be recycled, it does show the overall
importance of keeping waste streams segregated.
o Use of pigments in paste form instead of dry powders.
Pigments in paste form are dry pigments that have been wetted or mixed
with resins. Since these pigments are wet, less or no dust is generated
when the package is opened. In addition, most pigments in paste form are
supplied in drums (which can be recycled) and therefore would eliminate
the waste due to empty bags. While this method would increase the
amount of pigment handling occurring at the supplier's facility, it can be
argued that the overall number of handling/transfer points for dry powder
would be greatly reduced along with the probability of spills and dust
generation.
o Better operating practices.
Scheduled baghouse hopper emptying could be done to reduce the amount
of hazardous waste generated. The major portion of solid waste collected
by the air pollution equipment is titanium dioxide and extenders.
Hazardous pigments make up only a small fraction of the waste and would
only be present during colored paint production. Since many facilities
produce colored paints on a non-continuous basis (i.e., once a week or once
a month), scheduled cleaning of the equipment to coincide with colored
paint production would allow for segregating the waste that contained
hazardous pigments from that which did not.
9.1.3 Off-Specification Paint
Most off-specification paint is produced by small shops that deal in specialty paints.
Since these paints cost more to produce, and therefore sell at a premium price, most
off-spec paint is reworked into a salable product. Since elimination of off-spec paint
production has built-in economic incentives, the following techniques are widely used:
o Increased automation.
B8-13
-------
o Better operating practices.
Industry contacts indicated that better operating practices (better operator
training, closer supervision, and tighter equipment inspection and
maintenance), and increased use of automation were extremely effective
at reducing the amount of waste generated. Overall, better operating
practices are widely used by the paint industry while increased automation
is just starting to be implemented. As the cost of digital electronic
controls further declines, more of the paint industry is expected to take
advantage of automation.
9.1.4 Spills
Spills are due to accidental or inadvertent discharges usually occuring during transfer
operations or equipment failures (leaks). Spilled paint and the resulting clean up
wastes are usually discharged to the wastewater treatment system or directly
drummed for disposal. If the plant has floor drains, large quantities of water are used
to clean up water-based paint spills. Dry cleaning methods (using sawdust) are
employed for cleaning of solvent-containing spills or for water-based spills where floor
drains are not available. The average amount of paint spilled at each facility is
approximately 5 gallons per year, however, a few large spills can move it to 100
gallons per year (WAPORA 1975). Similar to the case of off-spec paint, waste
reduction methods include:
o Increased automation.
o Better operating practices.
The use of dry cleanup methods (sawdust, mopping) should be maximized
wherever possible. By closing floor drains, employees are discouraged from
grabbing a hose and washing down the area (WAPORA 1975, USEPA 1979).
In addition, a large decrease in wastewater is achieved since the floors
cannot be routinely washed.
B8-14
-------
9.1.5 Equipment Cleaning Wastes
For a general discussion of equipment cleaning in the process industries, the reader is
referred to a separate study (#B22) included in this appendix. The text below discusses
issues specific to the paint manufacturing process. Equipment cleaning is responsible
for most of the waste generated during the paint manufacturing process. Following
the production of either solvent or water-based paints, considerable waste or
"clingage" remains affixed to the sides of the preparation tanks. The three specific
methods of tank cleaning used in the paint industry are solvent washing for solvent-
based paint, caustic washing for either solvent or water-based paint, and water
washing for water-based paint.
Equipment used for preparation of solvent-based paint is rinsed with solvent, which is
then generally reused in the following ways:
Collected and used in the next compatible batch of paint as part of the
formulation.
Collected and re-distilled either on or off-site.
Collected and used with or without settling for equipment cleaning until
spent. When the solvent is finally spent, it is then drummed for disposal.
As of 1974, thirty-five percent of the solvent waste produced was recycled off-site
and eight percent was recycled on-site (WAPORA 1975). Recycling of spent solvent is
usually not practical whenever the solids concentration exceeds ten percent or the
level of inhibitors and stabilizers in the recycled solvent is low. Of all the solvent that
is recycled, seventy-five percent is recovered with the remaining portion disposed of
as sludge.
Caustic rinse is used for equipment cleaning of both solvent and water-based paints. It
is used more often with water-based paints, since water rinsing is usually insufficient
in removing paint that has dried in the mix tanks. Solvent rinsing can usually remove
solvent-based paint that has dried and therefore the need for caustic is reduced.
There are two major types of caustic systems commonly used by the paint industry. In
one type of system, caustic is maintained in a holding tank (usually heated) and is
B8-15
-------
pumped into the tank to be cleaned. The caustic drains to a floor drain or sump from
where it is returned to the holding tank. In the second type of system, a caustic
solution is prepared in the tank to be cleaned, and the tank is soaked until clean. Most
plants reuse the caustic solution until it loses most of its cleaning ability. At that
time, the caustic is disposed of either as a solid waste or wastewater with or without
neutralization.
Water wash of equipment used in the production of water-based paint is the source of
considerable wastewater volume. Wastewater resulting from rinsing is usually handled
as follows:
Collected and used in the next compatible batch of paint as part of the
formulation.
Collected and used with or without treatment for cleaning until spent.
Disposed with or without treatment as wastewater or as a solid waste in
drums.
Sludges from settling tanks are drummed and disposed of as solid waste. Spent recycle
rinsewater is drummed and disposed of as solid waste after the solubles content
prohibits further use.
The percentage of solvent-base and water-base paints produced is the most important
factor that affects the volume of process wastewater generated and discharged at
paint plants. Due to their greater use of water-wash, plants producing 90 percent or
more water-base paint discharge more wastewater than plants producing 90 percent or
more solvent-base paint. Additional factors influencing the amount of wastewater
produced include the pressure of the rinse water, spray head design, and the existence
or absence of floor drains. Where no troughs or floor drains exist, equipment is often
cleaned externally by hand with rags; when wastewater drains are present, there is a
greater tendency to use hoses. Several plants have closed floor drains to force use of
dry clean-up methods and discourage excessive water use.
Waste associated with equipment cleaning represents the largest source of waste in a
paint facility. Methods that reduce the need or frequency of tank cleaning or allow
B8-16
-------
for reuse of the cleaning solutions are the most effective. Waste reduction methods
considered include:
o Use of mechanical devices such as rubber wipers.
In order to reduce the amount of paint left clinging to the walls of a mix
tank, rubber wipers are used to scrape the sides of the tank. This operation
requires manual labor and hence the percentage of waste reduction is a
function of the operator*. Since the benefits will be offset by increased
labor, mechanization/automation should be considered. Many new mixers
are available that are designed with automatic wall scrapers (Weismantel
and Guggilam 1985). These mixers can be used with any cylindrical mix
tank (flat or conical bottom).
o Use of high pressure spray heads and limiting wash/rinse time.
After scraping the tank walls, high pressure spray hoses can be used in
place of regular hoses to clean water-based paint tanks. Based on studies
(USEPA 1979), high pressure wash systems can reduce water use by 80 to
90 percent. In addition, high pressure sprays can remove partially dried-on
paint so that the need for caustic is reduced. Tanks used for making
solvent-based paints normally employ a built-in high pressure cleaning
system. At Lilly, in High Point, N.C., a high pressure cleaning system was
installed in several mix tanks. By continuously pumping a fixed amount of
solvent into a tank until it was clean, the overall volume of solvent
required for cleaning was reduced (Kohl, Moses, and Triplett 1984).
o Use Teflon** lined tanks to reduce adhesion and improve drainage.
The reduced amount of "clingage" will make dry cleaning more attractive.
This method is probably applicable only to small batch tanks amenable to
manual cleaning.
* Confidential source 1985: Personal communication.
** Registered trademark of E.I. Du Pont de Nemours & Co.
B8-17
-------
Use a plastic or foam "pig" to clean pipes.
It was reported that much of the industry is currently using plastic or foam
"pigs" (slugs) to clean paint from pipes*. The "pig" is forced through the
pipe from the mixing tank to the filling machine hopper. The "pig" pushes
ahead paint left clinging to the walls of the pipe. This, in turn, increases
yield and reduces the subsequent degree of pipe cleaning required. Inert
gas is used to propel the "pig" so as to minimize drying of paint inside the
pipe. Piping runs and the equipment (launcher and catcher) must be
carefully designed so as to prevent plugging, spills, sprays, and potential
injuries.
Better operating practices.
At Desoto, in Greensboro, N.C., wash solvent from each solvent-based
paint batch is separately collected and stored. When the same type of
paint is going to be produced, waste solvent from the previous batch is used
in place of virgin solvent. In 1981, Desoto produced 25,000 gallons of
waste mineral spirits. In 1982, when the system was implemented, waste
solvent production amounted to 400 gallons. This same technique is
currently being applied to their latex paint production operation (Kohl,
Moses and Triplett 1984).
At Thiele-Engdahl, in Winston-Salem, N.C., a similar procedure of solvent
segregation and reuse was attempted. However, due to the large variety of
products made, segregation proved to be too complicated. In addition, the
high cost of raw materials (Thiele-Engdahl produces specialty inks for
rotogravure printing) and the risk of contaminating the entire batch far
outweighed the savings in virgin solvent cost (Kohl, Moses and Triplett
1984).
* E.I. Du Pont de Nemours & Co. 1986: Personal communication.
National Paint and Coating Association 1986: Personal communication.
B8-18
-------
Other waste reduction measures based on good operating practices would
be to schedule paint production for long runs or to cycle from light to dark
colors so that the need for equipment cleaning would be reduced. For
facilities using small portable mix tanks for water-based paints, immediate
cleaning after use would reduce the amount of paint drying in the tank and
hence reduce the need for caustic. Many times, dirty equipment is sent to
a central cleaning operation where it waits until a given shift (usually
night) to be cleaned*. While tanks wait to be cleaned, the residual paint
dries up often necessitating the use of caustic solution for cleaning.
Avoidance of accumulation of dirty tanks can be achieved by designing and
performing the cleaning operation to handle any peak load continuously.
This would reduce the need for caustic cleaning and promote water
cleaning instead.
For plants employing CIP (clean-in-place) recycle systems for wash/rinse operations,
the inventory replacement frequency (and therefore waste volume) can be minimized
by using these waste reduction methods:
o Use of a countercurrent rinsing sequence.
For facilities that have additional storage space available, countercurrent
rinsing can be employed. This technique uses recycled "dirty" solution to
initially clean the tank. Following this step, recycled "clean" solution is
used to rinse the "dirty" solution from the tank. Since the level of
contamination builds up more slowly in the recycled "clean" solution than
with a simple reuse system, solution life is greatly increased.
Countercurrent rinsing is more common with CIP systems, but can be used
with all systems.
•"•Confidential source 1985: Personal communication.
B8-19
-------
o Sludge dewatering by filtration or centrifugation.
Sludge removed from spent rinse water can be dewatered so as to minimize
total waste volume. Several paint plants have installed belt filters and are
assessing their effectiveness*. At some facilities, dewatered sludge (which
still contains solvents and resins) is mixed with paint sludge from solvent
recovery and incinerated. For facilities not producing solvent-based paint,
incineration would not be an attractive option since the heating value of
the sludge would be low. In addition, the capital cost and regulatory
requirements for building an incinerator could be prohibitive to all but the
largest producers.
o Provision for adequate solid settling time in spent rinse solution.
o Use de-emulsifiers in rinse water to promote emulsion breakdown and
organic phase separation.
Another item included under the heading of cleaning wastes are the used filter
cartridges produced during the paint loading operation. These cartridges are designed
to remove undispersed pigment from the paint during loading and will be saturated
with paint when removed. Hence, waste minimization and economy both call for as
small a cartridge as possible so as to reduce the amount of paint lost and the capital
spent for the filters. If frequent filter plugging is a problem, then it should be first
addressed from the standpoint of improving pigment dispersion, and not from the
standpoint of increasing filter area.
^Confidential source 1985: Personal communication.
B8-20
-------
9.2 Implementation Profile
The main source reduction methods discussed for use by the paint industry include
segregation of solid waste by prior content, reduction or elimination of hazardous
pigments (hazardous pigments that could not be eliminated should be used in paste
form), use of wipers on mix tanks, high pressure spray systems, use of countercurrent
rinse sequence on tanks, and sludge dewatering before sending it to a landfill. Reuse
of cleaning solutions as part of paint formulation is promoted by good operating
practices which include segregation. Incineration of sludges (especially from solvent-
based operations) destroys the organics present, but still may produce toxic inorganic
residuals requiring dispoal. Since incinerators are difficult and expensive units to
build, operate, and maintain, their use is only practical at large paint facilities.
While segregating solid waste by prior content and reducing the use of hazardous
pigments would not be difficult to implement, fitting all tanks with wipers would be
difficult. Square tanks would require custom modification and installation, since the
wiper could not- be installed on the mixer (the normal method of installation).
Retrofitting existing mixers with wipers is impossible since most mixers used by the
paint industry are single-shaft and a mixer/wiper unit requires a dual shaft. There-
fore,, mixer/wiper units would only be considered for new installations or as
replacements. Since much of the hardware in the paint industry is over 20 years old,
detailed return on investment analysis could be performed to determine if a unit
should be repaired or replaced. Because the cost of repairing a unit can range from
2,500 to 10,000 dollars, replacement could be an attractive alternative (Weismantel
and Guggilam 1985).
Installation of a high pressure spray system would require the installation of a pump
and some piping at each cleaning station. While the space requirements for a pump
are small, the layout of the additional piping would have to be given special attention
so as not to impede operations in the area. The economics of a high pressure spray
system has been worked out in detail (USEPA 1979), with the average cost running
about 20,000 dollars per system. Using a countercurrent rinse sequence and
dewatering sludges would require installation of at least two cleaning solution storage
tanks and a belt filter or similar unit. Besides the added space requirements for these
units and the significant capital cost, operators would need to be fully versed in the
B8-21
-------
theory and method of their operation. The economics of installing a filter unit and/or
sludge incinerator has been addressed in detail elsewhere (WAPORA 1975, USEPA
1979). For all but the largest of producers, the cost of installing, operating, and
maintaining the units was prohibitive.
When viewed in light of increasing hazardous waste disposal costs, many of the waste
reduction measures identified appear economically practicable. By segregating
wastes, savings in disposal costs could be achieved at little cost to the facility. No
special equipment would be required and the additional labor would be minor. While
reducing or eliminating the use of hazardous pigments would also be easily
irnplernentable, the economic costs would be much greater: either higher costs for
using less hazardous pigments would have to be incurred or the company would have to
forego competing in a given market, e.g. red lead primer.
9.3 Summary
Table 9-1 presents summary data on paint industry waste sources and control methods.
The ratings shown in the table for each of the noted methods were based on review of
the available literature and consultations with industry personnel. Based on the
measures already undertaken by the paint manufacturing industry, the waste appears
to have been minimized significantly to a level characterized by a current reduction
index of 2.2 (55 percent). The index is a measure of reduction relative to the waste
that would have been produced if none of the measures listed were applied at their
current level. By implementing additional waste reduction-measures or increasing the
use of existing measures, the amount of waste currently being generated can be
reduced to a level characterized by a future reduction index of 0.7 to 1.7 (18 to 43
percent) which is indicative of a moderate to significant potential for reduction.
Based on the extent of waste generated due to equipment cleaning, it appears that
installation of a high pressure wash system and further implementation of better
operating practices are the two most effective measures that can be taken. Of the
remaining control methodologies, use of "pigs" to clean lines, dewatering of spent rinse
sludge, and increased use of automation also show high promise, as evidenced by their
high individual future reduction indexes.
B8-22
-------
TABLE 9-1 SUWARY OF SOURCE CONTROL METHODOLOGY FOR THE PAIKT NANUFACTURING INDUSTRY
1
Haste Stream |
1 1
Leftover Inorganic |1.
Pigments in Sags |2
and Packages |3.
i._
1
Pigment Dust from |1.
Baghouses |3.
(3-
1 1
Off -Specification |1.
Paint |2.
1
I Spills |1.
03
-n
.
N3
12.
|3.
r
Equipment Cleaning |1
Wastes (*) |2.
13.
|4.
|5-
16
|7
18.
|9
1
All Sources 1
1
Control Methodology |-
1
Use water soluble bags and liners |
Use recyclable/lined containers |
Implement better operating practices |
Overall |
Install dedicated baghouse systems |
Use pigments in paste form |
Implement better operating practices |
Overall |
Increased use of automation I
Implement better operating practices |
Overall |
Increased use of automation J
Maximize use of dry clean-up methods |
Implement better operating practices |
Overall |
Use mechanical wipers on mix tanks I
Install high pressure wash system |
Install Teflon liners on mix tanks |
Use foam/plastic "pig" to clean lines|
Implement better operating practices |
Use a countercurrent rinse sequence |
Oewater spent rinse sludge |
Increase spent rinse settling time |
Use de-emulsifiers on spent rinse |
Overall |
All Methods
Found Documentation
Quantity | Quality
1 1
1 1
1 1
1.00 | 0
1 I
t 1
1 1
1.00 | 1
3 1
3 1
3.00 | 1
3 1
3 1
3 1
3 00 | 1
3 I
3 1
0 1
1 1
3 I
' I
3 1
0 I
0 1
1.56 { 1
1
... |
1
0 1
o 1
o 1
00 |
1 1
1 1
1 1
00 |
1 1
2 1
50 |
1 |
' 1
2 1
33 |
1 I
2 1
0 1
1 1
2 !
1 !
2 1
0 1
0 1
00 |
Waste I
Reduction |
Effectiveness I
3 1
2 1
3 1
2 67 I
2 1
3 1
2 1
2.33 i
4 I
4 1
4.00 !
4 1
3 1
4 1
3 67 |
2 1
3 i
1 1
3 1
4 1
3 1
3 1
1 1
1 1
2 33 1
Extent of |
Current Use |
1
1 1
1 1
2 1
t 33 |
1 1
1 1
1 1
1 00 i
1 !
3 1
2.00 |
1 |
1 1
3 1
1.67 |
1 1
i !
0 1
2 1
2 I
1 1
2 1
c I
0 1
1 00 [
Future | Fraction of |
Application | Total Haste |
Potential | |
0 1 1
' 1 1
2 1 1
t 00 1 0.02 |
1 1
3 1 1
t 1 1
1.67 | 0.02 |
2 1 1
2 1 1
2 00 | 0.05 |
2 1 1
3 1 I
2 1 1
2.33 | 0.06 |
< 1 1
3 1 1
' I I
2 I 1
2 1 1
' 1 1
2 1 1
1 I I
0 I I
1 44 | 0 85 j
i 1 00 |
Current |
Reduction |-
Index |
0.8 I
0.5 |
1.5 |
1.5 I
0.5 |
0.8 |
0.5 |
0.8 |
1.0 |
3.0 |
3.0 |
1 0 |
0.8 |
3 0 I
3.0 |
0.5 |
0 8 |
0.0 |
1.5 1
2.0 |
0.8 |
1.5 |
0.0 |
0.0 |
20 |
2.2 |
Future
Probable
0
0
0
0
0
t
0
0
1
0
1
)
I
0
1
0
1
0
0
1
0
0
0
0
0
0
Reduction Index
| Maximum
•0 1
•4 1
.8 | 0
.4 | 0
•4 1
.7 | 1
•4 I
8 I 1
.5 | 1
•5 1
.0 | 1
•5 I
.7 | 1
•5 1
.2 | 1
•4 I
.7 | 1
•3 I
.3 I
.0 I
•6 I
8 I
•3 I
•0 I
.6 | 1
.7 | 1
1
1
1
1
8 1
8 1
1
^ \
1
7 1
5 !
1
5 1
|
7 1
1
7 I
1
7 I
1
1
1
1
|
1
1
^ \
^ 1
(*) These waste streams include listed "F" and/or "1C RCRA wastes
-------
10. PRODUCT SUBSTITUTION ALTERNATIVES
3y altering the composition of paint, the amount and toxicity of hazardous materials
(and the overall toxicity of the paint itself) required to be landfilled can be reduced.
Another way of reducing waste associated with paint production would be to reduce
the demand and therefore the production of paint. By producing a paint with a longer
life span or by utilizing building materials that do not require painting (e.g., precolored
concrete, stucco, vinyl coated siding, brick), a reduction in the amount of
architectural coatings produced could be achieved. Reductions in the demand for
product coatings could occur by increasing the use of powder or plastic coatings. A
complete environmental assessment of each alternative is required before its overall
benefits and disadvantages can be determined. Such an assessment was not performed;
therefore, no specific product substitutions can be recommended.
10.1 Pigment Substitution
Two cases of pigment substitution were identified. The first case concerned the use of
•
lead pigment. While the paint industry has eliminated the use of lead to a large
extent, a market still exists for red lead primers (red lead makes an excellent primer
and few substitutes exist). Most plants produce a wide variety of paints, and typically,
red lead primers represent a small percentage of a plants' total output. A
commitment by management not to produce red lead primer would eliminate the need
to manage lead pigments (the loss of this market would have to be weighed against
handling and disposal costs for each facility). Because of the increasing regulations on
the use of all lead pigments, some plants have ceased production of lead primer*.
The second case involves chrome yellow. Chrome yellow is used mainly in traffic
paint and provides a very bright color. In order to avoid the use of chrome yellow in
other products, some companies have switched to organic pigments (expensive) or to
yellow iron oxide (dirty color). One such company is Environmental Inks and Coatings
Corporation in Morganton, N.C. (Kohl, Moses, and Triplett 1984). Again a
commitment by management is required in light of the added expense versus disposal
cost savings. Customer's specifications stand in the way of total elimination of
chrome yellow use, since it is required to be used in yellow traffic paint*.
Confidental source 1985: Personal communication.
B8-24
-------
In weighing the environmental pros and cons of toxic pigment substitution with less
toxic alternatives, it must be noted that the use of heavy metal pigments in product
coatings extends the durability of the product*. It is also important to note that
substitution of pigments in existing formulations will alter both the color and
performance of the coating; hence, it must be done with the approval of changed
specifications by a customer.
10.2 Increased Quality of Home Market Paints
While most paint sold is of very high quality (commercial architectural coatings,
product coatings, and special purpose coatings have very long lifetimes), paint sold to
the home market can vary widely. This is due, in part, to market conditions that
demand that a paint producer supply a low cost paint if he wishes to remain
competitive. If the home consumer showed preference for a higher quality paint, then
the production of lower quality paints would decrease. In addition, less paint would be
produced overall each year since the need for repainting would be lessened. Related
to this issue is the level of consumers' awareness as to the proper ways of selecting
and applying paint. If the level of awareness is found to be low, then consumer
education efforts could be conducted through various approaches, such as
dissemination of information about paint application economics. Additionally,
durability of coatings can be promoted by improvement of the general knowledge of
proper application techniques, specifically, surface preparation. Even the highest
quality paints will perform poorly if misapplied or if they are inappropriate for the job.
10.3 Architectural Coating Substitutes
Many substitutes exist for architectural coatings. For building exteriors, anodized
metal, brick, marble, glass, colored concrete, and vinyl coated sidings have been used.
For interiors, wood paneling, fabric coverings, and wallpaper are very popular. Since
the use of these materials is mostly for aesthetic rather than functional purposes, each
one's position in the market place will fluctuate based on demand.
* National Paint and Coatings Association 1986: Personal communication.
88-25
-------
10.4 Product Coating Substitutes
Unlike some architectural coatings, product coatings must provide functional
protection as well as an aesthetic function. Recent substitutions that are finding a
market place are powder and plastic coatings. Because powder coatings contain no
solvent and most of the waste generated can be recycled, equipment manufacturers
view powder coatings as a way of meeting EPA standards. Coupled with the savings in
energy and labor, there is increased productivity. It is expected that the current share
of the powder coating market (5 to 6 percent in 1983) will increase to 15 percent by
1993. While numbers were not reported for the U.S., some European finishers report
that over 70 percent of all new production lines being installed are powder (Church
1984).
11. CONCLUSIONS
While the paint industry has done much to reduce the amount of waste it produces, it
appears that further reductions are possible. Our estimates indicate that future waste
reduction potential can be characterized as significant with reductions ranging from
18 to 43 percent. Several methods that appear to be quite effective and feasible for
future implementation were identified and include high pressure spray systems for
cleaning of water-based paint manufacturing equipment, further implementation of
better operating practices, use of "pigs" for cleaning lines, dewatering of wastewater
sludge, and increased use of automation. Reductions will occur as more and more
facilities initiate and continue waste reduction programs and as older equipment wears
out and is replaced.
12. REFERENCES
Church, F. L., ed. 1984. Powder coating sales reach near-boom levels. Reprint from
Modern Metals. January 1984.
Haines, H. W., ed. 1954. Resin and paint production-1954 style. Ind. Enq. Chem.
46(10): 2010-22.
Huisingh, D., Martin, L., et al. 1985. Proven profit from pollution prevention.
Conference draft. Washington, D.C.; The Institute for Local Self-Reliance.
Kohl, J., Moses, P., and Triplett, B. 1984. Managing and recycling solvents. North
Carolina practices, facilities, and regulations. Raleigh, N.C.; North Carolina State
University.
B8-26
-------
Payne, H. F. 1961. Organic coating technology. 2 vols. New York, N.Y.; John Wiley &
Sons.
Ryan, William C. 1984. MASSPIRG, Massachusetts Public Interest Research Group.
Hazardous waste reduction potential in the paint manufacturing industry. Sponsored
by the Massachusetts Department of Environmental Management. Boston, MA.
Shreve, R. N., and Brink, J. A. 1977. Chemical process industries. 4th ed. New York,
N.Y.; McGraw Hill Book Co.
SRI 1981. Stanford Research Institute. Chemical economics handbook, 1982. Menlo
Park, Calif.: Stanford Research Institute.
USDC 1985. U.S. Department of Commerce, Bureau of the Census. Paint and applied
products. In 1982 Census of manufacturers. MC82-I-28E. Washington, D.C.: U. S.
Government Printing Office.
USEPA 1979. U. S. Environmental Protection Agency, Office of Water and Waste
Mangaement. Development document for proposed effluent limitation guidelines, new
source performance standards, and pretreatment standards for the paint formulating
point source category. EPA-440-l-79-049b. Washington, D.C.: U.S. Environmental
Protection Agency.
, 1980, U.S. Environmental Protection Agency, Office of Research and
Development. Treatability Manual Vol. 2; industrial descriptions. EPA-600-8-80-
042b. Washington, D.C.: U. S. Environmental Protection Agency.
WAPORA 1975. Wapora, Inc. Assessment of industrial hazardous waste practices,
paint and applied product industry, contract solvent reclaiming operations, and factory
application of coatings. EPA-530-SW-119c. Washington, D.C.: U. S. Environmental
Protection Agency.
Webber, D. 1984. Coating industry heading for record year. Chem. Eng. News.
62(40): 51.
Weismantel, G., and Guggilam, S. 1984. Mixing and size reduction. Chem. Eng.
92(13): 71-109.
13. INDUSTRY CONTACTS
Confidential sources
Dr. G.J. Hollod, Senior Environmental Engineer, Petrochemical Dept., E.I. du Pont de
Nemours & Co. Wilmington, DE.
R.J. Nelson, Associate Director, Environmental Affairs Technical Division, National
Paint and Coatings Association, Washington, D.C.
W.G. Vaux, Principal Engineer, Chemical and Process Engineering, Westinghouse
Electric Corp., Pittsburgh, PA.
B8-27
-------
-------
1. PROCESS: PETROLEUM REFINING
2. SIC CODE: 2911
3. INDUSTRY DESCRIPTION
•
The petroleum refining industry is comprised of establishments primarily engaged in
the manufacture of gasoline, kerosene, distillate and residual fuel oils, lubricants and
other products from crude petroleum and its fractionation products. Excluded from
this industry are establishments engaged in producing natural gasoline from natural
gas, the manufacture of lubricating oils and greases by blending purchased materials,
and those re-refining used lubricating oils.
3.1 Company Size Distribution
Table 3-1 presents the number of establishments and number of employees for the
petroleum industry in 1982. Output and employment in the industry is dominated by
the larger establishments. The largest 65 of the 434 petroleum refining establishments
accounted for more than 65 percent of the total industry employment and more than
63 percent of the total industry output in terms of value added by manufacture.
Table 3-1 1982 Company Size Distribution
No. of employees per facility
1-99 100-499 500-999 1000+ Total
No. of establishments 241 128 41 24 434
No. of employees 6,600 31,300 27,700 43,200^) 108,800
Source: 1982 Census of Manufacturers (USDC 1985).
(a) Excludes totals for two establishments with more than 2500 employees to avoid
disclosure of individual company data.
3.2 Principal Producers
In 1982, the following companies maintained the largest refining capacities in the
United States:
R9-1
-------
Exxon
Standard Oil of California
Standard Oil of Indiana
Shell (Royal Dutch) Oil
Texaco
3.3 Geographical Distribution
Gulf Oil
Atlantic Richfield
Mobil
Phillips Petroleum
Marathon Oil
Table 3-2 and Figure 3-1 shows the number of petroleum refining establishments by
EPA region. EPA Region VI contained 171 petroleum refineries in 1982, or
approximately 40 percent of all establishments. Region IX, containing 54 refinery
establishments, and Region V, containing 52 establishments, also contained large
numbers of refineries. The four leading states in employment were Texas, California,
Louisiana, and Pennsylvania, accounting for 64 percent of the industry's employment.
Table 3-2 Location of Facilities by EPA Region
EPA Region
I
II
III
IV
V
VI
VII
VIII
IX
X
Number of Establishments
. _
14
34
21
52
171
13
30
54
11
National
400(a)
Source: 1982 Census of Manufacturers (USDC 1985).
(a) Establishments in states with less than 150 employees in the petroleum refining
industry are not included.
P9-2
-------
VIII
OP
•ID
I
CO
11-20
2-5
21-50
6-10
51-100
Figure 3
Roman numerals show EPA regions
1 Petroleum Refineries in the U.S.
-------
4. PRODUCTS AND THEIR USE
The types of products shipped and the dollar value of these shipments in 1982 are
presented in Table 4-1. Gasoline, light fuel oils, jet fuel, and heavy fuel oil shipments
accounted for over 80 percent of the total value of product shipments from the
petroleum refining industry.
Table 4-1 1982 Product Shipments by the Petroleum Refining Industry
Shipment Value Percentage of
Product (millions) $Total Shipment
Value
Gasoline
Jet fuel
Kerosene
Light fuel oils
Heavy fuel oils
Lubricating oils and greases
made in refineries
Unfinished oil and lubricating
oil base stock
Asphalt
Liquified refinery gases
Aromatics
Other
Total
95,115
14,862
1,887
37,298
11,391
2,891
8,066
2,980
8,396
1,354
7,130
191,370
49.7
7.8
1.0
19.4
6.0
1.5
4.2
1.6
4.4
0.7
3.7
100.0
Source: 1982 Census of Manufacturers (USDC 1985).
5. RAW MATERIALS
The types of materials consumed as feedstock by the petroleum refining industry in
1982 are presented in Table 5-1. The delivered cost of each material type is also
shown. Crude petroleum is the major raw material consumed by the petroleum
B9-4
-------
refining industry, representing over 80 percent of the total delivered cost of all
materials consumed by the industry in 1982.
t
Table 5-1 Raw Materials Consumed by the Petroleum Refining Industry in 1982
Delivered Cost Percentage of
Material
Crude petroleum
Unfinished oils
Natural gas liquids
Benzol
Toluene and xylene
Additives
Chemical catalytic preparations
Caustic soda
Sulphuric acid
Containers
Other
Total
(millions) $
137,870
558
7,901
74
459
1,199
679
80
120
156
17,742
166,838
Total Delivered Cost
82.7
0.3
4.7
--
0.3
0.7
0.4
0.1
0.1
0.1
10.6
100.0
Source: 1982 Census of Manufacturers (USDC 1985).
6. PROCESS DESCRIPTION
Crude oil contains a wide variety of hydrocarbon compounds which range from light
gases to residual fractions that cannot be separated by distillation. Crude oil is made
up primarily of paraffin compounds, olefins, naphthenes, and aromatics in varying
proportions. Apart from carbon and hydrogen, the elemental analysis shows sulfur,
.oxygen, nitrogen and heavy metals present in small quantities.
Most refinery processes fall into one of three classifications: separation, conversion,
or upgrading. Detailed descriptions are available in the literature (Jahnig 1982, Shreve
1967). The following discussion highlights the main process activities. For
convenience, an overall block flow diagram is presented in Figure 6-1.
B9-5
-------
SULFUB
I NAPHTHA
CN
at
i p
EH
R T
'
«»»MTH»
«J US
ME»»Y
NiPHTHi
JL
'
N1PHTM1
i
CiTtlTTIC 1 RgroRNiTE^
Rl
[ilKTliTE
• OTOR
MS
US
(E*OSE*E
KE*OSE«C
MTDKO-
S2£
iTE
,
1
i LT »ro»o ' :RI:««TE
1 1 1
1 i 1
PETROCHE.ICiL
PU..T
tASOLlHC
PETRO - _
CNENICil.
KEROSENE
MOCEIS litre cmtomi
0 SEH04TIOM MOCCSS MSTIS (riLTEH CLiTS)
0 CONVCRSIOH MOCESS H5TES (ciTiLTST FINES/ SPENT CiTilTSTJ
0 UPHiOtNC PHOCESS liiTR [«»fNT CiTiLTtT/ SLUBIEJ
0 »uxiLU»T piocts* iisrn (SPENT »UNE/ srneTrono SOLUTION]
HOT! CSUIMtXT CLE»»I»S liSTIS liSTEliTEN THEiTICNT liSTCS iW
UTILJTT PRODUCTION liSTES »«E NOT SHOIN
Figure 6-1 Process Block Flo* Oiagrai for a Typical Refinery
B9-6
-------
6.1 Separation Processes
i
Separation processes are used to provide feedstocks of specified physical and chemical
properties to other refining unit processes. Typical separation processes are desalting,
primary distillation, deasphalting, solvent refining, and lube oil finishing. These
operations are discussed individually below.
The first step in crude oil refining is the desalting operation. Crude oil typically
contains salts, water, and suspended matter in the form of droplets or as a fine
emulsion. These materials are present naturally in the oil and also may be introduced
during shipping or storage. It is particularly important to remove these contaminants
so as to avoid corrosion or plugging and buildup problems in the refinery equipment.
The crude oil is pumped from storage into the heated desalter unit. Some of the larger
droplets settle out during storage. The remaining droplets and emulsified materials
are removed by the addition of water and flocculants. The oil/water emulsion is
separated by electrostatic precipitation with 95 percent removal efficiency.
Distillation follows the desalting process. Typically, the crude oil is separated or "cut"
into three categories of products, each cut differentiated by boiling range. These
products are: naphtha, middle distillate, and a heavy bottoms residual. Also from this
process, light fractions (Cj to C^ hydrocarbon) are recovered for further purification
in the gas treating operation.
The naphtha cut is the primary precursor to motor gasolines. It is upgraded by
hydrotreating (primarily to remove sulfur, oxygen, and metals) and then further
improved by the process of reforming. The middle distillate cut includes jet fuel,
diesel, and home heating oil. The end point of the middle distillate cut is about 300°C
(Jahnig 1982). Fractions boiling above this temperature come out in the bottoms cut.
The bottoms fraction provides feed-stock for the vacuum distillation process. The
vacuum stills remove the remaining heavy gas oil not recoverable in the atmospheric
stills, and provide asphalt and lube oil stocks and coker feed. The residue from the
vacuum stills goes to the coker where the volatiles are removed, leaving coke as the
final product.
Deasphalting removes asphalt or resins from the bottoms fractions, to produce stocks
suitable for subsequent lube oil or catalytic cracking processes. The asphaltic
-------
materials are extracted through the use of a solvent, such as propane. The process is
carried out in an extraction tower, where pipe still bottoms or other heavy stock are
mixed with propane. After the asphalt has been removed the propane is recovered.
Solvent refining includes a large number of alternative subprocesses designed to obtain
high-grade lubricating oil stocks or aromatics from feedstocks containing naphthenic,
acidic, organornetallic, or other undesirable materials. Basically, it is a solvent
extraction process dependent on the differential solubilities of the desirable and
undesirable components of the feedstock. The principal steps are countercurrent
solvent extraction, separation of solvent and product by heating and fractionation,
removal of trace solvent from the product, and solvent recovery. Recent advances in
solvent refining include a Residuum Oil Supercritical Extraction (ROSE) method
developed by Kerr-McGee Refining Corporation (Anon. 1981).
Lube oil finishing is used to further refine solvent-refined or dewaxed lube oil stocks,
and involves clay or acid treatment to remove color-forming and other undesirable
materials. The two methods most widely used by industry are continuous contact
filtration in which the oil-clay slurry is heated and the oil removed by vacuum
filtration; and percolation filtration, in which the oil is filtered by percolation through
clay beds. Percolation also involves naphtha washing and kiln burning of spent clay to
remove carbon deposits and other impurities from the clay so it can subsequently be
reused.
6.2 Conversion Processes
Many primary distillation products undergo some form of conversion operation.
Conversion processes are used to purify or change the chemical and/or physical
properties of the distillates, usually by breaking large molecules into smaller ones.
Typical conversion processes are hydrotreating, catalytic and thermal cracking,
hydrocracking, and coking.
In most refineries, usually only the naphtha and middle distillate cuts or low metals
content residuum are hydrotreated. The hydrotreating process converts the heavier
components of these cuts to lower boiling products by mild cracking and by adding
hydrogen to the molecules using a fixed bed catalytic reactor. The reactor products
are cooled, and the hydrogen with impurities and high grade product are separated.
B9-8
-------
Among the catalysts most commonly used in hydrotreating is cobalt molybdate with
various promoters on silica-alumina supports.
i
Hydrotreating was first used primarily on the relatively clean lighter feedstocks, but
with more operating experience and improved catalysts, hydrotreating has been
applied to increasingly heavier fractions. The process improves the quality of the feed
to subsequent operations by removing sulfur compounds, nitrogen, oxygen, and metals
carried over from primary distillation. Hydrotreating of kerosene, jet, and diesel fuel
components stabilizes and saturates the diolefins and other compounds which form
gums and deposits. Fixed catalyst bed hydrotreating of atmospheric or vacuum still
heavy oils has been limited to those feedstocks where the metals content is low. The
metals quickly degrade the catalysts by buildup and subsequent pore blocking. In
recent years, progress in applied catalysis has resulted in the development of longer-
life hydrotreating catalysts. Along with hydrocracking, hydrotreating is one of the
most rapidly growing refinery processes.
Cracking includes fluid catalytic cracking, thermal cracking, hydrocracking, vis-
breaking, and coking. The old process of thermal cracking, or now more generally
applied, fluidized-bed catalytic cracking, are processes used on the heavy distillates
from the vacuum still. In each of these operations, heavy oil fractions are broken
down into lighter fractions such as gasoline, domestic heating oil, etc. At this time,
about half of the gasoline sold in the United States is obtained from petroleum by
fluidized bed cat cracking of heavy oils. The reactors are so designed as to allow for
continuous catalyst regeneration. Thermal cracking, which was an important process
before the development of fluid catalytic cracking, is being phased out. Visbreaking
and coking units are installed in a significant number of refineries, and their
application is expected to increase.
Catalytic cracking, like thermal cracking, breaks heavy fractions (principally gas oils)
into lighter fractions, and is the key process in the production of large volumes of
high-octane gasoline stocks, furnace oils, and other useful middle distillates. The use
of a catalyst permits operation at lower temperatures and pressures than those of
thermal cracking and also inhibits the formation of undesirable by-products. Fluid
catalytic cracking processes have, in most cases, replaced the fixed-bed and
moving-bed processes which used a beaded or pelleted catalyst. The catalyst
presently used in the process is primarily zeolite.
B9-9
-------
Hydrocracking is a catalytic method of converting refractory middle-boiling or heavy
sour feedstocks into high-octane gasoline, reformer charge stock, jet fuel, and/or high
grade fuel oil. This process has a high degree of flexibility in adjusting production to
meet changing product demands. It is one of the most rapidly growing refinery
processes. The types of catalysts commonly used are tungsten sulfide-silica alumina,
iron-HP clay, nickel-silica alumina, and molecular sieves.
Coking is used to process the heavy asphaltic vacuum still residuals which do not pass
into the cat cracking or lube oil operation after vacuum distillation. The coking
operation involves destructive distillation of the very heavy low value residuum.
Coking this material produces lighter materials by means of thermal cracking. The
products of this operation include gas, naphtha, gas oils, and coke. Coke is the
carbonaceous material remaining after the volatiles are removed.
The coking process occurs either in a semibatch operation called delayed coking or is
performed continuously in a fluidized bed process. In the delayed coking, the feed is
heated and pumped into large drums. The material is then cooled to remove the heat
of reaction and is maintained at a fixed pressure which vaporizes the remaining
volatile materials in the feed. When the drum is full, the process is stopped and the,
drum is opened. The coke is removed using hydraulic cutters.
In the fluidized bed process, the residum feed is heated and sprayed into a hot
fluidized bed of coke. The coke particles grow and are removed from the system when
their size exceeds the maximum size required for fluidization. Seed particles of
smaller size are added continuously to the process to maintain the bed particle size
distribution in the required operating range.
6.3 Upgrading Processes
Many products and by-products obtained from conversion steps undergo upgrading to
generate a salable product which meets certain defined characteristics and increase
the yield. Typical upgrading processes are alkylation, polymerization, reforming,
isomerization, and drying/sweetening. The upgrading process used is dependent on
feedstock and desired end product.
B9-10
-------
Alkylation is the addition of an isoparaffin (usually isobutane) to an olefin (propylene,
butylene, etc.) in the presence of a catalyst to produce a high-octane alkylate, which
is one of the most important components of automotive fuels. Sulfuric acid is the
most widely used catalyst, although hydrofluoric acid and aluminum chloride are also
used. Alkylation process capacity is expected to continue to increase with the demand
for high-octane gasoline.
Polymerization converts olefin feedstocks (primarily propylene) into a higher
molecular weight polymer gasoline. While polymerization yields, per unit of olefin
feed, are less than alkylation yields, newer promoted cat cracker catalysts produce a
feed that favors polymerization. Because of the higher ratio of olefins to isobutane
produced, alkylation cannot convert all of the olefins to high-octane alkylate.
Therefore, while only a few refineries currently use polymerization, its use may
increase depending on the use of newer catalysts for cracking.
An example of a polymerization process is Dimersol, a process used to dimerize light
olefins such as ethylene, propylene, and butylene (Anonymous 1984). The main
applications of this process include dimerization of propylene to produce a high-octane
low-boiling gasoline called Dimate, and the dimerization of n-butylene to produce CB
olefins for plasticizer synthesis. Dimerization is achieved in a liquid phase at ambient
temperature by means of a soluble catalytic complex. The catalyst is removed by a
caustic wash process which generates an aqueous waste stream. After the catalyst is
removed from the reaction mixture, the products are separated by means of
distillation.
Reforming is a process of molecular rearrangement to convert low-octane feedstocks
to high-octane gasoline blending stock, or to produce aromatics for petrochemical
uses. The principal reactions are dehydrogenation, dehydrocyclization and isomeriz-
ation. The undesired hydrocracking also occurs to a small degree. Multi-reactor,
fixed-bed catalytic processes have completely replaced the older thermal processes.
There are many variations, but the essential difference is the composition of the
catalyst involved. The types of catalyst commonly used in this process are bimetallic
platinum-rhenium and various combinations of group VIII elements.
B9-11
-------
Isomerization is another molecular rearrangement process which is very similar to
reforming. The charge stocks generally are lighter and more specific (normal butane,
pentane, and hexane). The catalysts used are platinum supported om alumina, or
platinum supported on molecular sieves. The desired products are isobutane for
aikylation feedstocks and high-octane isomers of the original feed materials for motor
fuel.
Drying and sweetening are processes concerned primarily with removal of sulfur
compounds, water, and other impurities, from gasoline, kerosene, jet fuels, domestic
heating oils, and other middle distillate products. "Sweetening" is the removal from
these products of hydrogen sulfide, mercaptans, and other sulfur compounds, which
impart a foul odor and/or decrease the tetra-ethyl lead susceptibility of gasoline.
Drying is accomplished by salt filters or adsorptive clay beds. Electric fields are
sometimes used to facilitate separation of product from treating solution.
6.4 Offsites and Auxiliary Processes
In addition to the main process unitsj the refinery will have other processes or
operations which support the overall primary process. These processes include gas
recovery and treatment, sulfur recovery, hydrogen manufacturing, wastewater treat-
ment, steam and cooling water utilities, and blending/packaging.
Gas recovery and treatment facilities are provided to sweeten and recover the light
ends produced by the refining process. Typically, the light ends from the primary
distillation operation and tail gas from hydrotreating and cat cracking will contain Cj_
thru Cg hydrocarbons, hydrogen sulfide, and water. This material can be upgraded to
fuel gas and liquefied petroleum gas (LPG) in the. gas plant. Typical operations include
recovery of hydrocarbons via compression or lean-oil absorption, removal of hydrogen
sulfide by amine scrubbing and then separation of the C^ + hydrocarbons by
distillation. The 04 + hydrocarbons can also be converted to gasoline or used
elsewhere in the refinery.
Hydrogen sulfide from amine units and from the sour water stripper (which treats
nearly all aqueous process streams in the refinery), is then converted to elemental
sulfur in a Claus plant. Conversion to elemental sulfur is obtained by burning part of
the hydrogen sulfide to form sulfur dioxide. The sulfur dioxide is then reacted with
hydrogen sulfide to form elemental sulfur and water vapor in a series of vapor phase
B9-12
-------
catalytic reactors. Tail gas containing sulfur and sulfur compounds from the Claus
plant is further treated by Beavon or other processes before being released to the
atmosphere. The ammonia present in the feed gas to the Claus plant is oxidized to
nitrogen and water.
Hydrogen manufacturing is another auxiliary process. Some of the refinery hydrogen
demand is satisfied by catalytic reforming, but not all. It is therefore necessary to
manufacture hydrogen to meet the full process demand. The most common method of
manufacture is by reacting methane (natural gas) with steam over a nickel catalyst.
The products are hydrogen and carbon dioxide. Carbon dioxide is removed by amine
scrubbing and a hydrogen product of about 95 to 98 percent purity is obtained.
The required utilities, which include electric power, steam, and cooling water are
produced on site. Steam flows of up to 1000 tons per hour, at several pressure levels,
are generated to provide process power and heat. The fuel source is gas, oil, or
refinery byproducts. Electric power is purchased from utilities and supplemented by
cogeneration or generation onsite where feasible (e.g. through FCC power recovery
train). Cooling water is supplied as either once-through water or recirculated cooling
tower water.
Blending is the final step in the production of finished petroleum products. Blending is
required to meet quality specifications and market demands. The largest volume
operation is the blending of various gasoline stocks, including alkylates and other
high-octane components, with anti-knock (such as tetra-ethyl lead), anti-rust, anti-
icing, and other additives. Diesel fuels, lube oils, waxes, and asphalts are other
refinery products which normally require blending of various components and/or
additives. Packaging at refineries is generally highly automated and restricted to high
volume, consumer-oriented products such as motor oils.
7. WASTE DESCRIPTION
The primary specific wastes associated with petroleum refining, along with their
process sources, are listed in Table 7-1. While these wastes have been classified into
seven distinct groups (separation process wastes, conversion process wastes, upgrading
process wastes, wastes from auxiliary processes, equipment cleaning wastes, waste-
water treatment wastes, and waste from utilities production), many other more
R9-13
-------
Table 7-1 PetroSeurn Refining Process Wastes
No.
Waste
Classification
Process
Origin
Description
RCRA
Code
on
i
i—*
c-
1. Separation process wastes
2. Conversion process wastes
3. Upgrading process wastes
. Auxiliary process
wastes
5. Equipment cleaning wastes
6. Wastewater treatment wastes
7. Utility production
wastes
Kerosene or lube oil refining
Fluid catalytic cracker
Hydrotreating, hydrocracking
Alkylation, reforming,
isomerization
Alkylation
Dimersol
Amine scrubbing
Stretford process
Claus process, steam
reforming, sulfuric acid plant
Heat exchangers
Crude storage tank bottoms
Leaded gasoline tank bottoms
API separator
Dissolved air flotation
Slop oil tank
Vacuum filter
Biological treatment
Storm water settling basins
Filter clays
Spent catalysts
Spent catalysts
Acid sludge and lime
Spent caustic
Spent amine
Spent Stretford
solution
Spent catalysts
Sludge
Sludge
Leaded sludge
Sludge
Oil, water
Sludge
Sludge
Sludge
Sludge
Raw water treatment Sludge
Once-through cooling water Sludge
Cooling tower blowdown treatment Sludge
Boiler feedwater treatment Sludge
K050
K052
K051
K048
K049
-------
detailed classifications are possible due to the diverse and complex nature of the
industry. Additionally, the exclusion of RCRA codes from many of the waste streams
only means that no specific "F" or "K" RCRA code exists.
7.1 Separation Process Wastes
Principal wastes associated with separation processes are spent filter clays used for
lube oil and kerosene refining. Spent clay is produced in significantly greater
quantities from the clay contacting process than from the fixed bed process. In a
contacting process, the clay is mixed with the oil and subsequently removed with a
rotary vacuum filter. Treatment with fixed bed clay is presently more predominant
and is used to remove color, chemical treatment residues, and traces of moisture from
product streams such as gasoline, kerosene, jet fuel, light fuel oil, and BTX (benzene-
toluene-xylene) fraction from catalytic reformate. The spent clay from either of the
above processes is usually disposed of in a landfill.
7.2 Conversion Process Wastes
Conversion process wastes originate from operation of the fluid catalytic cracker,
hydrotreating operations, hydrocracking operations, and coking operations. Fluid
catalytic cracker (FCC) catalyst is continously regenerated by burning off the coke
formed on the catalyst during the cracking process. The flue gas from the regenerator
passes through a series of cyclones that recover most of the catalyst. This recovered
catalyst is then returned to the reactor vessel. Because of air pollution regulations,
refineries have installed electrostatic precipitators or equivalent tertiary separation
devices to significantly limit catalyst fines in the regenerator flue gas. These catalyst
fines are either landfilled, or in some cases sold. They are generated on a continuous
basis, but are generally disposed of on an intermittent basis.
A number of refinery processes require the use of a fixed-bed catalyst. These
processes include catalytic reforming, hydrodesulfurization, hydrotreating, hydro-
cracking, and others. These catalysts become inactive (viz six months to three years)
and are eventually replaced in the reactors with fresh catalyst during a unit shutdown.
Many of these catalysts contain valuable metals which can be recovered economically.
Some of these metals, such as platinum and palladium, represent the active catalytic
component; other metals are contaminants in the feed which are deposited on the
B9-15
-------
catalyst during use (e.g. nickel and vanadium). After valuable metals are recovered (a
service usually performed by the outside companies), the residuals are expected to be
disposed of as solid waste.
The major contaminating metals found on catalytic cracking catalysts are vanadium,
nickel, copper, chromium, and iron. Small amounts of these metals are present in the
crude petroleum and, except for some of the iron, all are in the form of metal-organic
compounds. Some of these compounds are volatile and when the vacuum gas oil feed
to the catalytic cracking units is prepared, they appear in the gas oil. A fraction of
the iron, and probably chromium, found on the catalyst is the result of erosion and
corrosion either in the lines or equipment.
Coke which is produced in the course of fluid coking and delayed coking, is sold as
solid industrial fuel or for use in electrode manufacturing. Coke fines are generated
intermittently and their quantity is a function of handling techniques. A certain
amount of spillage, and consequent contamination with dirt, results during the course
of loading operations onto trucks and railroad cars.
7.3 Upgrading Process Wastes
Process wastes from upgrading operation include primarily sludges resulting from
neutralization of acidic streams in alkylation process and spent dessicant clays. The
residuals from other upgrading processes, such as reforming, isomerization, poly-
merization and sweetening include wastes resulting from spent catalyst re-processing
(usually offsite), neutralized spent caustic from dimerization operation and spent salt
or clay from drying.
Alkylation sludges from the HP alkylation process are produced as a result of removal
of acid from organic streams (isobutane recycle and alkylate), vent scrubbing, tank
sludge and HP regenerator bottoms (acid oils). The wastes from the sulfuric acid
alkylation include residuals from acid regeneration (usually done by an offsite
contractor), caustic washes and acid blowdown stream. Alkylation sludges are
produced as a result of direct neutralization of acidic effluents using lime or indirectly
by using lime to reconstitute primary base, such as KOH. The sludges are usually
landfilled. Acid soluble oils from HP alkylation are either sold, burned or recycled to
other units in the refinery.
89-16
-------
7.4 Auxiliary Process Wastes
The amine scrubbing process, used to remove hydrogen sulfide and carboft dioxide from
refinery gases, generates a spent amine waste stream. Another waste stream, which
contains spent anthraquinone disulfonic acid (ADA) along with vanadium salts, results
from the Stretford section of the Beavon process that removes hydrogen sulfide from
the tail gases originating from a Claus plant. Both of these wastes are often drummed
and disposed of in a landfill. The wastes from auxiliary processes also include spent
catalyst from steam reforming, Claus plant, and sulfuric acid plant.
7.5 Equipment Cleaning Wastes
Equipment cleaning wastes are mainly generated from heat exchanger and storage
tank cleaning. Heat exchanger bundles are periodically cleaned during plant or unit
shutdown. Scale and sediment resulting from such cleaning is collected in sumps, from
which it is either flushed into the process sewer system or shoveled out and disposed
on land. The disposal practices for spent chemical cleaning solutions could not be
ascertained. The heat exchanger cleaning waste is considered to be a RCRA
hazardous waste.
•
Solid sediment from incoming crude oil accumulates at the bottom of the crude oil
storage tanks. These tanks are cleaned periodically to remove the sediment. The
cleaning frequency is dependent on the sediment content in the crude oil. In
refineries which use mixers in the storage tanks, this waste source is non-existent.
Constituents of the crude tank sludge vary with type of crude oil as well as with
handling and shipping methods employed prior to delivery to the refinery. Settled
sludge consists of a mixture of iron, rust, clay, sand, water, sediment, and occluded oil
and wax. Usually, this mixture is a tightly held emulsion which does not separate on
settling. The frequency of sludge removal varies from once a year to once every ten
years.
Solids from leaded or non-leaded products settle to the bottom of storage tanks, where
they remain until they are removed. This accumulated sludge is removed whenever
the tank service is changed, the sediment content of the stored product exceeds
specifications, or the tank itself needs inspection or repair. The characteristics of the
deposited sludge will vary with the type of product stored in the tank. Leaded tank
89-17
-------
bottoms are considered RCRA hazardous and special handling methods are used when
the sludge is removed. It is removed at intervals varying between once a year and
<
once every five to seven years. The sludge is spread on special concrete pads for
weathering. Weathered sludge is disposed of in a Class I landfill.
7.6 Wastewater Treatment Wastes
Waste due to wastewater treatment is produced by the API separator, dissolved air
flotation unit, slop oil tank, vacuum filter, biological treatment system (if present),
and the stormwater settling basins. Solids which settle in the API separator during
primary wastewater treatment are periodically removed with a vacuum truck, water is
discharged to the secondary treatment and oil is recycled back to the crude unit
through slop oil tank. Refinery API separators are usually connected to the oily water
plant sewer. The bottoms, therefore, are a mixture of settleable portions of all
sewered wastes, such as tank bottoms and desalter sludges, and, probably contain a
certain amount of all other compounds that are present in the wastewater. These
sludges are listed as RCRA hazardous wastes. In some refineries, additional oil and
solids are removed from the API water effluent by dissolved air flotation. The
flotation process takes place in a tank with or without the aid of chemicals. Air
bubbles form and bring the finely divided solids and oil particles to the surface, where
they are skimmed off. Oils are pumped to the slop oil tank. Sludge is sent to a
landfill. The waste is classified as RCRA hazardous.
Skimmed oil from the API separators is usually pumped into a slop oil tank where the
mixture is separated into three fractions: oil, water, and emulsion. The oil is returned
for reprocessing, and the water is recycled back to the API separator. The emulsion
layer may be disposed as a sludge, or it may be further treated, i.e., de-emulsified.
De-emulsification is carried out by chemical or by physical treatment. The former
process employs the use of special chemicals or agents, heat, and settling tanks. The
later involves removal of suspended solids by centrifugation or vacuum filtration and
the water and oils are separated in settling tanks. In either process, the oil is
reprocessed, the water is returned to the wastewater treatment system, and the solids
are disposed as a solid waste. This waste is classified as a RCRA hazardous waste.
In order to reduce sludge volume, some refineries concentrate certain waste streams
through use of a common dewatering system. The dewatered cake from these
B9-18
-------
processes is disposed to land, while the filtrate or centrate is returned to the
wastewater treatment system.
«
In" the process of biological treatment of refinery aqueous waste streams emerging
from the primary or secondary treatment, excess bio-sludge is created which, for
efficient operation, must be controlled by wasting. The waste bio-sludge has a very
high water content (99%) and is dewatered prior to disposal. This waste is generated
intermittently at a rate which is dependent upon activated sludge process variations,
the desired level of process efficiency, and the raw waste load.
Silt which collects in the stormwater settling basins in some refineries is periodically
removed, de-watered, and land disposed. The quantity of silt is usually a function of
the amount of rainfall and/or refinery paved area rather than of process complexity.
7.7 Utilities Production Wastes
Utility wastes are created by treatment of raw process water or once-through cooling
water conditioning, cooling tower water blowdown treatment, cooling tower sludge,
and boiler feedwater treatment. Raw water is usually strained, clarified using
coagulation, flocculation, and sedimentation, and sometimes softened using lime treat-
ment. Water pumped from a nearby source is passed through primary settling tanks
prior to usage for once-through cooling. Sludge is periodically removed from these
tanks.
The blowdown from the recirculating cooling water system is treated to precipitate
chromium. Chromium salts are added to the cooling water to inhibit corrosion of
carbon steel. Sludge which settles in the cooling tower basin is removed whenever the
cooling tower is out of operation. It is either washed into the process sewer system or
shoveled out and land disposed. This sludge also contains chromium.
Spent lime from cold or hot lime softening and from the clarification of boiler feed
water is continuously discharged, de-watered in a settling basin, and disposed on land
or re-used for neutralization of acidic waste streams (e.g. HF alkylation spent acid).
The quantities and composition of the spent lime sludges are dependent upon the
characteristics of the raw makeup water.
B9-19
-------
8. WASTE GENERATION RATES
In 1976, the EPA published a study on Hazardous Waste Practices in "the petroleum
refining industry (Jacobs 1976). One of the four basic objectives of the study was to
determine the source, nature, and quantity of potentially hazardous wastes generated
by industries engaged in the business of refining crude oil. Seventeen representative
refineries were selected for study. Potential sources of hazardous waste were
identified, sampled, and chemically analyzed for a range of constituents. These
analyses served as the basis for estimates of total waste emissions from the refining
industries. This information was subsequently reported in a 1979 study (Jacobs 1979).
Generation rates are summarized in Table 8-1 for various petroleum refining waste
streams. More recent waste generation data were not in evidence at the time of final
report preparation.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
Generation and minimization of waste from petroleum refining can be initially viewed
in the simple context of conceptualizing the complex process of refining as one in
which the undesirable impurities naturally occurring in petroleum (principally heavy
metals, water, sulfur, oxygen and nitrogen) are removed from the hydrocarbons. At
this general level, it is important to note that the wastes produced due to impurity
removal are unavoidable and can only be reduced by using lighter crude feedstock (an
issue of high economic and political complexity) or by a decrease in the use of
petroleum (an even more controversial and complex issue); neither one is suitable even
for a topical exploration in this study. Hence, waste minimization in petroleum
refining is more readily addressed in terms of what can be done given a current mix
and flow of refinery feedstock. Here, the following general areas of focus are
distinguished.
Maximize utilization (minimize loss) of auxiliary materials and supplies.
For example, extending the catalyst's life results in less frequent catalyst
replacement thus reducing possible waste associated with its disposal or
reprocessing.
89-20
-------
Table 8-1 1974 Waste Generation Rates from the Petroleum Refining Industry
DO
Waste
Kerosene filter clays
Lube oil filter clays
Neutralized alkylation sludge
Heat exchanger cleaning sludge
Crude oil tank bottoms
Leaded gas tank bottoms
API separator sludge
Slop oil emulsion solids
Biological treatment sludge
Storm water silt
Once through cooling water sludge
Cooling tower sludge
Boiler feedwater treatment sludge
Oil
(Weight %)
3.5
21.9
6.9
10.7
47.4
20.0
22.6
18.0
0.3
3.9
0.4
0.4
0.3
Water
(Weight %)
5.0
50.0
54.2
53.0
13.3
0.3
53.0
40.0
87.0
25.0
25.0
75.4
59.0
Others
(Weight %J
91.5
28.1
38.9
36.3
39.3
79.7
24.4
42.0
12.7
71.1
74.6
24.2
40.7
Total
(Metric Tons/Yr/
1,000 BPSD
Crude Capacity)
1.0
431. 2(a)
47.4
0.7
0.2
0.8
4.7
2.3
6.2
2.7
16.8
0.1
70.5
Source: Alternatives for Hazardous Waste Management in the Petroleum Refining Industry (Jacobs 1979).
(a) Metric tons per 1,000 BPSD lube capacity.
-------
Search for less toxic substitutes. An example of less toxic materials
substitution is found in the use of non-chromate corrosion inhibitors for
t
cooling water treatment.
Maximize slop oil recovery. This pertains to oil that is lost with sludge and
emulsion solids leaving primary or secondary wastewater treatment. In
this context, one should address the causes of slop oil generation and the
ways to improve its separation efficiency from aqueous streams.
Maximize energy and water conservation. This reduces waste generation
associated with treatment of boiler feedwater, plant water, and cooling
tower water.
Reduce equipment cleaning frequency to cut down on equipment cleaning
waste.
All of the above practices are encountered in one form or another at virtually every
refinery. The attempt to identify and characterize waste minimization options is
presented below. Most of the principal waste streams leaving a refinery were
addressed -some of which are exempt from federal regulations (but not from state
regulations), others'that are not considered to be hazardous, and still others that are
listed RCRA wastes. The rationale for including all streams was provided in the
introduction to this appendix.
9.1 Description of Techniques
The list of individual primary waste streams along with a list of possible source
reduction methods is presented in Table 9-1. Sections below deal with the description
of the listed methods. All proposed methods appear to be technically feasible;
however, the decision of whether or not to use a particular method is highly site
specific and is dependent on other factors such as economics, cost/benefits, configura-
tion of existing facility, and local regulatory constraints.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures titled "good
B9-22
-------
M SUHWRY or SOURCE CONIROI utiHoowocr ron THE PHROUUK REFINING INOUSIHT
CD
r-j
1
i
'
1
1
1
1
1
1
i
1
1
1
1
1
1
1
1
1
1
1
!
Seoarat tor. Process i t
niter Cla»s Onl,|,'
i
FCC Catal.st Onl,|2
1
Uogradlng Process |l
Alkylatlon Kastes|7
On1, 13
i
*u*1 Mary Process 1 1
jtretford |3
11
IS
1
[quipneni Cleaning | 1
Sludge (•). 13
Bctto»s. and IS
Gasoline lank 16
Bottoms 1'; i'
II
MO
!
i
ceo!a;e f'.ltrsfon «lth hydrotreatlng)
Regenerate and recycle clay |
Use fi>ed bed city filtration process I
Overall |
fmlaite aeration and purg? sl*a» |
Overall |
Proper operation of In* process |
Use HOH to neutrtliie HF acid uaste I
Upgrade feedstock, by hydrogenat Ion I
Overall |
Cholct of oroper a«lne solution |
Prevent deterioration of jtret so'n 1
tlte-nat'.es to .V-AO» I
Seco.er products froi tne »»it»s!rta«|
Oteinatives to Stretforo process 1
Overall |
Decrease fil» teaperatjre 1
Use proper corrosion inhibitors {
Use smooth Sett eicUnqt tube surface)
Reduction of Pb In gtinline |
IrsMll storage ttnk agitators |
Use liners/corrosion resist aaterlal I
Prevent omdation of crude oil |
Refine/recycle sludge fo' orgtnlcs I
Overall |
found Documentation
Quantity | Quality
1 1
1 1
1 1
1 00 | 1
0 1
0 30 | 0
7 I
7 1
t 1
1 67 | 1
7 1
1 i
; i
i :
i i
1 33 I 1
1 1
1 1
1 1
' 1
1 1
0 1
i I
1 1
0 80 | 0
1
1
30
0
00
1
7
It
7
1
7
13
1
1
1
1
0
1
1
90
1 Naste I
1 Effectiveness |
1 - 3 |
i 3 |
1 ? I' 1
1 7|
1 7 SO |
1 3 |
1 3 |
I 3 00 |
1 3 |
J 1
1 ) 1
1 * 1
1 Ml I
1 1 1
i y \
i 1 1
i 3 |
1 3 |
1 ' 1
1 3 i
1 ? 10 1
Extent of | Future t Iractlon of |
| Potential 1 |
1 1 3 | |
n n i
71 ' 1 1
1 It | 7 00 | 0 01 |
31 1 1 1
7 00 | 1 SO | unknown |
31 1 1 1
J 1 H 1
1 1 7| |
7 00 | 1 It | 0 31 |
31 M 1
71 M 1
71 1 1 1
M I) 1
M M 1
7 00 | 1 SO | unknom |
1 1 I 1 1
J 1 II 1
M M i
ji ii i
i ' i l
i i i i
i M i
1 M 1
1 M 1
1 3 | |
1 30 1 1 30 | 0 01 I
Current |
Inden |
01 |
1 5 |
1 0 I
1 S |
' 1
1 S |
1 5 1
I 3 |
1 5 |
0 1 1
7 3 |
» 3 |
1 0 |
1 S |
0 3 |
1 « 1
J 3 |
0 S |
1 0 1
' 5 |
1 0 1
0 3 |
7 3 |
0 1 1
0 5 I
0 1 1
7 ; ;
Future 1
Probable
1
0
0
0
1
0
0
0
1
0
0
0
0
0
0
0
g
0
o
0
0
1
0
1
0
eduction Indev
| Naiiaui
t 1 1
1 1
3 I
J | 1
1 1
• 1 '
7 1
' 1
1 1 1
t 1 1
3 I
1 I
7 I
S 1
• 1 '
1 1
} |
3 1
1 1
3 1
7 I
1 1
1 1
I 1 1
5 1 -1
1
7
1
1
1
1
7
7
(*} Jht?$«» $tr«?a«s include listed T* and/or "K" RCRA wastes
-------
7WU I-1 continued
'
1
1
i
1
1
I
1
1
1
1
1
1
1
1
1
1
i
1
1
1
1
1
Hlitf Strps"
1
1
1
Kaste>ater treat mg|l
1FI Separator
31ud,e >•).
0«f flcat (')
Sloo Oil lank
Siudge (•).
Biosludge. and
Si!t
Utility Production
Da> Hater
treatment Sludga
Cooling lo»er
Sludge.
Cl Blo.do.n
treatment Sludge
i;
i!
II
IS
16
|7
o
1 >
1
II
i2
13
II
IS
It
1'
and 6IM treaties 13
•11 Sou'C.S
19
no
!
.
I found Dori/.n«?nl jl ton | Haste | Intent of 1 future | fraction of |
io'nroi "pinooo'OQy | • -- — • "| Deduction | Lurrent usf | Application | i&tai Haste |
1 Quantity 1 Quality I effectiveness | 1 Potential I I
Segregate aqueous and ally Hastes | ' 1 1 I 2 | 3 | 1 | I
Kcco.er oil fro* «fl tludge | 1 1 1 I 3 | 1 | 3 I I
Install 1 loll ing roof cjvei s 0.1 unit I 1 I 1 | 1 | 1 | 31 |
Use pressurised air In air 'lotatton | 1 | 1 | 1 | ? | ? | j
Oversue unit to «a« tludge rteoval I 0 ; 0 | Ij 21 1 i I
Crtoue flocculant usage/addition | 0 | 0 I 1 | 1 | 1 | |
Us* alternate cleaning techniques | 1 | 1 | 2 | 1 | 21 I
Use sludge in the production of coke I 1 | 1 | 1 | 1 | 2 I I
Better operating practices I 1 I 1 | 2 | 3 | 1 | 1
Overall I 0 71 1 0 11 I 1 1) I 1 »7 | 1 1C | 0 10 I
Irstell air coolers | 01 0 | 3 | 31 1 | |
Meat e»changer tube leak prevention I 0 I 0 | 21 2 I 1 \ \
Proper cooling lo.er oater treatwnt I ? | 21 2 i 3 I 1 I I
Ninimie cooling to.er duly 1 0| 0| 2 I 3| I I |
Die cleaner »ak»-up «ater I 01 0 I 2 | 21 2 I |
Use non-chronate vater treatment | 2| 2 | 3| l| 2 I |
Add Insulation to pipes, tanks, ect I 01 0 | 1 | 3 I 1 | I
Con.ert stea« to rebelled strippers 1 1 | 1 I 2 | 2 I 1 i 1
«t.nli« heat recovery | ) I 3 | 1 | 3 | 21 I
Getter operating practice I 2| 2 | 2| 2 | 21 |
Ove'-al! | 1 00 1 1 00 | 2 00 | 2 10 | 1 SO | 0 57 |
ill HetMxts I | 00 |
Current I future keduCt • Qf«
Keduct ton
-. . - —
Index | Probable I *
1 S
0 1
0 3
0 S
0 S
0 3
0 S
I 0
1 S
1 5
7 3
1 0
1 S
1 5
1
0
0
1
0
1
0 1 |
1 t |
0 6 |
0 ) |
0 1 1
0 2 |
0 1 |
' S 1
0 1 1
0 t |
0 2 |
0 5 1
0 1 |
0 1 |
0 5 |
1 1 |
0 1 |
0 3 |
0 1 1
0 5 |
2 3 | 0 1 |
J 2 | 05|
!r,de- I
— 1
1,1.,,,. |
4
1
1 7 |
1
1
1
1
1
I
i
1 7 |
1
1
1
1
1
1 1 |
1
,
1
1
1 1 1
' 2 I
C) these ttreais Include listed T and/or '«' WS« wjstfi
-------
operating practices" have also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practices: *
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Separation Process Wastes
These wastes include kerosene and lube oil filter clays. Some uses of clays inclusive
final treating of lube oil base stocks to improve color and remove acidic compounds,
copper removal from kerosene filtrate, and removal of BTX from catalytic reformate.
The following methods allowing for elimination or reduction of this waste stream were
noted:
o Replace clay filtration with hydrotreating.
Where a need to remove color bodies and olefins exists, hydrotreating is
replacing clay filtration as the method of choice. Hydrotreating does a
better job of improving color, increasing stability against oxidation, and
has no yield loss (the oil measured in the spent filter clay represents a loss
of some of the most valuable products produced in a refinery). It also
eliminates the problem of spent clay disposal.
B9-25
-------
o Regenerate and recycle spent clay.
Spent clays from the contact process may constitute one of the largest
waste streams from major refineries. At least one refinery .reported that
it was attempting to recycle this clay (Jacobs 1979). It is not known what
type of regeneration process was being used, but their previous attempts to
regenerate contact clay by burning the residual oil with air, or by using
roasting kilns or similar equipment, were not satisfactory. Close
temperature control is necessary for the clay to maintain its activity, and
with the variations in the amount of residual oil left on the clay, this
becomes extremely difficult. The Socony-Mobil Oil Company has
developed their Thermofor kiln for the regeneration of non-activated clay.
The process consists of washing the spent clay with naphtha to free it from
oil and render it mobile. It is then dried by steam heating and fed to the
clay burning kiln where it is regenerated by controlled combustion of the
absorbed impurities. The unit is quite complex and a considerable amount
of auxiliary equipment is required. The process has been in use for many
years on a batch basis, but has, as yet, never been run as a continuous
operation.
89-27
o Use fixed bed clay filtration process.
The fixed bed (percolation') process uses more clay for a given duty as
compared to the fluid bed (contact) process but the spent clays can be
regenerated many times using kiln. In addition, clays used in the fluid bed
process must be treated with sulfuric acid before use. The main advantage
of the fluid bed process, however, is that it has a neutralizing effect on
acid treated oils.
9.1.2 Conversion Process Wastes
Conversion process wastes include spent catalysts from FCC, hydrotreating and
hydrocracking. FCC catalyst is continuously regenerated by burning off the coke
formed on the catalyst during the cracking process. Some of the metal compounds in
the feed are adsorbed on the catalyst. In the catalytic cracking unit regenerator,
where coke is burned off the spent catalyst, the organic portion of these molecules is
burned off and the metals are oxidized to an inorganic oxide that remains on the
B9-26
-------
catalyst. Corrosion and erosion products may be mixed with the catalyst as fine
particles or may also be deposited on the catalyst surface. The heavy metals,
vanadium and nickel, and to a lesser extent, iron and copper, act as dfthydrogenation
catalysts and produce excessive quantities of undesirable coke and light gases
(especially hydrogen). In many cases, these metal contaminants are the primary
reason for discarding part of the equilibrium catalyst. Fresh (uncontaminated)
catalyst is then added to maintain a desirable level of contamination.
The flue gas from the regenerator passes through a series of cyclones that recover
most of the catalyst which is returned to the reactor vessel. Fines are recovered in
electrostatic precipitators. Operators try to control the operation such that the fines
production rate about equals the poisoning rate in order to avoid removal of
equilibrium catalyst. Suggestions for reducing FCC catalyst losses are as follows:
o Demetalize gas oil charged to cat cracker.
Catalyst withdrawal rate can be reduced by decreasing the rate of catalyst
deactivation by removal of metals before gas oil is charged to the reactor.
A mild hydrogenation or hydrotreating process has been used on the
catalytic cracking feed in some units. This feed treatment removes some
of the metal compounds and provides other benefits to the catalytic
cracking operation, such as removing sulfur, which reduces the amount of
sulfur emissions from the regenerator, and increasing yields of desirable
products such as gasoline.
o Minimize use of aeration and purge steam.
Excess fines will be generated by overuse of aeration and steam purges
during cracker operation.
Minimization of spent catalysts from hydrotreating and hydrocracking operation was
not addressed in detail. Generally, catalyst losses can be reduced by proper loading
and pre-sulfiding procedures, good reactor temperature control, good flow distribution
and continuation of research into development of more active, stable and selective
catalysts.
B9-27
-------
9.1.3 Upgrading Process Wastes
The principal wastes of concern are the sludges resulting from neutralisation of acidic
effluents from alkylation operation. Wastes associated with the alkylation process
primarily result from carryover of acid into the organic phase (oil) and carryover of
organic phase into acid. The carryover in both directions is attributed to physical
entrainment (physical carryover) or to the formation of undesirable compounds
between components of the organic phase and the acids (chemical carryover). The
following methods were noted as potentially useful in reducing alkylation wastes:
o Proper operation of the process.
By reducing the degree of physical and chemical carryover that occurs, a
reduction in caustic waste can be achieved. Physical carryover of acid can
be caused by using excessively cold acid, improper acid settler level, low
reactor pressure, and an excessive acid recycle rate. Chemical acid
carryover can be due to low isobutane or acid concentration, high reaction
temperature, and improper mixing. Improvements in operation of the
alkylation process were previously discussed in more detail (NPRA 1977,
Libberman 1980, Hammershaimb and Shah 1985).
o Use KOH to neutralize HF acid waste.
In the past, lime was used to neutralize acid carryover and other wastes
produced by the alkylation process directly. The newer process alternative
relies on potassium hydroxide (KOH) to neutralize acid; this is followed by
reconstitution of KOH by reacting KF with lime outside of the neutralizer
or scrubber. With HF alkylation, the switch to KOH for initial
neutralization has achieved a reduction in the overall amount of lime
required and, thus, in the amount of sludge produced. Additionally, the
overall system efficiency and reliability is improved over calcium or
sodium based solutions (Hammershaimb and Shah 1985).
o Upgrade feedstock by selective hydrogeneration or hydroisomerization.
Because of increased demands for higher octane gasoline and the increased
conversion of heavier crudes to lighter products, there has been a relative
decrease in the quality of alkylation feed throughout the industry
89-28
-------
(Hammershaimb and Shah 1985). By selectively hydrogenating diolefins
(acid polymer precursors) in the feed, acid consuming impurities can be
removed and hence overall acid consumption reduced. Hydroisomerization
processes are commercially available, e.g. HPNIVB process.
9.1.4 Auxiliary Process Wastes
Refinery gases that contain CO2, H2S, and traces of COS and C$2, are typically
scrubbed by amines to remove these components. The amines form certain non-
regenerable compounds which must be removed by intermittent bleed of amine
inventory. The following source control methods were noted:
o Choice of proper scrubbing solution.
Presently, monoethanol amine (MEA), diethanol amine (DEA), digly-
colamine (DGA), diisopropanolamine (DIPA), and methyldiethanolamine
(MDEA) are in use. There are some advantages and disadvantages
associated with any of these solvents. Typically, the waste generation
aspect is addressed in terms of reducing costly amine losses due to
formation of non-regenerable compounds. This aspect usually receives
close attention of a process designer. The ultimate choice of solvent
depends on specific application, therefore it is difficult to generalize which
solvent is most preferable from the waste generation standpoint. Recently
MDEA has been more frequently applied than in the past. The use of non-
amine gas treatment processes (e.g. Purisol, Selexol or Benfield) should be
carefully considered from the waste generation aspect as an alternative to
amines.
o Reduction of amine loss.
The loss of amine solution through degradation can lead to many problems.
Besides the actual loss of solution, degradation products are corrosive and
can lead to equipment fouling. Studies have shown that the rate of DEA
degradation increases with increasing temperature and pressure (Melsen
and Kennard 1982). Therefore, the rate of DEA flowing through the
regenerator reboiler should be kept high and the steam or gas temperature
kept low so as to lower the film temperature on the process side. Efforts
B9-29
-------
to remove the degradation products by means of carbon filtration have not
been very successful (while the appearance of the DEA improves from dark
brown to light yellow, no major degradation products were removed).
Also, increased pressure was shown to accelerate degradation process.
Examination of tradeoffs between higher amine losses and operating the
regenerator at atmospheric pressure with amine gas re-compression
appears worthy of consideration.
The use of additives to inhibit amine deterioration is in common use and
was found effective in the case of "Amine Guard", an additive
manufactured by Union Carbide.
Tail gases from the Glaus sulfur recovery plant are treated to remove residual sulfur
and its compounds (CS2, COS, SO2, H2S) in order to meet sulfur emission standards.
Among the processes used to treat the tail gases is the Beavon process which relies on
a catalytic reactor to convert all sulfur compounds to H2S and a Stretford section
where H2S is selectively scrubbed and oxidized to sulfur in an aqueous phase. To
avoid excessive buildup of non-regenerable impurities (chiefly thiosulfate), the
Stretford solution inventory is bled, thus generating a potentially hazardous waste
stream containing sodium salts of anthraquinone disulfonic acid (ADA), vanadium
compounds, sodium bisulfite, thiosulfate, carbonate, and bicarbonate. The following
techniques are considered useful to minimization of this waste stream.
o Prevent excessive deterioration of Stretford solution.
If the absorber is overloaded with inlet tail gas, excessive amounts of
thiosulfate and insoluble vanadium oxysulfide are formed. When a Claus
plant produces big swings in the sulfur content of the tail gas, overloading
of Stretford solution is likely, which in turn leads to excessive chemical
losses. Adding small quantities of sodium tartarate to the solution helps
prevent the formation of vanadium oxysulfide and hence reduces losses
(Kohn and Riesenfeld 1979). The initial degradation of 2,7-ADA during
startup can be inhibited by adding sodium thiosulfate to virgin solution
prior to startup.
B9-30
-------
o Alternatives to 2,7-ADA.
Since 2,7-ADA is not very stable, the use of more stable substitutes is
desireable. Proprietary formulations, such as "Elvada" and,"Amada", have
been offered as possible substitutes. Elvada contains a mercury catalyst
and thus may not be an environmentally advantageous substitute. Amada
does not contain mercury and appears to be a better choice.
o Recover products from the waste stream.
Two processes, one developed by Nittetsu Chemical Engineering Ltd.
(Japan) and the other by Peabody-Holmes Ltd., are available to recover
useful products from Stretford process wastes (Kohl and Riesenfeld 1979).
Both processes concentrate the waste stream by evaporation of water,
followed by combustion of the residue under reducing conditions, where
vanadium, sodium carbonate, and hydrogen sulfide are recovered to be
returned to the process. Organic compounds, such as ADA and other
additives, are destroyed in the combustion process. These processes are
thus capable of reducing wastes from the Stretford process.
o Alternatives to Stretford process.
An alternate process (Seiectox) for sulfur recovery from Claus tailgas was
developed and licensed by Ralph M. Parsons Co. and Union Oil Co. (Beavon
et. al. 1979). This sulfur removal process uses a cobalt-molybdenum
catalyst to hydrolyze and hydrogenate COS, CS2, SO2, and elemental sulfur
to H2S which is then catalyticaly oxidized to SO2 using air. The H^S and
SO2 then form elemental sulfur by the Claus reaction. Since no chemicals
other than the catalyst are used in the process, no solution purges are
necessary. The process is said to have lower capital investment and
produce a slightly better quality sulfur than the Stretford process. At least
one industrial unit (Wintershall AG, Lingen, West Germany) has been
reported to use this process. New promising developments include Unocal's
Unisulf process which is yet to be tested on commercial scale.
Minimization of spent catalyst wastes from the Claus plant, sulfuric acid plant and
steam reformer was not addressed.
B9-31
-------
9.1.5 Equipment Cleaning Wastes
Heat exchanger tube bundle cleaning sludge is a major contributor, to equipment
cleaning wastes. Often these wastes are discharged into a pit which is then flushed
out with water. This sludge/water waste flows into the chemical or oily water sewer
(OW5) and on into the API separator for primary treatment. The following suggestions
for waste reduction were noted:
o Decrease film temperature and increase turbulence on heat transfer
surfaces.
A decrease in film (or surface) temperature and an increase in turbulence
(or velocity) will decrease the asymptotic fouling resistance, thus reducing
the cleaning frequency. This could be accomplished by not oversizing the
heat exchangers excessively or by provision of recirculation of cooling fluid
in order to maintain high velocity (Garrett-Price 1985). On the cooling
water side, excessive film temperatures often result from low flows of
cooling water during turndown operations. Where possible, the feasibility
of temperature control using process stream bypass as opposed to direct
throttling of cooling water should be addressed.
o Reduce deposit precursors in process fluids and cooling water.
Calcium and magnesium salts present in crude oil contribute to scale
buildup and corrosion in crude preheat trains. Reducing the allowable salt
content by water injection upstream of desalters, operating the crude
desalters at maximum efficiency, and addition of emulsifier downstream of
desalter will minimize the exposure of the process side of the exchanger to
scale forming salts (Van der Wee and Tritsmans 1966). If economically
feasible to do so, a softened water (i.e., clarified boiler blowdown) can be
used as cooling tower makeup. The reduced level of calcium and
magnesium salts in cooling water would reduce the exposure of the tubes to
scale forming salts.
o Use proper corrosion inhibitors in cooling water.
Selection of the proper type and dosage of corrosion inhibitors will cut
down on scale formation. Typical corrosion inhibitors currently being used
include chromium, zinc, phosphates, and free chlorine (USEPA 1978). Since
B9-32
-------
these additives pose a potential pollution problem, they need to be removed
prior to discharge or their use avoided. The use of organic chelating agents
as replacements for zinc and chromium compounds is a via51e alternative
(Gesick 1974, Zecher 1975). The use of a reacted phosphate product
developed by Hercules, Inc., as a substitute for heavy metal inhibitors has
been reported (Wilkes and Model 1984). Wites Chemical Co., reports the
use of a non-oxidizing biocide as a similar substitute (Hoblack and Kawlor
1985). A proprietary nonchromate treatment, Dianodic II, was found to be
effective in controlling scale formation at National Cooperative Refinery
Association, McPherson, Kansas (Nichols et al. 1980).
Substitute air coolers or electric heaters for heat exchanger.
Reduction in sludge generation due to bundle cleaning on utility side can be
accomplished by replacing heat exchangers with air coolers or electric
heaters. This should be done if economically sound.
Use smooth heat exchanger tube surfaces or on-stream cleaning devices.
Providing a smooth surface for heat transfer will minimize the sites
available for scale formation to start. The use of electropolished stainless
steel tubes in a black liquor forced circulation evaporator (paper and pulp
industry) resulted in a drastic reduction in cleaning frequency from once-a-
week to once-a-year (Uddeholm Co., Tubec Tubes Brochure). Smooth non-
stick surfaces can also be provided by Teflon*. Complete Teflon
exchanger designs are available (Dupont 1985). The additional advantage
of Teflon tubes is their ease of cleaning and corrosion resistance.
The onstream cleaning devices generally rely on an insert propelled by the
process fluid through the heat exchanger tube. Brushes propelled back and
forth by flow reversal accomplished with a diverter (Water Services of
American Inc.) or sponge balls recirculated by a special
retrieval/reinjection device (Amertap Inc.) are both commercially available
onstream cleaning systems.
Registered trademark of E.I. DuPont de Nemours & Co.
B9-33
-------
Sludge due to cleaning of crude oil, leaded gasoline, non-leaded gasoline, and other
product storage tanks is another major equipment cleaning waste. The following
methods for reducing the generation of sludge were noted: «
o Reduction of lead in gasoline.
The current trend of reducing the level of lead in gasoline, coupled with
the possibility of a ban on leaded gasoline, will lower the toxicity of leaded
gasoline tank sludges.
o Install storage tank agitators on crude oil tanks.
Agitation of the vessel contents will prevent the deposit of settleable
solids and hence reduce the need for cleaning. It must be noted that
agitation does not by itself reduce the amount of waste generated; it
simply transfers the solids downstream to the crude unit - they eventually
end-up either in asphalt or coke.
o Use corrosion resistant materials.
Some of the sludge generated is the result of corrosion or deterioration of
the storage tank internals. Installation of a liner or using materials of
construction which are more resistant to the corrosive elements of crude
oil will reduce sludge production.
o Prevent oxidation of crude oil.
Gums and resins form as the result of air oxidizing unsaturated compounds
in the crude oil. Air oxidation can be minimized in crude storage tanks by
providing a nitrogen blanket over the surface of the oil or, more commonly,
by use of floating roofs. The floating roof should preferably be of the
double cover type with liquid seals. Many storage tanks are already
equipped with floating roofs as a result of air emission regulations.
o Refine or recycle sludge for organic content.
Conventional cleaning of crude tanks often relies on mechanical removal of
sludge and its subsequent land disposal. The use of warm oil and
circulating cleaning techniques reduces the ultimate sludge volume and
allows for recovery of considerable quantities of valuable crude tied up in
the sludge (Barnett 1980). Here, a light gas oil, clean crude stock, or other
B9-3A
-------
available low viscosity straight run distillate is warmed, mixed with
dispersant additive, and circulated through the tank. This process re-
suspends the sludge and dissolves the crude that is entrapped in the sludge,
which amounts to 60 to 90 percent of oil by volume. The liquid can then be
sent to the refinery's slop oil recovery system. Again, as in case of
agitation, the solids are transfered to the part of the process where they
can be more efficiently separated from valuable entrained liquid.
The sludge volume can be reduced at small refineries by using the Victor
extraction process (API-NPRA Conference 1980). Here, the sludge is
dumped into a container and agitated for an extended period of time with
steam and air. This process separates the residual oil trapped in the sludge
which can then be taken off from the top of the container. The solids,
significantly smaller in volume by comparison to the original sludge, settle
to the bottom. This process works well for granular type sludges but is not
very effective for clay type sludges.
At large refineries, the sludge volume can be reduced by concentrating it
using vacuum, gravity belt, or automated plate and frame filters
(centrifugal filters were not found to be very effective). Also, the sludge
volume can be reduced significantly by using patented Chevron Recovered
Oil Process. This process is said to be more effective than conventional
filtration methods (API-NPRA Conference 1980). Other methods involve
thermal/chemical emulsion breakup using steam or indirect heating. One
such method was used at Vickers Petroleum refinery where the de-
emulsifier was added to the sludge, which was then pumped through a
steam heater into a conical bottom decanter (op. cit.). Other techniques
include ultrasonic de-emulsification, solvent extraction (e.g. using B.E.S.T.
process available from Resources Conservation Co.), and recently patented
electroacoustic dewatering (Battelle 1985).
9.1.6 Wastewater Treatment Wastes
Oily aqueous wastes which originate in the refinery are collected in chemical or oily-
water sewers (OW5) which usually discharge to an API separator. The feed to the
B9-35
-------
separator can vary widely in both flowrate and composition. The wastes are usually
comingled and consist of oil, water, suspended solids, and various emulsions. The
function of the separator is to perform the initial separation of solids from liquids and
oil from water. Virtually all dilute aqueous wastes from the refinery will pass through
the API separator.
Skimmed oil from the API separators is usually pumped to the slop oil tank where the
mixture is separated into three phases: oil, water, and emulsion. The oil is returned
for reprocessing, and the water is recycled to the API separator. The emulsion may be
disposed as a waste or may be processed further for recovery of oil.
Dissolved air flotation (DAF) units are usually installed after the API separator.
Water which is discharged from the API separator contains emulsified oil, water, and
suspended solids. Suspended solids are removed in the DAP unit either by flocculating
agents and settling, or by air coming out of solution and carrying the particles to the
surface along with the oil. The froth "float" which contains emulsified oil, water, and
solids is skimmed off and sent to a de-emulsifier.
Basically, the problem of reducing API sludge volume can be approached in two ways:
Reduce the raw waste loads upstream of the API separator.
Maximize recovery of oil from the sludge before disposal.
Raw oil-containing wastes entering the API separator can originate from heat
exchanger cleaning operations, crude tank cleaning operations, equipment and piping
drainage and steamouts prior to maintenance, maintenance parts cleaning, and spill
cleanup. Generally, most methods mentioned in Section 9.1.5 apply. Continuous
sources of aqueous streams entering the API separator include sour water stripper,
desalter, boiler blowdown, cooling tower blowdown, BFW treatment (ion-exchange
regeneration brine) effluent, and other refinery units. Source control techniques
allowing for a decrease in the carryover of oil are addressed in the appropriate
sections. Methods proposed for reducing waste associated with waste water treatment
are as follows:
69-36
-------
Reduce API separator inflow by waste segregation.
A major portion of the API separator load is water that originally may or
may not have been an oily waste. Efforts should be mafle to install a
separate sewer system which would convey oil-free aqueous wastes. Boiler
blowdown, some pump seal water, ion exchanger regeneration brines, and
other non-oil containing wastes should be diverted to this system. These
streams would then by-pass the API separator, thereby reducing its load
and thus improve separation efficiency in the existing separators.
Recover oil from API sludge.
Most techniques discussed in Section 9.1.5 under "refining or recycling of
sludge for organic content" for crude tank sludges also apply to API sludge.
These methods include filtering, thermal/chemical emulsion breakup and,
possibly, ultrasonic separation of oil from solids. Solvent extraction was
found to have high effectiveness and application potential (API comments).
Install floating roof covers on API separators.
At Conoco Inc., installation of floating covers on API separators was found
to reduce the oxidation of oil that results in the formation of heavy waste
material (API-NPRA Conference 1980). Floating roofs were installed
originally to reduce air emissions.
Use pressurized air technology.
DAF units using pressurized air generate less than one half of the float
volume compared to induced air units for a given amount of solids removal.
New facilities should incorporate this technology in the design if
technically and economically compatible with the overall refinery plan.
Facilities currently using induced air systems should convert to pressurized
air systems if the economics justify this change-over.
Use sufficient overdesign factor for new systems.
With oversized systems, the impact of process upsets and flow excursions
will be minimized. Reducing the impact of process upsets and flow
excursions leads to much more efficient and reliable operation of the
individual waste treatment units.
B9-37
-------
o Use proper flocculants.
Proper matching of flocculant to refinery waste will increase the
separation efficiency of the process and reduce the amount of, waste that is
carried downstream.
o Use alternate cleaning techniques.
All of the methods that reduce waste due to equipment cleaning (Section
9.1.5) will reduce waste associated with wastewater treatment.
Additionally, removal of sludge prior to its discharge to API separator may
reduce organic carryover into water and reduce the resulting biosludge. An
example is an alternate design of a heat exchanger bundle cleanup pit to
recirculate water through filter to remove sludge, prior to discharge into
API separator (NMERDI 1985).
o Use sludge in the production of coke.
For facilities with cokers, most of the sludge produced by the API
separator, the DAF unit, and the slop oil tank can be injected into the
coker and converted to coke (API comments).
o Better operating practices.
These include minimization of spills of process fluids and reduction of the
practice of hosing refinery wastes into the chemical sewer. Because spent
filter clays have high oil adsorption qualities it makes them ideally suited
for use on in-plant oil spills. The clays are typically from the filtration of
light-end distillates and lube oils. Rather than immediately disposing of
this spent clay, the refineries stockpile them within the diked areas around
the refinery until used. Stockpiling of the clay also allows, the hydro-
carbons to degrade by microbial action. For additional methods, the reader
is refered to a separate study of "good operating practices" contained in
this appendix (Study #819).
9.1.7 Utility Production Wastes
Sludge that settles in the cooling tower basin is removed whenever the cooling tower is
out of operation. It is either washed into the process sewer system or shoveled out and
R9-38
-------
disposed to a landfill. The sludge sources are residues from cooling tower biocides,
hardness scale from exchanger surfaces, cross contamination from process fluids,
biological materials, and airborne dirt. Suggestions for sludge volume reduction are as
follows:
o Install air coolers.
Replacing water-cooled exchangers, where practical, with air fan coolers
will reduce contamination of cooling water with process fluid as well as the
volume of cooling water in circulation.
o Heat exchanger tube leak prevention.
Cross contamination from the process side of heat transfer equipment is
one of the sources of sludge creating materials. Part of this cross
contamination is created by leaky exchanger tubes. Use of seal welded
tube joints in some cases, or double tubesheets will minimize or eliminate
process fluid leakage into the cooling water (and vice-versa). Tube
vibration analysis should be performed more routinely in the design stage.
o Proper cooling tower water treatment.
Cooling tower chemicals are responsible for sludge buildup. Operators
should refrain from overtreatment to avoid excess buildup due to chemical
addition. The heavy metal content in the cooling tower sludge can be
reduced or eliminated by methods such as the nonchromate treatment
discussed in section 9.1.5.
The cooling tower blowdown treatment sludge can be minimized by the following
methods:
o Minimize cooling tower duty
This is closely related to energy conservation in the refinery. The waste
(low pressure) stream, e.g. from leaky steam traps, is condensed using
cooling water which increases cooling tower duty. Reducing steam
consumption helps to reduce cooling tower blowdown.
o Use of a cleaner make-up water
This will reduce the blowdown stream and, thus, the toxic sludge generated
by chromate removal. Use of boiler blowdown as a make-up stream should
be considered. B9-39
-------
o Use of non-chromate treatment
As mentioned in Section 9.1.5, this will reduce the toxicity of the sludge
produced by treating the cooling tower blowdown. The* non-chromate
corrosion inhibitors are reliable, except in services where water film
temperatures are very high - in such cases the design, operation, and
controls of exchanger should be carefully re-examined anyhow.
Most refineries use a vast amount of steam in normal operation. Feedwater to the
boilers, particularly the high pressure boilers, must be very clean for safety reasons as
well as equipment longevity. Any projects undertaken by the refinery which would
seek to cut steam consumption would not only reduce spent lime production, but also
result in operating cost savings. Some suggestions for reducing steam consumption are
as follows:
o Add insulation to pipes, tanks, valves, etc.
Decreasing heat losses from pipes, tanks, valves and other process units
will reduce steam consumption. This, in turn, will reduce the load on the
boiler feedwater treatment system.
o Convert steam strippers to reboiled strippers.
All stripping steam eventually winds up in the waste treatment system. 3y
converting to reboiled strippers where possible, the spent lime required in
the treatment of the water used to produce stripping steam as well as the
wastewater load to API separator will be eliminated.
o Maximize heat recovery from process.streams.
Identify those areas in the refinery for installation of additional heat
exchanger networks to make maximum use of available heat energy.
o Better operating practices.
Many refineries have areas where steam demand may be reduced. It is to
the advantage of the refiner to promote the identification of areas and
develop projects to increase the overall plant thermal efficiency. Promote
an aggressive approach to steam system preventive maintenance. Maintain
steam traps, valves and piping runs to minimize steam or condensate leaks.
Provide strict guidelines to limit unneeded blowdown of boilers or
B9-AO
-------
automate it. Provide rigorous maintenance of the condensate return
systems and boiler feedwater treatment systems to minimize contaminants
which enter into the boiler feedwater. *
9.2 Implementation Profile
Good operating practices, while easy to implement, do not appear to have as high an
impact on waste minimization as do process modifications, such as replacement of
clay filtration with hydrotreating or recovery of oil from sludges. Process modifica-
tions are expected to be considered in the planning stages for unit replacement, major
overhaul, or during the conceptual design stage for any new grass-roots facility.
Considerable amount of engineering and economic analysis is usually required before a
decision can be made because of the high capital costs involved.
9.3 Summary
The ratings of every listed method shown in the table were based on review of
available literature, engineering judgement and input from the American Petroleum
Institute. Each method was rated on a scale of zero to four for its effectiveness,
extent of current use and future application potential. The current and future extents
of waste reduction were then derived using the methodology presented in the
introduction to this appendix.
A current reduction index of 2.2 (55 percent) is indicative of the significant level to
which the noted wastes have already been minimized (CRI is a measure of the
reduction of waste that would have been generated if none of the methods were
practiced at their current level of application). By implementing additional waste
reduction measures, or by increasing the use of existing measures, the amount of
waste currently being generated can be reduced to a moderate extent, as evidenced by
a future reduction index of 0.5 to 1.2 (12 to 30 percent).
The most effective measures currently applied to control hazardous (RCRA listed)
wastes include segregation of aqueous and oily wastes, oil recovery from sludges, re-
use of sludge as a coker feedstock and better operating practices.
B9-41
-------
The most effective future methods for minimizing hazardous wastes are those
characterized by high individual FRI value. These include oil recovery from sludges
and use of sludge as a coker feedstock.
10. PRODUCT SUBSTITUTION ALTERNATIVES
At this time there are no economic substitutes for petroleum products. Alternative
sources of energy such as solar, nuclear, coal or shale oil cannot currently compete
with petroleum to satisfy the energy needs at the level required in the United States.
Until the world supplies of crude oil are exhausted, or oil becomes unavailable for
other reasons, it is unlikely that it will be economically feasible for the alternatives
discussed above to significantly displace the refining of crude oil in the near future.
One area in which product substitution has been occurring is with the product gasoline.
Currently, there have been major efforts being undertaken to reduce and eventually
eliminate the use of lead in gasoline. Producing a non-leaded gasoline requires
additional processing to ensure the same level of octane as in leaded gasoline. Most
commonly, alkylation is employed to boost octane levels so that more acid wastes are
expected to be produced per gallon of gasoline sold. Offsetting this increase however,
is the elimination of wastes due to tetra-ethyl lead production, leaded gasoline storage
tank sludge, and the emission of lead into the environment by way of combustion.
Eliminating the use of lead in gasoline is expected to achieve a large reduction in the
amount of toxic metal contained in a refineries waste.
Another alternative which is worth mentioning is methanol which can be produced
from petroleum and which can be successfully used in place of gasoline (Othmer 1985).
Cited environmental advantages include low fire hazard, absence of carcinogenic BTX
compounds present in unleaded gasoline, and very low exhaust emissions of hydro-
carbons, carbon monoxide, nitrogen oxides and aldehydes.
11. CONCLUSIONS
While the petroleum refining industry has significantly reduced its waste volume
generation, it appears that further reduction is possible. Our estimates indicate that
it may be technically feasible to further reduce the volumes in the range of 12 to 30
B9-42
-------
percent. However, any further voluntary reduction will only be realized from in-plant
studies which show that changes, such as the ones suggested in this report, are
economical. *
12. REFERENCES
Anonymous, 1981. Novel solvent recovery enhances residuum upgrading. Chemical
Engineering, 88 (24): 69.
Anonymous, 1984. Dimersol, Hydrocarbon Processing, 63 (9):125.
API-NPRA Conference, 1980, American Petroleum Institute and National Petroleum
Refiners Association. Solid waste practices under RCRA in the hydrocarbon
processing industry, pp. 133-174. Dallas: API-NPRA.
Barnett, J.W., 1980. Better ways to clean crude storage tanks and desalters.
Hydrocarbon Processing. 60 (l):82-86.
Battele, 1985. U.S. Patent No. 4,561,953, issued to Batelle, Columbus Division,
Columbus, OH.
Beavon, D.J., Haas, R.H., and Muke, B. 1979. High Recovery, Lower Emissions
Promised for Claus-Plant Tail Gas. Oil and Gas Journal. 77(3):76-80.
Dupont, 1985. Chemical Processing. 48 (6):42-3.
Garret-Price B.A., et. al 1985. Fouling of heat exchangers; characteristics, costs,
prevention and removal. 1st ed. Noyes Publications Park Ridge, N.J.
Gesick, J.A. 1974. A comparative study of non-chromate cooling water corrosion
inhibitors. Presented at the 35th Annual International Water Conference of the
Engineers' Society of Western Pennsylvania.
Hammershaimb, H.U. and Shah, B.R. 1985. Trends in HP alkylation. Hydrocarbon
Processing, 64(6):73-6.
Hoblack, R. and Kawlor, L. 1985. Chemical Processing. 48(8):84-6.
Jacobs 1976, Jacobs Engineering Co. Assessment of hazardous waste practices in the
petroleum refining industry. EPA-SW-129C. Washington, D.C.: U.S. Environmental
Protection Agency.
1979, Alternatives for hazardous waste management in the petroleum refining
industry. EPA-530-SW-172C. Washington, D.C.: U.S. Environmental Protection
Agency.
Jahnig, C.E. 1982. Petroleum refinery processes, survey. In Kirk-Othmer
Encyclopedia of Chemical Technology. 3rd ed. Vol. 17, pp. 183-256. New York, NY:
Wiley.
Kohl, L.A., and Riesenfeld, F.D. 1979. Gas purification. 3rd ed. pp. 28-90 and pp.
476-487. Houston, TX: Gulf Publishing Company.
B9-43
-------
Libberman, N. 1980. Basic decision key to alky problems. Oil and Gas Journal. 78
(June 23): 141-4.
Melsen, A. and Kennard, M.L. 1982. DEA degradation mechanism. Hydrocarbon
Processing, 61(10):105-8.
Nichols, J.D., Clavin, J.S., and Blasdel, J.E. 1980. Nonchromate treatment performs
well. Hydrocarbon Processing, 59(lQ):75-8.
NMERDI 1985. New Mexico Energy Research and Development Institute. Water use,
conservation and wastewater treatment alternatives for oil refineries in New Mexico.
Prepared by Jacobs Engineering Group Inc. Albuquerque, NM.
NPRA. 1977. National Petroleum Refiners Association Q&A-5. Experience with alky
units traded. Oil and Gas Journal. 75 (May 23):66-75.
Othmer, D.F. 1985, Methanol: fuel for automobiles, Chemical Engineering Progress.
Robitaille, D.R., and Bilek, J.G. 1976. Molybdate cooling water treatments.
Chemical Engineering, 83(27): 77-82.
Shreve, R.N. 1967. Chemical process industries. 3rd ed. New York, NY: McGraw-
Hill Book Co.
Uddeholm Corporation (Sweden), Technical Brochure on Tubec Tubes.
USCD, 1985. U.S. Department of Commerce, Bureau of the Census. Petroleum
refining. In 1982 Census of Manufacturers. MC-82-I-28F. Washington, D.C.: U.S.
Governmental Printing Office.
USEPA, 1978. U.S. Environmental Protection Agency, Effluent Guidelines Division.
Draft development document for effluent limitations guidelines (batea), new source
performance standards, and pretreatment standards for the petroleum refining point
source category. Washington D.C.: U.S. Environmental Protection Agency.
Van der Wee, P., Tritsmans, P. A. 1966. Crude limit preheat exchanger fouling
Hydrocarbon Processing, 45(8):141-4
Van Matre, J. 1977. Clean heat exchange equipment on-stream. Hydrocarbon
Processing, 56(7):115-7.
Wilkes, T. and Hodel, A.E. 1984. Non-heavy metal inhibitor protects at pH 6.5-7.0.
Chemical Processing, 47(9):38-9.
Zecher, D.C. 1975. Corrosion inhibition by surface active chelants. The International
Corrosion Forum, Toronto, Canada.
13. INDUSTRY CONTACTS
P.J. Tussey, American Petroleum Institute, Washington, D.C.
B9-44
-------
1. PROCESS: PHENOLIC RESINS MANUFACTURE
2. SIC CODE: 2821(8)
3. INDUSTRY DESCRIPTION
Manufacturers of phenolic resins are included as part of the plastic materials and
resins industry. Large phenolic resin producers tend to produce the resins for sale to
other users. Smaller producers tend to produce resins for captive use.
3.1 Company Size Distribution
Relative to many other chemical producing industries, the phenolic resin
manufacturing industry contains a large number of producers. This may be because
the manufacturing processes involved are not highly capital intensive and because of
the large variety of final products.
In 1980, there were 50 companies producing phenolic resins at 102 establishments. Of
the 50 companies, 6 produced more than 100 million pounds of phenolic resins during
the year, 14 produced between 10 and 100 million pounds, and 31 produced less than 10
million pounds. Each establishment produced an average of 30 million pounds of
phenolic resins in 1980. The employment figures by EPA regions for the phenolic resin
industry were not available.
3.2 Principal Producers
The following principal producers manufacture in excess of 100 million pounds of
phenolic resins annually:
Borden Incorporated Occidental Petroleum Corporation
Georgia-Pacific Corporation Owens-Corning Fiberglass Corporation
Monsanto Company Reichhold Chemicals Corporation
B10-1
-------
3.3 Geographical Distribution
Table 3-1 displays the number of phenolic resin manufacturing establishments by EPA
Region. Region V contains 28 establishments, or over 25 percent of the establishments
in the U.S. However, relative to other chemical manufacturing industries, the
phenolic resins industry is geographically widespread. Figure 3-1 displays the
geographical distribution of phenolic resin facilities located in the U.S.
Table 3-1 Geographical Distribution of Phenolic Resins Manufacturing
Establishments by EPA Region
EPA Region
I
II
III
IV
V
VI
VII
VIII
IX
X
National
Number of Establishments
5
14
5
15
28
10
3
1
6
15
102
A. PRODUCTS AND THEIR USE
The binding quality of phenolic resins make them very versatile. Thus, they are used
in a wide variety of products including:
Molding Compounds Plywood
Insulation Fibrous or Granulated Wood
Foundry Binders Laminates
B10-2
-------
o
I
PO
0
Figure
I 2-5
d 6-10
Roman numerals show EPA regions
3-1 Phenolic Resins Plants in the U.S.
-------
Brake Linings
Abrasives
Protective Coatings
Clutch Facings
Rubber
For 1985, phenolic resin production was estimated to be 2.5 billion pounds (Greek
1985). Table 4-1 gives the use pattern for phenolic resins.
Table 4-1 1985 Use Pattern of Phenolic Resins
Use
Consumption
Bonding plywood and wood fiber board
Other adhesives
Molded items
All others
60%
28%
10%
2%
Source: Chemical and Engineering News (Greek 1985).
5. RAW MATERIALS
Reactants Formaldehyde (solution or solid), phenol, calcium hydroxide, sulfuric
acid, paraformaldehyde, hexamethyl tetramine, sodium hydroxide,
substituted phenols.
Solvents Cellosolve acetate, butanol, ethanol, methyl ethyl ketone,
cyclohexanone, xylene.
Additives Wood flour, amino resins, oils, plasticizers.
R10-4
-------
6. PROCESS DESCRIPTION
Phenolic resins are a broad variety of materials, all of which are based upon the
reaction between phenol, or a substituted phenol, such as cresol or resorcinol, and an
aldehyde, such as formaldehyde or acetaldehyde. Most significant resins are based
upon the reaction of phenol with formaldehyde. By proper control of the reaction, a
wide variety of phenol-formaldehyde resins can be manufactured.
In general, phenolic resins are produced batchwise in a batch kettle arrangement
schematically depicted in Figure 6-1. The batch reactor typically has a size varying
from 2,000 to 10,000 gallons. The reactor is equipped with a jacket to facilitate
cooling or heating at different phases of reaction. The following sections deal with
the manufacture of two principal types of phenolic resins: Resols and Novolacs.
6.1 Manufacture of Resols
Resols are thermoset phenolic resins used for making bonding resins, varnishes, or
thermosetting molding powders. Resols are formed by the condensation of phenol and
formaldehyde present in a molecular ratio of 1.0:1.5. An alkali, such as sodium
hydroxide, is used as a catalyst.
Phenol, in molten form, is charged to the kettle followed by addition of 37 percent
formaldehyde solution which washes away any residual phenol present in the lines
leading to the kettle. After the catalyst is charged to the reactor, the reactor
contents are heated to about 1AO°F by steam to start the condensation reaction. Once
initiated, the reaction generates heat, which is removed by the water-cooled kettle
jacket and a water-cooled overhead condenser. These are used to maintain the
temperature at 140-175°F for 1 to 5 hours, depending on the required properties of the
final product. The reaction is stopped at the desired point by cooling the reactor
contents to 95°F. At this point, the caustic may be neutralized by the addition of
sulfuric acid. The reactor contents are then heated to remove water and other
volatiles from the resin. The removed water contains unreacted phenol, formaldehyde,
and low molecular weight resins. This stream can be processed in a variety of ways
B10-5
-------
FORMALDEHYDE
PHENOL
RECEIVER
CRUSHINI
ANfi
•RIHDINS
SOLID
PRODUCT
RESIN
LUC/ID
PRODUCT
RESIN
I
! PROCESS HASTE CATEIORIES:
| 0 OFF-SHADE PRODUCTS
: 0 RECEIVER CONTENTS
! _
3} VASTEVATER
7) EQUIPMENT CLEANINS HASTES
Figure 6- i Process Flo* Diagrai for the Manufacture of Phenolic Resins
R10-6
-------
including re-use in compatible batches, phenol recovery followed by treatment,
dilution followed by treatment, incineration and land disposal.
The product resin may be washed several times, resulting in a waste water stream.
The solid products are dried and ground to a specific size prior to packaging. The
liquid products are filtered prior to packaging. This filtration produces a solid waste
of oversize resin particles which is disposed of by landfill or incineration.
By proper control of the reaction, three classes of products can be produced. If the
reaction is stopped when the polymer is still water-soluble, the product can be used for
bonding resins. If the reaction proceeds till the polymer precipitates, the product can
be used to manufacture a varnish. If the reaction is carried out till the resin reaches
the brittle stage, "one-stage" solid resin is produced and can be used as molding
powder. In the last case, the molten resin must be quickly discharged in order to
prevent solidification in the reactor.
6.2 Manufacture of Novolacs
The manufacture of Novolacs is similar to that of resols except that an acid such as
sulfuric acid is used as a catalyst. The ratio of phenol to formaldehyde is normally
1.0:0.75-0.9. The deficiency of formaldehyde produces a stable, low molecular weight,
thermoplastic.
Vapors generated by the heat of reaction are condensed and returned to the reactor in
order to maintain the reaction temperature at 185-195°F. After 3 to 6 hours of
reaction, the condensate is sent to a receiver instead of being returned to the reactor.
When the temperature rises to 250-300°F, vacuum is applied to remove water and
unreacted phenol. At this point, the receiver contains water, phenol, formaldehyde
and methanol, which is used as a stabilizer for formaldehyde. This stream can be
processed in a variety of ways including re-use in compatible batches, phenol recovery
followed by treatment, dilution followed by treatment, incineration and land disposal.
The reactor contents need not be cooled as the polymerization is complete. The final
product is purified by dehydration under vacuum before being dumped on cooling pans.
B10-7
-------
The reactors are periodically cleaned with caustic solution, generating an additional
waste stream.
7. WASTE DESCRIPTION
Table 7-1 shows the primary waste streams associated with the manufacture of
phenolic resins. The off-grade products, which amount to 0.8 percent of total resin
production (Snell 1978), are the result of bad batches. This waste stream contains
unreacted phenols, formaldehyde, methanol, and resins at various degrees of
polymerization. This waste may be drummed and landfilled or stored onsite in
concrete-lined lagoons. The material can also be incinerated using a fluid-bed
combustor (Lanouette 1977).
Losses from grinding, screening, packaging, and filtering are routinely collected and
reworked at most of the facilities. In a few cases, these handling losses generate a
waste stream which is disposed of along with the off-grade products.
The liquid in the condensate receiver amounts to 67 percent of total resin production
(Snell 1978). However, according to more recent information, this stream constitutes
30 percent of total resin production*. The stream is generated from product
dehydration in the case of resols, or in the case of Novolacs, the water is removed in
the final phase of reaction to accelerate polycondensation. This stream contains
water, phenol, formaldehyde, methanol, and low molecular weight resins. The stream
may be land disposed or may undergo recovery of phenol (hence producing a less toxic
and more easily treatable wastestream), controlled incineration, chemical or biological
oxidative treatment.
A wastewater stream, which amounts to 8 percent of total resin production (Snell
1978), is also generated when the final resin product is washed to remove traces of the
unreacted phenol, formaldehyde, and salts produced by neutralization of the reaction
mixture. This stream is usually filtered to remove any solids and then discharged to
the sewer, or combined with the receiver effluents, or treated separately using
chemical or biological oxidation, depending on the final organics content.
*Reichhold Chemicals Inc. 1985: Private communication.
B10-8
-------
Table 7-1 Phenolic Resins Manufacturing Wastes
2
o
I
No.
1.
2.
3.
Waste
Description
Off-grade
products
Receiver
contents
Wash water
waste
Process Origin
Grinding, screening
and packaging.
Filtration of the
liquid resin
Reaction water with
organics
Washing and/or neu-
tralization of the
Composition
Unreacted phenols,
low molecular weight resins
5-7 wt.% phenol
5-8 wt.% formaldehyde
5-7 wt.% methanol
Unreacted
phenol and salts
RCRA Codes
—
4.
5.
Spills
Equipment
cleaning
waste
product resin
Reactor, storage
tanks, valves,
pumps
Washing Novolak
reactor
Phenol and
formaldehyde, etc.
Na2SO4 and NaOH
in water
-------
Spills and leaks are inadvertent discharges occurring at various plant locations. The
liquid discharges are usually mixed with the receiver contents for disposal. It is
expected that the waste resulting from spills and leaks will be small.
A caustic solution is used to wash the reactor for the manufacture of novolacs,
generating a waste stream. This stream contains sodium sulfate, sodium hydroxide,
and traces of phenol and formaldehyde in water. A similar, but acidic waste stream
from resol manufacture is mixed with the receiver contents and disposed of as
described previously.
8. WASTE GENERATION RATES
The recent nationwide waste generation rates from phenolic resins manufacturing
process were not in evidence at the time of final document preparation. The most
recent estimates found through literature search give fractional waste generation
rates for 1974 (Snell 1978). Since the phenolic resin industry has made many process
improvements in the last decade, the current waste generation figures are expected to
be much lower than the 1974 figures. The 1974 estimates correspond to about 0.6
pounds of waste per pound of product, whereas current waste generation is estimated
at less than 0.2 pounds of waste per pound of product*. The relative waste generation
rates expressed as weight fractions of the total waste stream were computed based
upon the 1974 data (Snell 1978) and other available, more current information. These
relative rates are given in Table 8-1 and also in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The summary of all the waste streams together with source reduction methods is
shown in Table 9-1. The discussion of each method is given below along with the
rationale for its inclusion and application cases, if known.
Reichhold Chemicals 1985: Private communication.
B10-10
-------
Table 8-1 Fractional Waste Generation Rates from Phenolic Resins Manufacturing
Waste
Off-grade Products
Receiver Contents
Resin Wash-water
Spills
Equipment cleaning wastes
Total
Percent of Total(a)
(excludes added water)
1
64
15
1
19
100
Percent of TotaKb)
(includes added water)
> 1
10
36
> 1
52
100
Source: estimated by project staff.
^a' Water used for resin wash and equipment cleaning is excluded; only the organics
and salts present in these streams are counted.
(°) Includes water added for resin wash anrd equipment cleaning.
R10-11
-------
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" are defined as being procedural or institutional policies which
result in a reduction of waste. The following items highlight the scope of good
operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
preventive manintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Off-Grade Products
Off-grade products containing traces of unreacted phenol and formaldehyde are
obtained from bad batches and are usually landfilled offsite or stored on-site in
concrete lined lagoons. The following source reduction methods were noted:
o Proper control of reactor temperature.
In most cases, bad batches are the result of the reactor contents being
cooled too rapidly. This leads to the formation of an insoluble, infusible
mass. This can be prevented by proper control of the reactor temperature,
along with prompt attention to any rapid decreases in temperature.
Temperature excursions can be minimized by:
B10-12
-------
Frequent calibration checks of temperature sensors, and more
frequent maintenance of temperature control systems.
Frequent monitoring of set point drift.
Use of reactors with separate jackets or coils for cooling water and
steam. Such reactors are commercially available and will avoid the
deposition of dissolved cooling water solids on the heat transfer
surfaces when steam is applied. Buildup of these deposits leads to
inefficient heat transfer and subsequently poor temperature control.
o Increased automation.
Bad batches may also result from failure to charge proper quantities of
various reactants. This can be prevented by double-checking the quantity
of input materials to the reactor. Usually, feed materials are weighed
before being sent to the reactor. Weighing errors can be decreased by
using an automated batching system properly maintained by frequent
calibration. If installation of automated batching controls proves
impractical, manual batching can be improved by closer supervision, better
documentation procedures, better operator training, and frequent checks of
weighing equipment.
o Re-use of off-grade products.
Off-specification products can sometimes be blended with high-grade
product to yield an intermediate-grade resin which meets the required
specification. This practice is not always possible; however, it appears to
be widely used in the industry.
9.1.2 Receiver Contents
In resols manufacture, water is removed from the reactor to favor the poly-
condensation reaction. This water stream carries phenol, formaldehyde, and other
soluble organics and is stored in drums for land disposal, or treated to recover phenol.
In Novolaks manufacture, the reactor contents are maintained at constant
temperature in the initial phases by condensing water vapor and returning the
condensate to the kettle. Late in the reaction, the slowing polymerization rate is
810-13
-------
increased by diverting the condensate to a receiver rather than to the reactor. The
receiver contents are sent to waste storage. The following source reduction
techniques were noted:
o Use of improved catalyst.
The phenol and formaldehyde content in the receiver effluent can be
minimized by obtaining higher conversion yields for the polycondensation
reaction. This can be facilitated by use of more active catalysts (Brode
1982). One patented catalyst which is a combination of a divalent metal
and an acidic material, is an alternative to presently used catalysts
(Culbertson 1978). The use of improved catalysts should be examined for
their environmental advantages along with their potential ability to reduce
waste through better yields.
o Lower water content in the reactor.
A patented process for high molecular weight Ortho Novolaks using
exotherm control allows the use of higher concentration of aqueous
formaldehyde (Culberston 1978). In this process, 50 weight percent
formaldehyde can be used as opposed to 37 weight percent formaldehyde
used in conventional processes (Brode 1982). The lower water content in
the reactor should subsequently produce a smaller aqueous waste stream
and also should improve reactant's conversion. This process is claimed to
have an improved reaction rate because of less water and easier
temperature control.
o Use of azeotropic solvents.
Since phenol and water are totally miscible under reaction conditions, the
water which is removed carries phenol with it. Use of an azeotropic
solvent, such as xylene, can decrease the phenol content in water
(Culberston 1978, Brode 1982). As water is continually removed, the
organic phenol-rich phase can be separated and recycled. The environ-
mental advantages and disadvantages of this suggested technique will have
to be fully assessed, since introduction of entrainer or azeotropic solvent,
such as xylene, may not always be desirable.
BJO-14
-------
Re-evaluation of current temperature trajectory.
As explained earlier, in the final stage of reaction, water is removed from
the reactor to promote the equilibrium shift toward novolac products. In
the first stage of reaction, a vacuum reflux is used to maintain the reactor
at 185-195°F. After 3-6 hours of reaction time, the condensate from the
overhead condenser is sent to a receiver instead of being returned to the
reactor. This switch-over point is crucial from the standpoint of
phenol/formaldehyde losses through the overhead vapor. If this switch-
over is made too early, the losses of organic reactants will be high because
of their higher concentration in the reactor.
In addition to the cost of lost reactants, another factor to be considered in
the optimization is the cost of disposal and treatment of liquid waste. If
the organics content in the liquid is more than 12%, the fuel requirements
for incineration are minimal and phenol recovery by solvent extraction is
less costly. If the organics content is lower, then solvent extraction
becomes costly and the fuel requirements to support incineration may be
substantial (Snell 1978).
It is suggested that reassessment of the currently used temperature
trajectory may lead to the overall lowering of waste generation rates in
certain cases.
Re-evaluation of reactant and/or catalyst addition strategy.
Addition of reactants (phenol and/or formaldehyde) or catalyst (sodium
hydroxide, sulfuric acid) can be done in one step, in increments, or
continuously following a schedule designed for maximum conversion and
minimum losses during the evaporative mode of reactor operation.
Development of such a strategy can be assisted using a reactor modeling
approach. For example, staged addition of remaining formaldehyde before
switching to an evaporative mode from isothermal operation, reduces the
free phenol content in the reactor, which reduces phenol losses through the
vapor stream. As before, it is suggested that reconsideration of the
reactant or catalyst addition strategy may lead to the overall decrease of
waste generation in certain cases.
B10-15
-------
Inversion of dispersion mode.
The process involves a primary reaction cycle under basic conditions at
194-220°F and a secondary reaction cycle, at 176-194°F under neutral
conditons in the presence of a protective colloid used to stabilize the
dispersion (Brode 1982, Brode et al. 1974, Harding 1977). In the primary
cycle, condensation of formaldehyde and phenol occurs. In the secondary
cycle, inversion from a water-in-phenolic dispersion to a
phenolic-in-water dispersion takes place speeding up the poly-condensation
to the desired end point. Based upon preliminary review, the process seems
to offer a potential for higher conversion and thus lower waste loads
compared to the more conventional route.
Internal phenol recycle with treatment of receiver contents.
Recovery of phenols from aqueous streams and its recycle to the process
has been practiced. Phenol separation from water can be achieved by
several liquid solvent-extraction processes (Brode 1982, Lanouette 1977),
adsorption processes, or critical fluid extraction. A discussion of these
processes follows:
The benzene-caustic pherplization process uses benzene as the extraction
solvent in a packed tower with countercurrent flow. The caustic is used to
wash the phenols from benzene, converting them to sodium phenolates.
Using steam, carbonic acid, and additional caustic, phenols can be
separated from the aqueous phase to be recycled to the process. Removal
efficiency of 92-93% can be achieved. In some plants, it has been shown
that the removal efficiency can be improved to 95% by using a Podbielniak
centrifugal countercurrent solvent extractor instead of a packed column.
Efficiencies of 98.6-98.8% can be achieved by using a pulsed packed
column which improves scrubbing.
The Lurgi Phenosolvan process uses isopropyl ether as the solvent in a
multi-stage solvent extractor. This solvent has a lower boiling point than
phenol and can be distilled from phenol for further reuse. The Ifawol
dephenolization process uses a countercurrent packed column with a
solvent that has a boiling point higher than that of phenol and is practically
B10-16
-------
insoluble in water. The phenol can be recovered from the dewatered
solvent by vacuum distillation. After passing through a separator, the
solvent can be recovered for reuse.
Activated carbon adsorbent, in the form of granules, has a finite capacity
for removing phenol from waste water. Phenol can be removed from
activated carbon by chemical regeneration (Lanouette 1977). The waste
stream has to be filtered prior to contact with activated carbon. The
disadvantage of this process is that carbon removes other organics present
in the waste stream, thus the regenerated stream may contain other
organics besides phenol. In addition to carbon, some polymeric adsorbents
are found to be effective in removing phenol from aqueous waste (Fox
1975, 1978). The phenolic waste stream is passed through one or more
polymeric adsorbent columns yielding an effluent containing less than
1 ppm phenol. Column loadings of 0.6 to 11 Ibs of phenol per cubic foot of
this adsorbent are possible at raw waste phenol concentrations of 100 to
50,000 ppm, respectively. Regeneration of the resin with concentrated
formaldehyde allows convenient recycle of the resulting phenol
formaldehyde eluate directly to the polymerization reactor.
*
A critical fluid extraction method for recovering phenol from waste water
is commercially available*. The method uses a condensed gas, such as
carbon dioxide in the vicinity of its critical point. Critical fluids have
highly favorable solvent properties and they behave as a liquid in dissolving
significant amounts of organic substance. At the same time, they behave
like gases in that the rate of extraction is extraordinarily high compared to
normal liquid solvents. Commercial systems are available for capacities
ranging from 1 to 100 gallons per minute.
Each of the methods listed above has to be individually evaluated for a
specific application. General environmental attributes of each method
were not evaluated.
Critical Fluid Systems Inc. 1985: Personal communication.
B10-17
-------
o Altering resin molecular weight distribution and functionality to reduce
wastes.
Phenolic resins have different molecular weight distributions depending on
their end use. By making small changes in their molecular weight
distributions, the change in their properties may not be significant, but
could contribute toward waste reduction (Brode 1982). By increasing the
conversion slightly, the quantity of unreacted phenol and formaldehyde can
be decreased. Though this does not result in lowering the quantity of waste
water generated, it reduces the toxicity of this stream.
o Use of excess formaldehyde to reduce free phenol content.
Free phenol content in the reactor can be decreased by using higher
quantity of formaldehyde feed to the reactor. Excess formaldehyde can be
removed from the resin by stripping. Lower free phenol results in lower
phenol in the waste water stream, reducing its toxicity.
o Isolating the wash-water stream from receiver contents.
The wash water is sometimes mixed with the receiver contents before
treatment. If the wash water contains only traces of phenol, mixing it with
the receiver contents results in a larger and more dilute waste stream. By
isolating the wash water and treating it separately, the receiver contents
will not be diluted. This could facilitate their disposal by incineration or
use of phenol recovery options.
9.1.3 Resin Wash Water
In the manufacture of liquid resols, the final product may be washed with water
several times to remove inorganic salts and traces of unreacted phenol and
formaldehyde. This washwater is usually filtered to remove any solids present prior to
discharge to the sewer or treatment as a separate stream or in combination with other
aqueous streams. The following source reduction methods were considered for this
waste stream:
o Use of counter-current washing.
The amount of wash water necessary to remove a certain amount of
unreacted phenol or formaldehyde is lowest when counter-current washing
B10-18
-------
HASTEMATER
RESIN
-------
is used. Counter-current washing may be more favorable in a continuous
operation where washing can be done in counter-current extractors (e.g.
rotating disk type) if the viscosity of the liquid resin permits such an
operation. In batch operation, the liquid resin could be allowed to
accumulate in tanks and counter-current washing can be done periodically.
When the product resin needs to be washed several times, counter-current
washing (e.g. similar to the scheme shown in Figure 9-1) will generate low
quantities of waste water. Counter-current rinsing may offer a substantial
improvement over sequential rinse-decant sequences used to achieve the
same washing efficiency. It is possible that due to the increased
concentration of organics in the wash-water stream, it may then be
suitable for phenol recovery.
o Re-use of resin wash water.
The purpose of washing the resin with water is to remove inorganic salts
and residual water-soluble organics. In priciple, it is possible to re-use the
wash-water until the dissolved impurity concentration reaches prohibitive
levels. This type of wash-water re-use is common in the synthetic rubber
industry where the inhibitors present in monomers are washed using a
caustic solution (Snell 1978). If wash-water is reused for the first rinse,
the second (or consecutive) rinsing may require clean water to efficiently
remove the contaminants. The wash-water accumulated from the first
rinse may eventually be treated along with receiver contents to recover
phenol.
9.1.4 Spills and Leaks
Spills are due to accidental discharges of liquids during transfer operations or
equipment leaks. The spill cleanup wastes are stored in drums and are considered
hazardous. Only the implementation of better operating practices is suggested for this
waste stream, especially in the area of better operator training and preventive
maintenance.
B10-20
-------
9.1.5 Equipment Cleaning Wastes
After each batch, the reactors for resols and novolacs are cleaned with dilute acid or
caustic solution to remove resin particles that remain on the reactor walls. This
intermittent waste stream is filtered to remove the particles and the filtrate is mixed
with the receiver contents prior to disposal. The filtered particles are often mixed
with other non-recyclable off-grade products and disposed of in a landfill. The
equipment cleaning waste generated by the washing procedure contains traces of
phenol and formaldehyde. On occassions, it is necessary to use solvent for equipment
cleaning. In such cases, the resulting waste is generally incinerated, however, it may
be landfilled by facilities which do not use an incinerator. The following source
reduction techniques have a potential for reducing the amount of aqueous waste
produced:
o Reduction of resin buildup on the reactor walls.
The use of Teflon* coated reactors or reactors equipped with wipers that
continually clean the walls will reduce polymer buildup on the walls.
Teflon is known to possess low adhesion characteristics which may help to
relieve buildup and help drainage. It is also suitable for use in process
environments at temperatures up to 400°F which adequately covers the
range involved in production of phenolic resins. Teflon coating must be
closely scrutinized if there is an occasional need for mechanical cleaning
that involves chipping off accumulated material. However, the use of
Teflon may eliminate the need for mechanical cleaning altogether. The
designs of double shaft mixers equipped with wall and bottom wipers are
commercially available and are routinely used in applications requiring
mixing of high viscosity resins.
o More complete drainage of process lines.
Prior to rinsing with water, the equipment, along with process lines should
be properly drained. Such drainage can be accomplished by proper piping
layout (e.g. no pockets, sloping toward equipment) and also by "pigging" the
lines, which is an encountered practice in the paint manufacturing industry.
There, a plastic slug is propelled by an inert gas through the lines which
* Registered trademark of E.I. du Pont de Nemours & Co.
B10-21
-------
cleans them while increasing product yield. For phenolic resin process
piping it may be possible to use a similar approach.
Maximizing equipment dedication.
The frequency of equipment cleaning can be reduced by maximizing
equipment dedication to a single formulation. In the limit, a total
dedication is accomplished by conversion from batch to a continuous
process. Continuous processes for the production of phenolic resins have
been developed (Euleco 1975, Anon. 1965). Since the market calls for a
large diversity of products, it appears unlikely that these processes can be
used universally. Proper production scheduling, e.g., maximizing the size
of the product obtained from a single campaign, or scheduling compatible
batch campaigns adjacent to each other is probably a much more widely
practiced technique to lower the cleaning frequency and the associated
waste.
Filter resin recovery.
Prior to shipping, liquid resins are usually filtered. The resin remaining in
the filter casing can be easily recovered by blowing the resin out with the
use of plant air prior to screen cleanup. This procedure was effectively
used by Borden Chemical Company (Huisingh et. al. 1985).
Use of a two step rinse with recycle of the first rinse.
Some facilities use a "fill-and-empty" reactor cleaning technique, where
the reactor is filled with water, which is subsequently dumped after
agitation. This produces a large amount of wastewater contaminated with
phenolic resins. As an alternative, an initial low-volume rinse (e.g.
employing spray nozzles) can remove the bulk of the resins to produce a
small quanity of wastewater with a high concentration of phenol. This
material can be reused in the next compatible batch of resin. The tank can
then be washed with a full volume rinse to generate a waste with a much
lower resin content. This procedure, used at Borden Chemical Co.,
contributed to a 93 percent reduction in the amount of discharged organics
(Huisingh et. al. 1985).
B10-22
-------
9.2 Implementation Profile
While some of the methods proposed or identified require installation of additional
equipment, substantial reductions can be made through changes in operating
procedures. The use of a two step rinse method with the recycle of the first rinse, and
an increased awareness of waste on the part of plant personnel, contributed to a
significant reduction in the organics lost with equipment cleaning wastes at Borden
Chemicals Co. The procedure appears easy to implement at many phenolic resin
manufacturing facilities.
The use of countercurrent product washing and the recovery of phenol from aqueous
wastes for recycling to the process usually requires installation of additional
equipment. These options may not be implementable at all facilities. Recovery of
phenol using polymeric adsorbents or solvent extraction techniques was found feasible
by some manufacturers.
Because different manufacturers produce different grades of resin, great variation in
wastestream composition is encountered. Each facility must therefore choose the
control measures compatible with its own specific requirements. For facilities
already having suitable waste treatment plants or incinerators, many of the proposed
methods cannot be economically justified. Implementation will ultimately depend on
economic feasibility, which should take into account avoided treatment costs, avoided
disposal costs for treatment residuals, and raw material savings.
9.3 Summary
The summary of all noted source control techniques is given in Table 9-1. Each
technique was rated for its effectiveness, extent of current use, and future application
potential on scale of 0 to 4. The ratings were derived by project staff based on review
of the available data. The estimates of current level of waste reduction achieved
(current reduction index) and possible future reduction (future reduction index) were
obtained from the ratings in accordance with the methodology presented in the
introduction to this appendix. Fractional waste generation rates (including water)
were taken from Table 8-1. Current reduction and future reduction indices were also
R10-23
-------
TABLE 9-1 StMUftr OF SOURCE CONTROL METHODOLOGY FOR THE PHENOLIC RESINS MANUFACTURING INDUSTRY
DO
l—>
O
1
Haste Stream |
1
Off -Grade Products |1.
|2.
|3.
1
Receiver Contents 1 1.
12.
I'-
ll.
15
16
IT.
l«
19.
|10
1
Resin Hash Hater |1.
|2.
*"
1
Spills and Leaks |1.
1
Equipment Cleaning |t.
Hastes |2.
13.
I*.
15.
1
All Sources |
Control Methodology
Proper control reactor temperature
Increase use of automation
Re-use in product
Overall
Use improved catalyst
Lower water content In the reactor
Use of aieotropic solvents
Re-evaluate temperature trajectory
1
1-
1
1
1
1
1
1
1
1
1
Re-evaluate react/cat addition method|
Inversion of dispersion mode
Internal phenol recycle
1
1
Alteration of resins molecular weight]
Use excess formaldehyde
.Isolate washwater from rec. contents
Overall
Use counter-current rinse
Re-use of resin wash water
Overall
Better operating practices
Overall
1
1
1
1
1
1
1
1
Reduce resin build-up on reactor wall)
Use of a two-step rinse with recycle
Maximiie equipment dedication
Filter resin for recovery
More complete drainage/process Mnes
Overall
All Methods
1
1
1
1
1
Found Documentatior
Quantity 1 Quality
1 1
1 1
1 1
1 00 | 1
1 1
1 1
1
1
1
!
1
1
1
1
1.10 | 1
i 1
i 1
t.OO | 1
1 1
1.00 | 1
1 1
1 1
1 1
1 1
1 1
1.00 | 1
1
— 1
1
1 1
1 1
2 1
33 |
1 i
' 1
' 1
1 1
1 1
2 1
2 1
1 1
' 1
' 1
20 |
1 1
1 1
00 |
1 1
00 |
1 1
2 1
1 1
1 1
1 1
20 |
Haste | Extent of | Future | Friction of |
Reduction | Current Use | Application | Total Haste |
Effectiveness | | Potential | |
3 1 3| 1 | |
3 1 2| 2| |
31 3| 1 | |
3.00 | 2.67 | 1.33 | 0.01 |
21 2| 2| |
1 1 'I ' 1 1
2 1 1 I 2| |
' 1 3 | 1 | |
'1 3| 1 | I
21 H M 1
3 1 2| 3| |
2 1 2 | 1 | |
1 1 2| 1 | |
21 3| 1 | |
1.70 | 2.00 | 1.10 | 0.10 |
31 1 1 1 1 1
21 0 | 1 | |
2.50 | 0.50 | 1.00 | 0.36 |
2 1 3 | 3 | |
2.00 | 3.00 | 3.00 | 0.01 |
2 | 1 | 2 | !
3 1 1 1 3 | |
3 I 3 | 3| |
2 1 1 1 3 | |
21 2| 2| |
2.40 | 1.60 | 2.60 | 0.52 |
I 1.00 i
Current
Reduction
Index
2
1
2
2
1
0
0
0
0
o
1
1
0
1
1
0
0
0
1
1
0
0
2
0
1
z
1
1
1
1
3 1
5 1
3 I
3 1
0 1
3 1
5 1
8 1
8 1
5 1
5 1
0 1
5 1
5 1
5 1
8 1
0 1
8 1
5 1
5 1
5 I
8 1
3 I
5 1
0 1
3 1
8 1
Future Reduction Index
Probable
0
0
0
0
0
0
0
0
0
0
t
0
0
0
0
0
o
0
0
0
o
1
o
1
0
0
0
2
1
2
4
5
2
1
1
1
4
1
3
1
1
4
5
5
S
4
|
1
7
6
1
5
9
7
| Maximum
1
1 0
1
1 0
1
1
1
1
1
1
1 1
1
1
1
1 '
1 o
1
1 o
1 o
1 o
1
1 1.
1
1
1
1 1
1 1
1
1
1
1
1
1 1
1
8 1
1
1
1
1
1
1
1 1
1
1
1
1 1
6 1
1
« 1
4 1
< 1
1
7 1
1
1
1
7 1
2 1
(') These streams Include listed 'f and/or '1C RCRA wastes.
-------
computed using fractional waste generation rates that excluded water and the overall
results were very similar.
The current reduction index (CRI) is a measure of reduction of waste that would be
generated if none of the methods listed were implemented to their current level of
application. For the phenolic resin industry, CRI is 1.8 (45 percent) which is
indicative of the significant level of waste minimization that already has taken place.
The future reduction index (FRI) is an indication of the level to which the currently
generated waste can be reduced if all of the techniques noted were implemented
according to their rated potential. The FRI value of 0.7 to 1.2 (18 to 30 percent) is
indicative of the moderate extent of future waste reductions. Among the techniques
that were found most effective and applicable (as evidenced by high FRI value) for
control of wastes were the use of a two-step rinse with recycle, the use of a counter-
current rinse, and recycling of waste phenol.
10. PRODUCT SUBSTITUTION ALTERNATIVES
In 1980, about 27 percent of the phenolic resins were used as a binder for making
plywood. This demand depends on the extent of use of pine from the Pacific
Northwest instead of southern pine. Southern pine, due to its higher absorbency,
requires almost 20 percent more resin than northwest pine (SRI 1981). Phenolic resins
are also used as binders for fibrous and granulated wood to make waferboards.
Injection molded thermoplastics can be effectively used to make waferboards, thus
avoiding the use of phenolic resins.
In 1980, about 7 percent of phenolic resins were used for making decorative and
commercial laminates. Use of thinner boards for vertical use as opposed to thicker
horizontal boards will lower the amount of phenolic resin per square foot of the
decorative laminate. Low-pressure polyester and melamine laminates, which do not
use phenolic resins, or epoxy or silicone resin substitutes are some other possibilities.
B10-25
-------
11. CONCLUSIONS
Since phenolic resin manufacturing produces a relatively high quantity of waste,
waste reduction by source control deserves special attention. Our qualitative
estimates indicate that the possible waste reductions range from 18 to 30 percent.
Several methods that appear to be quite effective for the industry as a whole are the
two-step rinse method with the recycle of the first rinse, recovery and recycle of
phenol from aqueous wastes, segregating of wash water from the receiver contents,
and reuse of wash water.
12. REFERENCES
AESI. 1981. American Electroplater's society, Inc. Conference of advanced pollution
control for the metal finishing industry (3rd) held at Orlando Hyatt House Kissimmee,
Florida on April 14-16, 1980. EPA-600-2-81-028. Cincinnati, Ohio: U.S. Environ-
mental Protection Agency.
Anonymous. 1965. Planning it cool pays off. Chem. week. 97(19): 35-40.
Brode, G.L. 1982. Phenolic resins. In Kirk-Othmer encyclopedia of chemical
technology. 3rd. ed. Vol. 17, pp. 384-416. New York, N.Y.: Wiley.
Brode, G.L., Harding, J. Marrion, T. et al. 1974. Synthesis of phenolic dispersions. U.S.
Pat. 3,823,103 (July 9, 1974) to UCC.
Cameron, J.B., Lundeen, A.J., and McCulley, Jr., J.H. 1980. Trends in suspension PVC
manufacture. Hydrocarbon Processing (3): 39.
BTO-27
Culberston, H.M. 1978. U.S. Pat. 4, 113,700 (Sept. 12, 1978) to Monsanto Co.
Euleco. 1975. Euleco continuous process: technical bulletin. Milan, Italy: Euleco,
S.P.S.
Fox, C.R. 1975. Remove and recover phenol. Hydrocarbon processing 54 (7): 109-
11.
. 1978. Plant uses prove phenol recovery with resin. Hydrocarbon Processing
57 (11): 269-73.
Froment, G.F., and Bischoff, K.B. 1979. Chemical reactor analysis and design. New
York, N.Y.: John Wiley and Sons.
Gleason, M.N., Gosselin, R.E, Hodge, H.C., et. al. 1984. Clinical toxicology of
commercial products; acute poisoning. 4th ed. Baltimore: The William and Wilkins
Co.
Greek, B.F. 1985. Chem. Enq. News 63 (22): 38.
B10-26
-------
Harding, J. 1977. U.S. Pat. 4,026,848. (May 31, 1977) to UCC.
Hoshika, Y., and Muto, G. 1978. J. Chromato 157: 277-78.
Huisingh, D., Martin L., Hilger, H., et. al. 1985. Proven profit from pollution
prevention. Washington, D.C.: The Institute for Local Self- Reliance.
Johnson, H. 1973. A study of hazardous waste materials, hazardous effects and
disposal methods. Vol. 2. Booz-Allen Applied Research, Inc. EPA-670-2-73-15.
Washington, D.C.: U.S. Environmental Protection Agency.
Lanouette, K.S. 1977. Treatment of phenolic waste. Chem. Eng. 84 (22): 99-106.
Radian Corp. 1977. Industrial process profiles for environmental use; Chapter 10.
Plastics and resin industry. EPA - 600-2-77-023. Cincinnati, Ohio: U.S. Environ-
mental Protection Agency.
Sax, N.I. 1984. Dangerous properties of industrial materials. 6th ed. New York,
N.Y.: Van Nostrand Reinhold.
Snell, Inc. 1978. Assessment of industrial hazardous waste practices; rubber and
plastics industry. Plastic materials and synthetics industry. EPA-530-SW-163C-2.
Washington, D.C.: U.S. Environmental Protection Agency.
SRI. 1981. Stanford Research Institute. Chemical economics handbook. Menlo Park,
Calif.: Stanford Research Institute.
Steiner, T.E. 1984. Modern plastics encyclopedia. New York, N.Y.: Plastics
Catalogue Corp.
USEPA. 1974a. U.S. Environmental Protection Agency, Effluent Guidelines Division.
Addendum to the development document for the proposed effluent limitation guide-
lines and new source performance standards for the synthetic resins segment of the
plastics and synthetic manufacturing point source category. EPA-440-l-74-036a.
Washington, D.C.: U.S. Environmental Protection Agency.
_ . 1974b. U.S. Environmental Protection Agency, Effuent guidelines Division.
Development and new source performance standards for the synthetic resins segment
of the plastics and synthetics materials manufacturing point source category. EPA -
440-l-74-010a. Washington, D.C.: U.S. Environmental Protection Agency.
Wynstra, J., and Schultz, S.J. 1980. U.S. Pat. 4,206,295 (June 3, 1980) to UCC.
13. INDUSTRY CONTACTS
M.J. Kowalski. Manager, Environmental Compliance, Reichhold Chemicals, Inc.,
White Plains, NY.
B10-27
-------
-------
1. PROCESS: PRINTED CIRCUIT BOARD MANUFACTURE
2. SIC CODE: 3679 (052)
3. INDUSTRY DESCRIPTION
Manufacturers of printed circuit boards (PCB's) are included as part of the electronic
component manufacturing industry.
3.1 Company Size Distribution
As of 1984, the PCS manufacturing industry included a total of 585 plants* with a
total employment of 435,100 (NCO 1984). The industry consists of large facilities
which are totally involved with PCB's, large and small captive facilities, small job
shops doing contract work, and specialty shops doing low-volume and high-volume
precision work. Approximately half of the PCB's produced are by independent
producers, while the rest are by captive producers. Table 3-1 lists company size
distribution in the U.S.
Table 3-1 Company Size Distribution
Number of Employees per Facility
Total 1-50 51-100 101-500 501-1000 1001-2500 2501-5000 5000+
No. of
establish-
ments 585 177 109 151 42 49 14 43
No. of
employees 435,100 4,425 8,175 37,750 31,500 85,750 52,500 215,000
Source: Electronic Marketing Directory (NCO 1985).
* Industry contacts indicate that the actual number may be closer to 1000 plants.
Bll-1
-------
3.2 Principal Producers
The ten major PCB producers in the U.S. are listed below (PEI 1983):
TRW Cinch Graphite
Texas Instruments
Motorola Semiconductor
Microtran
GTE Sylvania
Western Electric
Rockwell International
Syn-thane-Taylor
Cincinnati Millacron
Chicago Etching
3.3 Geographical Distribution
Over 65% of all PCB manufacturing sites are located in the northeastern states and in
California. The remaining plants are scattered throughout the country as represented
by Figure 3-1 and Table 3-2 below.
Table 3-2 Geographical Distribution of Printed Circuit Board Manufacturing
Facilities by EPA Region
EPA Region
1-50
I 16
II 39
III 12
IV 18
V 42
VI 11
VII 2
VIII 1
IX 33
X 3
Number of Employees per
51-
100
12
14
8
13
37
4
1
1
18
1
101-
500
23
19
12
11
43
10
5
2
23
3
501-
1000
3
7
3
5
10
2
1
1
9
1
1001-
2500
7
8
4
4
12
2
2
0
8
2
Facility
2501-
5000
4
1
0
3
2
1
0
1
2
0
5001+
4
3
2
2
15
4
4
0
8
1
Total
69
91
41
56
161
34
15
6
101
11
National
177
109
151
42
49
14
43
585
Source: Electronic Marketing Directory (NCO 1985).
Bll-2
-------
VIII
i
C/J
11-20
TV^J 2-5
21-50
6-10
51-100
Roman numerals show EPA regions
Figure 3-1 Printed Circuit Board Plants in the U.S.
-------
4. PRODUCTS AND THEIR USE
PCB's can be classified into 3 basic types: single-sided, double-sided, and multilayered.
The total board production in 1983 was estimated at 14 million square meters (PEI
1983). Double-sided boards accounted for about 55 percent of the PCB's produced,
while the percentage of multi-layer board production was 26 percent (PEI 1983). The
type of board produced depends on the spatial and density requirement, and on the
complexity of the circuitry. PCB's are used mainly in the production of business
machines, computers, communication equipment, and home entertainment equipment.
5. RAW MATERIALS
The following raw materials are used by the industry (Stintson 1983, PEI 1983, Cox and
Mills 1985):
Board materials
glass-epoxy, ceramics, plastic, phenolic paper, copper foil
Cleaners
sulfuric acid, fluoroacetic acid, hydrofluoric acid, sodium
hydroxide, potassium hydroxide, trichloroethylene, 1,1,1-
trichloroethane, perchloroethylene, methylene chloride
Etchants
sulfuric and chromic acid, ammonium persulfate, hydrogen
peroxide, cupric chloride, ferric chloride, alkaline ammonia
Catalysts
stannous tin, palladium chloride
Electroless copper
bath
copper sulfate, sodium carbonate, sodium gluconate, Rochelle
salt, Versene-T, sodium hydroxide, formaldehyde
Screen
silk, polyester, stainless steel
Screen ink
composed of oil, cellulose, asphalt, vinyl or other resins
Resists
polyvinyl cinnamate, allyl ester, resins, isoprenoid resins,
methacrylate derivatives, poly-olefin sulfones
Bll-4
-------
Sensitizers
thiazoline compounds, azido compounds, nitro compounds, nitro
aniline derivatives, anthones, quinones, diphenyls, azides,
xanthone, benzil
Resist solvents
xylenes, toluene, benzene, chlorobenzene, celiosolve and
cellosolve acetate, butyl acetate, 1,1,1-trichoroethane,
acetone, methyl ethyl ketone, methyl isobutyl ketone.
Electroplating baths
copper pyrophosphate solution, acid-copper sulfate solution,
acid-copper fluoroborate solution, tin-lead, gold, and nickel
plating solutions
Resist stripping
solutions
sulfuric-dichromate, ammoniacal hydrogen perioxide,
metachloroperbenzoic acid, methylene chloride, methyl alcohol,
furfural, phenol, ketones, chlorinated hydrocarbons, non-
chlorinated organic solvents, sodium hydroxide
6. PROCESS DESCRIPTION
Three principal production methods have been employed by the industry to produqe
printed circuit boards. These include:
The conventional subtractive process
The fully additive process
The semi-additive process
Detailed descriptions of the process sequences are given elsewhere (Yapoujian 1982,
Coombs 1979, USEPA 1979, PEI 1983). Because of the limitations of the additive
processes, the subtractive method is currently the most widely used method. Figure
6-1 illustrates the procedure for the production of double-sided panels using the
subtractive method (Yapoujian 1982). Most of the operations shown are also common
to the production of other types of printed circuit boards such as single-sided or multi-
layered boards.
Bll-5
-------
• "1
L .«.
1
—
COPPER CLADDED
LAMINATE BOARD
© ©
STRIP
© ®
STRIP
® ®
REMOVE
BEVEL
~l
BAKE
ELECTROLESS
©
(D(D
©®
CLEAN
©®
1
-
DRILL
CD®
COPPER
INSPECT
NICKEL
INSPECT
f
>
SAND/ I
©
COAT
PHOTORESIST '»|
TIN /LEAD
PLATE | El
1
TAPE 'EXPOSE 1
FINGERS j *1
1
GOLD
PLATE | f|
i
1
1
!
PRINTED CIRCUIT
BOARD
PROCESS HASTE CATEGORIES
(?) AIRBORNE PARTICULATES
© SPENT SURFACE CLEANING SOLUTIONS
® SPENT CATALYST SOLUTIONS
® SPENT PLATING BATHS
® SPENT DEVELOPING SOLUTIONS
® SPENT RESIST REMOVAL SOLUTIONS
© SPENT ETCHANTS
® SPENT METAL STRIPPING SOLUTIONS
(?) RINSE HASTEHATER
Figure 6-1 Substractive Process for the Production of Double-sided Printed Circuit Boards
Bll-6
-------
The conventional subtractive process starts with a copper-cladded laminate board,
composed of a non-conductive material such as glass epoxy or plastic. PCB
manufacturers often purchase panels of board that are already copper cladded from
independent laminators to use as the starting material for the PCB's. The manu-
facturing process consists of the following operations:
Board preparation - The process sequence begins with a baking step to ensure that the
copper laminated boards are completely cured. Holes for the components are then
drilled through stacks of boards or panels, often 4 layers thick. The drilling operation
results in burrs being formed on one or both sides of the panel. These are removed
mechanically through sanding and deburring steps to create an even surface.
Electroless copper plating - The smooth copper-cladded board is subsequently
electroless plated with copper to provide a conducting layer through the drilled holes
for circuit connections between the copper-cladded board surfaces. Electroless
plating involves the catalytic reduction of a metallic ion in an aqueous solution
containing a reducing agent. The metal is therefore deposited without the use of
external electrical energy. The circuit board must be thoroughly cleaned before it is
electroless plated. The typical steps for electroless plating are:
Mechanically scrub and alkaline clean the boards to remove soil and
fingerprints. This is followed by a rinsing step usually involving spray or
counter flow rinsing with water.
Etch the copper-cladded surface. The board is immersed in ammonium
persulfate or a peroxide-sulfuric acid mixture to remove the oxidation
inhibitor in the copper foil. This provides for better adhesion of the
electroless plating catalyst. After etching, the board is again rinsed.
Catalyst application. This step is required for through-hole plating. A
catalyst must be applied for the initial metal deposition to occur on non-
metal surfaces, such as the interiors of the holes. Catalysis is done by a
stabilized reaction product of stannous chloride and palladium chloride
which is often sold as concentrated solution in hydrochloric acid. The
resulting board will consist of a surface layer of palladium nuclei, and
Bll-7
-------
stannous and stannic hydrous oxides and oxychlorides. Following catalyst
application, the board is rinsed with water.
Surface activation. The catalyzed board is immersed in a solution that
dissolves away the excess tin, exposing the molecular layer of palladium
metal on the board surface. The palladium then acts as the base for
electroless copper deposition. Following surface activation, the board is
rinsed.
Electroless plating. The board is immersed in an electroless copper plating
bath. The plating bath usually consists of a copper salt such as cupric
chloride or copper sulfate. The reducing agents most often used to reduce
the metal to its base state are formaldehyde and hypophosphate. In
addition, chelating agents are also added to the plating bath to hold the
metal in solution, preventing plate-out on the tank wall. The four most
common types of chelating agents used are amino acids, carboxylic acids,
hydroxy acids, or amines. Following plating, the board is rinsed and
mechanically scrubbed.
Pattern printing and masking - After the board is electroless-plated with copper to
provide a uniform conducting layer over the entire surface, the board can be panel
plated or pattern plated to produce the desired circuitry. Panel plating consists of
copper electroplating the entire board area, including holes, immediately following the
electroless copper plating. A metal etching resist is then plated onto the copper board
in a pre-determined pattern. The board is subsequently etched to provide the desired
circuitry. Pattern plating, on the other hand, consists of copper electroplating only
the holes and circuitry. Since pattern plating is more commonly used in the industry,
the steps involved in this operation are described below.
In pattern plating, a plating resist is applied to produce the circuit image on the board.
The resist is a resin whose function is to provide a protecting layer over areas of a
substrate that are not to be affected by a subsequent etching or plating process.
There are two main ways in which a resist can be applied onto the board. These are
screen-printing and photolithography.
611-8
-------
In the screen-printing method, ink is forced through unblocked spaces in a stainless
steel mesh or synthetic fabric screen and onto the board. The ink is then cured by
heating or by exposing the board to UV light, and thus produces the desired image of
the circuit pattern on the board.
The second method relies on exposing a light-sensitive resist to light through a
patterned glass or a photographic film master called a phototool (or photo mask).
There are many types of photoresist on the market. These include liquid or dry film,
which can be either negative or positive resists. A negative resist is a resin whose
solubility in the developer is greater for the areas which are unexposed to light. This
will lead to the formation of a pattern by the removal of unexposed zones when the
board is dipped into a developing solution. The common types of negative resist used
are epoxy, vinyl polymers, or halogenated aromatics.
A positive resist is more soluble in the developing solution in the light-exposed areas.
Therefore, a pattern will be formed by the removal of the exposed areas. The
different types of positive resist are: methacrylates, halogenated methacrylates,
cross-linked methacrylates, or polyolefin sulfones.
To transfer a circuit image onto the board, the board must be coated either with a
liquid resist or a dry film resist. Liquid resist can be applied to the panel by
immersing it into the solution and withdrawing it at a controlled rate. The coated
board is then dried, and the pattern is printed onto the board by exposing it to UV light
through a photo mask. Dry film resist is a photo-sensitive resin, sandwiched between
a polyethylene and a polyester layer. To apply the dry film resist, a polyethylene layer
is peeled off and the bare resin surface is laminated onto the board. The printing of
the circuit image can then be done by exposing the board to UV light with a photo
mask. The remaining polyester film is then peeled off before developing.
After the photoresist-coated surface has been printed by exposure to UV light, the
desired circuit pattern can be produced through a developing process. The developer
can be an organic solvent, or an alkaline aqueous-based solution, depending on the type
of resist being used. The developing solution will dissolve away the resist in unwanted
areas, leaving behind a pattern of copper-exposed areas for subsequent plating. This
operation is usually done by immersing the panels in a still developer solution, a spray
tank, or a conveyorized system.
Bll-9
-------
Electroplating - After a pattern is developed onto the board, the circuit is built up
through a copper electroplating process. The steps involved in the copper electro-
plating operation include:
Alkaline clean to remove any residual from the developing operation and
then rinse.
Light etch to ensure the removal of any residual photoresist not already
removed in the developing of the image. The board is usually immersed in
a light etching solution, such as peroxide-sulfuric acid or sodium
persulfate. Following this step, the board is rinsed.
Acid dip to remove oxides and then rinse.
Copper electroplating. The boards are immersed in a copper plating bath.
The bath contains copper salts dissolved in alkaline cyanide, acid sulfate,
pyrophosphate, or fluroborate solutions. The pattern or circuitry is thus
built up to the desired thickness. This step is then followed by rinsing.
After the circuit is copper electroplated, tin/lead is used as a resist for the subsequent
etching process. The procedure for tin/lead electroplating is similar to the copper
electroplating process described previously. However, the tin/lead electrolyte is very
sensitive to contamination by sulfates. Therefore, care must be taken to prevent
drag-in from the pre-cleaning operations. After the board is tin/lead plated, the
original plating resist is stripped off by immersing the boards in resist stripping
solutions such as caustic, methylene chloride, glycol ethers with caustic, or glycol
ethers with amines/ammonia. The boards are then thoroughly rinsed with water.
Etching - The removal of the original plating resist results in exposed areas of
unwanted copper. In batch etching, which is rarely used now, the panels are immersed
in a still tank containing an etching solution until all the unwanted copper is etched
away. In etching using a conveyorized system, panels are placed on a conveyor, and
etchant is sprayed on them. The most common types of etchant are: ferric chloride,
cupric chloride, chromic acid, ammonium persulfate, peroxide-sulfuric acid, and
ammoniacal etchants. Chromic acid used to etch tin/lead plated panels has been
virtually eliminated due to effluent guidelines; however, it is still used to etch epoxy
smears following the drilling of multilayer boards.
Bll-10
-------
Copper etching represents the last major operation involved in the manufacturing of
printed circuit boards. The boards are subsequently cleaned and electroplated with
different types of metals, such as nickel and gold, to provide solderability and
corrosion protection. The procedures for these operations, however, are similar to the
one described previously for copper electroplating.
The fully additive method differs from the subtractive method described above in that
it involves the deposition of the plating material onto the board in the pattern dictated
by the circuit, as opposed to the removal of the metal already deposited through
etching. The process begins with an uncladded board. Plating resist is then applied
onto the board in non-circuit areas. Electroless copper is subsequently deposited to
build up the circuit to the desired thickness. Since the board doesn't initially have any
copper in non-circuit areas, a copper etching step is thus eliminated.
7. WASTE DESCRIPTION
There are 5 principal operations common to the production of all types of printed
circuit boards. These include:
Cleaning and surface preparation
Catalyst application and electroless plating
Pattern printing and masking
Electroplating
Etching
Typical waste streams generated from the unit operations in the printed circuit board
manufacturing industry are listed in Table 7-1.
Airborne particulates generated from the cutting, sanding, routing, drilling, beveling,
and slotting operations during board preparations are normally collected and separated
using bag-house and cyclone separators. They are then disposed of, along with other
solid wastes, at landfills.
Acid fumes froTi acid cleaning and organic vapors from vapor degreasing are usually
not contaminated with other materials, and therefore are often kept separate for
subsequent treatment. The acid fume air stream is collected via chemical fume hoods,
Bll-11
-------
Table 7-1 Waste Streams from Printed Circuit Board Manufacturing
Waste Source
Waste Stream
Description
Waste Stream
Composition
RCRA
Code
Cleaning/Surface preparation
oo
Catalyst application/
Electroless plating
Pattern printing/masking
Electroplating
Etching
1. Airborne participates
2. Acid fumes/organic vapors
3. Spent acid/alkaline solution
4. Spent halogenated solvents
5. Waste rinse water
1. Spent electroless copper bath
2. Spent catalyst solution
3. Spent acid solution
4. Waste rinse water
1. Acid fumes/organic vapors
2. vinyl polymers,
3. Spent resist removal solution
4. Spent acid solution
5. Waste rinse water
1. Spent plating bath
2. Waste rinse water
1. Spent etchant
2. Waste rinse water
board materials,
sanding materials,
metals, fluoride,
acids, halogenated
solvents, alkali
acids, stannic
oxide, palladium,
complexed metals,
chelating agents.
F003, F005
chlorinated
hydrocarbons, organic
solvents, alkali
Copper, nickel, tin,
tin/lead, gold,
fluoride, cyanide,
sulfate.
ammonia, chromium,
copper, iron, acids.
F001, F002
F006, F007,
F008, F009
-------
and sent to a scrubber where it is contacted with water. The scrubbed air then passes
on to the atmosphere, and the spent absorbing solution is treated along with other
acidic waste streams. Similarly, organic fumes are often collected and passed through
a bed of activated carbon. The carbon bed is then regenerated with steam. In many
cases, the regenerative vapor is condensed and the condensate containing water and
solvents is drummed and land disposed. In a few cases, the regenerative vapor is
combusted.
The spent acid and alkaline solutions from the cleaning steps are either contract
hauled for offsite disposal or pH-treated to precipitate out the contaminants. The
precipitates are removed by clarification, filtration, or flotation, and the resulting
sludge is hauled away for disposal or recovery. Spent chlorinated organic solvents are
often gravity separated, and are recovered in-house or hauled away for reclaiming.
The remaining majority of the wastes produced are liquid waste streams containing
suspended solids, metals, fluoride, phosphorus, cyanide, and chelating agents. Low pH
values are often the characteristic of the wastes due to acid cleaning operations. The
liquid wastes may be controlled using end-of-pipe treatment systems, or a combination
of in-line treatment and separate treatment of segregated waste streams. A
traditional treatment system for the wastes generated is often based on pH adjust-
ment, or on the addition of chemicals that will react with the soluble pollutants to
precipitate out the dissolved contaminants in a form such as metal hydroxide or
sulfate. The solids are removed as a wet sludge by filtration or flotation, and the
water is discharged to the sewer. The diluted sludge is usually thickened before
dumping into landfills. Recent improvements in in-line treatment technologies such as
reverse osmosis, ion exchange, membrane filtration, and advanced rinsing techniques
increase the possibility for the recovery and reuse of water and metals.
8. WASTE GENERATION RATES
Very little data was available on specific waste generation rates from the printed
circuit board manufacturing industry. Typical concentration values of principal
pollutants present in wastewater were reported previously (USEPA 1979). Principal
pollutants included suspended solids, cyanide, copper, nickel, lead, chromium, fluoride,
phosphorus, and several noble metals. While no specific waste generation rates were
B1J-13
-------
reported, fractional rates were estimated by project staff based on the available
information and engineering judgement. These values are shown in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
A summary of the waste sources and the corresponding source reduction methods is
given in Table 9-1. This section deals with the description of the listed techniques.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
Bll-14
-------
9.1.1 Cleaning and Surface Preparation
The majority of the waste generated by this operation comes from various rinsing
steps. Since a substantial amount of water is often required, treatment of
contaminated rinse water presents a major problem in waste disposal for the printed
circuit board manufacturing industry. The rinse water is often pH treated, and the
resulting sludge hauled away for disposal or recovery. The less contaminated water is
then discharged.
Source reduction methods for the minimization of wastes generated from different
cleaning operations have been described in a separate study of metal parts cleaning
practice included in this appendix. Since a very high water usage rate is a major
characteristics of the printed circuit board manufacturing industry, reduction of the
amount of water used will greatly improve the waste treatment process along with
reducing the precipitated calcium and magnesium salts which are present in the
treatment sludges together with heavy metal precipitates. The reader is referred to
the study mentioned above for appropriate source reduction methods for the cleaning
and surface preparation operations in the manufacturing of PCB's.
9.1.2 Catalyst Application and Electroless Plating
The waste streams generated from this operation are liquids which are either spent
process solutions or waste rinse water. Generally, pH adjustment and clarification are
used to precipitate out the contaminants. The source reduction techniques include:
o Use combined sensitization and activation solution to eliminate one extra
rinsing step.
Some PCS manufacturers are using this technique in the manufacturing
process. Others prefer to use separate application and activation steps
since it seems to improve the activity of the catalyst. The reduction of
waste resulting from the elimination of one extra rinsing step, however,
should outweigh the disadvantage of reduced catalyst activity. The
opinions on the technical merit of this technique appear to be split.
o Use lower concentration plating bath to reduce the degree of subsequent
rinsing required.
Bll-15
-------
This approach has been tried mostly by large companies in the electro-
plating industry. No available information has been reported from PCS
manufacturers on the use of plating baths with lower concentrations.
Use differential plating instead of the conventional electroless plating
process.
By controlling the concentration of certain stabilizers in the electroless
copper bath, copper will be deposited three to five times faster on the
through-hole walls than on the copper cladded surface (Poskanzer and
Davis 1982). This process will reduce the amount of copper which must be
subsequently etched away in the subtractive method. The use of
differential electroless plating has not been reported by PCB
manufacturers and it may require significant developmental work before
commercialization is possible.
Use weak or biodegradable chelating agents.
The use of weak agents will allow more effective metal removal in
subsequent metal recovery operations, and thus reduce the amount of
waste discharged. Only a small number of PCB manufacturers have
reportedly been using weak chelating agents such as hydroxy acids in the
electroless plating bath.
Use in-line techniques for metal recovery.
Metal recovery units, such as an ion-exchange columns, should be installed
in-line to remove metals from spent plating baths and waste rinse waters.
The ion-exchange resin can be periodically regenerated to provide plating
solutions which can then be recycled to the plating bath. Large PCB
manufacturing plants have begun to install metal recovery units on-site to
recover metals from wastewaters. However, since these units are
generally used to remove different metals from a combined waste stream,
regeneration of the plating bath solution is not being practiced.
Use computerized/automated control systems.
Computerized process-control systems can be used for board handling and
process bath monitoring to prevent unexpected decomposition of the
plating bath. Since the use of a computerized control system not only
311-16
-------
requires a large capital outlay for initial installation but also increases the
demand for skilled operations and maintenance personnel, only very large
companies which manufacture both PCB's and other electronic components
are incorporating this change in their manufacturing process. For example,
Hewlett-Packard in Sunnyvale, California reported its successful use of
computers for plating operations on PCB's (Anonymous 1983). The majority
of PCS manufacturing plants, however, are small and therefore can only
utilize automated handling of the boards in the plating operations.
o Better operating practices.
Due to their simplicity and low cost, good operating practices are widely
used in the industry as the first effort to reduce the waste generated.
Some of the methods used are frequent inspection of plating racks for loose
insulation to prevent excessive dragout of process solution, distributing the
work load evenly to avoid dense loading which can cause localized
instability of the process solution, and stripping copper out of the plating
tank regularly to prevent continuous deposition of copper and palladium on
the tank walls.
Another approach is to segregate all waste streams. Chelated waste
streams should be separated from other streams to prevent problems in
precipitating out the metals during subsequent waste treatment.
Segregation of complexed copper streams is widely practiced in the
industry to reduce the volume of waste generated.
9.1.3 Pattern Printing and Masking
Depending on the chemicals used, air emissions consist of various organic compounds.
The fumes are usually collected and passed through an activated carbon bed. The
remaining liquid waste streams are spent chlorinated solvents, spent resist solution,
and waste rinse water. The organic solvents and resists are gravity separated and
collected for disposal or recovery. The following source reduction techniques were
noted:
Bll-17
-------
Use aqueous processable resist instead of solvent processable resist.
Aqueous processable resists, such as the Du Pont Riston photopolymer film
resists which allow for the use of caustic and carbonates as developer and
stripper*, can be used in place of solvent processable resists whenever
possible to eliminate the generation of toxic spent solvents. Hundreds of
facilities are now using these aqueous processable films for the manu-
facturing of PCB's.
Use screen-printing instead of photolithography to eliminate the need for
developers.
Screen-printing has conventionally been used only to produce PCB's which
require very low resolution in the width and spacing of the circuit lines.
Some companies have recently developed screen-printing techniques which
can provide higher degrees of resolution. For example, General Electric
has developed a method for screen-printing down to 0.01 inch resolution
which can be used to manufacture PCB's for appliances (Greene 1985). The
majority of PCB manufacturers, however, are still using the photolitho-
graphic technique for PCB's having circuitry finer than 12 mil lines and
spaces.
Use Asher dry photoresist removal method to eliminate the use of organic
resist stripping solutions.
Although this method is increasingly being used in the semiconductor
industry, its use has not been reported by PCB manufacturers, probably
because the PCB resists are usually much thicker than the corresponding
semi-conductor resist layers. Additional exploration is suggested.
* E.I. Du Pont de Nemours & Co. 1985: Personal communication.
Bll-18
-------
9.1.A Electroplating
Electroplating waste streams consist mainly of rinse water and contaminated or spent
dumps of plating solutions. The dissolved metals are often precipitated out by pH
adjustment, and are contract hauled for metal recovery or disposal. Cyanide-
containing waste must be oxidized using chlorine or other strong oxidants. The source
reduction techniques which are used to minimize the wastes generated from the
electroplating operation are given in a separate process study on electroplating (B3).
The source reduction method considered specific to the manufacturing of PCB's is the
use of pattern plating instead of panel plating:
o Use pattern instead of panel plating.
Since panel plating consists of copper plating the entire board area, while
pattern plating requires copper electroplating only the holes and circuitry,
the use of the latter technique will reduce the amount of non-circuit
copper which must be subsequently etched away. This practice can
therefore reduce the amount of waste generated from the etching
operation. The switch from panel to pattern plating has been made by a
large number of PCB manufacturers. Customers demanding applications
for a uniform cross section of circuitry in computer and microwave PCB's,
however, may dictate the use of panel plating to provide highly uniform
copper thickness.
9.1.5 Etching
Spent etchant and rinse water are the main waste streams generated from this
operation. As with electroplating, the metals are precipitated out of the waste stream
by pH adjustment. The following source reduction techniques were noted:
o Use dry plasma etching techniques.
The need for toxic etching solutions can be eliminated by using a dry
etching technique (Till and Luxon 1982). Etching can be done either
chemically using reactive gaseous radicals, or physically using non-reactive
ion bombardment. Radio-frequency sources can be used to ionize gaseous
molecules to create plasma. The etching products can then be removed by
vacuum pumps. The use of dry etching techniques has not been reported by
Bll-19
-------
PCB manufacturers. A separate study can to examine the applicability of
this technique in removing thick copper layers (1.4 mil) which are used in
PCB manufacture.
Use additive instead of subtractive method.
This change will eliminate the copper etching step, and therefore eliminate
the generation of substantial volumes of spent etchant as well as reduce
the amount of metal hydroxide sludges generated. Although subtractive is
still the most widely used method in the manufacturing of PCB's, the
additive method is gaining in popularity since its use results in less waste
and lower manufacturing costs (Brush 1983). A noted drawback to the
additive method, however, lies in the fact that additive processing requires
the use of solvent processable instead of aqueous processable photo-
resists*. Futhermore, spent additive plating bath often contains heavily
complexed copper which may result in waste treatment problems.
Use less toxic etchant.
Non-chromium etching solution has reportedly been used by PCB
manufacturers in an effort to reduce the toxicity of the waste generated.
Use in-line techniques for metal recovery.
Bend Research, Inc., has recently developed a metal recovery technique
using liquid membranes, which can be used to remove copper from PCB
etching solution (Basta 1983).
Use thinner copper foil to clad the laminated board.
This change will reduce the amount of copper which must be etched, and
thus will reduce the amount of waste generated from the etching process.
PCB manufacturers are switching to boards cladded with thinner copper as
their starting materials.
* E.I. Du Pont de Nemours &. Co. 1985: Personal communication.
Bll-20
-------
9.2 Implementation Profile
The amount of hazardous waste generated by the printed circuit board industry has
been and can be further reduced through implementation of the source reduction
techniques outlined above. Many of these techniques are currently practiced, which
demonstrates their technical and economic feasibility.
Basic improvements in operating practices can yield substantial savings in raw
material disposal and treatment costs at little expense to the manufacturer. For
example, frequent rack equipment inspection for loose insulation can prolong the
integrity of process solutions, thereby reducing the rate at which the solutions are
replaced. Segregating chelating agents from solutions containing metals enhances the
feasibility of recovery of these metals from the solutions. Improvements in the
process itself may require significant outlays of capital and increases in operating
cost. The economic feasibility of such proposed improvements should be examined
carefully and should include proper consideration of avoided costs.
Specific to the PCB manufacturing industry, implementation of major source reduction
techniques such as the use of the additive instead of the subtractive method, or the
substitution of materials and/or techniques associated with the electroless plating and
electroplating processes, might be promoted effectively through guidelines, technical
assistance, and information programs. A majority of the PCB manufacturers are small
companies who may have difficulties in conducting extensive research and testing to
assure the performance of modifications in the manufacturing process.
9.3 Summary
The summary of all noted source control techniques is given in Table 9-1. Each
technique was rated for its effectiveness, extent of current use and future application
potential on scale of 0 to 4. The ratings were derived by project staff based on review
of the available data and in consultation with the industry. The estimates of current
level of waste reduction achieved (current reduction index) and possible future
reduction (future reduction index) were obtained from the ratings in accordance with
the methodology presented in the introduction to this appendix.
Bll-21
-------
TABLE 9-1 SUWARY OF SOURCE CONTROL HETHOOOLOGY FOR THE PRINTED CIRCUIT BOARD NANUFACTURIN6 INDUSTRY
Found Documentation f Waste f Extent of f future f'Tracffon'of "f Current" "f Fufure~Reduc£fon~fndex f
-I
I
1 | | Quantity | Quality | Effectiveness | Potential
Cleaning/Surface |1. See study on Ketal Parts Cleaning | — | — | — |
I Overall | — | — | — |
Catalyst Activation|l. Use combined sensitization/activation| 2 | I | 2 | 2 2
| and Electroless |2. Use tower concentration plating bath | 1 | 1 | 2 | 1 2
Plating |3. Use differential electroless plating | 0 | 0 | 3 | 0 3
|4. Use weak/biodegrade chelating agents | 1 | II 2 | 2 2
|5 Use in-line metal recovery techniquesl 31 3 | 3 | 2 2
| |6. Use computerized/automated control | 1 | 1 | 2 | 2 3
|7. Better operating practices | 3 | 3 | 3 | 2 2 |
1 | Overall | 1 57 | 1 43 | 2 43 | 1.57 2 29
Pattern Printing/ |1. Use aqueous processable resists | 2 | 1 | 2 | 1 3
| Masking (») |2. Use screen print instead of photolith| 1 | 1 | 2 | 3 1
|3. Use Asher dry resist removal method | 1 j 1 | 1 | 0 I
1 Overall | 1.33 | 1.00 | 1.67 | 1 33 1.67
Electroplating (') |1 See study on Electroplating I — | — | --- |
|2. Use pattern instead of panel plating | 1 | 1 | 3 | 3 1
| | Overall | — | — | — |
Etching |1. Use dry plasma etching techniques | 01 0 | 3 | 0 2
|2 Use additive in place of sub method I 2 | 2 | 3 | 1 1
|3. Use less toxic etchants 1 1 | 1 | 2 | 2 i 2
|4. Use in-line metal recovery techniques] 3 | 3 | 3 | 2 2
|5. Use thinner copper foil for cladding | 1 | 1 | 2 | 2 1
| | Overall I 1 40 | 1.40 I 2 60 I 1 40 1 60
All Sources | All Methods
waste Keaucnon
Index Probable
2 0 1.2
0 15 2.0 1.2
1.0 0.5
0.5 0.8
0.0 2.3
1.0 0.5
1.5 0 8
1 0 0.8
1.5 0.8
0.21 1 5 0.9
0.5 1.1
1.5 0.1
0.0 0.3
0 05 1.5 0.5
1,3 0.3
2.3 0.2
0 44 2.3 0.5
0 0 1.5
0.8 0.6
1 0 0.5
1.5 | 0.8
1.0 0.3
0.15 1.5 0.7
1 00 2.0 0.7
Maximum
1 9
1 9
2.3
2 3
1 1
1.1
1 9
1.9
1 5
1.5
1 S
(*} These streams include listed T" and/or 'K" RCRA wastes
-------
Based on these ratings, it is estimated that the currently achieved level of waste
minimization is significant, as evidenced by the current reduction index (CRI) of 2.0
(50 percent). CRI is a measure of reduction of the potential waste that would have
been generated, if none of the techniques listed in Table 9-1 were used at their
present levels.
By additional implementation of the techniques listed, it is estimated that the extent
of future waste reductions can be characterized as moderate to significant, as
evidenced by the future reduction index (FRI) of 0.7 to 1.9 (18 to 48 percent).
The techniques that offer most promise to contribute significantly to waste reduction
in the future are the ones characterized by a high FRI. It appears that the use of more
efficient cleaning techniques (see study on metal parts cleaning), along with
differential electroless plating, use of aqueous processable resist, source control in
electroplating operation (see study on electroplating), and use of dry plasma etching
are the most promising for the entire industry. However, each installation is unique
and will evaluate only those options that have the highest application potential for
their specific needs.
-lO. PRODUCT SUBSTITUTION ALTERNATIVES
Improvements in the techniques used in the packaging of microchips can alter the
design of PCB's significantly and result in a decrease of waste associated with PCB
manufacturing. Presently, the dual-in-line package (DIP) accounts for 80% of all
packaging of integrated circuits (Bowlby 1985). More efficient packages, however, are
being developed which utilize a relatively new method of attaching packages to PCB.
One important me,thod is called surface mounting. The use of surface mounting
instead of the conventional through-hole insertion mounting allows for closer contact
areas of chip leads, and therefore reduces the size of PCB's required for a given
number of packages or DIPS. For a fixed number of packages, the PCB needs to be
only 35% to 60% as large as a PCB designed for the old style package (Bowlby 1985).
As the metal area on which cleaning, plating and photoresist operations are performed
is decreased, the wastes associated with these operations will also undergo a decrease.
In addition, the development of high-temperature, high-performance thermoplastics
has introduced the use of injection molding into the manufacturing of PCB's. In this
Bll-23
-------
process, heated liquid polymer is injected under high pressure into precision molds.
Since the molded substrates are unclad, semiadditive or fully additive plating is used
to produce metalized conductor patterns. (Eugelmaier and Frisch 1982). Injection
molding, coupled with a fast-rate electrodeposition (FRED) technique, such as that
developed by Batelie (LWVM 1985), can therefore be used to manufacture complex
three-dimensional PCB's with possible reduction in hazardous waste generation due to
the elimination of spent toxic etchants.
11. CONCLUSIONS
Waste reduction in the printed circuit board manufacturing industry has already been
accomplished to a significant extent through the implementation of the source control
techniques listed in Table 9-1. Additional waste reduction in the 18 to 48 percent
range, can be expected by wider adoption of such techniques as more efficient
cleaning, differential electroless plating, use of aqueous processable resist, source
control of the electroplating operation, and use of dry plasma etching. In the area of
product substitution, the use of surface mounting technology and the use of injection
molding coupled with fast-rate electrodeposition appear to offer high future potential
for waste reduction.
12. REFERENCES
AESI. 1981. American Electroplater's Society, Inc. Conference on advanced pollution
control for the metal finishing industry (3rd) held at Orlando Hyatt House, Kissimmee,
Florida on April 14-16, 1980 EPA-6QQ-281-Q28. Cincinnati, Ohio: U.S. Environ-
mental Protection Agency.
Anonymous. 1983. California-style circuit manufacturing using computerization.
Plat. Surf. Finish. 70:26-9.
Basta, N. 1983. Total metals recycle is metal finishers' goal. Chem. Enq. August 8.
pp.16-19.
BRSRTF. 1984. Boston Regional Source Reduction Task Force. Hazardous waste
source reduction potentials in the semi-conductor manufacturing industry.
Bowlby, R. 1985. The DIP may take its final bows! IEEE Spectrum. June 1985. pp.
37-42.
Brush, P.M. 1983. Fast track for PCBs. Prod. Finish. November 1983, pp. 84-5.
Coombs, C.F. 1979. Printed circuit handbook. 2nd ed. New York, N.Y.: McGraw-
Hill Book Co.
Bll-24
-------
Cox, D.5., and Mills, A.R. 1985. Electronic chemicals: a growth market for the 80's.
Chem. Enq. Prog. 81(1): 11-15.
Engelmaier, W., and Frisch, D.C. 1982. Injection molding shapes new dimensions for
boards. Electronics. December 15. pp 155-158.
Engles, K.O., and Hamby, J.T. 1983. Computerized controller for electroplating
printed wiring boards. Met. Finish. 81:95-100.
Greene, R., ed. 1985. Biotechnology and pollution control. Chem. Enq. March 4.
pp.85
Gunderson, R., and Holden, H. 1983. CAM techniques improve circuit board
production. Control. Eng. 30:141-2.
Harper, C.A. 1970. Handbook of materials and processes for electronics. New York,
N.Y.: McGraw Hill Book Co.
LWVM. 1985. The League of Women Voters in Massachussetts. Waste reduction the
untold story. Seminar Proceeding at the National Academy of Science, Conference
Center on June 19-21, 1985. Wood Hole, Mass.: the League of Women Voters in
Massachussetts.
Lyman, J. 1984. Surface mounting alters the PC board scene. Electronics. February
9, 1984.
NCO. 1984. National Credit Office. Electronic marketing directory. New York:
National Credit Office.
PEL 1983. Pedco-Environmental, Inc. Industrial process profiles for environmental
use. Chapter 30. The electronic component manufacturing industry. EPA-600-2-83-
033. Cincinnati, Ohio. U.S. Environmental Protection Agency.
Poskanzer, A.M. 1983. Plating printed circuit substrates: circuit topics. Plat. Surf.
Finish. 70:10.
Poskanzer, A.M., and Davis, S.C. 1982. An efficient electroless plating system for
printed circuitry. Plat. Surf. Finish. 69:95-7.
Stintson, S.C. 1983. Chemicals for electronics: new growth in competitive field.
Chem. Enq. News. 61(30):7-12.
Tills, W.C., and Luxon, J.T. 1982. Integrated circuit; materials, devices and
fabrication. Englewood Cliffs, N.J.: Prentice-Hall, Inc.
USDC. 1985. U.S. Department of Commerce, Bureau of the Census. Electronic
components and accessories. In 1982 Census of manufacturers. MC82-I-36E.
Washington, D.C.: U.S. Government Printing Office.
USEPA. 1979. U.S. Environmental Protection Agency, Office of Water and Hazardous
Materials. Development document for existing source pretreatment standards for the
electroplating point source category. EPA-440-1-79-003. Washington, D.C.: U.S.
Environmental Protection Agency.
Bll-25
-------
. 1983. U.S. Environmental Protection Agency. Office of Water
Regulation and Standard. Development document for effluent limitation guidelines
and standards for the metal finishing point source category. EPA-440-1-83-091.
Washington, D.C.: U.S. Environmental Protection Agency.
WAPORA. 1977. Wapora, Inc. Assessment of industrial hazardous waste practices;
electronic component manufacturing industry. PB-265532. Washington, D.C.: U.S.
Environmental Protection Agency.
Wynschenk, J. 1983. Electroless copper plating chemistry and maintenance. Plat.
Surf. Finish. 70:28-9.
Yapoujian, F. 1982. Overview of printed circuit board technology. Met. Finish.
80:21-5.
13. INDUSTRY CONTACTS
Dr. G.J. Hollod, Senior Environmental Engineer, E.I. Du Pont de Nemours & Company,
Wilmington, DE.
W.G. Vaux, P.E., Chemical and Process Engineering, Westinghouse Electric
Corporation, Pittsburgh, PA.
Bll-26
-------
1. PROCESS: PRINTING OPERATIONS
2. SIC CODE: 27
3. INDUSTRY DESCRIPTION
As defined in the Standard Industrial Classification Manual (USDC 1972), Major Group
27 - the printing, publishing, and allied industries include "establishments engaged in
printing by one or more of the common processes, such as letterpress, lithography,
gravure, or screen; and those establishments which perform services for the printing
trade, such as bookbinding, typesetting, engraving, photoengraving, and electrotyping.
This major group also includes establishments engaged in publishing newspapers, books,
and periodicals, regardless of whether or not they do their own printing". Of the
seventeen SIC subgroups under major group 27, about fifteen would involve substantial
printing operation activities.
The graphic arts industry, as SIC 27 is otherwise referred to in this report, is one of
the biggest in the United States, ranking highly among the twenty major manu-
facturing groups included in the SIC system. This is evidenced by the statistics shown
in Table 3-1.
Table 3-1 1982 Ranking of the Graphic Arts Industry
Rank
First
Sixth
Seventh
Eighth
Tenth
Tenth
Eleventh
Criteria Value,
in number of establishments
in number of employees
in total dollar payroll
in value added by manufacturing
in value of shipments
in dollars reinvested in capital expenditures
in average hourly earnings
$
$
$
$
$
U.S. dollars
53,500
1.3 million
22.7 billion
49.4 billion
85.8 billion
3.2 billion
8.58 per hour
Source: Printing Industries of America 1985: Personal communication.
B12-1
-------
3.1 Company Size Distribution
Of the 53,356 printing facilities located in the U.S., 80 percent employ less than 20
people each. Overall, the Bureau of the Census (USDC 1985a, 1985b, 1985c) estimated
that 1,291,000 people were employed by the printing industry in 1982. Table 3-2 lists
company size distribution as a function of total number of employees at a given site.
Table 3-2 1982 Company Size Distribution
No. of employees per
No.
No.
of establishments
of employees (x 1,000)
1-19
42,485
231
20-49
6,145
187
50-99
2,525
175
facility
100+
2,201
698
Total
53,356
1,291
Source: 1982 Census of Manufactures (USDC 1985a, 1985b, 1985c).
3.2 Principal Producers
The printing industry is composed of a small number of multi-plant firms and a large
number of single plant firms. The six largest firms, in terms of sales and number of
employees, are listed in Table 3-3 below.
Table 3-3 Principal Producers in the Graphic Arts Industry
Company
R.R. Donnelley & Sons
(Chicago, IL)
Hallmark Cards
(Kansas City, MO)
Moore Business Forms
(Glenview, IL)
American Greetings
(Cleveland, Ohio)
Deluxe Check Printers
(St. Paul, Minnesota)
Sales
(1984)
$1.8 billion
$1.5 billion
$1.4 billion
$945 million
$682 million
No. of Plants
17
1
38
7
61
Number of
Employees
17,700
20 , 100
15,000
23,000
10,900
Source: American Printer (Anonymous 1985).
B12-2
-------
3.3 Geographical Distribution
Graphic arts establishments are scattered all over the United States, with the largest
concentrations in California and New York. These two states share about 12 and 11
percent of the industry, respectively. Illinois, Texas, Florida, New Jersey,
Pennsylvania, and Ohio each share about 5 percent of the total. Distributions by EPA
region are shown in Table 3-4 and Figure 3-1. About 20 percent of the firms are
located in Region V, and about 15 percent in Regions II and IX.
Table 3-4 Location of Facilities by EPA Region
EPA
Region
I
II
III
IV
V
VI
VII
VIII
IX
X
Total No.
of Facilities
3,458
8,203
4,359'
7,305
10,586
4,985
3,153
1,827
7,530
1,800
Number of employees
1-19
2,625
6,420
3,203
6,097
8,174
4,111
2,485
1,514
6,237
1,519
per facility
20+
833
1,783
1,156
1,208
2,412
874
668
313
1,293
281
Source: 1982 Census of Manufactures (USDC 1985d).
4. PRODUCTS AND THEIR USES
The graphic arts industry produces a large number of products such as newspapers,
paper packaging containers, business forms, books, posters etc. (printed matter), and
printing plates which serve various needs. These products fall under any one of the
seventeen SIC subgroups. Table 4-1 lists the major sub-industries and the value of
product shipments.
B12-3
-------
VIII
DO
r-o
i
0
0-500
2501-5500
501-1500
over 5500
1501-2500
Roman numerals show EPA regions
Figure 3-1 Printing Establishments in the U.S.
-------
Table 4-1 Graphic Arts Industry: Value of Shipments in 1984
SIC
27
2711
2721
2732
2741
275
2761
2771
2795
-
Industry
Printing and allied products
Newspaper publishing and printing
Periodical publishing and printing
Book printing
Miscellaneous publishing and printing
Commercial printing
Manifold business forms
Greeting card publishing and printing
Lithographic platemaking services
Other printing and publishing
Value of Shipments^3'
(in millions of dollars)
40,256
9,890
5,932
3,585
2,310
11,600
2,318
1,228
830
2,308
Source: 1984 U.S. Industrial Outlook (USDC 1984).
(a)
Forecast estimate.
5. RAW MATERIALS
The principal raw materials used by the graphic arts industry are inks and substrates.
A substrate is any material upon which ink is impressed, such as paper, plastic, wood,
or metal. Table 5-1 shows the recent usage of ink and paper by the industry.
Table 5-1 Paper and Printing Ink used by the Graphic Arts Industry
Volume
Material (short tons)
Commercial printer 5,300,000
Magazines and other periodicals
(other periodicals include catalogs and
directories 4,000,000
Book 900,000
Printing Ink^b)
Lithographic and offset 190,150
Gravure 179,750
Letterpress 119,250
Flexographic 91,150
Source: 1984 U.S. Industrial Outlook (USDC 1984); American Paper Institute 1985:
Personal communication.
(a) Data are for 1984.
(b) Data are for 1982.
B12-5
-------
6. PROCESS DESCRIPTION
In the graphic arts industry, there are three major printing processes (offset
lithography, gravure, and flexography). Two additional methods, letterspace and
screen, are also important but represent a much smaller share of the market. Table 6-
1 presents current and future trends in the distribution of printing processes. Since
offset lithography and gravure are the most popular printing processes used, they will
be discussed in greater detail in this report. Detailed discussion of the other printing
processes may be found in the literature (USEPA 1976). Figure 6-1 is a block flow
diagram of offset lithography and gravure printing operations.
Table 6-1 Trends in Distribution of Printing Processes in the U.S.
Process
Lithography
Gravure
Letterpress
Flexography
Screen printing and other processes
Year
1977
38%
14%
32%
11%
5%
1985
45%
18%
15%
16%
6%
1990
47%
20%
8%
18%
7%
1995
46%
21%
4%
21%
8%
Source: Profit from Pollution Prevention (Campbell and Glenn 1982); Status of
Printing in the USA (Bruno 1985).
Except for the use of a plate or cylinder to transfer ink to a substrate (such as paper)
both off-set lithography and gravure are conceptually very similar. Art work or copy
is prepared and the image is transferred to the plate or cylinder. This operation makes
certain areas of the plate or cylinder receptive to ink (the image area). The substrate
is then passed by the plate (i.e. the ink is first tranferred from the plate to a rubber
blanket, and then onto the substrate) or cylinder. The substrate absorbs the ink, thus
reproducing the image. The substrate is then processed mechanically (cut, folded,
bound, etc.) to produce the final product. Since the only major difference between
lithography and gravure lies in the manner in which the plates or cylinders are made
and subsequently operated, the following discussion will be general except where
further details are warranted.
R12-6
-------
INK
SUBSTRATE
(PAPER)
ARTWORK
/TEXT
IMAGE
PROCESSIN6
[ON-SITE OR OFF-
FILM
PLATE OR CYLINDER
MAKING
(ON-SITE OR OFF-SITE]
FINISHED
PLATE/CYLINDER
DRYING
FINISHING
V
PRINTED
PRODUCT
PROCESS HASTE CATEGORIES!
(l) TRASH
(2) KASTEKATEH
(?) EQUIPMENT CLEANING
Figure 6-1 Block Flow Diagram for Offset Lithography/ Gravure Printing Operations
B12-7
-------
Most printing operations begin with art and copy (or text) preparation. Once the
material is properly arranged, it is photographed to produce transparencies and color
separations. Color separations are required to provide a single-color image or record
which can then be used to produce the single-color printing plate for lithography or the
cylinder for gravure (multi-color printing is done by passing the substrate through
several single-color printing operations). Once the film has been developed, checked,
and rephotographed (if necessary), it is sent on to the plate- or cylinder-making
operation.
Plate making consists of cleaning or counter-etching the surface of an aluminum plate
and then applying a photosensitive coating. Counter-etching involves the use of dilute
acid solutions such as phosphoric, acetic, hydrochloric, and sulfuric acids, although
alkaline counter-etches are sometimes employed. The solution is poured on the
surface and agitated with a bristle brush. The plate is then copiously rinsed with
water. The metal surface is then coated in the whirler with a photosensitive solution
consisting of an aqueous solution dichromate, ammonium hydroxide, and colloidal
protein.
The image from a photolithographic negative is transferred to the photosensitive plate
by exposure to light, and the coating \sf covered with a developing ink to increase the
affinity of the image area for ink during printing. The plate is developed by removing
the unhardened non-image areas of the coating with dilute ammonium hydroxide
solution and is then washed with water and dried. An acidified solution of gum arabic,
which may also contain ammonium dichromate, is subsequently used to desensitize the
developed plate. Finally, the plate is again washed with water, a plain solution of gum
arabic is applied, and the plate is allowed to dry.
Cylinder making begins with a steel cylinder plated with copper. The cylinder is
machined and polished so as to remove any imperfections in the copper plating. Next,
the surface is either engraved using a diamond stylus or chemically etched using ferric
chloride. Use of ferric chloride requires that a resist (in the form of the negative
image) be transferred to the cylinder before etching. The resist protects the non-
image areas of the cylinder from the etchant. After etching, the resist may be
subsequently stripped off. This operation is analagous to the manufacturing of
B12-8
-------
printed circuit boards. Following this operation, the cylinder is proofed or tested, re-
worked if required, and then chrome plated.
Following the plate- or cylinder-making operation, the next step is printing. In
lithographic printing, the plate is mounted to a rotating cylinder. As the cylinder
rotates, a dampening solution followed by ink is transferred to the plate's image area.
The inked image repels the solution and accepts the printing ink, while the non-image
area accepts the dampening solution and repels the ink. As the cylinder continues to
rotate, the inked image is transferred to a rubber roller or blanket and then onto the
substrate. The two major forms of substrates used in lithography are single sheets of
paper, (sheet-fed lithography) and continuous rolls of paper (web lithography).
In gravure printing, the cylinder is placed in the press and partially immersed in an ink
bath or fountain. Solvent is added to the ink to maintain the proper level and viscosity
of the bath. As the cylinder is rotated, ink coats the entire surface. Next, a metal
wiper (doctor blade) presses against the surface of the cylinder and removes ink from
the non-etched (non-image) areas. The substrate is then pressed against the rotating
cylinder and the ink is transferred.
After printing, the substrate may pass through a drying operation depending on the
type of ink used. Lithography can use heat-set and non-heat-set inks. In heat-set
lithography, the substrate is passed through a tunnel or floater dryer which utilizes
hot air or direct flare. With non-heat-set lithography, the ink is normally air dried.
Gravure uses heat-set inks, so a drying operation is required. Following this drying
operation, the printed substrate can be finished by slitting, cutting, trimming, folding,
binding, laminating, or embossing.
7. WASTE DESCRIPTION
Listed in Table 7-1 are the principal wastes associated with lithographic printing
operations. Gravure printing operations have been excluded since the major difference
between the two processes, from a waste generation viewpoint, is in the plate- and
cylinder-making operation. Gravure cylinder making is very similar to other metal
B12-9
-------
Table 7-1 Offset Lithography Printing Process Wastes
Waste
Stream
No
Waste
Description
Process Origin
Composition
RCRA Codes
1.
03
h-•
hO
2.
3.
Trash Image Processing
Plate Making
Printing
Finishing
Wastewater Image Processing
Plate Making
Equipment
Cleaning
Wastes
Printing
Printing
Empty containers, packages,
used film, out-dated materials
Damaged plates, developed film,
out-dated materials
Test production, bad printings,
empty ink containers, used blankets
Damaged products, scrap
Photographic chemicals, silver
Acids, alkali, solvents, plate
coatings (may contain dyes
photopolymers, binders, resins
pigments, organic acids),
developers (may contain
isopropanol, gum arable,
lacquers, caustics), and
rinsewater.
Spent fountain solutions (may
contain chromium).
Lubricating oils, waste ink,
clean up solvent (halogenated
and non-halogenated), rags.
F002
E003
F005
F002
F003
F005
-------
processing operations; therefore, the reader is referred to the process studies on metal
parts cleaning, electroplating, and metal surface treatment elsewhere in this appendix
for information regarding the types of waste that would be encountered. Additionally,
the inks used by the gravure industry tend to contain solvent and heavy metals, unlike
the inks used by the lithographic industry*. While this would affect the ways in which
certain wastes are classified, regulated, handled, and disposed, the nature of the inks
used for lithography does not detract from the desirability of minimizing all non-paper
waste.
The major wastestream encountered in the lithographic industry is trash. Almost 98
percent of the total waste generated by this segment of the industry is spoiled paper
and paper wrap*. This paper is normally recycled, incinerated, or disposed of. Other
trash, such as scrap photographic material, is sold for metal recovery. Empty
containers are normally scraped clean of ink (to an economically feasible extent) and
discarded.
Wastewater from image processing is normally treated to recover silver and is then
discharged. Plate-making wastes, such as acids and alkalis used to clean the plates,
must be either sent to wastewater treatment or drummed for disposal. For facilities
that use pre-sensitized plates, this waste is avoided (though the supplier of the pre-
sensitized plates would be producing this waste). Currently, only large-volume users
of plates (i.e. newspapers) still produce their own plates*. Photochemical wastes
would be handled in the same manner as image processing wastes. Wastewater
containing spent fountain solutions may go to a waste treatment unit but is normally
discharged into the sewer.
The final wastestream, equipment cleaning wastes, consists of dirty rags soaked with
solvent and waste ink. Dirty rags are laundered, are disposed as trash, or are
incinerated. If a facility employs a professional laundry, then the rags are picked up,
laundered, and returned. The fate of the solvent and ink contained in these rags would
depend on how the laundry is designed and operated (e.g. water vs. dry cleaning). Most
waste inks are either incinerated or discarded with the trash.
* Graphic Arts Technical Foundation 1985: Personal communication.
B12-11
-------
8. WASTE GENERATION RATES
The only published estimates of the nationwide waste generation rates from the
graphic arts industry were given for 1972 (Bohn 1976). It was estimated that the
industry as a whole generated 7,300 short tons of waste excluding waste paper and
photoprocessing chemicals. Separate listings by type of operation analyzed were not
provided and more recent estimates have not been reported. Fractional waste
generation rates for the lithographic printing industry were estimated by project staff
based on the available information, engineering judgement, and input from the
industry. These values, excluding trash attributable to scrap or waste paper, are
shown in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The list of individual primary lithographic waste streams and their sources along with
a list of source reduction methods is presented in Table 9-1. Recommended waste
reduction methods and identified procedures are discussed in the following sections.
The basis for identification came from published accounts in the open literature and
industry contacts.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
procedures or institutional policies which result in a reduction of waste. The following
items highlight the scope of good operating practices:
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
B12-12
-------
o Loss prevention practices
Spill prevention
Preventative maintenance
Emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Trash
Even excluding spoiled paper, trash represents a major wastestream in the lithographic
industry. This wastestream includes used photographic film, empty containers and
packages, and out-dated materials. The following waste reduction measures were
noted:
o Recycle empty containers.
Most ink containers are scraped free of ink and discarded in the trash.
Since the degree of cleanliness is a function of operator effort, the amount
of ink discarded can vary widely. By purchasing ink in recyclable bulk
containers, the container can be returned to the ink supplier for refilling
instead of being thrown away. In addition, the use of bulk containers also
cuts down on the amount of cleaning required since the surface area of the
container per unit volume of ink stored is reduced.
o Recycle spoiled photographic film and paper.
It is already a current practice of the industry to send used and/or spoiled
film to professional recyclers for recovery of silver*. However, this option
might not be practical to small scale producers or available to facilities
located far away from recyclers.
o Electronic imaging and laser platemaking.
Since all text and photos are edited on a video terminal, the need for
photographing, editing, and re-shooting is reduced. In addition, color
separations can be produced electronically and therefore may eliminate the
need for the many photoprocessing steps currently employed.
* R.R. Donnelley & Sons 1985: Personal communication.
B1Z-13
-------
o Installation of web break detectors.
The Oxy-Dry Corporation manufactures an electronic system that detects
web breaks in a non-contact fashion that will neither smear ink nor crease
the web, thereby reducing waste from these sources. Both McGraw-Hill
Publications in New York, New York, and the St. Petersburg Times in St.
Petersburg, Florida have installed web break detection systems and have
reduced waste (Campbell and Glenn 1982).
o Monitoring press performance.
Crosfield in Chicago, Illinois, markets the Pressdata 190 waste system
which provides a means of monitoring press performance.
o Better operating practices.
Photosensitive film and paper storage areas should be designed for
economical and efficient use. Some shops waste up to one-fourth of these
materials due to improper storage (op. cit.)
9.1.2 Wastewater
Wastewater comes from two main sources: image processing (including plate
developing) and the platemaking process. Other sources are clean-up operations and
spent formation solutions. Since these wastes are produced by many different
processes and therefore require different minimization procedures, each operation is
discussed separately.
Image processing (and plate developing) The photographic industry has made great
strides in pollution abatement and many of the methods developed by this industry are
applicable to the printing industry. The waste reduction methods are as follows:
o Use silver-free films.
Napp Systems, among several other companies, is marketing silver free
films for lithography (Campbell and Glenn 1982).
o Use water-developed lithographic plates.
3M is marketing its Hydrolith plate which requires only water to process
aluminum off-set plates (op. cit.).
B12-14
-------
Electronic imaging and laser platemaking.
Text and photos are read by an electronic scanner, edited with a display
monitor, and non-silver plates are made using laser beams*. Due to the
expense of this system, it is currently being tried by only the largest of
printers, e.g. newspapers. USA Today currently employs this system.
Recover silver and recycle spent chemicals.
Basically, photoprocessing chemicals consist of developer, fixer, and rinse
water. Keeping the individual process baths as uncontaminated as possible
is a prerequisite to the successful recycling of these chemicals. Silver is a
component in most photographic films and paper and is present in the
wastewaters produced. Various economical methods of recovering silver
are available (e.g. metallic replacement, chemical precipitation,
electrolytic recovery) and a number of companies market equipment that
will suit the needs of even the smallest printing shop. Technologies for re-
use of developer and fixer are available. These technologies make use of
various methods such as ozone oxidation, electrolysis, and ion exchange.
Employ counter-current washing.
Counter-current washing, as opposed to the parallel tank system, reduaes
processing solution contamination thereby increasing the ease of recycling
process baths and reducing the quantity of make-up chemicals required. In
a parallel system, fresh water enters each wash tank and effluent leaves
each wash tank. In counter-current rinsing, water from previous rinsings is
used in the initial film-washing stage. Fresh water enters the process at
the final rinse stage, at which point much of the contamination has already
been rinsed off the film. The main consideration in a counter-current
system is availability of space.
Use of squeegees.
Squeegees in non-automated processing systems can reduce chemical
carry-over typically by 50 percent on film and paper from one process bath
to the next by wiping off excess liquid. Minimizing chemical
* Printing Industries Association 1985: Personal communication.
B12-15
-------
contamination of process baths increases ease of recycling, enhances the
lifetime of the process baths, and substantially reduces the amount of
replenisher chemicals required. Most firms, however, use automated
processors*.
o Better operating practice.
Taking care to accurately add and monitor chemical replenishment of
process baths will cut down chemical wastage. Some process chemicals
have expiration dates. If there are not enough orders to consume these
chemicals before they expire, these would go to waste. Pre-mature
expiration of light-sensitive chemicals may be prevented by keeping them
in the dark.
Easily oxidizable process baths may be protected from quickly losing
potency by reducing their exposure to air. Small-scale photo developers
store their chemicals in closed plastic containers and use glass marbles to
bring the liquid level to the brim each time liquid is used. This way, the
amount of oxygen available to the chemical is reduced to a minimum,
thereby extending the chemical's useful life.
Wastewater from platemakinq Only the newspaper sector of lithography still widely
employs platemaking. Waste reduction can be accomplished using the following
technique:
o Removal of heavy metals from wastewater.
Hexavalent chromium can be reduced to the less toxic trivalent form by
lowering pH to about 2 with a strong mineral acid and then adding a strong
solution of reducing agent such as ferrous sulfate or sodium bisulfite. Upon
addition of caustic soda or lime, the heavy metals will precipitate out as
hydroxides (Latus 1976). This is applicable to older technologies that are
being phased out but would not apply to newer systems**. Moreover,
hooking up holding tanks to the sink and whirler will prevent the pollutants
from reaching the municipal sewage system without prior treatment.
* R.R. Donnelley and Sons 1985: Personal communication.
** Printing Industries Association 1985: Personal communication.
B12-16
-------
9.1.3 Clean-up Solvents and Waste Ink
The clean-up solvent waste stream consists of waste ink, ink solvents, lubricating oil,
and solvent. In many printing establishments, excess ink and solvent is collected in a
drip pan underneath the press. This waste is typically drummed and hauled away to a
landfill area. The following waste reduction methods are noted:
o Recycle waste ink and clean-up solvent.
Ink recovery machines are commercially available in a number of sizes.
For firms which choose not to recycle waste ink onsite, it can be sent back
to the ink manufacturer who may turn it into black newspaper ink
(Campbell and Glenn 1982, Huisingh et al. 1985).
If waste cleaning solvent is generated in substantial quantity, it may be
recycled on-site or sent to a professional solvent recycler. In many cases,
however, a printing shop does not generate enough solvent waste to justify
onsite recycling. Waste solvent can be reclaimed through simple batch
distillation or can be incinerated with recovery of heating value.
Clean-up is usually done by wetting a piece of rag with solvent and wiping
the equipment to be cleaned, or by pouring some solvent on the equipment
and wiping it. Drip pans are placed underneath equipment to collect waste
solvent. The dirty rags are then sent to professional cleaners for cleaning
or drummed for disposal. In many cases, the nature of the waste does not
lend itself readily to recycling efforts. However, as indicated by the
example of better operating practice' given below, waste handling
procedures can often be changed to facilitate recycling.
o Use of automatic cleaning equipment can promote a more efficient use of
cleaning solvent*. However, this equipment is very expensive to purchase
and maintain.
* R.R. Donnelly and Sons 1985: Personal communication.
B12-17
-------
o Use of an automatic ink leveller.
Oxy-Dry Corporation is one company that produces this equipment which
maintains the desired ink level in the fountain for optimum inking
conditions. This prevents ink waste and ink spoilage around the press
(Campbell and Glenn 1982).
o Substitution with less toxic solvent.
In some cases it may be possible to substitute highly toxic aromatic
solvents, such as benzene or toluene, with less toxic straight chain
paraffinic solvents, e.g. hexane.
o Better operating practices.
Rexham Corporation of Matthews, N.C., does high-tech printing and
coating, including film substrate, for the photographic industry. Toluene is
used to clean the ink from the press, and runoff toluene is collected as
waste. Rexham has nearly eliminated its toluene waste by segregating
used clean-up toluene according to the color and type of ink contaminant
and then reusing the collected wastes to thin future batches of the same
ink. The procedure has no effect on product quality and has resulted in
almost 100% reuse of the toluene solvent (Huisingh et al. 1985).
9.2 Implementation Profile
Most of the measures undertaken by the printing industry to reduce waste center
around recycling of materials and the implementation of good operating practices.
The technical and commercial viability of recycling waste ink is evidenced by the
commercial availability of ink recovery machines. Portable and non-portable models
are available to suit the different needs and preferences of printers. For waste ink
that is contaminated, turning it into black newspaper ink may be more practical than
preserving the original color. Implementation of good operating practices either
assures that materials can be more readily recycled or avoids the need for recycling
altogether.
B12-18
-------
The installation of silver, ink, and/or solvent recovery equipment (among other
methods) requires significant capital outlays and entails possible increases in labor
costs. The economic feasibility of making these changes depends in large measure
upon the quantity of recoverable/recycleable material generated by a facility. Small-
volume waste ink generators may find internal or on-site recycling less economical
than sending their ink runoff to ink manufacturers who can convert it into black
newspaper ink or to recyclers who collect from groups of small-volume generators.
Large-volume generators, who comprise approximately 20 percent of all waste ink
generators, will likely find on-site recovery systems to be more cost-effective. The
availability of greater financial and personnel resources serves to enhance the
potential for cost-effective on-site recovery and recycling for large-volume
generators.
9.3 Summary
The sources of waste from the conventional lithographic printing process and the
associated source control techniques are summarized in Table 9-1. Spoiled paper and
paper wrap were excluded from this summary so as to provide a better focus on the
wastes of concern. The ratings listed in the table are based on a scale of 0 to 4 and
are used to evaluate each technique for its waste reduction effectiveness, extent of
current use and future application potential. The ratings were derived by project staff
from the available information and from industry comments.
It appears that the current level of waste minimization in the lithographic printing
industry is high. This is evidenced by the current reduction index (CRI) of 2.5
(63 percent) which measures the extent of reduction in the waste that otherwise would
be generated if none of the listed methods were applied as they are currently.
The potential for future reductions appears modest to significant, as evidenced by the
future reduction index (FRI) of 0.7 to 1.4 (18 to 35 percent). The future reduction
index is the measure of waste reduction achievable through implementation of the
listed techniques to their full rated potential.
Ways in which these reductions can be achieved would be to recycle empty ink
containers and increase the use of counter-current rinsing and good operating
practices. Measures that have helped the industry to achieve a high degree of
812-19
-------
TABLE 9-1 SUNNUtr OF SOURCE CONTROL HETHOOOL06Y FOR THE PRINTING OPERATIONS INDUSTRY
Haste Stream
Trash
Hastewater
Clean-Up Solvent
and Haste Ink (<
All Sources
1 1
1 Control Methodology |-
i i
|t. Recycle empty containers 1
|2. Recycle spoiled photographic film |
|3. Electronic Imaging/laser platemaking |
|4. Install web break detectors |
|5. Monitor press performance |
|6. Better operating practices |
| Overall |
|1. Use silver free films |
|2. Use water developed litho plates |
|3. Electronic imaging/laser platemaking |
|4. Recover silver and recycle chemicals |
| 5. Use counter-current washing sequence |
| 6. Use of squeegees I
|7. Better operating practices |
|8. Remove heavy metals from Hastewater I
| Overall |
|t. Recycle waste ink and solvent I
)|2. Use of automatic cleaning equipment |
|3. Recovery of heating value from waste |
|4. Use of an automatic ink leveller |
|5. Use less toxic solvent |
|6. Better operating practices |
| Overall |
| All Methods
Found Documentation
Quantity | Quality
0 I
1 1
1 I
1 1
1 1
1 I
0 83 | 0
1 I
' 1
1
1
1
1
1
1 I
1.00 | 1
2 I
1 I
1 1
1 1
0 1
2 1
4
1.17 | 1
1
— !
1
I
o l
1
1
1
1
1
83 |
1 1
t I
1 1
t 1
2 1
1 1
t 1
2 1
4.
25 I
2 1
1 1
1 1
1 I
0 I
2 I
n |
Haste |
Reduction 1
Effectiveness |
3 1
3 1
2 1
' 1
1 1
2 1
2.00 |
1
1
1
1
1
1
2 1
3 I
2.63 |
2 1
2 1
« 1
1 I
0 I
• I
2 17 |
Extent of |
- .. i
current use |
1
0 1
3 1
1 I
1 1
1 1
1 1
1.17 |
1 1
3 1
1 I
3 I
2 I
1 1
2 1
1 I
1.75 |
2 1
1 1
1 1
1 I
2 I
2 I
1.50 |
Future | Fraction of I
11. ..J ITfclut 1
Application [ total waste |
Potential | |
2 1 1
1 1 1
2 1 1
2 1 1
2 1 1
« 1 1
2.17 | 0.36 |
' 1 1
2 1 1
2 1 1
' 1 1
2 1 1
1 1 1
3 1 1
2 1 1
1.75 | 0.24 |
3 1 1
1 1 1
1 1 1
2 1 1
2 1 1
3 1 1
2 00 | 0.40 |
1 i.oo |
Current |
n J
Reduction |--
Index |
0.0 |
2.3 I
0.5 i
0.3 |
0.3 |
0.5 |
2.3 |
0.3 |
3.0 |
0.5 |
3.0 |
2.0 |
0.3 |
1.0 I
0.8 |
3.0 |
1 0 |
0.5 |
1.0 |
0.3 |
0.0 |
2.0 |
20 |
2.5 |
Future Reduction Index
Probable | Maximum
1.5 | 1
0.2 |
0.8 |
0.4 |
0.4 I
1.5 | t
0.8 | 1
0 2 |
0.5 |
0.8 |
0.3 |
1.0 |
0.2 |
0.8 |
1.1 | 1
0.6 | t
0.8 |
0 4 |
0.8 |
0.4 |
0.0 |
1.5 | 1
1
0.6 | 1
0.7 | 1
1
1
4.
•5 1
1
|
5 1
5 1
,
|
I
1
1
1
.1 1
4-
•1 1
4
1
1
1
1
1
5 1
5 1
1
< 1
CD
(—>
NJ
I
(*) These waste streams include listed "F" and/or "K" RCRA wastes.
-------
reduction (but appear to have limited potential for further implementation) include
recovery of silver from film and process solutions and switching to water-developed
lithographic plates.
10. SUBSTITUTION ALTERNATIVES
10.1 Ink Substitution
Conventional inks used in lithography are heat-set solvent-base inks which contain
coloring pigments and about 30 to 60 percent low boiling organic solvents. Most of the
solvent is removed during the drying step and is either emitted into the air, recycled,
or incinerated. New types of ink are described below. These were formulated with air
pollution reduction (particularly solvent emissions reduction) and energy conservation
as a goal.
o Water-base inks.
Also called water-borne inks, these inks are usually pigmented suspensions
in water. These inks find their greatest application in flexographic printing
on paper substrates (Campbell and Glenn 1982) and their use is also
recommended fos gravure* (water-base inks are not available yet for
lithography).
One factor stifling the development of water-base inks is that they require
more energy to dry than do solvent-borne inks. Another difficulty results
from the necessity to shut presses down for short periods of time. During
this period the ink dries, and since water is not a solvent for the dried ink,
more frequent equipment cleaning is required. Other problems besetting
water-borne inks are low gloss and paper curl.
o UV-cured inks.
These inks consist of one or more monomers and a photosynthesizer that
selectively absorbs energy. Benefits of using UV inks are that the inks
contain no solvent. The paper is not heated above 50° C, and a minimum
of moisture is lost in the process (Carpenter and Milliard 1976, Shahidi and
National Printing Ink Research Institute 1985: Personal communication.
B12-21
-------
Rowanda 1975, Bassemir 1974). Since the inks do not "cure" until exposed
to UV light, and may therefore be allowed to remain in the ink fountains
(and plates) for long periods of time, the need for clean-up is reduced. UV
inks are particularly attractive for letterpress and lithography applications.
The following have been cited as advantages of using UV inks for sheet-fed
lithography (Carpenter and Milliard 1976):
- Elimination of "set-off", the unintentional transfer of ink to adjacent
sheets before the ink has dried up completely.
Elimination of the use of powders that are applied to protect an ink
film that is "set" but not "dry".
Elimination of the storage of printed sheets for ventilation required in
oxidative drying processes.
The disadvantages of the UV-cured inks include:
Cost (75 to 100 percent more expensive than conventional heat-set
inks).
Hazards of UV to operating personnel.
Formation of ozone by the action of UV light on oxygen.
Conventional commercial paper recycling process will not remove the
UV-cured ink which hinders recyclability of the paper printed with this
method.
Some chemicals used to formulate the ink are toxic.
Electron-beam-dried (EB) inks.
The "low-energy" EB system was developed by Energy Sciences Inc., of
Bedford, Massachusetts. Electri Graphics markets the Electricure system
B12-22
-------
which is claimed to operate at a much lower energy than the EB system.
Many advantages are claimed over the use of conventional heat-set ink and
other curing systems (Perino 1976, De Young 1976). Like UV curing, EB
curing causes resins to crosslink. Inks can be applied as fluid monomers
and rapidly converted to tough, solid polymers. The fluidity of the resin
eliminates the need for solvent. UV ink can be converted to EB-curable,
but little has been done to develop EB-curable coatings for use where UV
does not appear practicable. The disadvantage in EB systems is the
degradation of paper and the generation of x-rays that necessitates
elaborate and expensive operator protection (Carpenter and Hillard 1976,
Anonymous 1975).
Heat-reactive inks.
These inks contain a pre-polymer, a cross-linking resin, and a catalyst. The
catalyst activates at 350°F in the dryer and converts the liquid into a solid
polymeric film via condensation polymerization reactions. Reaction by-
products are principally C^-C^ alcohols, moisture, and small amounts of
formaldehyde. Overall volatile content of these inks is 20 percent or less
of that of the conventional heat-set inks. The smoking tendency was
reported to be practically nil. Because heat is required for drying, these
inks cannot be used on sheet-fed presses. Moreover, they are reported to
permit build up of static electricity in the folding operation (Carpenter and
Hillard, 1976).
10.2 Equipment Substitution
Electrostatic screen printing process.
Also known as pressureless printing, this process was developed by the
Electrostatic Printing Corporation of America and therefore is called the
EPC process (USEPA 1979b). A thin, flexible printing element, with a
finely screened opening defining the image to be printed, is used. An
electric field is established between the image element and the surface to
be printed. Finely divided "electroscopic" ink particles, metered through
the image openings, are attracted to the printing surface, where they are
firmly held by electrostatic forces until they have been fixed by heat or by
chemical means.
B12-23
-------
Other alternatives to conventional printing include "desk-top" publishing
using micromputers with laser printers - development seen by some as
presenting a serious future competition to small printing shops.
11. CONCLUSIONS
The industry has significantly reduced non-paper waste as evidenced by a current
reduction index of 2.5 (63 percent). Estimates indicate that future waste reductions
are expected to be moderate as characterized by a future reduction index of 0.7 to 1.4
(18 to 35 percent). Several methods that appear to be quite effective for the industry
as a whole include implementation of better operating practices, recycling of empty
ink containers, and the continued or increased recycling of silver, solvent, and waste
ink. For those waste reduction measures that could be used by the gravure industry to
reduce waste due to their electroplating operations, the reader is referred to the
process studies on electroplating, metal parts cleaning, and metal surface finishing.
12. REFERENCES
Anonymous. 1973. UV cure cuts pollution, energy use. Envr. Sci. Tech. 7(6).
. 1975. Curing ink with electron beams. Business Week.
March 24, 1975.
1977. Electronic ink curing can reduce costs improving
quality. Inland Printer/American Lithographer. March 1977.
1985. American Printer's top one hundred plus. American
Printer. 194(4): 61-76.
Bassemir, R.W. 1974. UV ink chemistry: paper and paperbook. Amer. Ink. Maker.
December 1974.
Berustein, M. 1977. From splitting atoms to curing inks. Screen Printing. August
1977.
Bohon, R.L. 1976. Lithography: laboratory evaluation of environmental risk. In
Environmental aspects of chemical use in printing operations. EPA-560-1-75-005.
Washington D.C.: U.S. Environmental Protection Agency.
Bruno, Michael H. 1985. Status of printing in the U5A--1985. IARIGAI, 18th
International Conference. Williamsburg, Virginia, U.S.A. (June 2-8, 1985).
Campbell, M.E., and Glenn, W.M. 1982. Profit from pollution prevention; a guide to
industrial waste reduction and recycling. Toronto, Canada: Pollution Probe
Foundation.
B12-24
-------
Carpenter, B.H., and Milliard, G. 1976. Overview of printing processes and chemicals
used. In Environmental aspects of chemical use in printing operations. EPA-560-1-75-
005. Washington D.C.: U.S. Environmental Protection Agency.
De Young, S. 1976. Electricure vs. ultraviolet printing. Amer. Ink. Maker. February
1976.
Huisingh, D., et al. 1985. Proven profit from pollution prevention. Washington, D.C.:
the Institute for Local Self-Reliance.
Latus, S. 1976. Platemaking and its effect on the environment. In Environmental
aspects of chemical use in printing operations. EPA-560-1-75-005. Washington D.C.:
U.S. Environmental Protection Agency.
Magee, B. 1985. Environmental control report. Graphic Arts Technical Foundation.
55:1-4.
Perino, D.A. 1976. Nonscanning thermoionic emission—what?. Paper, Film, and Foil
Converter. March 1976.
Shahidi, I.K., and Powanda, T.M. 1975. Ultraviolet curing: a review of the technology.
Amer. Ink. Maker. January 1975.
Stevenson, G.A., ed. 1968. Graphic art encyclopedia. New York, N.Y.: McGraw Hill
Book Co.
USDC. 1972. U.S. Department of Commerce. Standard industrial classification
manual. Washington, D.C.: U.S. Government Printing Office.
_. 1985a. U.S. Department of Commerce. Bureau of the
Census. Commercial Printing and manifold business forms. In 1982 Census of
manufacturers. MC82-I-27B. Washington, D.C.: U.S. Government Printing Office.
. 1985b. U.S. Department of Commerce, Bureau of the Census.
Greeting cards, bookbinding, printing trade services. In 1982 Census of manufacturers.
MC82-I-27C. Washington, D.C.: U.S. Government Printing Office.
. 1985c. U.S. Department of Commerce, Bureau of the Census.
Newspapers, periodicals, books, and miscellaneous publishing. In 1982 Census of
manufacturers. MC82-I-27A. Washington, D.C.: U.S. Government Printing Office.
_ . 1985d. U.S. Department of Commerce, Bureau of the
Census. In 1982 Census of manufacturers, geographic area series (various states).
Washington, D.C.: U.S. Government Printing Office.
USEPA 1976. U.S. Environmental Protection Agency. Environmental aspects of
chemical use in printing operations. EPA-560-1-75-005. Washington, D.C.: U.S.
Environmental Protection Agency.
. 1979a. U.S. Environmental Protection Agency. Compilation
of air pollutant factors. 3rd ed. Research Triangle Park, N.C.: U.S. Environmental
Protection Agency.
B12-25
-------
• 1979b. U.S. Environmental Protection Agency. Graphic arts;
an AP-42 update. EPA-450-4-79-014. Research Triangle Park, N.C.: U.S.
Environmental Protection Agency.
Zborowsky, J.L. 1976. Current status of Web heat set emission control technology. In
Environmental aspects of chemical use in printing operations. EPA-560-1-75-005.
Washington D.C.: U.S. Environmental Protection Agency.
13. INDUSTRY CONTACTS
G.J. Bender, Technical Director, Manufacturing Engineering, R.R. Donnelley & Sons
Co., Chicago, IL.
Dr. W.D. Schaeffer, Research Director, Graphic Arts Technical Foundation, Inc.,
Pittsburgh, PA.
H.F. George. Exec. Vice-President & Research Director. Gravure Research Institute,
Inc., Port Washington, NY.
Dr. J.W. Vanderhoff, National Printing Ink Research Institute, Lehigh University, PA.
B. Ryerson, Printing Industries Association, Los Angeles, CA.
S. Marshall, Printing Industries of America, Arlington, VA.
R12-26
-------
1. PROCESS: SYNTHETIC FIBERS MANUFACTURE
2. SIC CODE: 2824
3. INDUSTRY DESCRIPTION
By definition, SIC 2824 includes establishments involved in the production of synthetic
(or more commonly called man-made) noncellulosic organic fibers. Cellulosic fibers
such as acetate and rayon and the noncellulosic inorganic fibers such as boron,
fiberglass, and graphite are excluded. Included in SIC 2824 are acrylic and modacrylic
fiber, nylon, olefin (such as polyethylene and polypropylene), and polyester. These four
fiber types comprise more than 99 percent of the total man-made fiber produced in
the United States (Farr 1977).
The manner in which fiber production is carried out involves a small degree of
integration between fiber producer and user. Fiber production begins with the
production of a polymer solution and ends with the packaging and shipment of the fiber
to a converter or downstream processor. It is the converter or downstream processor
who twists the fiber into cord, dyes it, weaves it, and knits it. Fiber manufacturers
who do not produce their own polymer purchase it in flake or pellet form from a
supplier. - Hdwever, the majority of fiber producing operations are fully integrated
with the polymerization operation.
3.1 Company Size Distribution
In the U.S., synthetic fibers are currently produced by 28 companies which own and
operate a total of 71 facilities. The synthetic fiber industry employs approximately
60,000 people (USDC 1985).
3.2 Principal Producers
Table 3-1 lists the ten largest producers of synthetic fibers and their major products.
These ten companies account for about 85 percent of the total capacity of organic
man-made fibers in the United States.
B13-1
-------
Table 3-1 Major U.S. Producers of Man-Made Organic in Fibers in (1981)
Major Products and Capacities (millions Ib/yr.)
Producer
Akzona Inc.
Allied Chemical Corp.
American Hoechst Corp.
Badische Corp.
E.I. du Pont de Nemours
& Co., Inc.
Eastman Kodak Co.
Fiber Industries Inc.
Hercules Incorporated
Monsanto Co.
Standard Oil Co. (Indiana)
Total
Acrylic
--
--
70
317
40
--
--
315
--
742
Nylon
210
455
--
145
1375
--
--
--
560
--
2745
Olefin
15
--
--
--
--
--
--
130
__
215
360
Polyester
147
80
460
--
1623
540
1406
-_
120
--
4376
Source: Chemical Economics Handbook (SRI 1982).
3.3 Geographical Distribution
Of the 71 plants in operation in 1982, 44 were located in the states of North Carolina,
South Carolina, Virginia, and Tennessee (USDC 1985). The remaining 27 plants were
located in several adjoining states such as Georgia, Alabama, and Maryland. These
fiber producers are mostly concentrated in EPA regions III and IV. Figure 3-1 depicts
a geographical distribution of these producers.
4. PRODUCTS AND THEIR USE
Man-made fibers are broken into two major categories which can be further subdivided
as shown below:
acetate
cellulosic rayon
Man-made
noncellulosic - inorganic
organic
B13-2
-------
VIII
CO
1—"
CO
oo
0
izzu o-
2-5
6-10
11-20
Roman numerals show EPA regions
Figure 3-1 Synthetic Fiber Plants in the U.S.
-------
Noncellulosic organic fibers (SIC 2824) include acrylic and modacrylic fiber, nylon,
olefin, and polyester. Modacrylic fiber is composed of at least 85% by weight
acrylonitrile. Nylon is a polyamide in which less than 85% of the amide linkages are
attached directly to two aromatic rings. Olefin is composed of at least 85% by weight
ethylene, propylene, or other olefins. Polyester is composed of at least 85% by
weight of an ester of a substituted aromatic carboxylic acid (including but not
restricted to substituted terephthalate units and parasubstituted hydroxy-benzoate
units).
Fibers are generally marketed as yarn, staple, or tow. Other forms are monofilament,
split film, fiberfill, and nonwoven fabrics. The major end uses of man-made fibers are
shown in Table 4-1. About half of the synthetic fiber produced goes into the making
of knit and woven apparel, about 30 percent goes for home furnishings, and the
remainder for various industrial uses. Over 80 percent of the fiber used in home
furnishings is for rugs, carpets, and carpet backing. Most of the remainder is used for
draperies and upholstery. The major industrial use of man-made fibers is for
automobile tire cord. Other uses of these fibers are automobile seat covers, belting,
electrical wire insulation, hose, recreational surfaces, roofing, rope and twine, sewing
thread, tents, parachutes, sails, tarps, and webbing for outdoor furniture.
5. RAW MATERIALS
Relatively pure raw materials are required to produce man-made fibers because of the
deleterious effect of impurities on the properties of the fiber. The monomers
(described above) are derived from basic petrochemicals such as benzene, butadiene,
ethylene, propylene, and xylene. Table 5-1 presents a list of monomers and other
intermediate products from which the fiber-forming polymers are produced.
Many additives are blended with the polymer before fiber production. Examples are
delustrants, pigments, dyeing assistants, dye receptors, optical brighteners, heat
stabilizers, antioxidant stabilizers, and light stabilizers. Ordinarily, the total amount
of additives does not exceed five percent. Materials added to the fiber to enhance
product utility include lubricating agents, bacteriostats, humectants, anti-static
agents, and other additives. Table 5-2 contains a list of some additives added to man-
made fibers to increase their usefulness.
B13-4
-------
Table 4-1 Synthetic Fiber Product and Their Use in million tt>s./yr.
CD
i
vn
Acrylic/
Modacrvlic
Apparel
Knit
Woven
Home Furnishing
Carpets/Rugs
Curtains arid
Draperies
Upholstery
Other
Industrial
Rubber Industry
Nonwovens
Rope and Cordage
Fiberfill
Other
Total
Staple
and
Tow
467
461
6
111
n.a.
14
9
88
7
1
--
1
5
585
Filament,
Monofilament
383
324
59
740
715
1
24
1
366
232
1
28
„.
105
Nylon
Textured
Yard
149
147
2
12
12
6
2
1
..
3
Stable
and
Tow
17
3
14
659
632
25
2
35
13
--
33
Total
549
474
75
1,411
1,347
1
49
15
407
234
14
29
--
130
2,367
Filament,
Texturized
Yarn
1,667
1,054
613
152
--
62
13
77
310
191
1
13
--
105
Polyester
Stable
and
Tow
1,096
318
778
513
120
48
10
335
461
150
2
205
309
Total
2,763
1,372
1,391
665
120
110
23
412
771
191
150
15
205
414
4,199
Polyolefin
Yarn, Mono Stable
filament and and
Fiber Film Tow
11
11
-- - -
326 77
287 62
39 15
_ — — —
275 38
1
31 35
84
--
159 3
Total
11
11
*" ~
403
349
54
313
1
66
84
~~
162
727
Source: Chemical Fconomics Handbook (SRI 19B2).
-------
Table 5-1 Input Raw Materials for Production of Organic Man-Made Fibers in 1982
Raw Material
Consumption
million Ib/yr
Acrylonitrile
Caprolactam (nylon 6)
Adipic acid + hexamethylenediamine (nylon 6,6)
Glycols (ethylene, propylene, etc.)
Dimethyl terephthalate (DMT)
Terephtalic acid (TPA)
743
840
1860
1362
1560
1386
Source: 1982 Census of Manufacturers (USDC 1985).
Table 5-2 Typical Additives Used in Organic Man-Made Fiber Production
Delustrants
Optical brighteners
Antioxidant stabilizers
Light stabilizers
Dyeing assistants
Lubricants and other finishes
Usually titanium dioxide
Stilbene
Phenyl coumarin derivatives
Alkylated phenols
p-cresols mixed with sulfides
Thio compounds such as dilauryl or
distearyl thio dipropionate
Long-chain alkyl derivatives of
hydroxybenzophenones
2-methyl 5-vinyl pyridine
2-vinyl pyridine
p-vinyl-benzene sulfonic acid
sulphocinamic acid
Polyoxyethylene attached to
aliphatic hydrocarbon
Long-chain alkyl quartenary ammonium
salts
Hydroxyalkyl amine salts of fatty
acids
Aliphatic esters
Hydrocarbons
Fluid silicones
Source: Industrial Process Profiles for Environmental Use (Farr 1977).
B13-6
-------
For some fibers, solvents are mixed with the polymer in order to obtain a liquid
solution amendable to fiber formation. These solvents invariably are recovered and
recycled. Solvent is almost exclusively used for the production of acrylics and
modacrylics. The most widely used solvents are dimethylacetamide,
dimethylformamide, acrylonitrile, acetone, aqueous ZnCl£, aqueous NaSCN,
tetramethylene sulfane.
6. PROCESS DESCRIPTION
Man-made fibers are formed primarily by three different processes: melt spinning,
wet solution spinning, and dry solution spinning. Wet and dry solution spinning are used
for producing acrylic and modacrylic fibers. Melt spinning is used for producing nylon,
polyester, and polyolefin fibers. A summary of production methods for
acrylic/modacrylic fibers is given in Tables 6-1 and 6-2.
Table 6-1 Acrylic Fiber Processes
Spinning
process
Dry
Wet
Wet
Wet
Wet
Wet
Wet
Wet
Solvent
Dimethylformamide
Dimethylformamide
Dimethylacetamide
Aqueous NaSCN
Aqueous HNO3
Aqueous ZnCl2
Dimethyl sulfoxide
Ethylene carbonate
World
capacity, %
22
11
23
23
12
4
3
2
Source: Kirk-Othmer Encyclopedia of Chemical Technology, 3rd. Ed. (Roberts 1980)
B13-7
-------
Table 6-2 Acrylic and Modacrylic Fiber Production Routes
Company
Acrylic Fibers
American Cyanamid
Company
Dow Badische Company
E.I. du Pont de Nemours
& Company, Inc.
Monsanto Company
2 Modacrylic Fibers
OJ
i
00 E.I. du Pont de Nemours
& Company, Inc.
Eastman Chemical
Products, Inc.
Monsanto Company^8)
Typical Chemical
Registered Composition of
Trade Name Fiber
Cresland 89-90% Acrylonitrile
8% Methyl Methacrylate
Zefran 87-90% Acrylonitrile
8% Methyl Acrylate
2-5% Other
Orion 88-91% Arcylonitrile
7% Methyl Acrylate
2-5% Other
Acrilan 88-91% Acrylonitrile
8-10% Vinyl Acetate
2-4% Other
Orion FLR 66-70% Acrylonitrile
30% Vinyl Chloride
0-4% Other
Verel 37% Acrylonitrile
40% Vinylidene Chloride
20% Isopropylarcylamide
3% Methyl Acrylate
Elura 76-78% Acrylonitrile
20% Vinyl Acetate
2-4% Other
SEE 79-81% Acrylonitrile
8% Vinylidene Chloride
9% Vinyl Bromide
2-4% Other
Polymerization
Medium
Solution
Solution
Suspension
Suspension
Suspension
Solution
Suspension
Suspension
Type of Type of
Polymerization Spinning
Operation Process
Continuous Wet
Continuous Wet
Continuous Dry
Batch Wet
Continuous Dry
Batch Dry
Batch Wet
Batch Wet
Spinning
Solvent
Aqueous
NaSCM
Aqueous
ZnCl?
DMF
DMAc
DMF
Acetone
DMAc
DMAc
Source: (SRI 1982).
(a'In addition, Monsanto also has two other modacrylic fibers, known as type 65 and type 67.
-------
Fiber production begins with the production of polymer in a polymerization reactor.
This reactor can be operated either in a batch or continuous mode depending on the
type of polymer being produced. The major route for polymerization of acrylonitrile is
as a water emulsion. Most of the water formed is subsequently used in other
operations. By-products of polymerization also occur with the manufacture of nylon
6,6 and polyester. Most of these by-products also find use in other parts of the
process. For many other polymers, no by-products are formed.
The choice of spinning method employed depends primarily on polymer characteristics
such as melting point, melt stability, and solubility in different solvents. The fiber
formation step is accomplished by the extrusion of polymer in liquid form through fine
orifices called spinnerets. The manner in which the polymer is liquefied determines
the manner in which the extruded filaments are solidified. Figure 6-1 shows a
simplified pictorial representation of man-made fiber manufacturing operations,
excluding the polymerization reaction.
In keeping with its name, melt spinning begins with molten polymer. This process is
used for polymers which can be melted under reasonable conditions without
degradation, such as nylon, polyester, and olefin filaments.
Most melt spinning is integrated with polymerization units that supply the spinning
operation with molten polymer. The molten polymer is metered through accurately
machined gear pumps to filter assemblies consisting of either a series of metal gauzes
or layers of graded sand. The filtered polymer is then extruded under high pressure
and at a constant rate through a nickel or stainless steel spinneret. The extruded
liquid polymer streams are cooled using an air stream and the solid filaments thus
formed converge at a guide to yield spun yarn. For fibers such as nylon 6,6, the
filaments pass through a steam conditioning tube before converging.
Wet solution spinning is used to produce filaments from polymer by extrusion into a
liquid coagulation bath. This production method is usually reserved for the
manufacture of heavy tow, which requires slower spinning and processing speeds than
in melt or dry spinning. This method is used primarily to produce acrylic and
modacrylic tow.
B13-9
-------
MELT
SPIN
TRANSFER
MAAA/W\
©
1 HEATER '
DRY
SPIN
SPINNERET —
AIR STREAM
COOLINE
MET
SPIN
\
SOLVENT
>
DISSOLVER
FILTER
MAKE-UP
SOLVENT
OISSOLVER
FILTER
STEAM,HOT MATER
OR HOT INERT
LIQUID
LUBRICANT
•ASHING
©
LUBHICA -
TION
DRAWING
©
SOLVENT LEAN
VAPOR
SOLVENT RICHt
VAPOR
PROCESS HASTE CATEGORIES
© SOLID KASTE
(5) KASTEHATER
(7) SPENT SOLVENT
Figure 6- 1 Synthetic Fiber Hanufacturing Block Flow Diagram
B13-10
-------
Equipment required for wet spinning includes a solution vessel, a metering pump, a
filter, a spinneret, and a coagulant tank. Aging of the polymer solution (a mixture of
polymer and organic or inorganic solvent) before spinning requires a holding tank. All
wet spinning processes include a washing step immediately after extrusion to remove
solvent and other impurities, and a recovery system to separate the coagulant and
solvent. Four specific methods of wet solution spinning have been described in the
literature (Monorieff 1975). As an example, one of the four methods presented by
Monorieff is listed below:
An 18 percent solution of polymer in dimethylacetamide is spun into a
mixture of 2 parts dimethylacetamide and 1 part water; as the yarn
emerges from the bath it is washed with water, which flows counter-
currently into the spinning bath at such a rate that the bath composition is
kept constant; dimethylacetamide is continually being added by the
polymer as it is extruded; water is being added from the wash.
Dry solution spinning is the third process for converting polymer into filament.
Polymer is dissolved in solvent and the polymer solution is then extruded into a zone of
heated gas or vapor. The volatile solvent readily evaporates, leaving a solidifed
filament which is then further processed. This process is used for easily dissolved
polymers such as acrylonitrile, polyvinyl chloride, or polyurethane.
Gear pumps, filter packs, and spinnerets are used in much the same fashion as they are
in melt spinning. After leaving the spinneret, the solution passes through a spinning
cell which consists of a cabinet about 25 feet long. Hot, solvent-lean, gas or vapor
enters at one end and solvent-rich gas or vapor emerges from the other. An efficient
solvent recovery system is required.
The three basic finishing steps after fiber spinning are lubrication, drawing, and fiber
modification. The application of a lubricant immediately after filament formation
improves subsequent handling and processing. The three main functions of lubrication
are surface lubrication, plasticizing action, and static protection. Lubricants are
generally aqueous solutions, emulsions, or organic liquids. From a holding tank,
lubricant is circulated into feeding traps where it wets ceramic wheel applicators that
coat the fiber surfaces.
B13-11
-------
Drawing introduces molecular orientation to the fiber (this causes crystalization and
increases density), thus producing a stronger fiber. Fibers are drawn by stretching
between pairs of rolls, with the second set moving faster than the first in order to
collect the drawn fiber. Drawn lengths are from 2-7 times spun length. Many fibers
are drawn as an integral part of the spinning process. The drawing process may be
aided by heating the fibers either through direct metal-to-fiber contact, by passing the
fibers through a bath containing a heated liquid, or by passing the fibers through a
steam jet. In some cases, drawing may be combined with other processes such as
cleaning or lubrication.
The function of the fiber modification step is to add attributes necessary for a
marketable product. Modifications include twisting to produce interfilament cohesion,
heat setting or heat relaxation to produce dimensional stability, crimping to add bulk
and resilience, or cutting to produce staple products similar to natural fibers. Other
processing steps sometimes utilized are treating with water-repellant, fire retardant,
or other finishes.
7. WASTE DESCRIPTION
The primary specific wastes associated with the manufacturing of synthetic fibers are
listed in Table 7-1. Relatively little data was available on the environmental impact
of the man-made fibers industry (Farr 1977). The same remains true today. The
polymer materials are not toxic or otherwise hazardous unless heated to temperatures
at which decomposition can occur. Wastes from the fiber industry usually arise from
mechanical treatment of the polymer or are associated with auxiliary materials used
in processing such as solvents, additives, lubricants, or finishes. Companies which use
integrated polymerization spinning systems also produce waste which contains
unreacted monomer.
Most of the solid waste is waste fiber that is generated during startup of the process
or during times of upset. This waste fiber can be generated at the spinning, drawing,
and fiber modification steps. In the past, much of this waste was incinerated or
landfilled. Currently, much of this waste is reprocessed either on-site (converted back
into polymer and respun) or off-site (production of plastic bottles, for example). When
waste fiber contains highly degraded polymer or has become heavily contaminated
with machine oil or dirt, then it is incinerated or landfilled.
B13-12
-------
Table 7-1 Synthetic Fibers Manufacturing Process Wastes
No.
Waste Description
Process Origin
Composition
RCRA
Codes
CO
1.
2.
3.
4.
Solid wastes
Solvent vapor
Contaminated solvent
Contaminated coagulant
bath
Filtration; pump seals; waste
fiber; incineration residue
Evaporation from fiber;
scrubber loss
Solvent purification and
recovery
Wet spinning
Polymer, filter sand;
fiber finishers; Ti
others
Solvents (see Section 5)
Solvents (see Section 5)
Water; glycerol; CaCl2;
dimethylacetamide;
others
5.
6.
7.
8.
9.
Contaminated drawing
bath
Polymerization by-
products
Wastewater
Spills/leaks
Air emissions
Wet drawing
Nylon 6,6 polymerization,
polyester from TPA
Coagulant bath treatment;
polymerization by-product
Undefined
Scrubber; steam jets; tank
breathers; evaporation from
fiber
Water; glycerol; CaCl2;
dimethylacetamide; others
Contaminated water
Contaminated water
Water
Mostly solvent and in-
cineration products
-------
Wastewater is generated whenever the level of contaminants in the inorganic
coagulant, washing, or drawing batch becomes an unacceptable and purging is required.
Usually, drawing bath purges can be treated and used as make-up to the washing bath.
In addition, washing bath purges can also be treated and used as make-up to the
coagulant bath provided a water-based coagulant is being used. Many of the impurities
(solvent carried into the washing bath by the fiber, for example) can be recovered
during treatment and be recycled to the appropriate operation. When the need for
water in a previous step does not exist, the wastewater can be used to formulate
finishes. Since the fiber absorbs some water during processing, there is usually a
constant demand for water in the finishing operation. Only the purging of the
inorganic coagulant bath produces a major wastewater steam that requires treatment
and subsequent disposal.
Wastewater ia also generated as a by-product of certain polymerization reactions.
With the production of nylon or the polymerization of acrylonitrile, most of the water
produced is used elsewhere in the facility. With polyester, the by-products depend on
the starting monomer. If terephtalic acid is used, the by-products are water and
glycol. If dimethyl terephthalate is used, the by-products are methanol and glycol.
Glycol is recycled within the polymerization process while methanol is usually sold for
reprocessing back into DMT or used as fuel.
Fiber manufacturers try to limit the escape of organic solvent from their processes to
the fullest degree practical because of economic, safety and regulatory compliance
considerations. For this reason, most solvent losses are either accidental or occur at
very low levels of concentration (scrubber exhausts), where further recovery becomes
unfeasible. Bottoms from the solvent recovery operation used in wet spinning is
usually incinerated. When inorganic solvents are used, the contaminated solvent is
discharged to the facilities wastewater treatment system.
8. WASTE GENERATION RATES
As stated earlier, very little data was available on specific waste generation rates
from synthetic fiber production facilities. Waste stream information was given mainly
in the form of general comments related to production of a particular fiber. It
appears that aqueous emissions from wet spinning processes (coagulation bath purges)
represent the largest potential source of emissions. Next, is by-product water from
polymerization.
B13-14
-------
To a great extent, purge rates from coagulation baths depend on the type of bath and
the age of the manufacturing plant. Glycol baths are less likely to be discarded than
calcium chloride baths, due to the greater value of the glycol. The age of the plant
comes into play because early designs did not incorporate features for reusing
coagulation baths to the extent that those features would be incorporated today.
By-product water from polymerization, by its nature, must contain organic
contaminants. These contaminants would conist of the monomers used in producing
the polymer. Table 8-1 gives wastewater generation rates for nylon 6,6 and polyester
from TPA. The values are based on the stoichiometry of the chemical reactions. A
fair amount of this water is used where water containing small quantities of organic
matter is not objectionable (n some finishes, for example).
Table 8-1 Stoichiometric By-Product Water
By-Product Water
Fiber (gal/10-5 Ibs fiber)
Nylon 6,6 9.5
Polyester from TPA 22.7
*
Wastewater emissions from some operations have been classified according to waste
load and treatability (Farr 1977). This information is summarized in Table 8-2.
Analysis of samples from a settling pond at an acrylic fiber production facility
indicated the presence of acrylonitrile (100 mg/1), 2,3-dibromo-l-propanol (0.5 mgl/1),
an isomer of dibromopropene, and 2,4-dimethyl-diphenylsulfone.
Table 8-2 Summary of Wastewater Data for Selected Fibers
Wastewater Raw Waste Loads (Ib 103 product)
Suspended
Fiber (gal/103 Ib fiber) BOD COD Solids
Nylon 0.16-3.71 0.1-60 0.2-90 0.1-6.0
Olefin (polypropylene) 1.00-1.71 0.4-1.1 1.8-2.6 0.2-2.2
Source: Industrial Process Profiles for Environmental Use (Farr 1977).
B13-15
-------
Overall current waste generation rates for the fiber manufacturing industry were not
in evidence at the time of final document preparation. Fractional waste generation
(the fraction each waste represents of the total waste generated) was estimated by the
project staff based on the available data and engineering judgement. These values are
shown in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
A summary of the waste sources and the corresponding source reduction methods is
given in Table 9-1. This section deals with the description of the listed methods,
including known application cases.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
B13-16
-------
9.1.1 Solid Wastes
Fiber spinning operations generate solid wastes including filter residues and waste
fibers. The filter residue wastes are generally minor compared to waste fiber. Waste
fiber comes mostly from process upsets or during spinning machine startups.
Mechanical fiber modification operations, such as twisting, crimping, or cutting, also
produce solid waste in the form of particulates and waste fiber. Several methods for
waste reduction through source control were noted:
o Recycle waste fiber.
Waste fiber that does not contain surface finish materials is relatively easy
to reprocess. In the case of polyester, it can be sold to a PET
(polyethylene terephthalate) bottle reprocessor. In the case of all thermo-
plastics, this fiber can be remelted and respun, but a loss of quality
generally results due to thermal degradation. At least one company
markets a processor for chopping up thermoplastic waste and pelletizing it.
Clean acrylic waste is easily redissolved and respun; the economics of this
are so clearly in favor of it, in fact, that it is widely practiced.
Waste fiber containing surface finishes is often reprocessed by washing the
fiber in detergents, chopping the wet fiber, and flash drying it. In those
cases where waste fiber cannot be recycled due to surface contamination,
the causes for surface contamination should be identified and eliminated.
o Increased automation
The revolution now underway in electronics has spawned a new generation
of instruments such as tenacity gauges, optical inspection devices,
differential thermal analysers, etc. that can be employed to catch process
upsets at an early stage.
o Good operating practices
Good operating practices include determination of the causes for process
upsets and their subsequent minimization, examination of start up
procedures, identification of devices and methods that will cut waste
production, and commitment to an effort to develop additional methods of
B13-17
-------
salvaging off-grade fiber. An accounting system (or checklist) for
identifying origins of waste fiber appears worthy of consideration.
9.1.2 Wastewater
Fiber spinning also results in the generation of liquid wastes, primarily from wet
spinning. A large portion of liquid waste occurs when the coagulation bath is entrained
on fiber surfaces into subsequent wash stages. Liquid wastes are also generated in the
drawing and chemical fiber modification operations and during the polymerization
reaction of the polymer. For ways in which aqueous wastes associated with equipment
cleaning can be minimized, the reader is referred to the study on equipment cleaning
contained in this appendix. Methods for minimizing aqueous waste include:
o Redesign of washers.
The greatest potential for improved water economy stems from the use of
better washing methods (Masselli 1973). Washers should be designed to use
only the amount of water necessary for a particular step or operation.
Rodney-Hunt, manufacturers of the tersitrol washer, compared their
washer to two tight-strand washers and suggested that up to 85 percent
less water use may be achievable (USEPA 1974). •
o Use multistage counter-current wash system.
By using a multistage counter-current wash system, a large reduction in
the amount of water required for washing can be achieved. Since the
system produces a much more concentrated effluent stream, it is often
practical to concentrate the stream (by means of evaporation) and recycle
the material. In the wet spinning process, coagulant is entrained on the
fiber and carried over into subsequent wash baths. By producting a wash
water wastestream as concentrated as possible, the feasibility of recycling
this waste back to the coagulation bath can be improved.
o Use oil and lubricant substitutes.
Carding oils and antistatic lubricants can be replaced by mineral oils with
nonionic emulsifiers and other low-BOD substitutes (Masselli and Burford
1956). This change improves the treatability of the lubricant waste
generated. In addition, this method improves the treatability of the
B13-18
-------
In dry spinning, the recycled solvent does not build up objectionable impurities so that
it can be reused indefinitely. Solvent losses occur almost exclusively through
evaporation. These losses can be minimized by reducing the leakage of non-
condensibles during spinning.
In wet spinning, contaminated solvent becomes a problem only in isolated cases.
Usually, the preferred method for dealing with this material is to sell it back to the
solvent supplier or to install on-site distillation equipment.
9.2 Implementation Profile
Reduction of the waste load generated by fiber manufacturers can be accomplished to
a large extent through the employment of modern instruments and equipment,
advanced processing techniques, and careful identification of the origins of waste.
Water usage control techniques have been extensively studied and documented;
manufacturers of synthetic fibers are expected to have applied these to a high degree
methods to reduce their wastewater load. Additional waste reduction will come from
improvements in operating practices -these are the easiest to apply.
9.3 Summary
A summary of the source reduction techniques along with the associated ratings is
given in Table 9-1. The ratings represent the assessment of the relative usefulness of
the proposed techniques. Each method is rated on an integral scale of 0 to 4 with
respect to its waste reduction effectiveness, extent of current use, and future
application potential. Based on these ratings, current and future waste reduction
indices are derived to provide a measure of current and future extent of waste
reduction for each technique, each waste stream, and the entire process.
A current reduction index (CRI) of 2.3 (58 percent) is indicative of the high level of
waste reductions achieved by the synthetic fibers industry. CRI represents the ratio
of the amount of waste that was reduced to the amount that would be generated if
none of the measures listed were used at their current level of application. By
implementing additional source reduction techniques, the amount of waste currently
being generated can be reduced to a modest extent, as evidenced by a future reduction
index of 0.5 to 0.8 (13 to 20 percent). As seen from Table 9-1, the most effective
B13-19
-------
TABLE 9-1 SIMWRY OF SOURCE COHTROL HETHODOL06Y FOR THE SYNTHETIC FIBERS MNUFACTURIN6 INDUSTRY
'J3
. . ^
ho
1 I Found Documentation | Waste | Cxtent of future [ fraction of 1 Current
1 1 Quantity | Quality | Effectiveness | | Potential i I Index
Solid Wastes |1. Recycle waste fiber | 2| 2| 21 3 | 1| | 1
|2. Increased automation | 1 | 1| 3| 2 3| | 1
|3 Better operating practices 1 | 1| 3| 3 | 3 | | 2
I Overall 1.33 I 1 33 I 2 67 | 2 67 | 2.33 1 0.10 | 2
— +_
1
r
1
5 1
5 1
3 1
3 1
Wastewster |1. Redesign of washers 2 | 1 | 2 | 1 | 2 | | 0.5 |
|2. Use counter-current wash system | 2| 11 3 | 2 2| | 15|
|3. Use oil and lubricant substitutes 1 | 1 | 1 | 2 I t I | 0.5 |
|4 Fractionate by-product water | 0 i 0| t| 1 1 | | 03|
|5. Increased automation ! 1 | l| 2| 2| 21 | 10|
|6. Better operating practices | 1 | 1 | 3 | 3 3 I | 2.3 |
| Overall | 1 17 | 0.83 I 2.00 | 1.83 | 1 S3 | 0.85 | 2
Contaminated |1. Maintain solvent recovery / recycling) II !| t \ 3| 1| | 3
I Overall t.OO ( 1.00 I 4.00 I 3.00 1 1 00 1 0.05 1 3
All Sources | All Methods ' I 1.00 | 2
+ + 1 + . . . .
3 1
0 1
0 1
3 1
Future
Probable
0
1
0
0
0
0
0
0
0
0
0
0
0
0
Deduction Index
| Maximum
.' i
,ii 11
.6 1 1
6 | 1.1 |
3 | 0 8 |
8 | OS
1 1 1
2 1
.5 1 1
.6 1
.5 | 0.8
.3 | 03|
3| 03|
5 | 0.8 1
{*) These streams include listed T' and/or "K" RCRA wastes
-------
measures that can be used to achieve this reduction include better water usage e.g.
through redesign of washers, use of multistage counter-current washing, increased
automation, and implementation of better operating practices. Current measures that
have been very effective include recycling of waste fiber, recycling of solvent, and
good operating procedures.
10. PRODUCT SUBSTITUTION ALTERNATIVES
From the earliest days of rayon, the demand for man-made fibers has continually
increased. Combined production of polyester, nylon, and acrylic fibers was up about
14% from 1982 to 1983 (Greek 1983). This is a clear indication that man-made fibers
will continue to increase in demand.
v
No viable alternatives to man-made fibers have been identified; that is to say, man-
made fibers possess certain physical properties (e.g. wear and wrinkle resistance)
which have enabled them to compete effectively with certain natural fibers such as
cotton and wool. Presently, it would be unrealistic to assume that this trend will ever
reverse.
11. CONCLUSIONS
Generally, very little documentation was found concerning the extent to which fiber
producers have implemented waste reduction measures. Estimates indicate that
solvent recovery, waste reprocessing, and wastewater reductions are practiced widely
for economic reasons. It should be noted that each individual fiber production facility
may have its own specific production and finishing steps. Therefore, the extent of
reduction possible in one facility may differ greatly from that in another facility. For
the synthetic fiber industry, it is estimated that modest future reduction in waste
volumes are possible. The largest reductions in waste will center on ways of decreasing
wash water use. The industry has done much in the area of reducing solid and solvent
waste by recycling these materials.
12. REFERENCES
Dixit, M.D. 1972. Practical Reuse of Water in the Textile Industry. Colourage I-IV.
April 20, 1972.
B13-21
-------
Fair, J.L. 1977. Industrial process profiles for environmental use; chapter 11. the
synthetic fiber industry. Radian Corp. EPA-600-2-77-023k. Cincinnati, Ohio: U.S.
Environmental Protection Agency.
Greek, B.F. 1983. Chem. Enq. News. May 30, 1983. p.11.
Huisingh, D., Martin, L., Hilger, H., et al. 1985. Proven profit from pollution
prevention. Washington, D.C.: The Institute for Local Self-Reliance.
Masselli, J.W. 1973. Textile Waste Treatment, Past, Present, and Future. AATCC
Symposium. Washington, D.C.: AATCC.
Masselli, 3.W., and Burford, M.G. 1956. Pollution sources from finishing of synthetic
fibers. Boston, Mass.: New England Interstate Water Pollution Control Commission.
Monorieff, R.W. 1975. New Jersey Department of Environmental Protection, Division
of Waste Management. Source reduction of hazardous waste. Seminar Proceeding at
Douglas College, Rutgers University on August 22, 1985. New Jersey: N.J. Depart-
ment of Environmental Protection.
Roberts, W.J. 1980. Fibers: Chemicals. In Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed. Vol. 10, pp. 148-66. New York, NY: Wiley.
SRI, 1982. Stanford Research Institute. Synthetic Fibers. In Chemical Economic
Handbook. Menlo Park, Calif. Stanford Research Institute.
USDC. 1985. U.S. Department of Commerce, Bureau of the Census. Plastic
Materials, Synthetic Rubber, and Man-made Fibers. In 1982 Census of manufacturers.
MC82-I-286. Washington, D.C.: U.S. Government Printing Office.
USEPA. 1974. U.S. Environmental Protection Agency, Office of Technology Transfer.
Upgrading textile operation to reduce Pollution. Vol. 1. In-plant control of pollution.
EPA-625-3-74-004. Cincinnati, Ohio: U.S. Environmental Protection Agency.
13. INDUSTRY CONTACTS
Dr. G.J. Hollod, Senior Environment Engineer, Petrochemical Department, E.I. du Pont
Nemours and Co., Wilmington, DE.
BJ3-22
-------
1. PROCESS: SYNTHETIC RUBBER MANUFACTURE
2. SIC CODE: 2822
3. INDUSTRY DESCRIPTION
The industry consists of establishments primarily engaged in manufacturing synthetic
rubber by polymerization or copolymerization of monomeric feedstock. An elastomer,
for the purpose of this classification, is a rubber material capable of vulcanization,
such as copolymers of butadiene and styrene, or butadiene and acrylonitrile, polybuta-
diene, chloroprene rubbers, and isobutylene-isoprene copolymers. Butadiene copoly-
mers containing less than 50 percent butadiene are classified in SIC 2821. Natural
chlorinated rubbers and cyclized rubbers are considered as semi-finished products and
are classified in SIC 3069.
3.1 Company Size Distribution
The synthetic rubber industry consists of five major producers accounting for about 75
percent of the total production in the United States. More than 55 percent of the 77
synthetic rubber plants employ less than 20 employees each. Table 3-1 shows the
company size distribution of the synthetic rubber industry for 1982.
Table 3-1 Company Size Distribution
No. of employees per facility
Total 1-19 20-49 50-99 100-999 1000+
No. of establishments 77 43 4 6 22 2
No of employees 11,800 200 600 no data 11,000 no data
Source: 1982 Census of Manufacturers (USDC 1985).
3.2 Principal Producers
Table 3-2 lists the major producers of synthetic rubber along with their annual
capacity. The types of rubber considered are styrene-butadiene rubber (SBR),
polybutadiene rubber (PBR), neoprene or chloroprene rubber (CR), ethylene-propylene
B14-1
-------
Table 3-2 Principal Producers of Synthetic Rubber
Company and Annual Capacity (Thousands of Tons)
Plant Location SBR PBR CR EPR IIR IR NBR
American Synthetic Rubber Co.
Louisville, KY 70 50
Cities Service Co., Inc.
Lake Charles, LA 38
Copolymer Rubber & Chemical Co.
Baton Rouge, LA 125 5
Addis, LA 33
Denka Chemical Co.
Houston, TX 27
E.I. du Pont de Nemours Co.
Laplace, LA 36
Louisville, KY 134
Beaumont, TX 77
Exxon Co.
Baton Rouge, LA 60 85
Bayton, TX 105
The Firestone Tires Rubber Co.
Pottstown, PA 3
Akron, OH 50
Lake Charles, LA 310
Orange, TX 45 110
The B.F. Goodrich Co.
Orange, TX 74 22
Port Noches, TX 152
Akron, OH 14
Louisville, KY 28
The General Tire & Rubber Co.
Borger, TX 40
Mogadore, OH 15
Odessa, TX 83
The Goodyear Tire & Rubber Co.
Calhoun, GA 7
Houston, TX 395 16
Beaumont, TX 115 60
Akron, OH 3
Philips Petroleum Co.
Borger, TX 96 68
Polysar Resins, Inc.
Chattanooga, TN 30
Texas - U.S. Chemical Co.
Port Noches, TX 183
U.S. Steel Co.
Scotts Bluff, LA 15
Uniroyal, Inc.
Geismar, LA 44
Paineville, OH 16
ToT^I17624 420 197 236 228 125 82
Source: Chemical Economics Handbook (SRI 1980).
(a) Plant not operating at present.
B14-2
-------
rubber (EPR), butyl or isobutylene-isoprene rubber (IIR), polyisoprene or isoprene
rubber (IR), and nitrile or acrylonitrile-butadiene rubber (NBR).
3.3 Geographical Distribution
Since most synthetic rubber is used for tires and tire products, the industry is
concentrated near automobile assembly plants in Michigan or close to raw material
sources in Texas. There are twelve establishments in Texas, six each in Michigan and
Louisiana, and four each in New York, New Jersey and Ohio. The location of
establishments in the U.S. is shown in Figure 3-1 and Table 3-3 below.
Table 3-3 Location of Facilities by EPA Region
EPA Region
I
II
III
IV
V
VI
VII
VIII
IX
Number of
Establishments
--
8
1
8
10
18
--
--
—
National 77
Source: 1982 Census of Manufacturers (USDC 1985).
4. PRODUCTS
In 1977, it was estimated that about 65 percent of the synthetic rubber produced was
used for tires and tire products (Parr, Parson, and Phillips 1977). The annual
production rates for some significant types of rubber are listed in Table 4-1.
Table 4-2 shows the use pattern of these same types of rubber.
B14-3
-------
VIII
0
0-1
2-5
4 6-10
11-20
Roman numerals show EPA regions
Figure 3-1 Synthetic Rubber Plants in the U.S.
-------
Table 4-1 Annual Production Rates (TRY) of Various Synthetic Rubbers
Type
Styrene-Butadiene
Rubber
Polybutadiene
Rubber
Neoprene/Polychloroprene
Rubber
Ethylene-Propylene
Rubber
Butyl Rubber
Nitrile Rubber
Abbreviation
SBR
PBR
CR
EPR
IIR
NBR
Production
TPY
1,420,000
958,000
378,000
359,000
109,900
130,000
187,400
215,000
142,000
76,000
67,000
Year
1978
1984
1978
1984
1975
1984
1983
1984
1979
1979
1984
Source: Chemical Economics Handbook (SRI 1980), Chemical and Engineering News
(Stimson 1985).
Table 4-2 Use Pattern of Various Synthetic Rubbers
Application
Total
Percent Consumption
SBR
PBR
CR
IIR
100
100
100
100
100
IR
NBR
Tires and tire
products
Automotive
application
Mechanical goods
Latex applications
Seals and gaskets
Hoses
Footwear
Wire and cable
insulation
Impact modifier
Others
63
7
18
10
--
--
--
__
--
2
84
28
5 35
__
__
__
-_
13
11
24
9
34
3
--
3
6
1
7
13
24
75
--
--
--
10
--
_-
__
--
15
55
--
13
--
3
5
10
—
--
14
--
--
--
10
25
25
5
--
35
100 100
Source: Chemical Economics Handbook (SRI 1980).
B14-5
-------
5.
RAW MATERIALS
Monomers
Styrene, butadiene, chloroprene, isobutylene, isoprene,
acrylonitrile, ethylene, propylene, 1,5-cyclo octadiene,
dimethyl siloxane, diisocyanates, ethylene glycols,
ethylidene norborene, 1,4-hexadiene, dicyclopentadiene,
methylene norborene, etc.
Initiators
Potassium peroxy disulfate, benzoyl peroxide, azobis
isobutyronitrile, cumene hydroperoxide, p-menthane
hydroperoxide, butyl lithium, nitrogen hepta sulfuri-
mide, formamidine sulfinic acid, potassium persulfate.
Chain transfer agents
N-dodecyl mercaptan, tert-dodecyl mercaptan,
thiuranes, xylene solution of tetraethyl thiuram
disulfide.
Emulsifiers
Aqueous solutions of sodium stearate, sodium rosinate,
disodium salt of ethylene diamine tetracetic, acid, zinc
stearate, sodium formaldehyde sulfoxylate.
Terminators
Solvents
Sodium dimethyl dithio-carbamate, sodium nitrite,
sodium polysulfide, hydroquinone.
N-hexane, pentane, heptane, methyl chloride, naphtha,
cyclohexane, benzene, toluene, xylene, chlorobenzene,
propylene.
Catalysts
Butyl lithium, vanadium tetrachloride, vanadium oxy-
trichloride, diethyl aluminum chloride, diisobutyl
aluminum chloride, ethyl aluminum sesquichloride, tri-
ethyl or triisobutyl aluminum, TiCl4, Til4, CoCl2,
A1C13, BF3.
B14-6
-------
Activators Ferrous sulfate heptahydrate, disodium salt of ethylene
diamine tetracetic acid, sodium formaldehyde sulfoxy-
late, zinc stearate.
Fillers Carbon black, sulfur, aromatic (staining) oils, naph-
thenic (nonstaining) oils, paraffin wax, clay, whiting,
titanium dioxide, silica, hydrated alumina, polyvinyl
chloride.
Curing agents Tetramethyl thiuram disulfide, mercaptobenzothiozole,
zinc dibutyl diltuis carbamate, ferric dimethyl dithyio
carbamate, sulfur, dicumyl peroxide, zinc oxide, cal-
cined magnesia, etc.
Reagents Potassium chloride, potassium hydroxide, brine, sulfuric
acid, chlorine.
6. PROCESS DESCRIPTION
Synthetic rubber is manufactured mostly by emulsion polymerization, solution poly-
merizations, or slurry polymerization. Table 6-1 lists some synthetic rubbers and their
corresponding monomers along with the polymerization schemes used. Butyl rubbers
are produced by a slurry process which is considered to be similar to the solution poly-
merization process from a waste generation standpoint. For this reason, the slurry
process is not considered separately in this study. The following sections deal with the
emulsion and solution polymerization processes.
6.1 Emulsion Polymerization Process
In 1979, the emulsion polymerization process was used to produce 90% of the styrene-
butadiene rubber and almost all nitrile and neoprene rubber. Since styrene-butadiene
rubber accounts for more than 60 percent of all synthetic rubber produced in the U.S.,
emulsion polymerization is the most widely used method.
In emulsion polymerization, the monomers are emulsified in a medium, such as water,
along with emulsifying agents such as soaps and synthetic emuisifiers. The
B14-7
-------
Table 6-1 Production Methods for Various Synthetic Rubbers
oo
I—'
.p-
I
00
Rubber
SBR
PBR
CR
EPR
I1R
IR
NBR
Monomers
Styrene,
butadiene
Butadiene
Chloroprene
Ethylene,
propylene,
and dienes
Isobutylene,
Isoprene
Isoprene
Acrylonitrile,
butadiene
Polymerization
Scheme
Emulsion (90%)
Solution (10%)
Solution
Emulsion
Solution
Slurry
Solution
Emulsion
Temperature
<°F)
112 or 41
112
--
104
95
148
4
112 or 41
Percent
Conversion
60-75
98
70-90
91
--
70-95
--
75-90
Product
Composition
23%
styrene
98% cis
--
40-50%
Ethylene
3-9%
diene
--
98% cis
30-40%
Acrylonitrile
Residence Pressure
Time (atm)
8-15 hrs 4-5
1-2
--
..
--
30-60 min. 2-4
-.
5-12 hrs 8-9
Source: Kirk-Othmer Encyclopedia of Chemical Technology (McGrath et. al. 1979) and Assessment of Industrial Hazardous Waste
Practices: Rubber and Plastics Industry (Kushnir and Nagy 1978).
-------
polymerization reaction is started by the addition of a water soluble initiator, and can
be stopped at a desirable point by the addition of chemicals called shortstops (or
terminators).
The first step in emulsion polymerization, as shown in Figure 6-1, is the removal of
inhibitors. Inhibitors are added to some monomers to prevent premature
polymerization during transport and storage. This removal is achieved by washing the
monomer with a caustic solution. The caustic solution is usually recycled to the
washing process until it becomes saturated with inhibitors and is then transferred to
settling pits where it is mixed with other process effluents. The settling pits are used
to remove suspended solids present in some effluent streams. After sufficient settling
time, this stream is further treated to reduce its biological oxygen demand (BOD).
The uninhibited monomers are fed to a polymerization reactor containing water and
emulsifying agents. Typical reactors have sizes varying from 5,000 to 30,000 gallons.
The emulsion is created by agitators or by recirculating the mixture through
centrifugal pumps, as in the case of neoprene rubber production. Initiators, catalysts,
activators, and stabilizers are added to the reactor to start the polymerization
reaction. The polymerization reaction is highly exothermic and proper temperature
control is achieved by heating/cooling coils through which a heat transfer fluid is
circulated. Additional cooling is achieved by condensing the vapors emitted from the
reactor and returning the condensate to the reactor.
Polymerization is stopped at the desired degree of conversion by the addition of
chemicals, called short stops, in aqueous solution. The polymerized monomer of latex
is then sent to a vacuum flash unit where light monomer, along with water vapor, is
separated out as overhead. This vapor stream is condensed and the light monomer is
recovered by decantation to be recycled to the polymerization process. The water
stream generated is sent to wastewater treatment. The noncondensible impurities are
sent to a flare.
Unreacted heavy monomer present in the latex from the vacuum flash unit is removed
by steam stripping the latex. Again, the vapor stream from the stripper is condensed
to remove the heavy monomer by decantation, which is recycled to the process.
Another aqueous waste stream is generated in this separation step.
B14-9
-------
ACTIVATOR
4 MONOMEH RECYCLE
CAUSTIC
| INHIBITED l
; MONOMER |INHIB^.
j STORAGE IHOMOH.H
1 i
*
MASH
DRUM
Cz}
UNINHIBITED
MONOMER
STORAGE
MODIFIER ! A ^
CATALYST SOAP j j [
SOLNS SOLN.
1 1 1
r^ i
• POLYMER- ~
~*| IZATION
: REACTORS
i ,
v .,,_.. .0
I
SHORT
STOP
LIGHT | HEAVY '
PARATION T | SEPARATION
T' i
' ^VACUUM l^\ STE*K
j FLASH 1 STRIPPER
|
H I
J4 STEAM
1
^ V !
i I i . .TEY LATEX
V ^ V,, „,, ,„ o LATEX i &
A STORAGE 1 PRODUCT
ANTIOXIOANT | j
KATER
! '" 1
j RINSINS
{ OEHATERIN8 1
I ©i
DRYIN6
(3)
A
BALING AND
RUBBER |
PRODUCT ,
STORAGE i
PROCESS HASTE CATEGORIES!
(T) OFF-GRADE PRODUCTS j
•ASTEMATER
VENT 6ASES
I
Figure 6-1 Eaulsion Polyserization R'ocess for Synthetic Rubber Manufacture
814-10
-------
The latex, thus freed of monomers, is mixed with antioxidants and sent to latex
storage, where it can be sold as a product. When the product is to be sold as bales, the
rubber particles dispersed in the latex need to be solidified. This is done by
coagulating the rubber particles, which causes them to settle faster due to their
increased size. To achieve this separation, the latex is sent to a coagulation tank
where it is mixed with brine. The brine causes the latex to "cream," which is a partial
flocculation of the rubber particles. This flocculation is caused by the breaking of the
emulsion, resulting from the altered ionic interaction between emulsion components.
The emulsifiers are thus separated from the rubber particles, though the particle size
of the rubber is still too small for rapid settling. Addition of dilute sulfuric acid
causes these particles to agglomerate. Sometimes extender oils in aqueous emulsion
and carbon black in slurry form are added to the latex prior to coagulation. The
particles from the coagulation liquid are separated on a shaker screen. The liquor is
recycled with fresh brine and acid make-up. The liquor overflow (blowdown) is a
significant waste stream and is sent to settling ponds before being sent to waste
treatment. The screened rubber particles from the coagulation tank are slurried in
water and sent for further processing.
The screened rubber particles slurried in water are rinsed with more water and are
then filtered and dewatered in rotary vacuum filters. The filtrate is sent to a reslurry
tank where some of the liquid is discharged to waste treatment and the remainder is
recycled to the rinsing unit. The dewatered crumbs are dried using hot air in
continuous or screen dryers. The dry product is weighed and formed into bales using
hydraulic balers and sent to product storage.
6.2 Solution Polymerization Process
Solution polymerization is the other major process for synthetic rubber manufacture,
used mostly for the production of polybutadiene, polyisoprene, ethylene-propylene
rubbers, and part of styrene-butadiene rubbers. The use of solution polymerization for
SBR (the major synthetic rubber produced), is expected to grow because SBR made by
solution polymerization has better rolling resistance and other desirable qualities
(McNaughton 1983).
In solution polymerization, the monomers are dissolved in a solvent, such as an
aromatic, aliphatic, or alicyclic hydrocarbon. Addition of catalysts of the Zeigler-
B14-11
-------
Natta or Friedel-Crafts type starts the polymerization reaction. As the polymer
forms, it precipitates out since it is insoluble in the solvent. The reaction can be
stopped at any time by the addition of shortstops.
The solution polymerization process is shown schematically in Figure 6-2. The
removal of any inhibitor present in the monomer is achieved in a manner similar to
that used in the emulsion polymerization process. The uninhibited monomers are then
mixed with the solvent and the mixture is dried in a desiccant column. The solid
desiccant used for drying is regenerated or discarded. The regeneration step may
produce some aqueous waste and the discarded desiccant is a solid waste.
The dried monomer is sent to the polymerization reactor where a catalyst, which can
be a complex alkyl of metals such as titanium or aluminum, is added to start the
polymerization. The reaction produces heat, which is removed by cooling coils in
which a suitable refrigerant is circulated. The reaction is terminated at the desired
point by the addition of shortstops.
The polymer slurry from the reactor is then treated with liquids such as aqueous
alcohol. The alcohol allows the transfer of the catalyst to the aqueous phase, leaving
the now catalyst-free polymer slurried in the solvent. The aqueous alcohol/catalyst
solution is treated to precipitate the catalyst as metallic oxides, and recover solvent
and alcohol to be recycled to the process. The metal oxides, suspended in water, are
mixed with other effluents and sent to waste treatment. The polymer slurry is mixed
with antioxidants and extender oil and is stored in a suspension form called rubber
cement.
The rubber cement is sometimes mixed with carbon black slurry and then contacted
with hot water (for butyl rubbers) or steam (for others). This operation removes the
solvent and unreacted monomers while coagulating the rubber at the same time. The
vapor stream from this process is distilled to recover solvents and monomers which are
recycled to the polymerization process. Distillation produces an aqueous waste stream
and an organic stream containing oily slops. The aqueous stream is either sent directly
to wastewater treatment or is steam-stripped before discharge. The oily wastes are
used as a fuel or processed for further purification. The polymer crumb slurry thus
generated is dewatered, screened, dried, and baled by processes similar to those used
for producing bales in the emulsion polymerization process.
B14-12
-------
SOLID DESICCANT
MONOMER
RECYCLE
NONON.
©
r
em veur . &
f <
L
SOLVENT
LI6HT
MONOMER
REMOVAL
OLVENT
t
SOLVENT
REMOVAL
SOLVE
RECYC
NT
IE
MONONE
RECYCl
HEAVY j
MONOMER
REMOVAL I
ANTI- EXTENDER
HATER — — fr
CARBON ,,
HATER
CD
~i
r-*
/ N
V
SLURRY ^
STEAM STRIPPING
AND
COASULATZON
1 STEAM
[ 1
, , A 1 i
DEKATERIN6 JQ pRODUCT ,
jv tun . ,„ .Jv UH'lNb *„ oil Tiir i r»J '
*H »HD " ,0v DALIN6 e- STORA6E 1
SCREENING Cl> | 1
& \ \
PROCESS HASTE CATEBORIES'
0 OFF-8RAOE PRODUCTS
(?) HASTEKATEH
(a) HEAVY ENDS AND HASTE OILS
® SPENT CATALYST
(D SPENT ADSOftBEWT
© VENT BASES
Figure 6-2 Solution Polynerization Process for Synthetic Rubber Manufacture
B14-13
-------
7. WASTE DESCRIPTION
Table 7-1 lists the primary wastes associated with synthetic rubber manufacturing
both by emulsion and solution polymerization.
Off-grade products, which amount to 1 to 3 percent of total production, are the result
of bad batches and product changes. This stream contains rubber particles with
different degrees of contamination suspended in water (in the case of emulsion
polymerization), or a solvent (in the case of solution polymerization). This stream is
usually mixed with other effluents and sent to settling pits where the solid particles
separate out. These solids are removed periodically and are reused, landfilled, or
incinerated depending on the degree of contamination.
Periodic cleaning of the equipment results in aqueous and organic wastes. The solid
particles suspended in water are usually recovered at the settling ponds, as are the
heavy ends and waste oils present in the waste water. Since solution polymerization
involves the use of solvents (and hence requires a water-free environment) very little
water would be used for cleaning equipment used in the solution process. Solvents
used for cleaning can be collected and reused in the process.
Other aqueous wastes result from inhibitor removal, monomer and solvent recovery,
coagulation dewatering, rinsing, screening and carbon black slurry preparation opera-
tions. The inhibitor removal achieved by caustic wash generates an aqueous stream
containing phenols. As this stream is usually reused until it is completely saturated,
its discharge is intermittent. Solvent and monomer recovery operations consist of
condensation, decantation, and distillation of the vapor overheads from the reactor.
Since this aqueous stream contains a significant amount of dissolved organics, it is
usually steam-stripped to remove the volatiles before being mixed with other
effluents. Coagulation of the latex in emulsion schemes is achieved by the addition of
brine and sulfuric acid solutions. The coagulation overflow is the major portion of all
the aqueous wastes. Overflow from carbon black slurry preparation is an aqueous
waste containing suspended carbon particles. All the above streams are mixed and
sent to settling pits where the solids and oils are removed. The aqueous effluent from
these pits is sent to treatment plants where the biological oxygen demand (BOD) is
lowered to acceptable levels.
814-14
-------
Table 7-1 Synthetic Rubber Manufacturing Wastes
Waste
Description
Process
Origin
Composition
RCRA
Codes
Off-grade products
Equipment cleaning
wastes
Other aqueous waste
03
Heavy ends and waste
oils
Spent catalyst
Spent adsorbent
Spills
Gaseous emissions
Bad batches, production
changes
Washing operations on
various equipment
Monomer solvent separations,
rinsing, filtering, de-
watering, coagulation,
screening, inhibitor removal,
carbon black slurry making
operations
Monomer and solvent recovery
operations in solution
polymerization, oils
separated from aqueous
wastes
Catalyst removed in solution
polymerization
Feed drying in solution
polymerization
Various process equipment
and piping
Leaks, drying and stripping
operations, adsorbent
regeneration in solution
polymerization
rubber particles containing
traces of unreacted monomers,
solvents, catalysts and
other additives
rubber particles in water
containing organics
rubber particles, carbon
black, and various
organics in water
Monomers, solvents,
extender oils, etc.
Oxides of titanium, aluminum
Traces of monomers, solvent
in solid adsorbent
Monomers, solvents, oils, etc.
Monomers, solvents, etc.
-------
Heavy ends and waste oils are generated by the solvent and monomer recovery
operations, and by the entrainrnent of extender oils in the coagulation overflow. The
heavy ends generated in the distillation operations contain significant hazardous
organic materials. This stream is mixed with waste oils recovered from the settling
pits and incinerated or drummed for land disposal.
Solution polymerization requires the removal of catalyst. This catalyst is usually
removed by an undisclosed process in the form of metallic oxides suspended in water.
This stream is expected to be small and is probably mixed with other aqueous streams
and sent to settling pits, where the catalyst particles along with other suspended
material settle out. The solids are periodically removed and disposed of as described
previously.
The monomer and solvent dehydration operations in solution polymerization are
accomplished by using solid desiccant. This adsorbent is regenerated periodically and
discarded eventually as a solid waste. This waste is expected to be small and
information on the disposal methods was not found.
Spills that occur in various parts of the plant due to accidental discharges generate
another waste stream. This stream is usually mixed with other streams and disposed
of as explained before. The gaseous waste streams are generated by fugitive emissions
at various parts of the plant, air drying of rubbers, and non-condensables vented from
flashing and stripping operations. The gas streams are generally vented to the
atmosphere in high-level stacks or sent to a flare for incineration.
8. WASTE GENERATION RATES
Current waste generation rates were not in evidence at the time of final document
preparation. Table 8-1 lists 1974 nationwide waste generation rates for various
synthetic rubbers. Based on a review of the other process studies, it is likely that
these rates have changed dramatically due to RCRA.
B14-16
-------
Table 8-1 1974 Nationwide 1984 US. Waste Generation
Rates for Various Synthetic Rubbers in metric tons per year
to.
1)
2)
3)
4)
5)
6)
7)
8)
Waste Source
Off-grade products
Equipment cleaning wastes
Other aqueous wastes
Heavy ends and waste
oils
Spent catalyst
Spent adsorbent
Spills
Fugitive emissions
Emulsion
SBR
28,600
5,700
incl.(2)
2,900
NA
NA
incl .(4)
NA
Solution
SBR
2,200
2,200
incl.(2)
2,200
NA
NA
incl. (4)
NA
PBR
1,000
NA
NA
50
NA
50
incl .(4)
NA
CR
5,300
NA
NA
10,600
NA
NA
incl. (4)
NA
EPR
Q(a)
NA
NA
Q(b)
30
100
incl. (4)
NA
IIR
O(b)
NA
NA
NA
NA
100
NA
NA
IR
NA
NA
NA
NA
NA
NA
NA
NA
Source: Assessment of Industrial Hazardous Waste Practices (Kushnir and Nagy 1978) and
Industrial Process Profiles for Environmental Use (Parr, Parson, and Phillips 1977).
(a) Material was sold for special use.
(b) Use was found for the material at the production site.
B14-17
-------
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The list of individual waste streams and their sources along with the source reduction
methods are presented in Table 9-1. In addition to the waste reduction measures
classified as being process changes or material/product substitutions, a variety of
waste reducing measures labeled as "good operating practices" has also been included.
Good operating practices are defined as being procedural or institutional policies
which result in a reduction of waste. The following items highlight the scope of good
operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
• - scheduling
o Loss prevention practices
spill prevention
preventive maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
•
Separate listings have been provided whenever case studies were identified.
9.1.1 Off-Grade Products
Off-grade products are the result of operator error or system malfunction. The
product lost due to this waste stream amounts to 1-3 percent of the total production.
This solid waste is either sold as an off-grade product or buried in landfills, depending
on the degree of contamination. These solids contain traces of unreacted monomers
B14-18
-------
and other additives and may pose an environmental hazard. The following source
reduction methods are noted:
o Reduction of polymer buildup on the reactor walls.
During the polymerization reaction, the polymer that forms may build up
on the reactor walls. This impairs heat transfer resulting in poor
temperature control and thus contributes to lower product quality. It may
even result in production of a bad batch. Preventing polymer buildup on
the reactor wall can be done by spraying or rinsing the reactor walls with
certain chemicals as is done with polyvinyl chloride reactors (Wilkins 1977),
using Teflon* liners on the reactor walls to reduce adhesion, or installing
mechanical wall wipers inside the reactor (Wilkins 1977).
Teflon* liners are suggested because the anticipated low adhesion between
Teflon and the polymeric material should reduce buildup. Tank designs
employing dual shaft mixers, one serving the mixer and another one for the
wiper blade, are available commercially and have been used extensively for
viscous fluid mixing applications.
o Using continuous versus batch operation.
From a waste generation standpoint, continuous operation is preferable to
batch operation because it requires less equipment cleaning, and because
the probability is operator error leading to production of a bad batch is
lower. This is due to the higher degree of human involvement required for
a batch operation. Neoprene rubber is usually made by a batch operation
and its manufacture suffers from 3 percent product loss. Continuous
operation for Neoprene rubber and for SBR, have been developed and can
cut down both product loss and waste generation (Huisingh et. al. 1985; Aho
1958; SIR 1968). Most SBR is already produced by continuous operation.
Application of a continuous process concept to Neoprene rubber can be re-
examined in the context of waste generation and reduction.
^Registered trademark of E.I. du Pont de Nemours & Co.
814-19
-------
Increased automation.
By increasing the degree of automation, operator errors that result in bad
batches could be avoided. Automation is typically simpler and less costly
in a continuous process as opposed to a batch, or discrete, process.
Backup systems.
Emergencies, such as power failures or equipment failures, should be
considered in the design or production planning stage and contingency
procedures should be worked out. An interruption of cooling water flow
can cause a run-away reaction which, in addition to being a safety hazard,
may generate large waste quantities. By using diesel or steam turbine-
driven pumps to provide cooling water during a power failure, this can be
avoided. Automatic back-up power systems and chemical injection systems
that spray short stops to stop polymerization are some alternatives. Such
systems are used in the manufacture of polyvinyl chloride (Wilkins 1977).
Finding use for off-grade products.
By blending off-grade products with other rubbers or rubber wastes, a
useful rubber product can be made (Zimmerman 1981; Anonymous 1982a).
This is done extensively for both butyl and ethylene-propylene rubber.
French researchers report a procedure whereby raw materials can be
recovered from the off-grade products (Anonymous 1982b). The process
consists of dissolving the off-grade products in an oil (primarily composed
of aromatic compounds) at 572-716°F. Carbon black and rubber (free of
monomer) can be recovered from this solution.
In another process, off-grade products are depolymerized in a liquid
hydrocarbon medium by means of agitation, heat, free radical initiators
and molecular oxygen (Scott 1972 and USEPA 1975). The rubber-modified
hydrocarbon that results can be vulcanized to produce a moisture barrier or
insulation coating. It can also be blended with asphalt to provide
rubberized asphaltic composition, or thermally decomposed to produce
carbon black.
814-20
-------
9.1.2 Equipment Cleaning Wastes
Equipment cleaning wastes are the result of various washing operations performed on
the reactors and other process equipment. The reactors in the emulsion scheme are
usually washed with water, thereby generating an aqueous waste. In solution schemes,
since the reactors should be kept free of water, organic solvents are used to wash the
reactors. This organic stream can be treated to remove the solvent and the dissolved
rubber can be precipitated to be reused or discarded as a waste. The following
techniques for reducing wastes from equipment washing operations were noted:
o Reduction of polymer build-up on the reactor walls.
Reduction of polymer build-up on the reactor walls by using Teflon* liners
or spraying chemicals on the walls reduces the required equipment washing
frequency and therefore results in less wastewater being produced. Use of
reactors equipped with wipers that continuously clean the walls is another
practical alternative.
o Altering product composition.
Changing the product composition can reduce the concentration of
contaminants in the aqueous waste. For example, SBR typically contains
23 percent styrene and the increasing cost of styrene has prompted some
manufacturers to lower the styrene content to 15 percent (Anonymous
1985). This results in a lower styrene content in the waste wash water.
By increasing the degree of polymerization allowed, rubbers with higher
molecular weights could be produced. This higher degree of conversion
would result in less unreacted monomer ending up in the wastewater
(McGrath et. al. 1979). While these rubbers would be harder to process,
their processability could be improved by adding extender oils. This
practice is widespread in the industry.
o Increased automation.
Increased use of automation results in the reduction of errors and bad
batches. This means a lower wash frequency and thus less waste.
''Registered trademark of E.I. du Pont de Nemours &. Co.
B14-21
-------
o Increase batch size to minimize cleaning frequency.
Reducing equipment cleaning frequency by increasing batch size would
result in less cleanup wastes. Expansion of production capacity is usually
done by adding more batch reactors. By using a larger batch reactor, as
opposed to two smaller batch reactors, the amount of waste generated
from cleaning would be reduced. This trend, observed in the polyvinyl-
chloride (PVC) industry, might prove economical in the synthetic rubber
industry (Cameron, Lundeen, and McCulley Jr. 1980).
o Better operating practices.
Usually, equipment is rinsed with large volumes of water to remove
residual rubber. By initially rinsing the reactors with a small volume of
water (about 5 percent of equipment volume), waste water with a high
concentration of rubber particles is generated. This stream can then be
recycled to the process. The equipment can then be washed with a full
volume rinse and generate wastewater with a lower solid content. This
two-step rinsing procedure was adopted by Borden Chemical Co. and
reduced off-product waste by 95 percent (Sittig 1975b).
A similar procedure was adopted by a plant for cleaning latex tanks. The
plant used a rinse procedure that generated a rinse water containing more
than 2 percent solids. This stream was then used for latex blending (Riley
1974). Rinse water containing less than 2 percent solids is not useful for
blending.
9.1.3 Other Aqueous Wastes
These aqueous wastes are the result of inhibitor removal, coagulation and screening,
rinsing and dewatering, carbon black slurry preparation, and monomer, solvent, and
catalyst recovery steps. The removal of inhibitors from the monomers by caustic wash
generates a waste containing traces of phenol. Since this stream is usually recycled to
the washing process until it becomes saturated, the volume of this waste stream is
quite small. The overflow from carbon black slurry preparation is also small and
contains only the suspended carbon particles.
B14-22
-------
The bulk of the aqueous wastes are the result of coagulation and screening, rinsing and
dewatering, and monomer, solvent, and catalyst recovery operations. All of these
streams are usually sent to settling pits where suspended solids and heavy oils are
removed. Since some of these aqueous streams contain significant amounts of
organics (unreacted monomers, etc.), they may be steam stripped to remove volatile
organics, incinerated, or stored in drums for land disposal. The following methods
could reduce the generation of these aqueous wastes:
o Use of alternate coagulation procedures.
Usually, coagulation of the latex produced by emulsion polymerization is
achieved by the addition of brine and sulfuric acid solution. The overflow
from the coagulation tank contains solid particles which are recovered in
settling pits. This waste stream can be reduced by using alternate means
of coagulation such as spraying the latex into a heated chamber and
thereby accomplishing coagulating and drying the latex in one step.
Another approach is to deposit the latex as a frozen film on rolls kept at
low temperature followed by washing, using ultrasonic sound, or by forcing
the latex through jets for emulsion breaking (Riley 1974). Since all of the
above alternatives achieve coagulation without using increased quantities
of water (brine and sulfuric acid solutions), less aqueous waste is produced.
Using acid polyamide, a coagulation procedure that reduces the quantity of
total dissolved solids in the overflow has been developed. This procedure
was adopted by a synthetic rubber plant which reduced coagulation
overflow wastes significantly. Also reported was a process used by another
plant that converted rubber cement from solution polymerization directly
into baleable crumbs using an extruder. This process was used for
polybutadiene and reduced waste significantly since the traditional method
of coagulation (using hot water or steam) was not required.
o Recycling of vacuum pump seal water.
Vacuum pumps are used for generating vacuum in the light monomer
recovery operation. The water used to seal the vacuum pump accumulates
organics and is usually discharged as a waste stream. Studies have been
done on recycling this aqueous waste to the polymerization process (op.
cit.). Though the quantity of this stream is not known, this procedure could
contribute to overall waste reduction.
B14-23
-------
o Different mode of carbon black slurry preparation.
Usually, carbon black (in the form of a water slurry) is added to the
polymer latex from the reactor. The overflow from the slurry mixer
results in a wastewater stream. 3y using a different slurrying procedure,
some plants have effectively reduced this loss (Riley 1974).
o Altering product composition.
As discussed in section 9.1.2, changing the product composition could
reduce the concentration of contaminants in the aqueous wastes.
o Alternatives to solution polymerization.
Polyisoprene and polybutadiene are produced mostly by solution poly-
merization. Enoxy Chimica 5.P.A. of Italy has developed a polymerization
process which does not use solvents (Zimmerman 1981, Lipowicz 1982).
The process uses a rare earth-based catalyst system at conventional
temperatures and pressures. Currently, a pilot plant is in operation in
Italy. The process has the advantage of eliminating the need for steam
stripping the product which reduces the amount of wastewater generated.
Additional advantages include energy savings up to 80 percent, a much
smaller reactor, and reduction of heavy slops formation.
9.1.4 Heavy Ends and Waste Oils
Heavy ends are removed by the various distillation operations used for the recovery of
monomers and solvents. This stream, sometimes referred to as "heavy slops", is
usually disposed of by drumming for landfill or incineration. This potentially
hazardous stream can be reduced by the following methods:
o Replace the slurry process for butyl rubber with a solution process.
The slurry process for butyl rubber synthesis uses toxic methyl chloride as
a carrier fluid, which ends up in the heavy slops (Parr, Parson and Philips
1977). Though it was suggested that a use could be found for this waste
stream (Kushnir and Nagy 1978), this has not been supported by other
sources. By employing a solution process using a solvent such as hexane or
pentane, the toxicity of this stream could be reduced. A solution process
of this type is believed to be in commercial operation in the USSR (Parr,
Parson and Phillips 1977).
B14-24
-------
o Alternatives to solution polymerization processes.
As mentioned in Section 9.1.3, a non-solvent process developeed by Enoxy
Chimica S.P.A. can reduce formation of heavy slops in production of
polyisoprene and polybutadiene.
o Avoiding the use of aromatic solvents.
Aromatic solvents are usually more volatile and toxic in comparison to
aliphatic solvents. By using aliphtic solvents, the toxicity of the heavy
slops could be reduced (Parr, Parson, and Phillips 1977).
9.1.5 Spent Catalyst and Adsorbent
The catalysts used in emulsion polymerization usually stay in the final rubber product.
In solution polymerization, the catalysts are removed by contacting the polymer slurry
with an alcohol solution and then precipitating the catalyst from the alcohol (Kushnir
and Nagy 1978). The separated catalyst, present as a suspension in an aqueous stream,
is usually mixed with other aqueous wastes and sent to settling ponds. In these ponds,
the catalyst (in the form of oxides of titanium or aluminum), separates out along with
other suspended particles (rubber, carbon black, etc.). All of these solids are removed
periodically and disposed of in a landfill. Since the spent catalyst waste is reported to
be equal to about 0.01 percent of the total product, and since reducing this waste will
not have a significant impact on overall waste generation, no waste reduction
measures are proposed.
In solution polymerization processes, care is taken to keep water out of the system.
This is accomplished by drying the monomer-solvent mixture with a solid desiccant.
After several regeneration cycles, the desiccant is discarded as a waste. The actual
disposal methods for this solid waste were not reported. Elimination of this stream
can be accomplished, in principle, by using a distillation of the feed to remove water,
as is done in the case of EPR. However, it is not clear that the resulting wastewater
stream is environmentally preferable to the existing alternative. Further exploration
of this approach is needed.
B14-25
-------
9.1.6 Spills and Leaks
Spills and leaks are the result of accidental discharges of liquids and solids in various
parts of the plant. The exact amount of this stream is not known. The liquids are
usually mixed with the heavy slops and disposed of as explained previously. The
following suggestions could reduce this waste stream:
o Use of continuous operation instead of batch operation.
Since there are many sequential processing steps in a batch operation,
chances of operator errors that result in inadvertent spills and emissions
are higher. Thus, a continuous operation is preferable.
o Better operating practices.
Latex spills are usually washed down with water and are sent to the waste
treatment plant. Washing with water makes coagulation of this stream
difficult. By using alum to coagulate the latex in-situ and removing the
coagulated rubber solids with scrapers, lower quantities of water are
required for washing (Riley 1974). This reduces the quantity of wastewater
generated due to washing of latex spills.
Losses from loading and unloading procedures can be reduced by the
following method. Two hoses are usually connected to the track of a
railroad car for loading or unloading. The bottom hose transfers material
while the top hose maintains pressure. Material left in the hoses may be
lost or spilled upon disconnection. This can be avoided by more complete
drainage and by purging the lines with inert gas to a recovery system or a
control device prior to disconnect.
9.2 Implementation Profile
Some of the noted source control methods appear to be economically attractive. For
example, by reusing off-grade products or by selling them at a lower price, off-grade
product waste can be reduced, thereby cutting disposal costs and perhaps increasing
revenue at the same time. Segregating wastes (e.g. keeping the spent catalyst stream
separate from the coagulation overflow) involves little or no cost to the facility but
B14-26
-------
can mean substantial savings in disposal costs. Modifying equipment cleaning
procedures results in recycleable waste streams generated with only minor increases in
operating costs.
All other methods identified may require a significant expenditure of capital and an
increase in labor costs. The economic feasibility of these methods depends upon their
potential to produce a savings in disposal and treatment costs and reduce future
liability for landfilled waste. Some of these methods may even be feasible for firms
which have already installed wastewater treatment units, though managers at these
firms are generally reluctant to consider further environmental expenditures. Even so,
source reduction methods are certainly worth considering when planning for plant
expansions or replacements.
Because of the diversity encountered in the synthetic rubber industry, source control
methods feasible at one facility may not be feasible at another. In addition, one
facility may produce many different grades of rubber, making it difficult for a general
and yet meaningful control scheme to be profiled. In practice, managers of each
facility must independently determine their own best course of action with regard to
waste minimization.
9.3 Summary
The summary of all noted source control techniques is given in Table 9-1. Each
technique was rated for its effectiveness, extent of current use and future application
potential on scale of 0 to 4. The ratings were derived by project staff based on review
of the available. The estimates of current level of waste reduction achieved (current
reduction index) and possible future reduction (future reduction index) were obtained
from the ratings in accordance with the methodology presented in the introduction to
this appendix.
The current reduction index (CRI) is a measure of reduction of waste that would be
generated if none of the methods listed were implemented to their current level of
application. For the entire synthetic rubber industry, CRI is 2.1 (53 percent) which is
indicative of the significant level of waste minimization that already has taken place.
B14-27
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY FOR THE SYNTHETIC RUBBER MANUFACTURING INDUSTRY
1
Waste Stream |
1
Off-Grade Products |1
12.
|3
|4.
15
1
Equipment Cleaning | 1
Hastes |2.
|3
|4.
|5
1
Other Aqueous | 1 .
Hastes |2.
|3.
|4.
|5.
I
Heavy Ends and | 1 .
Waste Oils |2.
|3.
1
Spills and Leaks |1.
12.
1
All Sources |
1
Control Methodology |-
1
Reduce polymer build-up on react wal)|
Use continuous versus batch operation)
Increased automation |
Provide emergency backup systems |
Find use for off-grade product |
Overall )
Reduce polymer build-up on react wallj
Alter product composition |
Increased automation |
Increase batch size / reduce cleaning)
Better operating practices |
Overall |
Use alternate coagulation methods |
Recycle vacuum pump seal water |
Different carbon slurry preparation j
Alter product composition |
Use alternatives to solution process |
Overall |
Replace slurry process - Butyl rubber)
Use alternatives to solution process |
Avoid the use of aromatic solvents |
Overall 1
Use continuous versus batch operation)
Better operating practices |
Overall |
All Methods
Found Documentation
Quantity | Quality
1
1
1
1
1
1.00 | t
t 1
1 1
1 1
1 I
1 1
1.00 | 1
2 1
1 1
1 1
1 1
1 I
1 20 | 1
1 I
' I
1 I
1.00 | 1
1 1
1 1
t 00 | 1
1
.... 1
1
1 1
1 1
1 1
t 1
2 1
20 |
|
1
1
1
1
00 |
2 1
1 1
1 1
1 1
2 1
40 |
1 1
1 1
t 1
00 |
1 1
1 I
00 |
Waste |
Reduction I
Effectiveness |
1 1
2 1
2 1
2 1
3 1
2.00 |
2 1
1 |
1 1
2 1
3 1
1.80 |
3 1
1 |
t |
1 |
' 1
1.40 |
2 I
1 1
2 1
1.67 |
2 1
3 1
2 50 |
Extent of |
Current Use |
1
1
1 1
1
3 1
3 1
t 80 |
' 1
3 1
1 1
2 1
2 1
1.80 |
1 1
t 1
1 1
3 1
0 1
1 20 |
1 1
0 1
1 1
0.67
1 I
2 1
1 50 |
Future | Fraction of |
Application I Total Waste 1
Potential | |
2 1 1
1 1 1
2 1 1
1 1 1
2 1 1
1.60 | 0.77 |
2 I I
1 1 1
' 1 1
1 I I
3 I I
1.60 | 0 04 |
2 1 1
2 1 1
' 1 !
t 1 !
1 i I
t 40 | 0.12 |
1 1 1
' 1 i
' 1 1
1.00 | 0 06 i
1 1 1
3 1 1
2 00 ) 0 0! |
i '.DO |
Current
Reduction
Index
0
0
0
t
2
2
0
0
0
1
1
1
0
0
0
o
0
0
0
0
0
0
0
I
I
•3 I
5 !
5 I
5 I
3 I
3 I
.5 I
8 I
3 I
0 I
5 I
5 I
8 I
3 I
3 I
3 I
0 I
8 I
5
0 I
5
5 I
5 I
1.5 |
1
2
5 I
1 1
Future
Probable
0
0
0
0
0
0
0
0
0
0
1
0
,
0
0
0
0
0
0
0
0
0
0
1
o
0
deduction Index
| Maximum
.4 I
4 I
.3 | 0
•1 1
4 1
.4 | 0
•S 1
.1 1
2 1
3 1
.1 | 1
.5 | 1
, , ,
4 1
•2 1
1 1
.3 1
4 | 1
4 | 0
3 1
.4 | 0
3 | 0
4 1
' 1 1
8 | 1
4 | 0
1
1
1
8 1
1
1
.3 1
1
1
1
1 1
1 1
1
1
1
1
1
1 1
4 i
1
4 1
4
1
1
1
8
03
I
hO
CD
(*) These streams Include listed "F* and/or 'IT RCRA wastes.
-------
The future reduction index (FRI) is an indication of the level to which the currently
generated waste can be reduced if all of the techniques noted were implemnted
according to their rated potential. The FRI value of 0.4 to 0.8 (10 to 20 percent) is
indicative of a low to moderate extent of future waste reductions. Among the
techniques that were found most effective and applicable (as evidenced by high FRI
value) for the control of waste generation were increased automation, use of alternate
coagulation methods, and further implementation of better operating practices.
10. PRODUCT SUBSTITUTION ALTERNATIVES
Synthetic rubbers are predominantly used in the manufacture of tires and tire
products. It is estimated that tire replacement amounts to about 75 percent of total
tire sales (Zanetti 1984a). By increasing the lifetime of passenger car tires, the
demand for synthetic rubber can be decreased. This could be done by better consumer
tire maintenance (e.g. by preventing tire under-inflation and by maintaining proper
wheel-alignment), and by increasing car-pool use.
It is estimated that 92% of the passenger tires made in 1990 will be radial tires, up
from 61% in 1980. Since radial tires use more natural rubber, the demand for
synthetic rubber could decrease.
Natural rubber is an accepted substitute for synthetic rubber (the roles were reversed
prior to 1940). Most of the natural rubber is imported; however, domestic rubber
production from guayule plants (found in the southwestern U.S.) has received recent
attention (McNaughton 1983, Lipowicz 1982). The concept of guayule rubber was
explored in the U.S. during the Second World War when natural rubber imports declined
drastically. Though guayule rubber lacks economic feasibility at present, it is
expected to be economical by the end of this decade.
The increasing cost of styrene prompted manufacturers to investigate possible substi-
tution of SBR by medium vinyl polybutadiene rubbers (Parr, Parson, and Phillips 1977).
The waste reduction implications of this substitution are not clear.
EPR can be used for tire-making purposes, replacing other synthetic rubbers. Various
problems associated with using EPR for tire applications have been solved (Parr,
B14-29
-------
Parson, and Phillips 1977). The apparent low waste loads associated with EPR makes
this substitution worthy of exploration.
11. CONCLUSIONS
While the synthetic rubber industry has done much to reduce wastes, it appears that
moderate reductions characterized by a future reduction index of 0.4 to 0.8 (10 to 20
percent) are possible. Several methods that appear to be quite effective would be to
expand the reuse or recycle of off-grade products, increase the use of automation, use
alternate coagulation procedures, and extend the application of good operating
practices. Source reduction methods already in application at some plants need to be
examined at others.
One area that can have an impact is improved tire maintenance by consumers which,
in turn, could increase passenger car tire life and thus reduce the demand for synthetic
rubber. By increased use of radial tires, the amount of synthetic rubber used per tire
could be lowered since radials use more natural rubber than other types of tires.
Natural rubber is the best substitute for synthetic rubber; however, full environmental
assessment of such a substitution must be made, along with an evaluation of
socioeconomic impacts, before its desirability is fully established.
12. REFERENCES
Aho, C. E. 1958. U.S. Pat. 2,831,842. (Apr. 22, 1958) to E.I. duPont de Nemours &
Co., Inc.
Anonymous. 1982a. Chementator. Chem. Eng. 89 (2): 17.
. 1982b. Chementator. Chem. Enq. 89 (14): 17-8.
. 1985. Chem. Enq. News. 63 (23): 26.
Cameron, J.B., Lundeen, A.J., and McCulley Jr., J.H. 1980. Trends in suspension PVC
manufacture. Hydrocarbon Processing. 59 (3): 39.
Greeme, R. 1981. Chem. Enq. 88 (17): 101.
Huisingh, D. et. al. 1985. Proven profit from pollution prevention. Washington, D.C.:
The Institute for Local Self-Reliance.
Johnson, H. 1973. A study of hazardous waste materials, hazardous effects and
disposal methods. Vol. 2, Booz-Allen Applied Research, Inc. EPA-670-2-73-15.
Washington, D.C.: U.S. Environmental Protection Agency.
B14-30
-------
Kushnir, J.M., and Nagy, S.F. 1978. Assessment of industrial hazardous waste
practices; rubber and plastics industry. Snell (Foster D.), Inc. EPA-530-SW-163C-2.
Washington, D.C.: U.S. Environmental Protection Agency.
Lipowicz, M. 1982. Organic and inorganic chemicals see major advances. Chem. Enq.
89 (3): 109.
McGrath, J.E., Vial, T.M., Baldwin, P.P., et. al. 1979. Elastomers, synthetic in
Kirk-Othmer Encyclopedia of Chemical Technology. 3rd ed. Vol. 8, pp. 446-640. New
York, N.Y.: Wiley.
McNaughton, K.J. 1983. Surge in search for synfuels. Chem. Eng. 90 (2): 108.
Parr, J., Parson, T.B., and Phillips, N.P. 1977. Industrial process profiles for
environmental use: chapter 9. the synthetic rubber industry. Radian Corp. EPA-600-
2-77-023L Cincinnati, Ohio: U.S. Environmental Protection Agency.
Riley, J.E. 1974. Development document for effluent limitations guidelines and New
source performance standards for the tire and synthetic segment of the rubber
processing; point source category. EPA-440-l-74-013a. Washington, D.C.: U.S.
Environmental Protection Agency.
Scott, C.E. 1972. U.S. Pat. 3,700,615. (Oct. 24, 1972) to Cities Service Co.
SIR. 1968. Shell International Research. Brit. Pat. 1,136,189 (Dec. 11, 1968) to Shell
International Research.
Shreve, R.N., and Brink, J.A. 1977. Chemical process industries. 4th ed. New York,
N.Y.: McGraw Hill Books Co.
Sittig, M. 1975a. Environmental sources and emission handbook. New Jersey: Noyes
Data Corp.
. 1975b. Pollutant removal handbook. New Jersey: Noyes Data Corp.
SRI. 1980. Stanford Research Institute. Chemical economics handbook, 1978. Menlo
Park, CA: Stanford Research Institute.
Stimson, S.C. 1985. Chem. Eng. News. 63 (15): 33-60.
USDC. 1972. U.S. Department of Commerce, Office of Management and Budget.
Standard industrial classification manual. Washington, D.C.: U.S. Government Printing
Office.
. 1985. U.S. Department of Commerce, Bureau of the Census. Plastics
materials, synthetic rubber, and man-made fibers. In 1982 Census of manufacturers.
MC 82-I-28B. Washington, D.C.: U.S. Government Printing Office.
USEPA. 1975. U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Standard support environmental impact document. Vol. 2. Research
Triangle Park, N.C.: U.S. Environmental Protection Agency.
B14-31
-------
Wilkins, G.E. 1977. Industrial process profiles for environmental use, chapter 10;
plastics and resin industry. Radian Corp. EPA-6QO-2-77-023; Cincinnati, Ohio: U.S.
Environmental Protection Agency.
Zanetti, R. 1984a. Battles for tire markets feature two SBR types. Chem. Enq. 91
(12): 29-33.
_. 1984b. Guayule makes progress toward commercialization. Chem. Enq.
91 (9): 27-31.
Zimmerman, A. 1981. U.S. rubber - flat growth is buoyed by specialities. Chem. Enq.
88 (24): 34-5.
B14-32
-------
1. PROCESS: 1,1,1 -TRICHLOROETHANE MANUFACTURE
2. SIC CODE: 2869(2)
3. INDUSTRY DESCRIPTION
Manufacturers of 1,1,1-trichloroethane (1,1,1-TCE) are included as part of the organic
chemicals manufacturing industry. The industry is composed of relatively large plants
producing a wide variety of chlorinated compounds. The manufacturing processes are
often interrelated, since products from one process can be used as feedstock for
another.
Three major companies are responsible for all of the U.S. production of 1,1,1-TCE.
Dow Chemical is the largest producer of 1,1,1-TCE, with two plants providing 57
percent of the total U.S. production capacity. PPG Industries Inc. and Vulcan
Materials Co., each with one plant, account for 27 percent and 16 percent of the total
production capacity, respectively.
Three of the plants producing 1,1,1-TCE are located in Louisiana. The remaining
operational site, owned by Dow, is located in Texas.
4. PRODUCT AND THEIR USE
Annual U.S. production of 1,1,1-TCE was 334,000 short tons per year in 1984
(Anonymous 1985a). 1,1,1-TCE is primarily used as solvent for oils, grease, and wax in
metal cleaning operations. Commercial grades of 1,1,1-TCE are available with or
without oxidation inhibitors added. Solvent degreasing grades often contain trace
amounts of organic inhibitors, such as acid acceptors or metal stabilizers, which are
used to neutralize any HC1 formed during solvent use, and to deactivate metal
surfaces for the prevention of adverse reactions during the cleaning process.
1,1,1-TCE is also used as raw material for the manufacture of other chlorinated
chemicals, in aerosol propellant formulations, and as a solvent in adhesive and coating
formulations. The relative consumption rates of 1,1,1-TCE are given in Table 4-1
based on 1977 data. The relative distribution of 1,1,1-TCE among the different end
uses appears to have remained unchanged since that date.
B15-1
-------
Table 4-1 End-Use Profile for 1,1,1-TCE in the U.S.
End Use Relative Consumption Rate
Percent
Metal Cleaning 67.0
Chemical Intermediate 2.0
Miscellaneous 31.0
Total 100.0
Source: Organic Chemicals Manufacturing (Key et al. 1980).
5. RAW MATERIALS
The raw materials used to produce 1,1,1-TCE are summarized below (Van Gemert
1982, Stevens 1979, Key et al. 1980):
Reactants hydrogen chlorine, chlorine, vinyl chloride, 1,1-dichloroethylene
(VCM route).
ethane, vinylidene chloride (ethane route).
Catalysts FeCl3
Inhibitors • butylene oxide, dioxane, nitro methane, tert-amyl alcohol,
methyl ethyl ketone, isopropyl nitrate
Neutralizes ammonia, sodium hydroxide
6. PROCESS DESCRIPTION
Production of 1,1,1-TCE in the United States is based primarily on the two-step
chlorination process using vinyl chloride as a raw material. This method accounts for
85 percent of total production. An alternative production method, based on the direct
chlorination of ethane, is responsible for the remaining 15 percent. The ethane
process, however, is also capable of using ethylene dichloride derivatives as feedstock.
Detailed descriptions of the two processes mentioned are given elsewhere (Boozalis
and Ivy 1978, Berrie, Chaler, and Chuffart 1978, Key et al. 1980). Due to its wider
applicability, only the vinyl chloride route will be discussed.
B15-2
-------
In this process, vinyl chloride is first catalytically hydrochlorinated to give
1,1-dichloroethane (1,1-DCE). The 1,1-DCE intermediate is then thermally
chlorinated to give 1,1,1-TCE as the final product. Yield from vinyl chloride often
exceeds 95 percent. The overall reactions include:
1. Hydrochlorination of vinyl chloride to 1,1-DCE
FeCl3
catalyst
CH2 = CHC1 + HC1 ^CH3-CHC12
2. Chlorination of 1,1-DCE to 1,1,1-TCE:
CH3-CHC12 + Cl2 >CH3CCl3 + HC1
Figure 6-1 represents the flow diagram for the production of 1,1,1-TCE from vinyl
chloride. For this process, vinyl chloride, HC1, and recycled chlorinated hydrocarbons
are fed into a hydrochlorination reactor, operating at 95°F-105°F. The reaction is
homogeneously catalyzed by dissolved ferric chloride. Ammonia is then added to the
reactor effluent stream to form a solid complex with the residual HC1 and the FeCl3
catalyst.* The complex is removed through a spent catalyst filter as solid waste. The
filtered hydrocarbon stream then passes through a heavy ends column. Bottoms from
this column, containing tars and heavy chlorinated hydrocarbons, are removed as a
liquid waste. The overhead is transferred to a light ends column for the separation of
1,1-DCE. Overhead from the light ends column, containing mostly unreacted vinyl
chloride, is recycled to the hydrochlorination reactor. The 1,1-DCE intermediate is
removed as bottom from the distillation column and is stored for further processing.
In the second stage of the process, 1,1-DCE and chlorine are combined and fed to a
non-catalytic fluidized bed chlorination reactor, operating at 750°F. The exothermic
reaction is thermally induced and subsequently self-sustaining. The reactor effluent
then passes through an HC1 column, where HC1 and light chlorinated hydrocarbons are
removed as overhead. This stream can be used directly to supply HC1 for the
manufacturing of other chlorinated chemicals, or is further purified to recover HC1 as
a by-product.
* Other means of FeCl3 removal, which do not involve NH3 addition, are also
practiced.
Dow Chemical Co. 1985: Personal communication.
B15-3
-------
HCL
VINYL
CHLORIDE^
FECL3
CATALYST
HYDRO-
CHLORINATION
REACTOR
CD
NH-q
SPENT
CATALYST
FILTER
HVY
ENDS
COL
I
HCL TO
RECOVERY
A
LIGHT
ENDS
COL
CHLOHINATION
REACTOR
HCL
COL.
1,1,1-
TCE
COL
CHLORINE
Cg CHLORINATED
FEEDS TO OTHER
PROCESSES
PROCESS WASTE CATEGORIES
0 HEAVY ENDS
0 SPENT CATALYST
(3) VENT GASES
Figure 6-1 Production of \\\\ i-TCE from Vinyl Chloride
-------
The bottom stream from the HC1 column is fed to the 1,1,1-TCE column, where the
purified 1,1,1-TCE is removed as overhead. Inhibitors are then added to provide
different commercial grades of the final product. The bottom stream from the
1,1,1-TCE column, containing mostly C2 - chlorinated hydrocarbons, is often
transferred as feed to other chlorination processes. Storage tanks and light ends off
distillation column overheads are vented to a scrubber to remove residual HC1 and
chlorinated hydrocarbons, although alternate means of processing are also used.
7. WASTES DESCRIPTION
The primary specific wastes associated with the 1,1,1-TCE manufacturing industry are
listed in Table 7-1. The waste streams generated from this industry include liquid
organic wastes, semi-solid spent catalyst, and scrubbing system wastewater.
Since liquid organic effluents, such as heavy tars, are often used as a feedstock to
other processes in a highly integrated chlorohydrocarbon facility, the term "waste" is
not applicable universally. Organic heavy ends account for a large portion of the
residuals generated. This liquid stream comes from the distillation of the
hydrochlorination reactor effluent stream to remove heavy chlorinated hydrocarbons
that cannot be directly recycled or further processed. The resulting tars are often
oxidized for recovery of HC1 and heat or sent to land disposal.
Another major source of waste comes from the removal of spent catalyst which is used
for the hydrochlorination reaction. Ammonia is added to the reactor effluent stream
to form a solid complex with the catalyst. The solid is then filtered out, and the
resulting sludge is sent to land disposal. Again, it must be noted that certain facilities
use alternate methods of catalyst removal which do not rely on NH3 addition.
The scrubber effluent, containing mostly water with condensed or dissolved
halogenated organics, is normally sent to a central wastewater treatment system for
the entire facility.
B15-5
-------
Table 7-1 1,1,1-TCE Manufacturing Process Wastes
No.
1.
Waste
Description
Heavy ends
Process Origin
Hydrochlorination
reactor
Composition
tars, heavy
chlorinated
hydrocarbons
RCRA
Code
K096
2.
3.
4.
5.
Spent catalyst Spent catalyst
filter
Scrubber
effluent
waste
Emission control
operations
Equipment Tanks, towers,
cleaning wastes heat exchangers
Spills & leaks
FeCl3, ferric K028
ammonium chloride
complex,
dichloroethene
chlorinated
hydrocarbons,
water
F024
8.
WASTE GENERATION RATES
The survey of the available literature did not yield waste generation rates' from the
production of 1,1,1-TCE. Fractional waste generation (the percentage each stream
represents of the total waste generated) was estimated by the report authors based on
the available information and industry comments. These values are shown in Table 9-
1.
9.
WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" have also been included. Good operating practices are defined as
procedural or institutional policies which result in a reduction of waste. The following
items highlight the scope of good operating practice:
B15-6
-------
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified. A
summary of the waste sources and the corresponding source reduction methods is given
in Table 9-1. This section describes the listed methods, including known specific
applications.
9.1.1 Heavy Ends
A major portion of the waste generated from the manufacturing of 1,1,1-TCE comes
from the distillation operation required for the removal of chlorinated compounds
containing more than two carbon atoms. These compounds are formed by various
polymerization reactions during_the hydrochlorination of vinyl chloride to form the
1,1-DCE intermediate. Since these compounds cannot be directly recycled, or readily
used as feedstock to other manufacturing processes, they must be chemically
converted prior to reuse or produced in minimum quantities prior to disposal. The
following minimization techniques were noted:
o Heavy ends hydrocracking.
Similarly to the method proposed in the process study for TCE/PCE
manufacturing, an additional reactor might be used to hydrocrack the
heavy chlorinated hydrocarbons into lighter products. The light chlorinated
B15-7
-------
organics generated from this step can then be recycled to the hydro-
chlorination reactor. The experiments at the Illinois Institute of
Technology revealed severe catalyst fouling problems. Further research is
planned to explore other operating conditions such as the effect of
increasing H2/feed ratios to alleviate the fouling problem*.
o Lowering hydrochlorination reactor temperature.
It is possible that the heavy ends formation reaction has a higher activation
energy than the principal 1,1-DCE formation reaction. If this is true, then
the selectivity ratio would increase with a decrease in temperature. This
would lead to higher yields of 1,1-DCE and, consequently, lower generation
of heavies. Examination of the reaction kinetics along with the economic
feasibility of this proposed method appears worthwhile.
o Thermal oxidation with recovery of HC1 and heat.
This technique is used by at least one facility**. The heavy ends are
combusted and both heat and HC1 are recovered. Recovered hydrochloric
acid can either be sold or used to generate anhydrous HC1 for use in other
processes.
9.1.2 Hydrochlorination Catalyst Waste
Waste is produced from the disposal of spent catalyst from the hydrochlorination
reactor. For the facilities where FeCl3 catalyst (dissolved in the 1,1-DCE liquid
effluent from the reactor) is treated with ammonia to form solid complexes, the
following source reduction methods were noted:
o Elimination of filter aid.
The type of filters used to separate the precipitated salts from the organic
stream was not described in the available literature. In cases where pre-
coat type pressure filters are being used, a filter aid is first deposited to
improve the filtration rate. Waste volumes can be significantly reduced by
switching to bag or leaf type filters which do not require the use of a filter
aid (LWVM 1985).
* Illinois Institute of Technology 1985: Personal communication.
** Dow Chemical Co. 1985: Personal communication.
B15-8
-------
o FeCl3 catalyst precipitation and recycle.
Decreasing the temperature of the 1,1-DCE liquid effluent stream may
cause FeCl3 to precipitate because of the decrease in solubility. The
precipitated catalyst can then be separated, redissolved, and recycled back
to the reactor. Residual catalyst in the 1,1-DCE stream can then be
removed conventionally using ammonia.
9.1.3 Equipment Cleanup Wastes
Usually, the wasteloads associated with equipment cleaning are small and periodic in
nature (once every 1 or 2 years). Further reductions may be obtained through:
o More complete drainage of process piping or equipment prior to
cleaning.
o Lower process film temperatures and increase turbulence at the heat
exchange surfaces to reduce fouling rates. This can be accomplished
by avoidance of overdesign and using recirculation during turndown
operations.
o Use of electropolished or Teflon* heat exchanger tubes to reduce
deposit clingage (Anonymous 1985c).
o Use of in-process heat exchanger tube cleaning devices (Anonymous
1985b).
The reader is also referred to the study of process equipment cleaning contained in
this appendix.
9.1.4 Spills and Leaks
As mentioned before, spills and leaks constitute a very minor waste stream owing to
the extensive implementation of preventive maintenance measures in facilities dealing
Registered trademark of E.I. Du Pont de Nemours & Co.
B15-9
-------
with hazardous materials. Further source reduction is possible, in principle, through
better operating practices (see also process study entitled "Good Operating
Practices"). Additionally, some consideration should be given to:
o Replacing single mechanical seals with double mechanical seals on
punnps or using canned seal-less pumps.
o Using leak detection systems and portable monitors.
o Using enclosed sampling and analytical systems.
o Using vapor-recovery systems for loading, unloading, and equipment
cleaning.
9.2 Implementation Profile
The facilities that currently combust their organic wastes with attendant recovery of
HC1 and heat value will probably have little or no incentive to pursue waste
minimization options, some of which may require considerable investment of effort for
engineering and economic analyses before implementation.
9.3 Summary
Table 9-1 represents a summary of source control methodologies for the 1,1,1-
trichloroethane manufacturing industry. Based on the measures currently taken to
minimize waste, it is estimated that waste generation has been reduced to the level
characterized by a current reduction index of 3.0 (75 percent), due to the
implementation of methods listed in this section (the 75 percent value represents the
amount of waste that current waste reduction measures have reduced compared to the
waste volumes that currently would be generated without these measures.) By
implementing additional waste reduction measures, it appears that the amount of
waste currently being generated can be further reduced to a level characterized by an
index of 0.7 to 0.8 (18 to 20 percent reduction from current waste generation levels.)
The most effective potential measures for achieving this reduction are those with high
future reduction index shown in Table 9-1. These include: hydrocracking the heavy
chlorinated hydrocarbons, precipitation and recycling of the FeCl3 catalyst. Among
the measures that do have current industrial precedent, thermal oxidation of the heavy
ends with attendant HC1 and heat recovery has the highest application potential.
315-10
-------
TABLE 9-1 SUHURY OF SOURCE CONTROL METHODOLOGY FOR THE 1,1.1 - TRICHLOROCTHANE MANUFACTURIN6 INDUSTRY
1
1
I Heavy Ends (*)
1
1
i
1
I Spent Catalyst (*)
1
1
1
| Equipment Cleaning
1 Wastes («)
1
1
i
1
| Leaks and Spills
1
1
1
1
1
| All Sources
1
1
1
|1
It-
|3.
1
|1.
I
I'-
|2.
|3-
I
11.
|2
|3.
|4.
I
I
Hydrocrack heavy chlor. hydrocarbons
Use lower hydrochlor. reaction temp.
Recover HCl/heating values
Overall
Eliminate use of filter aid
Precipitate & recycle FeC13 catalyst
Overall
Increase equipment drainage time
Lower heat exchanger film temperature
Electropollsh heat exchanger tubes
Use in-process H.X. cleaning devices
Overall
Use double mechanical seals on pumps
Use leak detectors
Enclosed sampling and analy. systems
Use of vapor recovery systems
Overall
All Hethods
Found Documentation
Quantity | Quality
1 I
0 1
2 1
1.00 | 1
0 1
0 1
0.00 | 0
1 1
1 1
2 1
2 1
1.50 | 1
1 I
3 I
3 I
2 I
2.25 | 2
I
" I
I
n
o I
2 I
00 |
o I
o I
00 |
1 I
1 1
1 1
1 1
00 |
1 I
3 I
2 I
2 I
00 |
Waste I
Effectiveness |
3 I
2 I
4 l
3.00 |
2 1
3 1
2 50 1
3 1
2 1
3 1
2 1
2.50 |
3 1
2 1
2 1
4 1
2.75 |
Extent of |
1
0 1
0 1
3 1
1.00 |
1 1
0 1
0.50 1
3 1
1 I
0 I
1 1
1.25 |
4 I
3 I
3 I
4 I
3.50 |
Future | Fraction of |
Potential | [
' i i
' i i
3 1 1
1.67 | 0 89 |
1 1 1
' 1 1
1.00 | 0.05 I
2 I I
' 1 1
2 1 1
2 1 1
1.75 | 0.05 |
2 I I
1 1 1
1 1 1
1 1 1
1.25 | 0.01 |
1 '.oo |
Current I
Index |
0.0 |
0.0 |
3.0 |
3.0 |
0.5 |
0.0 |
0.5 |
2.3 I
0.5 |
0.0 |
0.5 |
2.3 I
3 0 |
1.5 |
1.5 |
3.9 |
3.9 |
3.0 |
Future Reduction Index
Probable I Maximum
0.8 |
0.5 |
0.8 |
0.7 |
0.4 I
0.8 |
0.6 |
0.» |
0.4 I
1.5 |
0.8 |
0.8 |
0.0 |
0.1 |
0.1 I
0.0 |
0.1 |
0.7 I
0
0
0
0
0
1
1
0
0
0
0
1
1
1
8 1
1
8 1
8 1
1
8 1
8 1
1
1
5 1
1
5 1
1
1 1
1 |
I
1 1
8 1
CD
(*) These waste streams include listed T" and/or "K" RCRA wastes.
-------
10. PRODUCT SUBSTITUTION ALTERNATIVES
The growth in industrial use of 1,1,1-TCE as a solvent for vapor degreasing and cold
cleaning operations has been substantial in the 1970's and the early 1980's. Due to its
relatively low toxicity, low level of photochemical reactivity, its adequate solvency,
and the absence of the need for equipment changeover, 1,1,1-TCE was used
extensively to replace the more highly criticized trichloroethylene as a solvent for
metal cleaning operations. Environmental concerns over the fact that 1,1,1-TCE may
threaten the earth's stratospheric ozone layer, however, has placed the use of 1,1,1-
TCE under stricter scrutiny (SRI 1982).
Substitution of water-soluble synthetic cleaners for organic solvents has also been
practiced in industry. Improvement in cleaning techniques can enhance the ability of
weaker cleaning substances, such as alkaline solutions or soaps, to remove oily
residues. These less toxic cleaners can then be used to replace chlorinated organic
solvents such as 1,1,1-TCE.
Production demand for solvents (including 1,1,1-TCE) is largely controlled by the
consumers' effort for conservation and the extent of solvent recovery being employed.
Increasing costs of waste disposal, coupled with stricter air emission regulations, are
likely to result in increased recycling and tighter control of solvent evaporation losses.
Conservation and recovery, together with availability of less toxic substitutes, may
result in a future reduction of the production rate of 1,1,1-TCE.
11. CONCLUSIONS
The principal waste stream from the 1,1,1-TCE production process consists of
chlorinated organic by-products (heavy ends), which are formed during the synthesis
stage. Past modifications of the manufacturing process to eliminate the formation of
by-products were not in evidence; instead the trend has been to reprocess this stream.
In facilities where heavy ends are currently dealt with satisfactorily by thermal
oxidation combined with HC1 and heat recovery, manufacturers of 1,1,1-TCE do not
have the incentive to change their manufacturing process. Further extension of this
option (thermal oxidation) appears to be a major practical step toward minimizing
waste disposal. Nevertheless, source reduction techniques, such as heavy end
hydrocracking with recycle, may deserve further investigation primarily as yield
promoters and not necessarily as waste minimizers.
B15-12
-------
It appears that, in the case of 1,1,1-TCE, the issue of reducing wastes related to
manufacturing process is subordinate to the issue of product substitution or use
alteration relying on more extensive use of improved cleaning techniques, employment
of less toxic cleaning substitutes and increased recovery and conservation efforts by
the users. This conclusion is based on the observation that nearly two-thirds of the
total 1,1,1-TCE produced is consumed in cleaning operations and subsequently disposed
of as cleaning sludges or lost as atmospheric emissions. By comparison, the amount of
wastes produced during the manufacturing process is very minor.
12. REFERENCES
Anonymous. 1979. Key chemical: 1,1,1-Trichloroethane. Chem. Eng. News. 57(40):
130.
Anonymous. 1980a. Clean-up for solvents market. New Scientist. 86: 315.
. 1980b. ICI commissions safer solvent plant. Chem & Ind.
June~21, 1980. p.A72.
. 1980c. Solvent processing gets a second look. Textile
World. 130; 67-76.
f . 1983. Solvent switch cuts cost of vapor degreasing. Met.
Prog. 123: 17.
. 1985a. Aliphatic hydrocarbon output. Chem. Market.
Rep. July 15, 1985. p.15.
. 1985b. Chemical Engineering Progress, 81(7):7.
. 1985c. Chemical Engineering Progress, 81(7):104-5.
Archer, W.L., and Simpson, E.L. 1977. Chemical profile of polychloroethanes and
polychloroalkenes. Ind. Eng. Chem. Product Res. & Dev. 16: 158-62.
Archer, W.L., and Stevens, V.L. 1977. Comparison of chlorinated, aliphatic, aromatic,
and oxygenated hydrocarbon as solvents. Ind. Eng. Chem. Product Res. & Dev. 16:
319-25.
Berrie, J.S., Charles, R., and Chuffant, H. 1978. Manufacture of 1,1,1-
Trichloroethane. Brit. Pat. 1,500, 136. (Feb. 8, 1978).
Boozalis, T.S., and Ivy, J.B. 1978. Process for production of 1,1,1-Trichloroethane and
vinylidene chloride. U.S. Pat. 4,119,674 (Oct. 10, 1978). To Dow Chemical Co.
B15-13
-------
Clark, J.B., Steven, J.C., and Pevetti, D.J. 1984. Laser initiated free-radical
reaction. Proc. SPIE - Int. Soc. Opt. Enq. 458: 82-8.
Emig, G., Hoffman, V., and Ruppert, W. 1979. Investigation of kinetics for
photosynthesis of methyl chloroform. Chem. Eng. Sci. 34(6): 801-9.
Forsht, E.H. 1983. Development document for proposed effluent limitation guidelines
and new source performance standards for the organic chemicals, plastics, and
synthetic fibers industries. EPA-440-1-83-009B. Washington, D.C.: U.S. Environ-
mental Protection Agency.
Johnson, J.C., and Wedmore, L.K. 1983. Metal cleaning by vapor degreasing. Met.
Finish. 81(9): 59-63.
Key, J.A., Stuewe, C.W., Standifer, R.L., et al. 1980. Organic chemicals manu-
facturing, Vol. 8. selected processes. IT Enviroscience. EPA-450-3-80-028C.
Research Triangle Park, N.C.: U.S. Environmental Protection Agency.
Khan, Z.S., and Hughes, T.W. 1979. Source assessment; chlorinated hydrocarbons
manufacture. Monsanto Research Corp. EPA-600-2-79-019G. Research Triangle
Park, N.C.: U.S. Environmental Protection Agency.
LWVM. 1985. The League of Women Voters in Massachusetts. Waste reduction, the
untold story. Meeting materials, National Academy of Science Conference Center on
June 18-21, 1985. Woods Hole, Mass.: The League of Women Voters of Massachusetts.
Monahan, R. 1977. Vapor degreasing with chlorinated solvents. Met. Finish. 75: 26-
31.
PACE. 1983. PACE Company Consultants & Engineers, Inc. Solvent recovery in the
United States; 1980 - 1990. Houston, Tex.: PACE Co.
SRI. 1982. Stanford Research Institute. C2-Chlorinated solvents. In Chemical
economic handbook, 1982. Menlo Park, Calif.: Stanford Research Institute.
Stevens, V.L. 1979. 1,1,1-Trichloroethane. In Kirk-Othmer encyclopedia of chemical
technology. 3rd ed. Vol. 5, pp. 728-31. New York, N.Y.: Wiley.
Stevens, V.L., and Hansen, T.F. 1982. Non VOC solvents for air quality compliance.
Met. Finish. 80: 41-2.
Van Gemert, B. 1982. Role of stabilizer: aluminum reaction in methyl chloroform
stabilization. Ind. Enq. Chem. Product Res. & Dev. 21: 296-9.
13. INDUSTRY CONTACTS
S.L. Arnold, Manager, Environmental Information Clearinghouse, Dow Chemical Co.,
Midland, MI.
R. Samelson, PPG Industries, Pittsburgh, PA.
815-14
-------
1. PROCESS: TRICHLOROETHYLENE/PERCHLOROETHYLENE
MANUFACTURE
2. SIC CODE: 2869(2)
3. INDUSTRY DESCRIPTION
Manufacturers of trichloroethylene (TCE) and perchloroethylene (PCE) are included as
part of the organic chemicals industry. A variety of feedstocks and processes are used
to manufacture TCE and PCE as co-products.
3.1 Company Size Distribution
Four major manufacturers are currently responsible for all of the U.S. production of
TCE and PCE. These are large chemical companies producing a great variety of
organic compounds at each of their manufacturing plants. A company usually owns
one or two plants, with combined annual production capacity of TCE and PCE between
25 to 200 thousand short tons/year (TPY).
3.2 Principal Producers
The five principal producers of TCE and/or PCE are listed below*:
Diamond Shamrock Corp. - PCE only
Dow Chemicals USA - TCE/PCE
E.I. du Pont de Nemours & Company, Inc. - shut down
PPG Industries - TCE/PCE
Vulcan Materials Company - PCE only
3.3 Geographical Distribution
Nationwide, there is a total of six plants producing TCE and/or PCE (SRI 1982).
Among these, one plant is located in Texas, three in Louisiana, one in Kansas, and one
in California.
PPG Industries 1985: Personal communication.
B16-1
-------
4. PRODUCTS AND THEIR USE:
TCE and PCE are used mainly as solvents for dry-cleaning and metal degreasing
operations. PCE is also used as an intermediate for the production of other chemicals
such as C2-fluorocarbons. The end-use profile of TCE and PCE in 1981-1982 is given
in Table 4-1. Due to stricter government regulations on the use of toxic solvents,
consumption of TCE in metal cleaning was expected to drop by an average of 3%
annually in the 1980's. While annual production rates of TCE in recent years could not
be found, significant reduction from the value of 133,100 TPY produced in 1975 is
expected due to the effect of increasing application of solvent re-use, recovery, and
recycling. Demand for PCE peaked in 1979, but has been declining since. Annual U.S.
production of PCE was 254,364 TPY in 1984 (Anonymous 1985).
Table 4-1. 1981-82 End-Use Profile for TCE/PCE in the U.S.
Relative Consumption Rates, %
End Use TCE PCE
(a) (b)
Metal Cleaning
Dry Cleaning/Textile Processing
Chemical Intermediate
Miscellaneous
66
--
7
27
21
59
6
14
Total 100.0 100.0
Source: Hughes, et al., 1985
5. RAW MATERIALS
The raw materials used in the production process (Liepins et. al., 1977) are:
Reactants ethylene chloride, ethylene dichloride, Ci~Cj hydrocarbons,
C2-chlorinated hydrocarbons, chlorine, hydrochloric acid, oxygen
(air)
Catalysts CuCl2, BaC^, KC1, Aid}, graphite, activated carbon.
Stabilizers trimethylamine, pyrrole-based compounds, hydroquinone mono-
methyl ether, p-tertiary annyl phenol.
B16-2
-------
Neutralizes sodium hydroxide, ammonia
Drying materials sulfuric acid, molecular sieves, activated carbon, alumina gel,
silica gel, and others
6. PROCESS DESCRIPTION
A variety of feedstocks and processes can be used to produce TCE and PCE. The
principal processes of current commercial importance in the U.S. are:
The coproduction of PCE and carbon tetrachloride (CC^) from the thermal
chlorination of Ci to C4 paraffinic hydrocarbons, or their partially
chlorinated derivatives.
The coproduction of TCE and PCE from the thermal chlorination of
ethylene dichloride (EDC), or other C2~chlorinated hydrocarbons.
The coproduction of TCE and PCE from the oxychlorination of EDC, or
other C2-chlorinated hydrocarbons.
*
ThesS processes allow for re-use of C3 and higher chlorinated hydrocarbon by-
products, and thus enable the upgrading of chlorinated hydrocarbon residuals from
other processes, with the exception of chlorinated tars*.
The thermal chlorination process can be performed either with or without a catalyst,
and results in an output of HC1 as a co-product. The catalytic oxychlorination process,
on the other hand, avoids a net production of HC1 and even provides an outlet for
unwanted HC1 from other processes. The oxychlorination reaction produces water
along with TCE/PCE and hence requires the removal of water from the final product.
Descriptions of the above manufacturing processes have been given in the literature
(McNeill Jr., 1979; Key et al., 1980; Liepins et. al., 1977). In this report, only the
oxychlorination route will be described and discussed in detail.
E.I. Du Pont de Nemours & Co. 1985: Personal communication.
B16-3
-------
TCE and PCE are coproduced from the oxychlorination of EDC or other
C2-chlorinated hydrocarbons. The raw material ratios determine the proportions of
TCE and PCE being produced. Conversion of chlorinated hydrocarbons to TCE and
PCE was reported as 75-85%, with 10-15% chlorinated by-products, and 5-10% loss as
CO and CC>2 (McNeill 1979). The overall reaction for this process is given by:
C2-chlorinated hydrocarbons + C>2 + C12 TCE + PCE + H2O (1)
HC1 can also be used as a feedstock since C12 can be produced from HC1 by a Deacon
reaction:
2HC1 + 1/202 > C12 + H20 (2)
Figure 6-1 represents the flow diagram for the production of TCE and PCE by
oxychlorination. In this process, EDC, recycled chlorinated hydrocarbons, chlorine,
and oxygen are fed into a fluidized bed reactor. For this process, C2-chlorinated
hydrocarbon wastes from other processes can also be consumed by introducing them
into the organic recycle storage tank which feeds the fluid bed reactor. The reactor
operates at slightly above atmospheric pressure and about 800°F (Keil 1979). The
most common catalyst used is CuCl2.
The reactor effluent gas, containing chlorinated organics, water, a small amount of
HC1, CO, CO2, and traces of other inert gases, is condensed using water-cooled and
refrigerated condensers. The condensed crude product stream is drained through a
decanter to remove entrained catalyst fines. The non-condensed vent gas stream is
sent through a HC1 recovery unit where the gas stream is water scrubbed to remove
HC1 as by-product.
The crude hydrocarbon stream is separated in the decanter. This stream is then
azeotropically distilled to remove water. Waste water from the azeotropic distillation
is combined with the aqueous phase containing catalyst fines from the decanter. The
combined stream is sent to waste water treatment. An alternative approach relies on
concentrated sulfuric acid (98% H?SO4) to remove water from the crude TCE/PCE.
The acid and crude TCE/PCE are contacted and then allowed to separate. The acid is
then sent to a vacuum flash unit where the absorbed water is removed. The water can
then be steam- or air-stripped of organics (which are recycled to the reactor), pH
B16-4
-------
VENT BAS
TCE
C2 CHLORINATED
ORCANICS FROM
OTHER PROCESSES
PROCESS HASTE CATEGORIES!
(7) HEAVY ENDS
(7) SPENT CATALYSTS
(7) SPENT ALKALINE SOLUTION
0 SPENT ORYINB MATERIALS
(T) NAETENATER
Figure 6-1 Production of TCE and PCE by Oxychlorination
B16-5
-------
adjusted and discharged to NPDES permitted outfall. The recovered sulfuric acid
(*> 80% h^SO^) is then used in other processes or sold, e.g. as a raw material to
produce fertilizer.
The crude chlorinated hydrocarbon mixture is further separated into TCE and PCE
crude streams in a TCE/PCE column. The TCE crude is purified in a TCE column.
Overhead from this distillation column is sent to an organic recycle storage tank. The
bottom stream is neutralized with dilute caustic, ammonia or washed with water, and
subsequently dried to give the desired TCE product. Wastes from the neutralization
and drying operations are sent to waste-water treatment or disposal.
The bottom stream from the TCE/PCE column is separated in a heavy-ends column.
The PCE-rich overhead stream from the heavy-ends column is sent to the PCE column
where final PCE purification takes place. The overhead from the PCE column is
recycled to the process and the purified PCE is withdrawn as bottoms, then
neutralized, washed and dried.
The bottom stream from the heavy-ends column is flashed to separate the tars and
carbon in the organic recycle system. The flash overhead stream is recycled back to
the process. The separated heavies, called "hex-waste", contain hexachlorobenzene,
hexachlorobutachrene plus various chlorinated G£ compounds in addition to tars and
carbon.
7. WASTE DESCRIPTION
The primary specific wastes associated with TCE/PCE manufacturing using the
oxychlorination process are listed in Table 7-1. Organic heavy ends account for a
large portion of the waste generated. This liquid stream comes from the distillation of
the hydrochlorination reactor effluent stream to remove heavy chlorinated
hydrocarbons that cannot be recycled or further processed. The resulting tars are
often oxidized for recovery of HC1 and heat or sent to land disposal.
A second source of waste involves spent catalyst. Entrained catalyst fines are
removed from the reactor effluent stream when the condensed stream is drained
through the decanter. The liquid waste stream is then combined with wastewater from
B16-6
-------
the azeotropic distillation or sulfuric acid drying operation, and is sent to wastewater
treatment. Spent sulfuric acid is then used elsewhere in the facility or sold.
Neutralization and drying operations produce streams containing aqueous caustic
solutions and spent absorbents. The aqueous waste streams are sent to waste water
treatment, and the spent absorbent is regenerated (if a regenerable type is used) or
land disposed.
The remaining portion of the waste is the scrubber effluent wastewater. This liquid
waste stream is generated from the use of emission control equipment to limit the
amount of gases being released to the atmosphere. The main sources for emission
include vents from drying and neutralizing columns, vents from the HC1 absorbent
unit, and vents from organic storage tanks. The gas streams are often water scrubbed
to remove hazardous materials. The resulting aqueous waste streams are combined
and sent to wastewater treatment.
Table 7-1. TCE/PCE Manufacturing Wastes from the Oxychlorination TCE/PCE Process
No.
1.
2.
Waste
Description
Heavy ends
Spent catalyst
Process
Origin
Distillation from
organic recycle
system
Decanting of
entrained
catalyst
Composition
Tars, heavy
chlorinated
hydrocarbons
Metals, tars,
chlorinated
hydrocarbons
RCRA
Code
K030
3. Spent caustic
solution
4. Drying waste
5. Spent scrubber
wastewater
6. Equipment
cleanup
wastes
7. Spills &. leaks
Neutralization
Chlorinated
hydrocarbons,
spent caustic
solution
Drying operations Spent absorbents
Emission control
operations
Cleaning of tanks,
towers, and heat
exchangers
Chlorinated hydro-
carbons, acids,
water
F024
B16-7
-------
8. WASTE GENERATION RATES
Very little data was available on waste generation rates from TCE/PCE manufacturing
facilities. In 1975, Catalytic Inc. reported waste generation rates for the production
of PCE from a generalized model plant. These values, however, were given for the
manufacturing of PCE based on the direct chlorination process alone. Due to the lack
of information, estimation of current annual waste generation volumes by the
TCE/PCE manufacturing based on oxychlorination process could not be made.
Fractional waste generation (weight fraction each waste represents as compared to all
waste generated) was estimated by the project staff. These values are shown in
Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" has also been included. Good operating practices are defined as
being procedural or institutional policies which result in a reduction of waste. The
following items highlight the scope of good operating practices:
o Waste Stream Segregation
o Personnel Practices
Management Initiatives
Employee Training
o Procedural Measures
Documentation
Material Handling and Storage
Material Tracking and Inventory Control
Scheduling
o Loss Prevention Practices
Spill Prevention
Preventive and Corrective Maintenance
Emergency Preparedness
316-8
-------
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified. A
summary of the waste sources and the corresponding source reduction methods is given
in Table 9-1. This section deals with the description of the listed methods, including
known application cases.
9.1.1 Heavy Ends
A major portion of the waste generated from the manufacturing of TCE and PCE
comes from distillation and flashing operations required for the removal of chlorinated
hydrocarbon heavy ends. Since only light chlorinated hydrocarbons can be recycled to
give TCE and PCE as products, reduction of the waste load can be done either through
modifications of the existing manufacturing process to minimize the formation of
heavy chlorinated hydrocarbons, or from the upgrading of these heavy by-products.
The following source reduction methods are noted:
o Rapid cooling of reactor effluent gas.
It is postulated that the extent to which the formation of the undesirable
byproducts can be inhibited depends, in part, on how quickly the reactor
offgas is cooled. This postulate is based on the argument that after leaving
the fluid bed, the gas mixture is no longer in intimate contact with the
catalyst. Therefore, side reactions can now take place without the benefit
of selectivity control of the catalyst.
Rapid quench of the reactor offgas leaving catalytic reactors is a feature
often encountered in chemical processes, such as acrylonitrile or TCE/PCE
production through thermal chlorination. It is suggested that the control of
cooling (or quench) rate be examined in the context of reducing formation
of the compounds found in the "hex-waste", along with the effects on
TCE/PCE distribution.
o Control of chlorine-to-hydrocarbon ratio in the feed.
Formation of 03 and heavier compounds is strongly dependent on the
chlorine-to-hydrocarbon ratio in the feed, as chlorine presence moderates
carbon-to-carbon radical reactions responsible for the formation of C3+
byproducts. Examination of Cl2/hydrocarbon ratio as a waste generation
control parameter is suggested.
B16-9
-------
Re-examination of fluidization conditions.
In the fluidized beds operating in a bubbling regime, the effective size of
bubbles controls the degree of mixing, which, in turn, influences the
catalyst-gas contacting, represented by an interchange coefficient
(Froment & Bischoff 1979). It is postulated that by decreasing the bubble
size (e.g. by decreased tube pitch of the cooling coil bundle) the catalyst-
gas contacting is improved resulting in a lower rate of formation of
undesirable byproducts. It is suggested that this aspect be examined for its
waste minimization potential.
Application of improved catalysts.
Use of the oxychlorination catalysts with higher selectivity toward
TCE/PCE is an effective method of minimizing waste due to formation of
undesirable byproducts. Research efforts have been made to improve
currently used catalysts.
Heavy ends hydrocracking.
An additional reaction may be used to break the heavy chlorinated
hydrocarbons into lighter products. The cracking can be done using
catalytic hydrogenation processes. The light hydrocarbons generated from
this operation can then be combined with other light chlorinated organic
streams to be recycled to the reactor. Research work at the Illinois
Institute of Technology indicates that the process remains, so far,
technically infeasible due to severe fouling of the catalyst*. However,
further research is planned to explore the effect of increasing H2/feed
ratio on reactor performance.
Replacement of oxychlorination with a combination of direct chlorination
and an HC1 oxidation unit.
Since the product yields from oxychlorination process (r>> 80%) are usually
much lower than those from direct chlorination processes (^95%), the use
of chlorination instead of oxychlorination to manufacture TCE and PCE
should reduce the amount of byproducts formed. Also, subsequent product
purification operations are simpler because of the absence of water being
formed as byproduct in an organic environment in oxychlorination step.
* Illinois Institute of Technology 1985: Personal communication
316-10
-------
A major disadvantage in using the chlorination process is the large amount
of HC1 that is formed as a byproduct. The HC1 stream, however, can be
electrolytically oxidized into H2 and Cl£. The chlorine generated in this
process can then be recycled to the feed. Conversely, HC1 can be directly
oxidized using oxygen to H£O and C\2 (Kel-chlor process). Both HC1
oxidation routes have been applied commercially. A big problem is that
the direct HC1 oxidation route is very costly and, at the present, cannot
compete with the conventional oxychlorination route. The only operating
unit in the U.S. (Du Pont's Kel-Chlor facility in Corpus Christi, TX.) is
planned to be shut down and replaced with oxychlorination due to
unacceptably high cost. However, the electrolytic route is still under
consideration*.
o Thermal oxidation with recovery of HC1 and heat.
The heavy ends are often incinerated and both heat and HC1 are recovered.
Recovered hydrochloric acid can either be sold or used to generate
anhydrous HC1 for use in the oxychlorination process.
9.1.2 Spent Catalyst
Hazardous waste can be produced from the periodic disposal of spent catalyst, or from
the loss of catalyst due to entrainment. To reduce the waste load generated, the
following steps can be taken:
o Use dry dust collectors or filters.
High temperature ceramic fiber** or sintered metal can be used as filter
elements to remove entrained catalyst fines from the effluent gas. The
catalyst can be partially recycled back to the process or collected
separately from other waste streams. The advantages include nearly
complete removal of solids in the dry phase, which decreases waste
generation associated with liquid filtration. The disadvantages may include
longer residence time of reactor offgas in the stand-alone filter as opposed
to currently used cyclones and the potential for plugging.
* Du Pont 1986: Personal communication.
** Acurrex Corporation 1985: Personal communication.
B16-11
-------
o Use of a tougher catalyst support.
In fluidized bed catalytic reactors, it is evident that the catalyst particles
break into smaller fragments due to attrition and are consequently swept
out of the reactor with the effluent gas. This results in the loss of
expensive catalyst material and additional waste residual that must be
disposed of. A substantial reduction in these losses might be obtained using
a tougher catalyst support and/or by reducing the velocity of gas jets out
of the distributor.
9.1.3 Spent Caustic Solution
Caustic wastes are generated in the TCE and PCE neutralization process where large
amounts of water and ammonia are used. A method to reduce the toxicity of this
stream as opposed to its volume was identified*:
o Use of NaOH instead of
The usual method of treating the spent caustic stream is to neutralize it
using spent sulfuric acid and to then send it to wastewater treatment.
Depending on the type of treatment system employed, the presence of
copper in the stream can have adverse effects. Facilities using copper
chloride catalysts would be advised against using ammonia since it will
solubilize copper from the catalyst and create potential upsets
downstream. This effect also points out the importance, of efficient
catalyst recovery back at the reactor and decanter units. According to
industry contacts, most facilities are already using NaOH.
9.1.4 Drying Waste
The waste generated from the disposal of spent absorbent from the drying operation
can be reduced by using regenerabie drying materials which include activated carbon,
aluminum gel, and silica. Regenerabie drying agents are currently in wide use*.
* E.I. Du Pont de Nemours & Co. 1985: Personal communication.
B16-12
-------
9.1.5 Equipment Cleanup Wastes
Usually, the wasteloads associated with equipment cleaning are small and periodic in
nature (once every 1 or 2 years). Further reductions may be obtained through:
o More complete drainage of process piping or equipment prior to cleaning.
o Lower process film temperatures and increased turbulence at the heat
exchange surfaces to reduce fouling rates.
o Use of electropolished or Teflon* heat exchanger tubes to reduce deposit
clingage (Anonymous 1985b).
o Use of in-process heat exchanger tube cleaning devices (Anonymous 1985a).
All suggestions listed above will have only a minor impact on overall waste generation,
since equipment cleaning wastes are but a small fraction of th'e total waste.
9.1.6 Spills and Leaks
«
As mentioned before, spills and leaks constitute a rather minor waste stream owing to
extensive implementation of preventive maintenance measures in facilities dealing
with hazardous materials. Further source reduction is possible, in principle, through
better operating practices (see practice study entitled "Good Operating Practices").
Additionally, some consideration should be given to:
o Replace single mechanical seals with double mechanical seals on pumps or
use canned seal-less pumps.
o Use of leak detection systems and portable monitors.
o Enclosed sampling and analytical systems.
o Use of vapor-recovery systems for loading, unloading, and equipment
cleaning.
* Registered trademark of E.I. Du Pont Co.
B16-13
-------
9.2 Implementation Profile
Since the current common practice for the disposal of organic waste by incineration is
environmentally viable and successfully complies with government standards and
regulations, most manufacturers do not have a strong incentive to investigate
alternative waste reduction techniques. However, current permit difficulties to site,
build and operate an on-site hazardous waste incinerator, combined with increasing
land disposal costs and restrictions, could induce source control efforts in those
facilities which currently landfill their waste.
9.3 Summary
A summary of the source reduction techniques, along with the associated ratings is
given in Table 9-1. The ratings represent the assessment of the relative usefulness
among the proposed techniques. It is estimated that the potential waste that could be
generated has been reduced to a level characterized by a current reduction index of
2.3 (58 percent). By implementing the proposed source reduction techniques or
expanding the use of those already in place, the amount of waste currently being
generated can be reduced to a level characterized by a future reduction index of 0.4 to
0.9 (10 to 22 percent). Among the most effective measures are the increase in the
recovery of HC1 and heating values from the heavy ends and, potentially, control of
the quench rate to reduce the byproduct formation.
10. PRODUCT SUBSTITUTION ALTERNATIVES
TCE and PCE have been the principal solvents used for cleaning due to their
outstanding ability to dissolve a great variety of oily substances. TCE is used mainly
in the vapor degreasing process for metal surface preparation operations. PCE is more
widely used in dry-cleaning, textile processing, or cold cleaning of metal surfaces.
TCE can be replaced by other solvents such as 1,1,1-trichloroethane or
chlorofluorocarbons (CFC) in metal degreasing operations. In dry-cleaning
applications, petroleum-based Stoddard solvent is used as an alternative to PCE.
Chlorofluorocarbons, such as l,l,2-trichloro-l,2,2-trifluoroethane, which are both
non-flammable and very low in toxicity, are not used widely mainly due to cost.
816-14
-------
TABLE 9-1 SUWARY OF SOURCE CONTROL METHODOLOGY FOR THE TCE/PCE HMHIFACTURIK6 IHDUSTRY
1
Waste Stream
1
| Heavy Ends (*)
1
1
1
1
1
| Spent Catalyst
1
I Spent Caustic
1
| Drying Waste
1
i
| Equipment Cleaning
| Wastes (»)
1
1
I
| Leaks and Spills
1
1
1
1
I All Sources
1
1
11
12
13
14
15.
16.
|7
I
|2.
1
M
I
I
|1-
12.
13.
|4
1
I'.
12
13.
14
1
1
1
Control Methodology I
Found Documentation I Waste | Extent of 1 future | Fraction of | C
__ Deduction I Current Use | Application | Total Waste | Re
| Quantity | Quality | Effectiveness |
Cool reactor effluent gas rapidly |
Control chlorine/hydrocarbon ratio |
Improve fluidization conditions |
Improve catalyst |
Hydrocrack heavy chlor hydrocarbons |
Replace Oxychlorination w/HCl oxid. . |
Oxidize waste w/HCl and heat recovery!
Overall |
Use dry dust collectors and filters |
Use of a tougher catalyst support I
Overall |
Use NaOII in place of NH3 |
Overall |
Use regenerable drying materials |
Overall |
Increase equipment drainage time I
Lower heat exchanger film temperature!
Electropolish heat exchanger tubes |
Use in-process H.X. cleaning devices 1
Overall |
Use double mechanical seals on pumps I
Use leak detectors [
Enclosed sampling and analy systens 1
Use of vapor recovery systems |
Overall |
All Methods
2 1
1 1
1 1
1 1
1 1
2 1
3 1
1 57 |
1 I
1 1
1.00 |
2 1
2 00 |
1 1
1 00 |
1 1
1 1
2 1
2 1
1.50 |
1 1
3 1
3 1
2 1
2.25 |
1
1
1
2 1
2 I
1 29
1 I
1 I
1.00 |
2 1
2 00 |
1 I
1.00 |
1 1
t 1
1 1
1
1.00 I
1 1
3 1
2 1
2 1
2 00 |
3 I
2 1
1 |
2 1
2 1
3 1
4 1
243 |
1 1
3 1
2 00 |
2 1
2.00 |
3 1
3.00 I
3 1
2 1
3 1
2 1
2 50 |
3 I
2 1
2 1
4 1
2.75 |
| Potential | |
2 1
3 1
0 1
3 1
0 i
1 I
2 I
1 57 I
0 1
3 1
1 50 I
4 1
4.00 I
3 1
3.00 !
3 1
1 1
0 1
1 I
1.25 |
4 1
3 1
3 1
4 !
3 50 |
2 1 1
1 1 1
1 1 1
3 1 1
1 1
1 1 1
2 1 1
1.57 I 0.73
2 1 1
1 I I
1.50 | 0.01 |
0.00 ! 0.10 |
1 1 1
1.00 | 0.05 |
2 1 1
1 I I
2 I I
2 I I
1 75 I 0 05 |
2 1 1
' 1 1
' 1 1
1 I I
1.25 | 0 81 |
1 LOO |
jrrent | Future Reduction
Index | Probable
1.5 |
1-5 1
0.0 I
1.5 |
0 0 |
0 8 |
2.0 |
2.0 I
0.0 I
23 |
2.3 |
2.0 I
2 0 |
2.3 |
2.3 |
2.3 |
0.5 |
0.0 |
0.5 |
2.3 |
3.0 |
1.5 |
1.5 |
3.9 |
3.9 |
2.3 |
0
0
0
0
Index I
-|
| Maximum |
8 1
J |
3 1
4 I
1
1
1
1
0.5 | |
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
6 1
0 1
5 1
5 1
2 1
3 1
0 1
0 1
2 1
2 1
4 1
4 I
5 1
8 1
8 1
0 1
t 1
1 1
0 1
t 1
» 1
1
1.0 I
1 0 I
0.5 !
1
0.5 |
1
0.0 |
0.2 |
0 2 |
1
1
1.5 |
1
1.5 1
1
0.1 |
0.1 |
1
0.1 |
0 9 !
CD
h-'
I
t—i
\-n
(*) These waste streams include listed "F' and/or "K" RCRA wastes
-------
Improvement in equipment and techniques which make emission control more
effective, may contribute to an increased use of these solvents to replace TCE or PCE
in the future. However, it must be noted that since PCE is used as a raw material in
the manufacture of several CFCs, their substitution for TCE/PCE may have a limited
impact on TCE/PCE production.
Changes in cleaning techniques can be adopted to allow for the use of less powerful
cleaning substances in place of TCE and PCE. For example, a combined ultrasonic-
alkaline cleaning step would greatly improve the ability of alkaline cleaning solution to
remove oily residues, and thus would eliminate the need for using organic solvent
cleaners (the reader is also referred to the study of metal parts cleaning contained in
this appendix).
The dominant factor controlling the production demand for any solvent (including TCE
and PCE) is the extent of the recovery and conservation effort by the users. The
increasing costs of waste disposal coupled with tougher and more vigorously enforced
air emission standards are likely to result in increased recycling and tighter control
of solvent evaporation losses. These, in turn, may lead to the decrease of production
demand.
i
In summary, the likelihood of future decreases in the production rate of TCE/PCE
remains strong in view of such factors as the enhanced solvent conservation and
recovery efforts by consumers and the availability of less toxic substitutes.
11. CONCLUSIONS
Those TCE/PCE manufacturers who use the oxychlorination route appear to have
minimized their waste generation considerably. The reduction is on the order of 60
percent with respect to the waste that would have been produced if none of the noted
methods were implemented. Further reductions appear possible, albeit low, on the
order of 10 to 20 percent with respect to current waste.
The most effective techniques identified were oxidation of heavy chlorinated waste
with attendant HC1 recovery and, potentially, control of quench rate to reduce the
byproduct formation.
B16-16
-------
Waste generation from TCE/PCE manufacturing is directly related to the demand for
these solvents which, in turn, is influenced by consumer conservation and recycling
efforts and substitution of TCE/PCE with other media in cleaning applications.
12. REFERENCES
Anonymous. 1985. Aliphatic hydrocarbons output. Chem. Market Rep. July 15, 1985.
p. 15.
. 1985a. Chemical Engineering Progress, 81(7):7.
. 1985b. Chemical Engineering Progress. 81(7):104-5.
Arcoya, A., Cortes, A., and Seoane, X.L. 1980a. Tri- and perchloroethylene 1. fluid
catalytic oxyhydrochlorination of ethylene. Ind. Eng. Chem. Prod. Res. Dev. 19: 77-
82.
. 1980b. Tri- and perchloroethylene 2. fluid catalytic oxy hydro-
chlorination of dichloroethane. Ind. Eng. Chem. Prod. Res. Dev. 19: 82-6.
Catalytic, Inc. 1975. Water pollution abatement technology; organic chemicals
industry. PB 244544. Springfield, VA.: National Technical Information Service.
Forsht, E. H. 1983. Development document for proposed effluent limitation guidelines
and new source performance standard for the organic chemicals, plastics and synthetic
fibers industries. EPA-440-1-83-009B. Washington, D.C.: U.S. Environmental
Protection Agency. '
Froment, G.F., Bischoff, K.B., 1979. Chemical reactor analysis & design, J.Wiley &
Sons, 1 ed.
Genser, J. M., Zipperstein, A. H., Klosky, S. P., et al 1977. Alternatives for hazardous
waste management in the organic chemicals, pesticides, and explosives industries.
Process Research, Inc. EPA-530-SW-151C. Washington, D.C.: U.S. Environmental
Protection Agency.
Gruber, G.I. 1975. Assessment of industrial hazardous waste practices; organic
chemicals, pesticides, and explosives industries. TRW Systems Group EPA-530-SW-
118C. Washington, D.C.: U.S. Environmental Protection Agency.
Hughes, T.H., et. al. 1985. A descriptive survey of selected solvents, Open file report
No. 1, Environmental Institute for Waste Management Studies, University of Alabama,
Tuscaloosa, Alabama.
Keil, S. L. 1979. Tetrachloroethylene. In Kirk-Othmer Encyclopedia of Chemical
Technology. 3rd ed. Vol. 5, pp. 754-62. New York, N.Y.: Wiley.
Key, J.A., Stuewe, C.W., Standifer, R.L., et al. 1980. Organic chemicals
manufacturing, vol. 8. selected processes. IT Enviroscience. EPA-450-3-80-028C.
Research Triangle Park, N.C.: U.S. Environmental Protection Agency.
816-17
-------
Khan, Z.S., and Hughes, T.W. 1979. Source assessment; chlorinated hydrocarbon
manufacture. Monsanto Research Corp. EPA-600-2-79-019G. Research Triangle Park,
N.C.: U.S. Environmental Protection Agency.
Liepins, R., Mixon, F., Hudak, C, et. al. 1977. Industrial process profile for
environmental use: chapter 6. the industrial organic chemicals industry. Research
Triangle Institute. EPA-600-2-77-023f. Cincinnati, Ohio: U.S. Environmental
Protection Agency.
Marsman, C. J., and Bleich, B. J. 1982. Pollution control practice: HC1 emission
reductions from reactor vent system. Chem. Eng. Prog. 78(6): 40-2.
McNeill, Jr., W.C. 1979. Trichloroethylene. In Kirk-Othmer encyclopedia of chemical
technology. 3rd ed. Vol. 5, pp. 745-53. New York, N.Y.: Wiley.
Miller, S. 1983. Chlorinated hydrocarbon waste. Environmental Sci. Tech. 17(7):
290A-1.
Payer, S. 1974. Recover chlorine from HC1. Hydrocarbon Processing. 53(11): 147-50.
PPG Industries, Inc. 1981. Perchloroethylene-trichloroethylene. Hydrocarbon
Processing. 60(11): 195.
Reich, D.A., and Cormany, C.L. 1979. Dry cleaning. In Kirk-Othmer encyclopedia of
chemical technology. 3rd ed. Vol. 8, pp. 50-68. New York, N.Y.: Wiley.
Scharein, G. 1981. Recover products from chlorohydrocarbon residues. Hydrocarbon
Processing. 60(9): 193-4.
Schneiner, W.C., Cover, A.E., Hunter, W.D., et al. 1974. Oxidize HC1 for chlorine.
Hydrocarbon Processing. 53(11): 151-6.
SRI. 1982. Stanford Research Institute. C2~chlorinated solvents. In Chemical
economic handbook, 1982. Menlo Park, Calif.: Stanford Research Institute.
Vail, S.L. 1983. Textiles: finishing. In Kirk-Othmer encyclopedia of chemical
technology. 3rd ed. Vol. 22, pp. 769-802. New York, N.Y.: Wiley.
13. INDUSTRY CONTACTS
S.L. Arnold, Manager Environmental Information Clearinghouse, Dow Chemical Co.,
Midland, MI.
G.J. Hollod, Sr. Environmental Engineer, Petrochemical Department, E.I. Du Pont de
Nemours & Co., Wilmington, DE.
R. Samelson, PPG Industries, Pittsburgh, PA.
816-18
-------
1. PROCESS: VINYL CHLORIDE MONOMER MANUFACTURE
2. SIC CODE: 2869
3. INDUSTRY DESCRIPTION
Because the manufacture of vinyl chloride involves a capital-intensive process,
virtually all of the U.S. vinyl chloride manufacturing capacity is provided by a small
number of producers with 14 plant locations (see Table 3-1). Individual plant annual
production capacities of major vinyl chloride manufacturers range from 150 million
pounds to 1,250 million pounds. In 1984, U.S. plants produced 7,513 million pounds of
vinyl chloride (Anonymous 1985a).
Vinyl chloride production units are generally part of large integrated chemical
production facilities. Of the 14 major vinyl chloride producing plants, eight are
located in Louisiana and four are located in Texas. California and Kentucky each
contain one plant.
4. PRODUCTS AND THEIR USE
Most vinyl chloride production is consumed in the manufacture of polyvinyl chloride
(PVC) and its copolymers. The small remainder is either exported or used in the
manufacture of adhesives and specialty chemicals.
5. RAW MATERIALS
Chemical Feedstocks; ethylene route - ethylene, chlorine, air, or oxygen; acetylene
route - acetylene (C2H2), hydrogen chloride (HC1).
Drying Agents and Absorbents: silica gel, solid potassium hydroxide (KOH),
concentrated sulfuric acid (H2SO4), activated charcoal, bauxite, methanol.
Catalysts: FeCl3 (direct chlorination), KC1 or CuCl2 (oxychlorination), HgCl2
(acetylene hydrochlorination), on various supports and promoters.
B17-1
-------
Table 3-1 U.S. Producers of Vinyl Chloride Monomer
Annual Capacity
Company and Plant Location (million pounds)
Borden Chemical Divison
Geismar, LA 610
Dow Chemical U.S.A.
Freeport, TX 150
Oyster Creek, TX 750
Plaquemine, LA 1,250 (a)
Ethyl Corporation
Baton Rouge, LA 330 (°)
Formosa Plastics Corporation
Baton Rouge, LA 300
Georgia-Gulf Corporation
Plaquemine, LA 1,000
B.F. Goodrich Company
Calvert City, KY
La Port, TX 1-,000
Calvert City, KY 1,000
PPG Industries
Lake Charles, LA 500
Shell Chemical Co.
Deer Park, TX 840
Norco, LA 700
Stauffer Chemical Co.
Carson, CA 174
Vista Chemicals Company
Lake Charles, LA 700
Total 9,304
Source: Chemical Economics Handbook (SRI 1982); PPG Industries 1985: Personal
communication.
(a- Source at Dow Chemicals indicates that capacity is lower.
(b) Plant not operating at present.
B17-2
-------
6. PROCESS DESCRIPTION
Detailed process descriptions for vinyl chloride monomer (VCM) production have been
presented in open literature (Cowfer and Magistro 1983, McPherson, Starks, and Fryar
1979, Sittig 1978, Liepins et al. 1977). The following paragraphs are based on a review
of the literature and on industry comments. The two main methods of production of
vinyl chloride, the acetylene route and the ethylene route, are discussed. The
descriptions presented below are general and may not account for all different
configurations encountered in the actual plant design.
Prior to the 1950's, the acetylene process for the manufacture of VCM was dominant.
As ethylene became plentiful and the demand for VCM increased, the commercial
processes shifted to the ethylene route. Currently, most of VCM production in the
U.S. is from ethylene. The acetylene route is still practiced only in one installation
owned by Borden Chemicals.
In the first route, the acetylene feed is reacted with HC1. The acetylene feed is first
dried by passing it through a bed containing silica gel, KOH solid or concentrated
H2SO4. The HC1 feed is dried by contacting it with concentrated h^SO^. The
dehydrated feed streams are mixed in a chamber containing activated charcoal to
adsorb small quantities of Cl2 present in HC1; this is done to prevent explosive
chloroacetylene formation. The drying agents and activated carbon are periodically
regenerated and discarded eventually as wastes.
The mixture is then passed through packed tubes containing mercuric chloride on
activated charcoal as a catalyst. Thorium, cerium, cadmium, and/or potassium
chlorides are used in conjunction with mercuric chloride. The products are vinyl
chloride, ethylidene chloride, acetaldehyde, unreacted HC1 and C-2^2- The spent
catalyst containing mercuric chloride is a potentially hazardous waste stream. The
inorganic impurities from the product are removed by alkaline and water scrubbing.
Water is removed by freezing the scrubbed gas stream or passing it through a methanol
column or a bauxite drier. Secondary drying of VCM is accomplished by contact with
solid KOH. The first distillation column removes low molecular weight gaseous
impurities at low pressure. In the second column VCM is further purified. Yields as
high as 99 percent are possible.
B17-3
-------
The more predominant ethylene route, shown schematically in the block flow diagram
in Figure 6-1, consists of 3 basic reaction steps:
1. Direct chlorination of ethylene to ethylene dichloride (EDC):
CH2 = CH2 + C12 »CH2C1 - CH2C1
2. Oxychlorination of ethylene to EDC using HC1 and O2:
CH2 = CH2 + 2HC1 + i 02 } CH2C1 - CH2C1 + H2O
3. Pyrolysis of EDC to VCM and HC1:
2 CH2C1 - CH2 Cl neat> 2CH2 = CHC1 + 2 HC1
Direct chlorination is conducted by passing ethylene and chlorine gas through liquid
EDC at 80-250°F (depending on whether a sub-cooled or boiling reactor is used) and at
pressures close to atmospheric or higher. The reaction is homogeneously catalyzed by
ferric chloride dissolved in EDC. The heat of reaction is dissipated either by cooling
with water coils or by the vaporization of EDC, with the reactor operating at the
boiling point.
EDC is produced at very high yield and can be withdrawn either as a liquid (sub-cooled
reactor) or as a vapor (boiling reactor). Where liquid withdrawal is used, FeCl3 must
be removed from EDC using an acid wash, or by adsorption on a solid such as activated
carbon. With vapor withdrawal, no FeCl3 separation step is necessary.
Oxychlorination also produces EDC for use in pyrolysis, except that it uses HC1
recycled from the pyrolysis step, thus resulting in zero net HC1 production from the
entire process. This was precisely the reason why Oxychlorination was added to the
direct ethylene chlorination step in the 1950's, when HC1 produced by the direct
chlorination and pyrolysis steps could not be marketed. Oxychlorination is typically
conducted at 430-490°F and 20-270 psig in fluid bed reactors, or at 450-570°F and 20-
200 psig in tubular fixed bed reactors in the presence of cupric chloride catalysts.
Temperature control of this highly exothermic reaction is of paramount importance to
waste generation, as temperatures in excess of 620°F result in an increased byproduct
formation. Other penalties for excessive temperatures include catalyst deactivation
and loss of ethylene through burning.
B17-4
-------
CL,
DIRECT
CHLORINATION
REACTOR
A. SUBCOOLEO
B. BOILINS
VCM
ETHYLENE
Oa OR AIR
-=
OXYCHLORINATION
A. FIXED BED
B. FLUID BED
EDC
HCL
EDC
PROCESS HASTE CATEgORIES!
(T) LIBHT/ HEAVY ENDS FROM EDC PURIFICATION
(?) HEAVY ENDS FRON VCN PURIFICATION
(3) PYROLYSIS COKE /TARS
(7) SPENT CATALYST
© AQUEOUS STREANS
(?) VENT BASES
Figure 6- 1 Vinyl Chloride Manufacture via Ethylene Route
B17-5
-------
The excellent temperature control characteristics of fluid beds led to the development
of the fluid bed reactor, which successfully competed with the older fixed bed tubular
design. Fluid beds offered improved EDC yield over the fixed bed reactor. In recent
years, however, the introduction of new catalysts increased the yields and made the
fixed bed reactor technology more competitive with the fluid bed reactors.
Either air or oxygen can be used in oxychlorination; however, the trend has been
toward use of oxygen which drastically reduces vent gas volume and associated losses.
The hot oxychlorination reactor effluent gas is condensed and the condensate
containing EDC and water is sent to the EDC treating and purification section. The
offgas is treated with chlorine to convert unreacted ethylene to EDC, which then is
removed through solvent absorption or refrigerated condenser. For the oxygen-blown
reactor, it is possible for excess ethylene to be recycled back to the oxychlorination
unit.
Comparison between direct chlorination and oxychlorination shows that the latter
produces more impurities. These include ethyl chloride, vinylidene chloride, 1,1-
dichloroethane, chloral and others. EDC from direct chlorination, oxychlorination and
pyrolysis (after treatment) is washed with water and caustic to remove water soluble
impurities (e.g., chloral), and then distilled in two steps to yield 98-99.9 percent p'ure
EDC for pyrolysis. Distillation also yields light and heavy end waste streams.
The endothermic pyrolysis step is conducted at 900-1025°F and usually in the 50-500
psig pressure range. After pyrolysis, the products are rapidly quenched to avoid
increased heavy ends and tars formation. HC1 is recycled back to the oxychlorination
section. The VCM product is then separated from the unreacted EDC via distillation
and EDC is recycled back to the purification section, usually following chlorination to
convert chloroprene and trichloroethylene. The pyrolysis step produces a large portion
of byproducts, most of which are removed when unconverted EDC is washed and
distilled in the EDC purification/treatment section ahead of pyrolysis.
7. WASTE DESCRIPTION
The primary wastes associated with VCM production are shown in Table 7-1. The
wastes consist of liquid, solid, and gaseous streams and, when economically feasible,
are processed on-site to recover some of the chemicals. In highly integrated
B17-6
-------
Table 7-1. Process Wastes from VCM Manufacturing Process
No.
Waste
Description
Process Origin
RCRA
Codes
Light and heavy ends
from EDC purification
2 Heavy ends from VCM
purification
3 Pyrolysis coke/tars
4 Spent catalyst
5 Aqueous streams
6 Vent streams
7 Leaks & spills
9 Equipment cleaning
Direct chlorination reactor K019*
Oxychlorination reactor
Pyrolysis reactor
Pyrolysis reactor K020
Pyrolysis reactor K024
Direct chlorination (FeCl3)
Oxychlorination
Oxychlorination water
EDC wash solutions
Misc. equipment
Misc. equipment & piping
Tanks, towers, heat exchangers, F024
piping
RCRA code covers heavy ends only.
B17-7
-------
manufacturing facilities, some of these streams are used as feedstocks to other
manufacturing units, and hence are not classified as "wastes". The wastes shown in
Table 7-1 can be further grouped into the following categories.
Organic Liquid Wastes
The EDC purification step yields light and heavy ends chlorinated hydrocarbons which
are removed in two consecutive distillation steps prior to pyrolysis. Both streams
undergo reprocessing in many facilities in the following ways (USC 1983):
High efficiency thermal oxidation followed by water or caustic scrubbing
of the flue gas (SCC 1981). This process may be preceded by an additional
distillation step to effect further recovery of useful chemicals.
Catalytic oxidation with heat recovery and direct recycle of HCl-bearing
flue gas to oxychlorination (Benson 1979).
Chlorination at elevated temperature and pressure. The process ultimately
yields carbon tetrachloride and/or perchloroethylene.
•
Other disposal methods include landfilling, deep well injection and at-sea incineration,
which is currently practiced on a developmental basis (Benson 1979, Scharein 1981).
Land disposal is practiced only by a few manufacturers; others use the recycling
techniques mentioned above. Ocean dumping was discontinued in the United States.
Crude VCM is purified, yielding a bottoms stream which contains heavy ends from the
pyrolysis section. In an integrated plant, VCM heavies can be routed to the EDC still;
in non-integrated plants, they are stripped of EDC which is recycled to pyrolysis.
Ultimately, this stream is expected to be disposed of by incineration, landfill or
subjected to further chlorination to yield other chlorinated organic products.
Solid Wastes
Pyrolysis tars and coke form as a result of thermal reactions inside the tubes of the
cracking (pyrolysis) furnace and are periodically removed to restore heat transfer and
to lower the hydraulic resistance of the equipment. The solids can be either
incinerated or landfilled.
B17-8
-------
Spent catalyst waste stream is comprised of FeCl3 catalyst from the direct
chlorination section and spent CuCl2-based catalyst from the oxychlorination section.
In direct chlorination, the subcooled reactor generates a considerably larger FeCl3
stream compared to the boiling reactor, owing to the withdrawal and subsequent wash
of the liquid EDC stream containing dissolved catalyst (by comparison, the boiling
reactor generates a vapor EDC stream with virtually all FeCl3 'e^*- behind).
In oxychlorination, the catalyst is CuCl2 on various supports (e.g., alumina or silica-
alumina). The catalyst waste is generated due to deactivation and due to attrition (in
the fluidized beds only).
Aqueous Wastes
Aqueous streams originate from various EDC washing and vent scrubbing operations
involving alkaline solutions and from the formation of water in the oxychlorination
reaction. The wastewater is typically steam-stripped to remove volatile organics,
neutralized and treated in an activated sludge system prior to discharge.
Vent Streams
Vent streams originate mainly from the non-condensibles present in the offgas from
the direct chlorination and oxychlorination. The air-blown process generates a
considerable vent stream. The treatment includes high efficiency incineration or
catalytic combustion and scrubbing with alkaline solutions.
Spills & Leaks
Leaks and spills along with equipment cleanup are minor effluents in comparison with
the waste streams mentioned above. Due to the highly toxic nature of process fluids
and the resulting regulation of emissions, the quantity of leaks and spills from the
VCM process has been greatly reduced in the last ten years.
Equipment Cleaning Wastes
The equipment cleaning wastes include periodic wastestreams originating from the
steamout of process vessels and the chemical or mechanical cleaning of heat
exchangers and piping. These do not include pyrolysis coke and tars from the cracking
B17-9
-------
furnace, which were considered separately. Disposal methods for equipment cleaning
wastes probably vary from facility to facility and it is likely that land disposal is
practiced.
8. WASTE GENERATION RATES
The current waste generation rates were not in evidence at the time of the final
document preparation. The 1974 specific waste generation rates found (Liepins et. al.
1977, USEPA 1975); however, according to industry contacts these rates have
undergone a substantial decrease and are outdated.
The approach used to estimate the achieved and projected reductions in waste
generation requires that relative (or fractional) waste generation rates be known.
Such were compiled by the project staff using information obtained from the industry
and are shown in Table 8-1.
Table 8-1 Waste Generation Profile for the Vinyl Chloride Industry
Waste/ Total
Residual Percent (by weight)
Organic Liquids 10
Spent Catalyst 1
Solids 1
Aqueous Streams 87
Leaks and Spills 1
Equipment Cleaning 1
Total 100
Source: Estimated by project staff
9. SOURCE REDUCTION TECHNIQUES
9.1 Description of Techniques
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" have also been included. Good operating practices are defined as
B17-10
-------
being procedural or institutional changes which result in a reduction of waste. The
following items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel practices
management initiatives
employee training
o Procedural measures
documentation
material handling and storage
material tracking and inventory control
scheduling
o Loss prevention practices
spill prevention
prevention maintenance
emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
The primary waste streams and their source reduction methods are summarized in
Table 9-1. The following is a description of the source control techniques considered
for each of the waste sources.
9.1.1 Liquid Organic Wastes
As mentioned previously, these wastes are composed of distillation light and heavy
ends separated out during EDC and VCM purification processes in separate distillation
steps. At highly integrated manufacturing facilities, these streams are often used as
feedstocks to other units and, as such, are not classified as wastes. The primary
process origins of liquid organic waste include oxychlorination and pyrolysis reactors.
Direct chlorination is not a significant byproduct generator, as catalyst selectivity is
excellent. Sections below describe source control methods that have actual or
potential application to reduce byproduct formation.
B17-11
-------
9.1.1.1 Oxychlorination Step
Minimization of waste generation from the oxychlorination step could conceptually be
approached in two ways. First, improvements to the oxychlorination process could be
made to reduce the formation of the undesirable byproducts. Second, the
oxychlorination step can be avoided altogether by replacing it with an alternative low-
waste process. Byproduct generation is related to the following five main aspects of
the oxychlorination reaction: temperature uniformity and control, feed purity, use of
oxygen versus air, catalytic selectivity, and gas-catalyst contact.
The following practiced and proposed source reduction methods have been identified:
o Use of fluidized bed reactors as opposed to fixed bed reactors.
Fluid bed reactors are characterized by better temperature uniformity and
lower operating pressures and temperatures in comparison to fixed bed
designs. Fluidized bed reactors are more widely used than older fixed bed
designs (Leddy et al. 1983). However, recent improvements in the catalyst
performance make the fixed bed technology competitive again with the
fluid bed designs.
o Modifications to tubular fixed bed reactor design.
Currently practiced modifications include: an increase in catalyst
concentration along the reaction path to avoid hot spotting; minimization
of radial temperature gradient by optimizing tube diameter; and staging of
consecutive oxygen or air injection.
o Use of oxygen instead of air.
This decreases EDC vent losses and byproduct formation. The EDC
obtained from the C^-based oxychlorination can be comparable in purity to
EDC obtained from direct chlorination*. The manufacturers have been
shifting to the oxygen-based process for oxychlorination because of product
purity and lower vent gas loads (Leddy et al. 1983). The opinions on
whether the C>2-based process is economically advantageous over the air-
based process are split.
817-12
-------
o Selective hydrogenation of acetylene in the feed.
This has been practiced to reduce subsequent byproduct formation in the
oxychlorination reactor (Leddy et al. 1983).
o Use of improved catalysts.
The efforts to improve the selectivity, stability and attrition resistance of
oxychlorination catalysts yielded improvements, especially for the fixed
bed applications. New fixed bed catalysts offer significant lifetime
improvements; one catalyst offered by a Japanese manufacturer since 1983
can extend the lifetime from 9 to 18 months*.
o Re-examination of reactor conditions.
Additional means of improving the oxychlorination process to reduce
byproduct formation can be postulated. These include: a thorough
investigation of the effects of reducing the effective bubble size in fluid
beds (e.g., using shorter pitch of the cooling coil bundle) to promote gas-
catalyst contact; variations of the ethylene recycle rate; a decrease in
operating pressure; and quick post-reactive cooling.
The identified alternatives to oxychlorination include an HC1 oxidation step, the use of
an ethylene/acetylene route, and the use of the Akzo-Zout Chemie process.
o Replacement of oxychlorination with an HC1 oxidation step.
In this scheme, the formation of the undesirable byproducts due to
oxychlorination is avoided by oxidation of the purified pyrolysis HC1 and
subsequent C12 recycle back to the direct chlorination EDC production unit
(see Figure 9-1).
HC1 oxidation can be accomplished in one of two ways:
Direct oxidation (Schreiner et al. 1974, Bostwick 1976)
2HCl+iO2 >C12+H2O
Electrolytic oxidation (Payer 1974)
2HC1
* Conoco Inc. 1986: Private communication.
B17-13
-------
ETHYLENE
DIRECT
CHLORINATION
REACTOR
A. SUBCOOLED
B. BOILINS
VCM
CL,
FOR
OXIDATION
HCL OXIDATION
OR
ELECTROLYSIS
HCL
HATER OR H,
EDC
PROCESS MASTE CATE60RIES.
(T) LIGHT/ HEAVY ENDS FROM EDC PURIFICATION
(D HEAVY ENDS FRON VCN PURIFICATION
© PYROLYSIS COKE /TARS
(7) SPENT CATALYST
(?) AQUEOUS STREAMS
(T) VENT BASES
Figure 9-1 Vinyl Chloride Manufacture via an Alternate Process Scheie
with HCL Oxidation Replacing Oxychlorination
817-14
-------
Both schemes employ commercially proven processes such as the Deacon,
Shell, or Kel-Chlor process for direct oxidation, or the Westvaco or Bayer-
Hoechst-Uhde process for electrolytic oxidation (Leddy et al. 1983, Versar
1979). Shell Oil operates a 19,000 TPY pilot plant at Pernis in the
Netherlands using a modified Deacon process for direct oxidation (Versar
1979). Du Pont operates a 200,000 TPY plant at Corpus Christi, Texas
using the Kel-Chlor process (Versar 1979). The Hoechst-Uhde process is
practiced at 14 plants in various countries with a total capacity of 1400
tons per day using electrolytic oxidation*. Mobil Chemical has operated
two electrolytic HC1 oxidation units in their toluene diisocyanate facility
in Baytown, Texas.
The postulated advantages of the HC1 oxidation route over the currently used
oxychlorination route include:
Waste byproducts of oxychlorination are totally avoided; instead, a high-
quality EDC stream from direct chlorination can be further purified at
minimum expense.
Purer EDC feed to pyrolysis results in a decreased waste byproduct
generation during the pyrolysis step.
Waste water is produced outside of the organic phase, thus avoiding costly
waste treatment and the disposal of ultimate treatment residuals
associated with organics carryover.
Reaction water generation can be largely avoided in an electrolytic step.
The principal disadvantages are the current high capital and operating costs.
Numerous examinations of the HC1 oxidation processes indicate that they were not
economically competitive with oxychlorination*. Excessive costs are the reason for
the planned shutdown of the only U.S. installation of the direct HC1 oxidation process
* Hoechst-Uhde Corporations 1986: private communication.
** Dow Chemical Company 1985: Personal communication; PPG Industries 1985:
Personal communication.
817-15
-------
(Kel-Chlor) at Du Pont's Corpus Christi facility*. The economics of these processes
was discussed in detail (Versar 1979). The Kel-Chlor process is economical only for
large scale units (600-1000 TPD of HC1 loads) using the oxygen-based process. The
high capital cost of the Kel-Chlor process is partly due to the tantalum construction
required because of the corrosive process environment. The Hoechst-Uhde process,
though suitable for any capacity due to modular construction of electrolysis units, may
not be economically competitive with oxychlorination even though the electricity
requirements were reduced from 1900 kWh per metric ton of Cl2 in 1964 to 1400 kWh
presently. Further reductions in electricity consumption can be explored in the area
of H2 fuel cell technology. Even if the HC1 oxidation alternative is viable for a grass
roots plant, it may not be affordable to existing plants which are currently handling
their wastewaters in an environmentally acceptable manner.
o Mixed feedstock option.
Another process option is the use of a mixed feedstock of ethylene and
acetylene (Cowfer and Magistro 1983). HC1 produced in the pyrolysis step
can be reacted with acetylene to produce VCM, thus avoiding
oxychlorination of ethylene. The mixed feed stock option depends on
economic and geographic conditions, and the development of lower cost
routes to the production of acetylene. At the present, acetylene routes
generally lack economic feasibility. The environmental disadvantage of
this process is the generation of spent catalyst (mercuric chloride on
activated carbon) which is potentially hazardous.
o Akzo-Zout Chemie process.
An alternate route to VCM, which is still in its experimental stages,
converts ethylene, sodium chloride, oxygen, and carbon dioxide to EDC and
sodium bicarbonate (Anonymous 1983). EDC can then be pyrolyzed to yield
VCM, and sodium bicarbonate can be calcined to soda ash. This route
eliminates the direct chlorination and oxychlorination routes to produce
VCM. This process is being patented by Huls (Germany) and Akzo-Zout
* Du Pont 1986: Personal communication.
817-16
-------
Chemie B.V. (Hengelo, the Netherlands). The process uses a homogeneous
catalyst of complexed ions of trimethylamine hydrochloride, copper
chloride and iodine in adiponitrile solvent. The development work on this
process was halted*. The advantages and disadvantages of the process
over the conventional routes from a waste minimization point of view
could not be assessed due to the lack of available data.
9.1.1.2 Pyrolysis Step
Generation of byproducts from the pyrolysis reaction has been an area of considerable
investigative effort. The following is a listing of both practiced and postulated
techniques for reducing byproduct formation:
o A laser-induced EDC cracking technique.
Under development by the Max Planck Institute in West Germany, laser-
induced EDC cracking is claimed to have a considerably lower byproduct
yield than thermal cracking (Wolfrom 1979, 1980). At the present, the
process is not a commercially proven technology; the problem of the
availability of a large size laser still persists.
o Use of additives to EDC feed.
Additives such as chlorine or carbon tetrachloride, have been reported in
patent literature to suppress methyl chloride formation (Cowfer and
Magistro 1983). According to Dow Chemical, methyl chloride is only a
minor impurity (less than 40 ppm in VCM). The presence of nitromethane
was reported to increase EDC- conversion to 92.5 percent (McPherson,
Starks and Fryar 1979).
o More stringent control of EDC feed purity.
This widely practiced technique reduces fouling and byproduct formation
during pyrolysis.
* Vinyl Chloride Institute 1985: Personal communication.
B17-17
-------
o Rapid quench rate.
Byproduct formation depends to a large measure on how quickly the
pyrolysis offgas is cooled. If cooling is done too slowly, the VCM yield is
substantially decreased with attendant formation of undesirable byproduct.
The technique of direct quench of the hot pyrolysis offgas using cold
recycled EDC condensate is widely practiced.
9.1.2 Solid Wastes
Solid wastes from the manufacture of VCM consist of pyrolysis tars and coke produced
in the cracking furnace, spent catalysts, catalyst fines from oxychlorination, and spent
drying reactants. Possible measures to reduce these wastes are:
o Modification and/or proper control of the pyrolysis process.
Coking can be reduced through a reduction of the trichloroethylene
concentration in the feed to the cracking furnace (opinions are split on this
issue*), temperature profile monitoring along the pyrolysis path, and the
use of an alternate laser-induced EDC cracking process (Wolfrom 1979,
1980). Of these measures, only temperature monitoring appears to have
the largest immediate utility for further application.
o Convert pyrolysis tars into valuable byproducts.
Chlorinated tars and coke can be blended with other chlorocarbon streams
and treated in a combination of thin-film evaporators and decomposition
screws to produce HC1, carbon black (low in chlorine), and some useful low
boiling chlorohydrocarbons. This process is in commercial operation at
Chemische Werke Huls AG, West Germany (Scharein 1981).
o Reduction of the oxychlorination catalyst attrition rate.
The oxychlorination catalyst fines generation rate can be decreased by
reducing the attrition rate through use of a more stable catalyst support.
Catalyst attrition can largely be avoided by using fixed bed reactors,
though they may suffer from temperature maldistribution (hot-spotting)
problems.
Vinyl Chloride Institute 1985: Personal communication.
B17-18
-------
o Prevention of oxychlorination catalyst deactivation.
Excessive temperatures result in volatilization and subsequent loss of
CuCl2 from the catalyst leading to its deactivation. Avoidance of upsets
resulting in excessive temperature episodes will contribute to curtailment
of the catalyst loss and the associated generation of waste. Also, as
discussed previously, new fixed bed catalysts offer significant lifetime
improvement over older ones.
9.1.3 Aqueous Wastes
Process related aqueous waste streams originate mainly from:
Water produced in oxychlorination.
Miscellaneous washing operations.
Oxychlorination water is a necessary reaction product and cannot be avoided without
drastic revamping of the entire process scheme. A process option where water is
produced outside of the organic phase—or is not produced at all—is to substitute the
oxychlorination step with an HC1 oxidation step as discussed before. In concept, the
elimination of oxychlorination water formation by electrolytic HC1 oxidation is
possible, but the attendant reduction in the hazardous waste loads attributable to
water will probably be very small (possibly through avoidance of organics carryover
into wastewater treatment and an associated reduction of the treatment sludge
volume).
Other sources of wastewater are washing operations to remove water-soluble
impurities. Washing operations are performed on EDC to remove FeCl^ when using a
subcooled reactor for direct chlorination and on EDC from oxychlorination, to remove
water soluble impurities (mainly chloral). The washing operations are often performed
on combined EDC feed to the pyrolysis section prior to distillation and drying. Water
use and the associated treatment waste can be minimized by the following practiced
source reduction techniques:
o Use of solid absorbent for FeCl3 catalyst removal from EDC liquid effluent
from the direct chlorination reactor.
817-19
-------
o Use of boiling vs. subcooled reactors.
As explained in sections 6 and 7, the boiling reactor for direct chlorination
produces EDC in vapor form, thus obviating the need for continuous FeCl3
removal in large quantities from the effluent. The original catalyst charge
is replaced much less frequently by comparison to the subcooled reactor
design.
Boiling reactors, pioneered by Stauffer Chemical Company, are already in
widespread use. Boiling reactors eliminate the need to wash FeCl3 out °f
the EDC produced and thus eliminate the generation of related treatment
residuals (McNaughton 1983). Also, steam consumption is drastically
reduced, which results in limiting the boiler feedwater treatment wastes.
o Use of multi-stage countercurrent contactor for EDC pyrolysis feed
washing. When washing is to be performed, a multi-stage countercurrent
contactor will produce a smaller, more concentrated wash-water stream
than will a single stage.
9.1.4 Spills and Leaks
Due to the toxic/carcinogenic nature of VCM, a great deal of attention has been paid
to reducing fugitive emissions from VCM plants. In 1974, EPA estimated that a
typical VCM plant lost through leakage 0.1215 kg of VCM per 100 kg of VCM produced.
The EPA formulated various regulations to reduce these emissions by 94 percent.
Since 1974, the VCM industry has reduced leakage by a factor of ten*. The following
source reduction techniques are considered as good operating practices, with many
being actually mandated by the 1976 VCM standards (NESHAP 1984):
o Replace single mechanical seals with double mechanical seals on pumps or
use canned seal-less pumps.
o Use of bellows sealed valves to limit leakage around stem packing.
o Use of leak detection systems and portable monitors.
Dow Chemical Co. 1985: Personal communication.
817-20
-------
o Enclosed sampling and analytical systems.
o Use of vapor-recovery systems for VC loading, unloading and equipment
cleaning.
9.1.5 Equipment Cleaning Wastes
Generation of this minor waste stream can be minimized by:
o More complete drainage of process piping or equipment prior to cleaning.
o Lower process film temperatures and higher turbulence at the heat
exchange surfaces to reduce fouling rates.
o Use of non-stick (electropolished or Teflon*) heat exchanger tubes to
reduce deposit clingage (Anonymous 1985c).
o Use of in-process heat exchanger tube cleaning devices (Anonymous 1985b).
All the suggestions, listed above will have only a minor impact on overall waste
generation, since equipment cleaning wastes are but a small fraction of the total
waste.
9.2 Implementation Profile
VCM plants tend to be an integral part of large chlorohydrocarbon production facilities
where significant capital has been invested to handle process wastes and effluents in
an environmentally acceptable manner. This makes waste minimization infeasible,
unless significant economic benefits can be demonstrated. In this context, a basic
process modification requiring substantial capital investment would need extensive
economic evaluation, including consideration of the incremental profit due to the
increased VCM yield, the incremental decrease in waste treatment and disposal costs,
and the avoidance of future costs associated with liability for environmental
impairment.
* Registered trademark of E.I. Du Pont de Nemours Co.
B17-21
-------
Most waste minimization measures identified in this work are capital-intensive, hence
their implementation appears more likely for the grass-roots plants than for the
existing ones. Other measures identified in this report are still developmental in
nature (laser-induced pyrolysis, mixed feed operation, or Akzo-Zout Chemie process),
lack economic feasibility at their current stage of development (HC1 oxidation route),
or would have only a relatively small impact (use of non-stick heat exchanger
surfaces).
9.3 Summary
The summary of all noted source control techniques is given in Table 9-1. Each
technique was rated for its effectiveness, extent of current use and future application
potential on scale of 0 to 4. The ratings were derived by project staff based on review
of the available data and in consultation with the industry. The estimates of current
level of waste reduction achieved (current reduction index) and possible future
reduction (future reduction index) were obtained from the ratings in accordance with
the methodology presented in the introduction to this appendix.
The current reduction index (CRI) is a measure of reduction of waste that would be
generated if none of the methods listed were implemented to their current level of
application. For the entire VCM process, CRI is 1.5 (38 percent) which is indicative of
the moderate level of waste minimization that already has taken place. It must be
noted, however, that reduction of waste was not a primary objective behind most of
these measures; rather, it was the effort to increase VCM yield or energy efficiency of
the process.
The future reduction index (FRI) is an indication of the level to which the currently
generated waste can be reduced if all of the techniques noted were implemented
according to their rated potential. The FRI value of 0.1 to 0.3 (2 to 8 percent) is
indicative of low extent of future waste reductions. Among the techniques that were
found currently most effective and applicable (as evidenced by high FRI value) for
control of liquid organic wastes was selective hydrogenation of C2H2 in the feed, use
of improved catalysts and fluid beds in oxychlorination, and use of additives to the
EDC feed to pyrolysis. Use of boiling reactors for direct chlorination was found
effective in reduction of solid waste (FeCl3 catalyst), washwater and water treatment
sludges.
B17-22
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY FOR THE VINYL CHLORIDE MONOMER MANUFACTURING INDUSTRY
f
Waste Stream
Liquid Organics
Oxychlorlnatlon
Reactor (•)
1
I Liquid Organ (cs
I Pyrolysis
1 Reactor (»)
1
| Solid Waste
I Pyrolysis Coke
I and Tars (*)
1
1 Solid Waste
| Spent Catalysts
I
1
| Aqueous Haste
1
I Aqueous Waste
EDC Washing
1
1
| Spills and Leaks
1
1
1
1
1
Equipment Cleaning
| Wastes (')
1
1
I All Sources
1 1
I Control Methodology |-
1
|1 Use fluid-bed over fixed bed reactor |
|2. Modify fixed bed reactor design |
|3. Use oxygen Instead of air |
II. Selective hydrogenatfon, C2H2 in feed|
1 5 Use Improved catalysts |
|6. Re-examine reaction conditions |
|7 Replace OxychloMnation w/HCl oxld. |
|8. Use mixed C2H4/C2H2 feed option |
19. Use Akzo-Zout Chemie process |
1 Overall |
|1. Use of laser Induced pyrolysis |
|2. Use of additive to EDC pyrolysis feed]
|3. Tighter control of EOC purity |
|4. Quench products rapidly |
| Overall |
|). Proper control of pyrolysis |
|2 Recover products from solid waste |
| Overall |
|1. Reduce oxychlor cat attrition rate |
|2. Prevent catalyst deactlvatlon |
1 3. Use boiling versus sub-cooled reactor |
| Overall |
|1 Use electrolytic IIC1 oxld vs oxychlor|
I Overall |
|t. Use solid FeC13 absorbent |
|2 Use boiling versus sub-cooled reactor |
|3. Use multistage countercurrent wash |
I Overall |
|1. Use double mechanical seals on pumps |
|2. Use bellows seal valves |
|3. Use leak detectors and monitors |
|4. Enclosed sampling and analy. systems |
|5 Use of vapor recovery systems I
| Overall |
1 1 Increase equipment drainage time |
|? ElectropoHsh heat exchanger tubes 1
|3 Lower heat exchange film temperature |
| Overall |
1 All Methods
Found Documentation
Quantity | Quality
3 1
1 1
3 1
1
1
1
1
1
1
1.56 | 1
2 I
t I
1 1
1 1
1.25 | 1
' 1
3 i
2.00 | 2
1 1
1 1
2 I
1.33 | 1
2 1
2.00 | 2
1 1
2 1
1 1
1.33 | 1
1 1
3 1
3 1
3 1
2 !
2.40 | 2
1 1
2 1
1 1
1 33 | 1
1
— -1
1
3 1
1 1
3 1
1 1
1 1
1 I
3 I
1 1
2 1
78 I
2 1
1 1
1 I
1 1
25 |
1 1
3 1
00 |
1 1
1 1
3 1
67 |
2 1
00 I
' 1
3 1
1 1
67 |
1 I
3 I
3 1
2 1
2 1
20 |
1 1
1 1
1 1
50 1
Waste |
Reduction |
Effectiveness I
2 1
2 1
1
i
1
1
1
0 1
1 89 |
2 1
1
2 1
3 1
2.00 |
^»
3 I
2.50 |
2 1
1 1
3 1
2 00 |
3 1
3 00 I
' 1
2 1
2 1
• ^
1.67 |
3 1
2 1
0 1
3
0 1
1 60 |
2 1
2 1
1 i
1 67 |
Extent of I
Current Use |
1
2 1
3 1
3 1
3 1
2 1
3 1
0 1
' 1
0 1
1 39 |
0 1
2 1
3 1
4 1
2.25 |
3 1
0 1
1.50 |
3 1
3 1
3 I
3 00 |
0 1
0 00 |
o 1
2 1
4 1
2.00 |
3 1
1 1
3 1
3 1
4 1
2.30 |
3 1
0 1
3 1
2 00 |
Future I
Application |
Potential |
1 I
1 1
1 1
3 1
2 1
1 1
0 1
1 1
0 1
1 11 |
Q j
2 1
1 1
0 1
0 75 |
t 1
1 I
1.00 |
1 I
1 1
2 1
1.33 |
1 1
0 00 |
1 1
3 1
0 1
1.33 |
3 1
1 1
3 1
3 1
4 1
2 80 |
« I
2 1
1 1
J 33 |
1
Fraction of I
Total Waste I
1
I
1
1
1
1
1
1
1
1
0.05 |
1
1
!
1
0 05 |
1
1
0.01 |
1
1
1
0.01 |
I
0 60 |
1
1
1
0.26 |
1
1
1
1
!
0.01 |
1
1
1
0.01 |
t 00 |
Current |
Reduction |-
Index |
1 0 I
1.5 |
0 8 |
30 |
05 |
1.5 |
0.0 |
0 3 |
0.0 |
3.0 |
0.0 |
0.5 |
1 5 |
3.0 |
3.0 |
1.5 |
0 0 |
1 5 |
1.5 |
0 8 |
2.3 |
2.3 |
0.0 |
0.0 |
0.0 |
1.0 |
2.0 |
2 0 |
2 3 |
0 5 |
0.0 |
2 3 |
0 0 |
2 3 |
1.5 |
0 0 |
0 8 |
1.5 |
1.5 1
Future Reduction Index
Probable
0
0
0
0
0
0
0
o
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
| Maximum
3 1
t 1
1 1
8 ! 08
3 1
1 1
0 I
2 1
0 1
2 | 0.8
0 1
3 | 0.3
' 1
0 I
1 | 0.3
1 1
8 | 08
4 | 0.8
1 1
1 1
4 | 04
2 | 0.4
0 1
0 | 0.0
3 1
8 I 0.8
0 1
3 | 0.8
6 | 0.6
4 1
0 1
6 | 0.6
0 1
3 | 0.6
5 1
0 | 1.0
1 1
5 | 10
1 | 03
1
-I
1
1
1
1
1
1
1
1
1
1
1
I
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
CD
ro
(') These waste streams Include listed T and/or 'K1 RCRA hastes.
-------
Among the techniques that were found potentially most effective were the use of
electrolytic HC1 oxidation in place of oxychlorination and the use of laser-induced
pyrolysis. Both will probably require substantial development effort before both
economic and technical feasibility are fully established.
10. PRODUCT SUBSTITUTION ALTERNATIVES
Since most VCM output is consumed by PVC production, the principal product
substitution routes must be explored primarily in the area of PVC applications. Long-
term demand for PVC is expected to remain stable, although in the short term the
operating capacity has steadily been decreasing. In 1981, about 50 percent of PVC was
consumed in construction-related uses (SRI 1982). Clay, cast iron, or ductile steel can
replace PVC in piping uses. This substitution is economical only for pipe diameters
larger than 16 inches. Aluminum is an effective substitute for PVC in the irrigation
pipe market.
Other PVC uses include consumer goods, electrical, and packaging uses. Product
substitution in these areas, even if possible, would have only a minor impact on the
total PVC demand.
11. CONCLUSIONS
While the VCM manufacturing industry appears to have appreciably minimized their
waste (current reduction index of 1.5), it is also apparent that very modest further
reductions are possible (future reduction index of 0.1 to 0.3). The most effective
control methods appear to be selective hydrogenation of acetylene in the ethylene
feed to oxychlorination, use of fluid bed reactors and improved catalyst in
oxychlorination, use of additives to the EDC feed to pyrolysis and use of boiling
reactors.
The technical and economic details of replacing oxychlorination with direct or
electrolytic HC1 oxidation appear to be worth exploring, since the measure would be
very effective in eliminating the organic liquid wastes from oxychlorination. The
currently prohibitive economics of the HC1 oxidation route is a problem; however, the
potential benefits may justify the search for cost-effective solutions. Laser-induced
pyrolysis, while in the developmental stage, offers a substantial potential for reduction
in pyrolysis byproduct formation.
817-24
-------
The identified PVC product substitution alternatives include clay, cast iron, steel or
aluminum piping.
12. REFERENCES
Anonymous, 1983. Chemical Engineering, 90(5):10.
, 1984. Output makes a strong recovery. Chemical Engineering News,
62(24):34.
, 1985a. Production increased only weakly in 1984. Chemical Engineering
News, 63(23):27.
, 1985b. Chemical Engineering Progress, 81(7):7.
, 1985c. Chemical Engineering Progress, 81(7):104-5.
Benson, J.S., 1979. Catoxid for chlorinated byproducts. Hydrocarbon Processing,
59(10):107-8.
Bostwick, L.E., 1976. Recovering chlorine from HC1., Chemical Engineering,
83(21):86-7.
Cowfer, J.A., and Magistro, A.J. 1983. Vinyl chloride. In Kirk-Othmer Encyclopedia
of Chemical Technology. 3rd ed., Vol. 23, pp. 865-85. New York, N.Y.: Wiley.
Hardie, D.W.F., 1964. Vinyl chloride. In Kirk-Othmer Encyclopedia of Chemical
Technology. 2nd ed. Vol. 5, pp. 171-8. New York, N.Y.: Wiley.
Johnson, H., 1973. A study of hazardous waste materials, hazardous effects and
disposal methods. Vol. 2, Booz-Allen Applied Research, Inc. EPA-670-2-73-15.
Washington, D.C.: U.S. Environmental Protection Agency.
Leddy, J.J., Jones, Jr., I.C., Lewry, B.5., et al., 1983. Alkali and chlorine products. In
Kirk-Othmer encyclopedia of chemical technology. 3rd ed., Vol. 1, pp. 826-7, 844-5.
New York, N.Y.: Wiley.
Liepins, R., Mixon, P., Hudak, C., et al., 1977. Industrial process profiles for
environmental use; Chapter 6. The industrial organic chemical industry. Research
Triangle Institute. EPA-600-2-77-023f. Cincinnati, Ohio: U.S. Environmental
Protection Agency.
McNaughton, E.J., 1983. Ethylene dichloride process. Chemical Engineering.
90(25):54-8.
McPherson, R.W., Starks, C.M., and Fryar, G.J., 1979. Vinyl chloride monomer ...
what you should know. Hydrocarbon Processing. 58(3):75-88.
NESHAP 1984. "Survey of EDC Emission sources", EPA-450/3-84-018.
Payer, S., 1974. Recover chlorine from HC1. Hydrocarbon Processing. 53(11):147-50.
B17-25
-------
SCC, 1981. Stauffer Chemical Co., HC1 recovery. Hydrocarbon Processing.
60(11):170.
Scharein, G., 1981. Recover products from chlorohydrocarbon residues. Hydrocarbon
Processing. 60(9):193-4.
Schreiner, W.C., et. al., 1974. Oxidize HC1 for chlorine. Hydrocarbon Processing.
Sittig, M., 1978. Vinyl chloride and PVC manufacture, process and environmental
aspects. Parkridge, N.J.: Noyes Data Corp.
SRI, 1982. Stanford Research Institute. Chemical economics handbook, 1982. Menlo
Park, Calif.: Stanford Research Institute.
DSC, 1983. U.S. 98th Congress, Office of Technology Assessment. Technologies and
management strategies for hazardous waste control. Washington, D.C.: U.S.
government printing office.
USEPA, 1975. U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Standard support and environmental impact statement: emission
standard for vinyl chloride. EPA-450-2-75-009. Research Triangle Park, N.C.: U.S.
Environmental Protection Agency.
Versar, Inc., 1979. A survey of chlorine/caustic production processes. Virginia:
Argonne National Lab.
Wolfrom, J.,'1979. Ger. Pat. 2,938,353 (Sept. 21, 1979), to Max Planck Institute.
, 1980. Ger. Pat. 77,008,848 (March 7, 1980) to Max Planck Institute.
13. INDUSTRY CONTACTS
S.L. Arnold, Manager, Environmental Information Clearinghouse, Dow Chemical Co.,
Midland, MI.
R. Samelson, PPG Industries, Pittsburg, PA.
Dr. W.C. Holbrook, Vinyl Chloride Institute, Cleveland, OH.
G.J. Fryar, Conoco Inc., Ponca City, OK.
M.W. Packbier, Hoechst-Uhde Corporation, Englewood Cliffs, N.J.
B17-26
-------
1. PROCESS: WOOD PRESERVING
2. SIC CODE: 2491
3. INDUSTRY DESCRIPTION
As defined in Standard Industrial Classification (SIC) 2491, the wood preserving
industry "comprises establishments primarily engaged in treating wood, sawed or
planed in other establishments, with creosote or other preservatives to prevent decay
and to protect against fire and insects. This industry also includes the cutting,
treating, and selling of crossties, poles, posts, and piling."
3.1 Company Size Distribution
Of the 524 wood preserving facilities located in the U.S., 65 percent employ less than
twenty people each (USDC 1985). Overall, the Bureau of the Census estimated that
10,900 people were employed by the wood preserving industry in 1982. Table 3-1 lists
company size distribution as a function of total number of employees per facility.
Table 3-1 1982 Company Size Distribution
No. of employees per facility
Total , 1-19 20-49 50-99 100+
No. of facilities 524 340 131 40 13
No. of employees 10,900 2,400 4,200 2,800 1,500
Source: 1982 Census of Manufacturers (USDC 1985).
3.2 Principal Producers
The wood preserving industry is composed of a small percentage of multi-plant firms
and a large number of single plant firms. The six major wood preserving companies in
the U.S. are:
Koppers Co., Inc. J.H. Baxter & Co.
Kerr McGee Chemical Corp. Wyckoff Co.
Southern Wood Piedmont Co. International Paper Co.
B18-]
-------
Koppers operates the most facilities, 20, and controls about 20 percent of the industry
capacity. Over fifty percent of the industry capacity is held by the ten largest firms
(USDC 1985).
3.3 Geographical Distribution
Due to the expense of transportation, wood preservers tend to be located close to
timber cutting operations. Over seventy-five percent of all facilities are located in
two regions: a) from east Texas to Maryland, and b) the northern Pacific Coast
(USEPA 1980). Approximately 33 percent of all wood preserving sites are located in
five states (Georgia, Alabama, Louisiana, Mississippi, and Florida), with 57 percent
being located in ten states. Distributions by EPA region are shown in Figure 3-1 and
Table 3-2 below.
Table 3-2 Location of Facilities by EPA Region
No. of employees per
EPA Region
I
II
III
IV
V
VI
VII
VIII
IX
X
Total
5
10
51
170
71
83
26
21
43
44
1-19
5
10
26
102
49
50
19
21
37
21
facility
20+
0
0
25
68
22
33
7
1
6
23
National 524 339 185
Source: American Wood Preservers Association Statistics for 1981 (AWPA 1982)
adjusted to reflect data in the 1982 Census of Manufacturers (USDC 1985) and
industry comments.
B18-2
-------
VII!
CD
h->
CO
GHZ] o-s
6-10
11-20
21-50
Roman numerals show EPA regions
Figure 3-1 Wood Preserving Plants in the U.S.
-------
4. PRODUCTS AND THEIR USES
Historically, railroad ties, utility poles, and pilings treated with creosote or
pentachlorophenol (PCP) have been the major products of the industry. In recent
years, lumber and plywood treated with chromated copper arsenate (CCA) has
experienced rapid growth. Table 4-1 lists the major products and use distribution of
the wood preserving industry in 1981.
Table 4-1 1981 Preserved Wood Products and Use Distribution
Item
Crossties
Switchties
Pilings
Marine
Foundation
Not Specified
Poles
Utility
Construction
Not Specified
Crossarms
Lumber and Timber
Fence Posts
Other
Preservative, Percent Treated Volume
Creosote PCP
97.3 2.7
95.2 4.8
65.0 11.7
30.0 48.7
4.4 93.3
5.4 6.4
22.3 31.7
-
CCA million cu. ft.
103.9
6.6
23.3 13.1
33%
48%
19%
21.3 ' 63.7
78%
10%
12%
2.3 2.2
88.2 129.0
46.0 16.5
14.8
Source: American Wood Preservers Association Statistics for 1981
(AWPA 1982).
5. RAW MATERIALS
In the U.S., the major raw materials used to preserve wood are creosote, creosote/coal
tar, or pentachlorophenol (PCP) dissolved in a petroleum based solvent, or inorganic
arsenicals such as chromated copper arsenate (CCA) dissolved in water. During the
last 15 years, use of CCA treated wood for construction has increased by about 800
percent (AWPA 1982). However, the use of creosote and PCP has only lessened by
B18-4
-------
about 10 percent during this period. Annual nationwide consumption rates of raw
materials used to preserve wood are shown in Table 5-1 below.
Table 5-1 Raw Materials Used by the Wood Preserving Industry in 1981
Materials Usage
Creosote
Straight 45.9 million gallons
with Coal Tar 65.2 million gallons
with Petroleum 38.5 million gallons
Pentachlorophenol
PCP 33.2 million pounds
Petroleum 52.2 million gallons
Waterborne Preservatives
Ammoniacal Copper Arsenate 1.2 million pounds
Chromium Copper Arsenate 46.4 million pounds
Flame-Retardants 22.3 millions pounds
Source: American Wood Preservers Association Statistics for 1981
(AWPA 1982).
Also of importance are the types or species of wood used to produce a given product.
Items made from Douglas fir or southern pine (pilings, poles, lumber and timber,
crossarms, and fence posts) can be treated with any one of three preservatives. Items
made of oak or mixed hardwood (crossties and switchties) are commonly treated with
creosote or PCP only. Use of CCA tends to make wood surface harder or more brittle,
so that its use may be inappropriate for some applications. Table 5-2 lists the
breakdown of products by wood type and preservative used.
B18-5
-------
Table 5-2 Percentage Market Shares of Wood Products and
Wood Treatments by Type of Wood
Product
Crossties
Switchties
Pilings
Poles
Crossarms
Douglas
Fir
0.6
-
0.7
1.9
0.5
Lumber and Timber 3.5
Fence Posts
TOTAL
0.1
7.3
Southern
Pine
1.2
0.1
3.2
15.4
0.2
31.9
4.4
56.4
Wood Type
Other Mixed
Pine Oak Hardwood Other
0.3 16.8 10.5 1.6
1.0 0.7 0.2
-
1.1 - - 0.6
-
0.4 0.4 - 2.3
0.3 - - 0.1
2.1 18.2 11.2 4.8
Total(a)
31.0
2.0
3.9
19.0
0.7
38.5
4.9
100.0
Treatment Type
Creosote
PCP
CCA
Douglas
Fir
30.3
20.1
49.6
Southern
Pine
17.4
18.3
64.3
Wood Type
Other Mixed
Pine Oak Hardwood Other
14.3 94.0 99.1 38.9
32.2 5.2 0.9 20.6
53.5 0.8 - 40.5
Total(a)
42.4
14.5
43.1
Source: American Wood Preservers Statistics for 1981 (AWPA 1982).
(a) Market share based on Table 4-1; excludes "Other" products.
B18-6
-------
6. PROCESS DESCRIPTION
Descriptions of the wood preserving processes can be found in the open literature
(Wilkinson 1979, USEPA 1973, Thompson 1973a, Hunt and Garratt 1953). However, it
should be mentioned that none of these descriptions were viewed as being complete.
Wood preserving is a complex process and the nature of the wood treated, preservative
used, and product produced can have a large influence on the ways in which the wood
is treated. Very few descriptions accounted for these factors. In addition, process
changes that have been made by the industry in the last few years to comply with air
and water discharge regulations have had a large impact on the ways a facility handles
its waste. The following description relies on the available information, engineering
judgement, and inputs from industry. The description is generic and it does not
necessarily represent the operations at any one given facility. Figures 6-1 and 6-2
present block flow diagrams of pressure treatment facilities using oil-based and water-
based preservatives.
Before wood can be preserved, it must be preconditioned. Preconditioning of wood
reduces the wood's inherent moisture content and improves treatability. The methods
used to pretreat or condition wood usually depend on the type of wood being treated.
These conditioning methods are:
Ambient drying of the wood in an open yard (oak and mixed hardwoods).
Kiln drying (Douglas fir and southern pine).
Steaming the wood at elevated pressure in a retort (pressure vessel)
followed by vacuum (southern pine).
Heating the wood in a preservative bath under reduced pressure.
Heating the wood by means of a solvent vapor.
In the U.S., most wood is preconditioned by ambient drying in open yards, kiln drying,
or by steaming. Steaming helps to dry out the wood and makes it more permeable to
oil-based preservatives. For southern pine, steaming is usually performed if the wood
is to be treated with creosote or PCP. Douglas fir, oak, and mixed hardwoods or
woods treated with water-borne preservatives are seldom steamed.
Steaming is usually performed in the same retort that is used for treating the wood
with preservative. Wood is placed or stacked onto a tram and the tram carries the
wood into the retort. Three different methods are used to steam wood. They are:
B18-7
-------
UTILITIES-STEAM
VACUUM GENERATION
(5]
CD
^—•
CO
oo
CREOSOTE
PROCESS NASTE CATEGORIES!
0 RAM MATERIAL HASTES
@ STEAM CONDENSATE FROM
CONDITIONING AND CLEANING
(D NORK TANK HASTE
(f) STORM RUNOFF AND AREA HASHDOMNS
(5) COOLING HATER
(?) HASTEHATER TREATMENT SLUDGE
Figure 6-1 Block Flow Diagram for Oil- Based Pressure Treatment System with Closed Steaming
-------
00
I—I
CD
I
vo
UTILITIES-STEAM
VACUUM GENERATION
©
RETOBT —v
HOOD
DISCHAF
CCA
STORA6E
YARD
1GE
PRE - CON -
DITIONING
HASTE
TREATMENT
©
RAH
MATERIALS
1
1
1
l_
00
i—"
00
!-•
O
f*
t
TREATME
i
HO
i
^
i
NT
t
RK TANK
SYSTEM
©
i
k
MIX
TANK
(*.
i
-|
I
£ 1
CLEA
i
! DRIP TRACK STORAGE TnEATED
HIH6 ' t- AND PAD t- (OPTIONAL) r-
© 1 © ' ©' "OOD
I
7 V
qrOQABF 1 HOLDING
STORAGE ^ ^ BA5IH
TANK (OPTIONAL)
1
I
SLUDGE
F
PROCESS HASTE CATEGORIES!
0 RAH MATERIAL HASTES
© STEAM CONDENSATE FROM CLEANING
(5) HORK TANK HASTE
(7) STORM RUNOFF AND AREA HASHDOHNS
© COOLING HATER
(i) HASTEHATER TREATMENT SLUDGE
Figure 6-2 Block Flow Diagram for Hater-Based Pressure Treatment System
-------
Open Steaming - The wood is placed in a retort and steam is applied for up
to fifteen hours, during which time the condensate is collected and
discharged from the retort to an oil/water separator. The separator
recovers the preservative washed from the retort and sends the water on to
wastewater treatment.
Closed Steaming - The wood is placed in a retort with internal heating
coils and the retort is filled with water until the heating coils are covered.
Steam inside the coils causes the water outside the coils to boil, thereby
generating the steam used to condition the wood. Upon completion, all of
the water is drained from the retort to an oil/water separator and then
back to a storage tank for reuse. Since the levels of wood sugar and acids
in the water build up over time, some water is periodically removed from
the tank and treated as wastewater.
Modified Steaming - The wood is placed in a retort and steam is applied
until the condensate covers the heating coils. When the coils are covered,
the flow of steam into the retort is diverted to the heating coils. Upon
completion, all condensate is drained from the retort and sent to the
oil/water separator. Preservative is recycled to the process and the water
is sent to wastewater treatment. *
Following the preconditioning operation, the wood is treated using either pressure or
non-pressure methods. Non-pressure methods are usually applied to wood that has
been air or kiln dried only and are intended for use where deep penetration of the
preservative into the wood is not required. The non-pressure methods used for
applying preservatives include:
Brushing or spraying
Dipping
Soaking
Steeping
Pressure methods are used when a deeper penetration of preservative is required. In
the past, many facilities operated production lines that utilized creosote PCP, and
CCA in all possible permutations. While the production line utilizing water-based
B18-10
-------
preservative was segregated from the oil-based production line, some waste handling
problems did occur. Today, most facilities employ only one type of preservative.
Depending on the degree of penetration required and the preservatives used, two
different pressure treatment processes are practiced.
The full-cell, or Bethall process, maximizes impregnation of the wood with oil-based
preservative and therefore is used on wood exposed to extreme environmental
conditions (e.g., marine pilings). The wood is first subjected to a high initial vacuum,
causing air trapped within the cell walls of the wood to expand and diffuse out of the
wood. Next, without allowing air to enter the retort, the retort is filled with
preservative solution until the pressure reaches 125 to 200 psig. Because of the initial
vacuum, the wood cells fill with preservative as the pressure is increased. If needed,
steam may be applied to the heating coils and the temperature maintained at 180°F
to 200°F for several hours. Following this heating operation, the retort pressure is
released and the preservative is drained back to the storage tank. A final steaming
and vacuum cycle may be applied to maximize the recovery of unabsorbed
preservative and to produce a "clean" wood product. This steaming and vacuum
operation helps remove preservative ("kick-back") that may drip from the wood during
its removal from the retort. As an added precaution, wood removed from the retort is
allowed to remain on the drip pad for several hours following treatment. The drip pad
provides a means of recovering preservative or "kick-back" that would otherwise be
lost in the yard. Following this operation, the wood is either held for storage in the
treated wood storage yard or shipped immediately to the customer.
For wood preserved with CCA, the full-cell process is normally employed. Wood is
placed inside the retort and the retort is filled with preservative. Steam is applied to
the heating coils and the temperature and pressure are maintained for several hours.
During this time, preservative solution diffuses into the wood and reacts with the acids
formed from wood sugars that have broken down (the CCA is said to be "fixed" in the
wood). After a given period of time, the solution is drained back to a storage tank (for
reuse) and a short steaming and vacuum cycle may be applied. The wood is then
removed from the retort and allowed to remain on the drip-pad for several hours or
more while the process of fixation continues. From the pad, the wood is either stored
in the treated wood storage yard or shipped directly to the customer.
-------
For wood subjected to less extreme environmental conditions or requiring less
penetration of oil-based preservative, the empty cell, or Rueping process is used.
After conditioning the wood, the retort is pressurized with air. This causes
contraction of the air trapped within the cell walls of the wood. Preservative is added
to the retort and the pressure is increased to about 150 psig for two to eight hours.
Following this operation, the pressure is released and the preservative is drained from
the retort back to a storage tank. Because of the initial pressurization with air,
releasing the pressure causes the air within the cell walls to expand and hence leave
the cell "empty" of preservative. While wood treated with the empty cell process has
a lower tendency to produce "kick-back", a final steaming and vacuum may still be
applied to the retort. Following this operation, the wood is handled in the same
manner as before.
7. WASTE DESCRIPTION
The primary wastes associated with wood preservation are listed in Table 7-1, along
with their process sources. Most of the streams generated are high volume water
streams with various levels of preservative concentration. While many of the specific
process streams only contain water, wood products (primarily acids and sugars), and
the preservative in use, streams such as storm runoff may contain both oil-based and
water-based preservatives. The source of this cross contamination is due to rainwater
flowing over areas where leaks, spills, and kick-back of different preservatives from
freshly treated wood has occurred. The widespread installation of drip tracks and pads
during the last ten years and the current trend of using only one type of preservative
in a plant are viewed as important steps towards reducing and eliminating the
occurrence of cross-contamination.
Many wood treating facilities receive their raw materials in bulk. CCA is normally
sold as a 50 percent concentrate shipped by tank car, with a minor portion sold in 55
gallon drums. This concentrate is usually mixed with recycled process water and
pumped to the work tank as needed. Creosote is purchased in bulk and may either be
mixed with coal tar and petroleum at the facility or by the supplier. PCP may also be
purchased in bulk liquid form (must be heated during transport) but is more commonly
supplied in 1000 to 2000 pound blocks or in solid prill or flake form. Solid PCP is
dissolved in heated fuel oil, and sent to the work tank as needed.
B18-12
-------
Table 7-1 Wood Preserving Process Wastes
No.
Waste
Description
Process Origin
Composition
RCRA
Code
03
I—I
Vj4
leftover raw materials
in original containers
process steam
condensate
sludge from work tank
system
storm runoff and
area washdowns
cooling water
wastewater treatment
sludge
unloading of materials
into mixing tanks
preconditioning
cleaning
filtering of preservative
spills from around retort
cylinder door, drippings
from treated wood, equip-
ment leaks
vacuum generation
primary treatment
secondary treatment
containers with a few
ounces of CCA or PCP
remaining inside
water with creosote, PCP,
wood fiber, silt, sugars,
acids, and oil
water with preservative,
wood fiber, silt, sugars,
acids and oil
preservative with wood
fiber, silt, and dirt
water with preservative
water with preservative
due to direct contact
or leaks
lime or polymer with
emulsified oils, wood
fiber, silt, and
preservative
activated carbon, bio
logical sludge, etc.
K001
-------
For facilities that use small amounts of preservative, CCA and PCP can be supplied in
bags or drums. Once emptied, these containers would still contain traces of
preservative. From the industry comments obtained, it appears that current use of
bags or drums is extremely small (though use was more common in the past).
However, since a recent settlement agreement between the AWPI and the USEPA
(USEPA 1985) spelled out the ways in which CCA and PCP supplied in bags must be
handled, it is viewed that the use of bags (and/or drums) has not been entirely
eliminated.
Steam condensate from conditioning operations would contain high levels of oil-based
preservative, wood fiber, silt, wood sugars, acids, and oil. The preservative washed
from the retort is present because the same retort is used to condition and treat the
wood. Depending on the method used (open, closed, or modified), the total stream
volume can vary widely. Closed steaming, now the most widely used method, takes
advantage of recycling the water used for conditioning. Since the level of wood sugars
in the condensate builds up over time, some water is periodically purged or bled to
wastewater treatment. The total volume of this purge stream, however, is
considerably less than the volume of water sent to treatment by open steaming.
Regardless of the steaming method employed, all condensate passes through an
oil/water separator to provide for recovery of preservative.
For a facility using an oil-based preservative, steam condensate from the cleaning
operation would be very similar to the condensate produced during conditioning. The
major difference would be the lower level of acid and wood sugar present since the
cleaning operation is not as severe as conditioning. This stream is handled in the same
ways as before. Steam cleaning condensate from CCA or water-based facilities would
contain preservative, wood sugar, and acid. This stream is usually sent to a recycle
water storage tank and used for solution make-up.
Work tank waste is the result of dehydration (for oil-based preservative) and filtration
operations. Oil-based preservatives recovered from the oil/water separator are heated
to drive off water. The preservative, free of water, is then filtered to remove dirt,
silt, and wood fiber and is stored in a heated work tank. Water-based preservative is
also filtered to remove dirt, silt and wood fiber. While the amount and disposal
method used for this waste stream is unknown, it is expected that the volume of this
B]8-14
-------
stream would be comparatively small and that it would be either treated and/or
disposed of along with the wastewater treatment sludge.
Storm runoff and area washdowns was considered by industry personnel to be the
largest wastestream generated at any facility. Rainfall flowing across the drip track
and pad picks up preservative that was "kicked-back" from the wood. For facilities
using oil-based preservatives, this stream is sent to oil/water separation (for recovery
of preservative) and then on to wastewater treatment. Runoff at facilities using
water-based preservatives direct the flow into the recycle water storage tank for
subsequent use.
Depending on local rules and regulations, runoff from the treated yard storage area
must also be collected and treated or recycled. While the amount of preservative that
may leach from the wood is very low, all of the preservatives exhibit a certain degree
of solubility in water. In addition, the ever present potential for "kick-back" occurring
in the yard and the past practice of not providing drip tracks and pads means that
some areas of the treated wood storage may be contaminated with preservative.
Cooling water blowdown from open circulating systems should always be treated
whenever the potential for contact with preservative exists or whenever chromate
corrosion inhibitor is used. In the past, steam jets with single-pass barometric
condensers were used for vacuum generation. Single pass barometric condensers used
large volumes of cooling water and permitted direct contact between cooling water
and process vapors. Today, most facilities employ recycling of barometric cooling
water (only a small amount of blowdown would be necessary to purge the system) or
have replaced barometric condensers with surface condensers (where mixing of cooling
water with process vapors is avoided). While cooling water waste is normally sent to
the process water recycle tank for reuse in a facility using water-based preservatives,
cooling water contaminated with oil-based preservatives is sent to wastewater
treatment.
Wastewater treatment generally can consist of primary, secondary, and tertiary
treatment systems. For oil-based preservatives, primary treatment consists of
flocculation with lime (used very little now) or with polyelectrolytes. The sludge
produced from this operation is RCRA waste "K001" and is disposed of in a secure
landfill. Wastewater from a water-based preservative operation contains copper,
B18-15
-------
chromium, arsenic, and other heavy metals. Treatment can be performed by adding
sulfuric acid to the water and lowering the pH to 4.0 or less. Sulfur dioxide is then
used to further lower the pH of the water and to reduce the hexavalent chromium to
its trivalent form. Next, the pH is raised to 7.0 or 8.0 using caustic or lime to
precipitate chromium and copper. Arsenic will also precipitate if lime is used. The
remaining chromium, copper, and arsenic are precipitated using hydrogen sulfide or
sodium sulfide. The produced sludge is disposed of in a RCRA permitted landfill.
The objective of secondary treatment is removal of any dissolved organic matter
remaining in the water. Common methods are bio-oxidation (activated sludge),
chemical oxidation, and activated-carbon adsorption. Other methods include use of
sand bed filters, spray fields, and evaporation in cooling towers. Methods used in the
past have included lagoons and evaporation ponds but these methods (including many
being currently used) have come under close scrutiny and regulation. Water from
these operations normally undergoes tertiary treatment or discharge to a POTW
(Publicly Owned Treatment Works).
8. WASTE GENERATION RATES
The wood preserving industry generates 5.5 million gallons of wastewater per day
(Thompson 1973a). Depending on which preservative a facility uses, the wastewater
generated can contain high levels of creosote, PCP, oil, CCA, wood sugars, and acids
(from the wood). Since the quantity of wastewater generated is a function of
conditioning method, moisture content of the wood, rainwater drainage, and other
various waste streams (cooling water, boiler blowdown, etc.), the quantity of waste-
water generated by different facilities can vary drastically. Typically, the amount of
wastewater generated by a facility using CCA is much less than a facility using
creosote or PCP. This is because much of the wastewater generated can be used in
subsequent formulations of CCA solution.
Based on a review of the available literature, no overall generation rates for each
waste stream were in evidence at the time of final report preparation. Fractional
waste generation rates were estimated by project staff based on review of the data,
engineering judgement, and inputs from industry and are shown in Table 9-1.
B18-16
-------
9. WASTE REDUCTION THROUGH SOURCE CONTROL
Waste reduction has been conceptually addressed using three approaches: (a) how to
reduce the amount of toxic material entering a waste stream, (b) how to maximize the
separation and subsequent recycle of toxic material back to the process, and (c) how to
avoid excessive water use. This last point requires some clarification. Reducing the
amount of water contacting the preservative will reduce the amount of sludge
precursors that dissolve in it and may also lower their entrainment. As a result, the
total organic or inorganic carry-over into the treatment system is reduced and sludge
volume is minimized.
The following sections discuss waste reduction methods both identified and suggested
for the main wood preserving process waste streams. These streams include raw
material wastes, wood conditioning condensate, storm runoff and area washdown, and
cooling water blowdown. Sludge due to wastewater treatment has not been included
since its generation is directly to these other wastes. By reducing the amounts of
aqueous waste produced, and hence treated, the amount of wastewater treatment
sludge will decrease accordingly.
9.1 Description of Techniques
The list of individual primary waste 'streams and their sources, along with a list of
source reduction methods, is presented in Table 9-1. The following sections discuss
noted waste reduction methods and procedures. The basis for identification came
from published accounts in the open literature and communications with the industry.
In addition to the waste reduction measures classified as process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" have also been included. Good operating practices are defined as
procedural or institutional policies which result in a reduction of waste. The following
items highlight the scope of good operating practices:
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
B18-17
-------
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency preparedness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Raw Material Wastes
In the U.S., most creosote, PCP, and CCA is sold, transported, and stored in bulk.
However, some small scale wood preservers purchase PCP and CCA in 50 pound bags
and drums. Since an ounce or two of preservative remains inside the bag after being
emptied, the entire bag may be classified as being hazardous. If these bags come in
contact with non-hazardous material, then all of the material might be considered
hazardous. In addition, spills of preservative in the mix tank area can lead to area
contamination, increased wastewater generation (due to rainwater and area clean-up),
and undue worker exposure. While most liquid formulations are handled in automated,
closed systems, many powder handling operations are still manual. This includes
facilities receiving powder or dry preservatives in bags, drums, and in bulk. The
following source reduction methods for facilities purchasing powder or dry
preservatives (prilled or flaked PCP and inorganic arsenicals) have been noted:
o Purchase of materials in bulk.
This is by far the most commonly used method in the industry. By
receiving raw materials in bulk, no container waste is generated and the
degree of manual handling (and hence the degree of spillage) is reduced.
o Use of closed systems for powders.
After September 1, 1987 the use of closed systems for unloading and
mixing prilled or flaked PCP must be installed at all facilities using
B18-18
-------
powdered PCP (USEPA 1984). The system would consist of pneumatic
equipment design to convey the PCP directly from a hopper truck or
railroad car to a closed (dissolving) mixing tank. The cost for this system
was estimated to be $10,000. Closed systems are also required for handling
dry inorganic arsenicals but since their use is considered to be widespread,
no deadline was set.
For facilities purchasing preservative in bags or drums, closed systems are
also available. For CCA, use of in-drum mixers to prepare solutions has
been documented in the open literature (Wilkinson 1979). The unit consists
of a small pump and a concentric pipe that is screwed into the drum's bung
hole. Water is pumped into the drum and then into the treatment cylinder
or retort. Final rinse water is also pumped into the drum and used as
make-up solution. A pneumatic system was also described for use with
powder formulations supplied in bags.
o Use of rinseable/recyclable drums with plastic liners instead of paper bags
o Better operating practices.
Through industry contacts in the paint manufacturing industry (an industry
characterized by a high degree of powder handling from paper bags), it was
established that the most effective way of reducing hazardous waste
associated with bags and packages was to segregate the hazardous
materials from the non-hazardous materials. As an example, empty
packages that contained hazardous materials should be placed into plastic
bags (so as to reduce or eliminate dusting that leads to contamination of
non-hazardous materials) and should be stored in a special container to
await collection.
9.1.2 Wood Conditioning Condensate
Since the same retort is used for steam conditioning and treatment, preservative
remaining in the retort from the last treating operation is washed out with the steam
condensate. While much of the preservative is recovered by sending the wastewater to
an oil/water separator, some of the preservative is lost because of oil/water emulsions
that are difficult to separate. For facilities using CCA, steam condensate is not
B18-19
-------
produced since the wood is seldom steam conditioned. The source reduction methods
considered for facilities using creosote or PCP are listed below:
o Reduction of steaming by use of air or kiln drying.
By increasing the use of air or kiln drying and thereby reducing the need
for steaming, generation of a major wastewater stream can be avoided.
Because the effectiveness of air drying is limited to only a few geographic
locations, many facilities have installed drying kilns instead (Thompson
1973a). While kiln drying has several advantages over air drying (increased
production rate and reduced yard inventory), its use is limited to mainly
items made of Southern pine such as pilings, poles, and timber. Products
such as crossties and switchties do not readily lend themselves to kiln
drying, since the hardwoods used to produce these items will crack and
warp when kiln dried.
o Use of separate retorts for steam conditioning and treating.
Another way of reducing the waste associated with steam conditioning is
by physically separating the conditioning operation from the treatment
process (Van Frank and Eck 1969). By preventing the preservative from
mixing with the steam condensate, this waste stream would no longer
contain preservative. Apart from added capital cost, the major
disadvantage of this method is the increase in time it takes to transfer the
wood from one retort to another. During this time, the wood will cool and
therefore some steaming of the wood will be required during treatment
(southern pine must be treated while hot to prevent formation of sludge).
In addition, many facilities have only two or three retorts and thus would
have to buy additional equipment.
o Conversion from open steaming to closed steaming.
By far the most popular method of reducing waste associated with
conditioning has been the conversion from open to closed (or modified)
steaming (Thompson 1973b). Converting from open to closed steaming
requires only the installation of a storage tank for the recycled condensate.
Conversion to modified steaming, while not as effective, requires only
some piping modifications. In addition to the reduction in wastewater
generated, (it is estimated that facilities using open steaming generate
B18-20
-------
2.43 gallons of wastewater per cubic foot of wood treated, closed steaming
facilities generate 0.45 gal/ft^, and modified steaming facilities generate
about 1.0 gal/ft^ (USDC 1983)), it is also possible that the resulting
wastewater is easier to treat. Studies have shown that emulsification of
oil in water apparently results from violent agitation of the oil and water
when passing through a steam trap (Thompson 1973a). When closed or
modified steaming is employed, the water in the retort can be drained
through a discharge pipe as opposed to a steam trap orifice. By proper
piping design and using a full port ball valve for on-off level control so as
to minimize agitation, the amount of oil in water emulsion formed can be
reduced.
o Use of air-flotation on the oil/water separator effluent.
By installing an air-flotation (dissolved-air or induced-air type) unit on the
effluent from the oil/water separator, an increase in recovery of the
preservative can be achieved. The float material produced (preservative,
water, and air) would be mixed with the preservative recovered from the
separator and sent to the work tank system for dehydration, filtering, and
reuse. API separators, followed by air-flotation units, are quite common at
many oil refineries.
o Use purer preservatives.
By using purer preservatives, the total amount of toxics entering a wood
preserving facilities wastewater could be lowered. As recently reported,
the USEPA wants suppliers of pentachlorophenol to lower the level of
hexachlorobenzo-p-dioxin (HXCDD) present in the preservative from
15 ppm down to 1 ppm (USEPA 1984). The authors of this report believe
that further investigation of preservative purification by the suppliers is
warranted. While increasing the degree of purification required may lead
to increased waste production at the supplier facilities, it is believed that
they are in a much better position to properly handle the resulting waste.
9.1.3. Sludge from Work Tank System
Before preservative can be reused, it usually undergoes some form of treatment. For
oil-based preservatives, treatment usually includes dehydration and filtering. For
818-21
-------
water-based preservatives, treatment can consist of pH adjustment followed by
filtration. The ways in which sludge from the work tank system can be reduced are:
o Clean the wood prior to insertion into retort.
By ensuring that the least amount of dirt, silt, and loose wood fiber enters
the retort before treatment, the least amount of sludge will be produced.
Treatment of dirty wood should never be allowed (Wilkinson 1979).
o Better operating practices.
Also to be avoided is treatment of unseasoned wood, long soaking,
contamination of treating solution, and incomplete mixing of preservative.
Since treatment and disposal of sludge represents a loss of valuable raw
material (preservative), efforts are made to minimize its production. Dust
entrainment into solution can be avoided by constructing windwalls around
the drip-pad area. Frequent cleaning of drip-pads and use of covers will
help to prevent dirt from getting into solution.
9.1.4 Storm Runoff and Area Washdowns
Water that contacts freshly treated wood or flows across plant areas contaminated
with preservative constitutes part of the wastewater generated by a facility. For
most plants, the major source of area contamination is due to preservative dripping
from treated wood. In the past, this preservative would drip in the treated wood
storage yard and contaminate the area. Today, most facilities have installed drip
tracks and pads so that the drippings can be collected and reused. For facilities using
CCA, area contamination used to occur when the retort door was opened and some of
the solution spilled out. Nowadays, spill basins have been installed so that spilled
solution can be recovered and reused. Other sources of area contamination are due to
equipment leaks and spills. The following source reduction measures were noted for
reducing wastes associated with storm runoff and area washdowns:
o Install drip tracks, pads, and spill basins.
This method is widely used by the industry and was among one of the first
waste minimization measures undertaken (Van Frank and Eck 1969; Morgan
and Burdell 1978).
B18-22
-------
Covering the pad and treated wood storage area.
By covering the pad and treated wood storage area, less contaminated rain
water would have to be collected and treated. In addition, the collection
of dirt and silt would also be reduced so that less work tank sludge would
be generated. While this method has been identified as having been used by
the industry (Thompson 1973a, Wilkinson 1979), the cost of covering the
yard could be prohibitive.
Divert storm water away from contaminated areas.
For facilities with large areas of contamination (old holding ponds and
lagoons, for example), diversion of clean storm water away from the area
reduces the amount of runoff that must be collected and treated (Morgan
and Burdell 1978). In addition, reduction of these flows also allows for
more reliable and consistent operation of the wastewater treatment
system, since it is less subject to overloading.
Better operating practices.
According to obtained information, most wood is left on the pad for several
hours following treatment. During this time, oil-based preservatives "kick-
back" and drip onto the pad, while wate^-based preservatives become
"fixed" and their leachability diminishes. Following this period, the amount
of preservative that drips or leaches from the wood is extremely small.
However, it was reported that water-based preservatives can leach from
wood for up to one week after treatment (Wilkinson 1979). Therefore,
maximizing the amount of time the wood is allowed to remain on the pad is
essential for maintaining a clean treated wood storage area.
Other related methods include proper stacking of the wood in the retort so
that the preservative can drain freely, use of a higher degree of vacuum to
remove unabsorbed preservative, use of dry clean-up methods for drips and
spills occuring off the pad, minimizing water use for cleaning the pad, and
always ensuring that the wood produced is clean. While the USEPA tried
to specify approaches which assure production of clean wood, the AWPI
recommended that particular methods not be specified, due to the diverse
and complex nature of the process (USEPA 1984). Instead, it was
recommended that facilities adhere to clean wood standards (AWPA
B18-23
-------
standard C-l, for example) which only set the degree of cleanliness
required and not the ways to achieve it.
9.1.5 Cooling Water
For facilities that use single-pass systems, barometric condensers associated with
vacuum generation via steam-jets can be a large source of wastewater. Barometric
condensers permit direct contact between cooling water and process vapors. While the
resulting level of water contamination is lower than the retort effluent, this stream
still represents a major wastestream requiring treatment. The following source
reduction methods for reducing wastes due to contaminated cooling water were noted:
o Replace barometric condensers with surface condensers.
o Construct cooling water reuse systems.
In 1974, 41 percent of all plants employed surface condensers (in which
there is no contact between cooling water and the process vapors), 41
percent employed barometric condensers, and 18 percent did not use
condensers (Thompson 1975). Compared to a 1972 survey (Thompson
1973b), the number of plants using barometric condensers was decreasing,
and 84 percent were recycling their cooling water. Therefore, it appears
that industry has widely implemented the above two measures.
o Use vacuum pumps instead of steam jets.
o Use of non-chromate cooling water treatment.
For facilities using oil-based preservatives, use of non-chromate cooling
water treatment for corrosion inhibition would eliminate the need for
treating cooling tower blowdown for chromium. For facilities where
water-based preservatives are used, the amount of chromium in the
treatment sludge would be reduced.
o Better preventive maintenance.
Another method that deserves special attention is the effect that better
preventive maintenance can have on the contamination of cooling water.
Wilkinson (1979) and Thompson (1973a) stress the importance of detecting
B18-24
-------
leaks in process equipment so that preservative loss can be minimized. By
reducing the amount of preservative that leaks into the cooling water, the
overall requirements for treatment can be reduced.
9.2 Implementation Profile
The wood preserving industry has implemented many measures to reduce the amounts
of waste produced. Some of the most effective measures include: purchasing of raw
materials in bulk; converting from open to closed steaming; installation of drip tracks,
pads, and spill basins; and replacement of barometric condensers with surface
condensers. Additional measures that could result in future reductions include the use
of closed systems for powder handling, use of air flotation on oil/water effluent,
covering of the pad and treated wood storage area, and use of non-chromate water
treatment chemicals. While the costs for some of these methods could be quite low
(the USEPA estimated that the cost for a closed powder handling system would be
10,000 dollars), others could be very expensive (covering of the treated wood storage
yard). The cost of each method would have to be weighed against the cost savings in
waste treatment and disposal. Since operating facilities have already installed waste
treatment systems to achieve regulatory compliance, some of the proposed methods
might be hard to justify economically. For facilities under design, incorporation of the
proposed methods could be more cost-effective.
9.3 Summary
Table 9-1 presents summary data on wood preserving waste sources and control
methods. The ratings of. each listed method shown in the table were based on review
of the available literature and consultation with industry personnel. Each method was
rated on a scale of zero to four for its effectiveness, extent of current use, and future
application potential. The current and future extents of waste reduction were then
derived using the methodology presented in the introduction to this appendix.
A current reduction index of 3.0 (75 percent) is indicative of the very high level to
which the waste has already been minimized (CRI is a measure of the reduction of
waste that would have been generated if none of the methods were practiced at their
current level of application). By implementing additional waste reduction measures or
increasing the use of existing measures, the amount of waste currently being
B18-25
-------
TABLE 9-1 SUHHARY OF SOURCE CONTROL KTHODOLOGY FOR THE HOOD PRESERVING INDUSTRY
03
h-1
CO
I
INJ
ON
1 1
Waste Stream |
1 1
1 Raw Material Wastesll.
1 12
i 1 3
|4
1
| Wood Conditioning [1
| Condensate |2.
i 3.
(*•
I |5.
1
| Sludge from Work |1
) Tank System |2.
i
Storm Runoff and | 1
1 Area Washdowns |2
i 13
! 14.
1 1
| Cooling Water |1
I 12.
I 13.
I l«
I |5
I *
1 !
| All Sources |
Control Methodology |-
1
Purchase raw materials in bulk )
Use of closed systems for powder
Use of rinseable/recyclable drums
Getter operating practices |
Overall
Increase use of air/kiln drying j
Use separate retort for conditioning |
Convert from open to closed steaming |
Use air-flotation on effluent |
Use purer preservatives I
Overall 1
Better operating practices
Clean the wood prior to treatment 1
Overall j
Install drip track, pad, spill basin |
Cover pad i treated wood storage area
Divert storm water runoff |
Better operating practices '
Overall
Replace ba^oirietnc condensers 1
Construct cooling water reuse system |
Use vacuum pump instead of steam jet I
Use non-chromate water treatment
Better preventive maintenence |
Overall
Al? Methods
Tound OocU'T
Quantity 1
1 I
3 1
0 1
' i
< 25 i
3 i
1 1
3 !
1 1
t 1
1 30 I
3 1
1 1
2 00 !
3 :
3 i
3 i
2 1
2 75 |
3 1
3 I
3 1
2 i
3 1
2 50 j
t'uUtion i
1
Quality |
• |
2 1
0 1
1 I
1.00 |
2 I
1 i
3 I
1 !
0 I
1 40 1
2 1
1 i
1 50 |
J 1
(. \
2 !
2 !
2 25 |
3 1
•> I
3 1
2 1
3 1
2 SO I
Waste i
Reduction !
Effectiveness 1
4 \
3 1
2 1
J 1
3 00 I
4 1
2 1
3 1
2 1
1 i
2 40 !
3 i
3 1
3.00 1
3 i
3 1
3 1
3 1
3 00 1
4 !
3 !
3 1
2 1
3 1
3 00 I
Extent of | future 1 fraction of !
Current Use j Application 1 Total Waste |
i Potential \ \
31 1 !
2 I 4 | |
0 ! 1 I 1
3 1 I
2 00 1 1.75 ! 0 01 I
2 ! 11 1
1 i 11 !
3 21 I
11 31
01 21 I
1 40 | 1 80 1 0 10 |
3 i i : i
31 1 1 1
3.00 1.00 | 0 01
4 1 ! I
11 3 i
3 11 |
3 1 1 1 I
2 75 1 1 50 I 0 75
3 | 1 ! 1
3 I 11 !
21 21 |
0 1 3 |
31 11 |
2 20 i 1.60 | 0 13 |
1 '-00 |
Current |
Reduction |-
Index |
3.0 |
1.5 |
0 0 I
2 3 |
3.0 |
2.0 |
0 5 |
2 3 I
0.5 |
0.0 I
2.3 |
2 3 |
2.3 |
2 3 |
3.0 |
0.8 I
2 3 |
2.3 I
3.0 |
3 0 |
2.3 I
1.5 |
0 0 !
2.3 I
3.0 |
3.0 |
future Reduction Index
Probable Maximum
0 3 i
1.5 | 15
0 5 |
0.2 I
06 15
0 5 |
0 4 |
0 4 |
I.! | 11
05 |
0.6 | 11
021 0.2
0 2 | 0.2
02| 02
0 0
1 7 | 17
0.2
0 2 |
051 17
0 3 |
0.2 |
0 8 |
1.5 1.5
0.2 |
06| 15
0 5 | 16
1
1
|
1
1
1
1
I
1
1
|
1
1
1
1
1
1
1
1
(*) These streams include listed T" and/or "K" RCKA wastes
-------
generated can be reduced by a moderate amount, as measured by a future reduction
index of 0.5 to 1.6 (13 to 40 percent). The most effective measures for achieving this
reduction would be to cover the pad and treated wood storage area, use non-chromate
cooling water treatment chemicals, install air-flotation units on oil/water separator
effluent, and install closed systems for powder handling.
10. PRODUCT SUBSTITUTION ALTERNATIVES
10.1 Steel and Concrete
Very few substitutes exist for treated wood products. Substitution of metal or
concrete pilings for creosote treated wood pilings has been investigated (USDC 1983).
The major conclusions of this study were that:
Although they offer considerably longer life spans, concrete and steel
pilings cost three to five times as much as wood. The overall economic
analysis favors treated wood.
While producing less hazardous waste than the wood treatment process,
production of steel pilings also produces waste.
Due to increased environmental/regulatory pressure, many wood preservers
have stopped producing creosote-treated wood pilings.
Subsequent inquiries with industry revealed that environmental/regulatory pressure
was not a factor for stopping the use of creosote. A subsequent study (USEPA 1984)
showed that a large annual cost penalty ranging from 1.3 billion to 2.1 billion dollars
would have to be incurred by consumers if treated poles were unavailable and concrete
or steel had to be used. For pilings, the total annualized installed cost would increase
by 33 percent for concrete and 67 percent for steel. Use of concrete for railroad ties
would also result in adverse cost increases. Overall, halting the use of all wood
preserving chemicals would have major economic impact nationwide.
B18-27
-------
10.2 Trends in Preservative Usage
Investigation of wood preservation statistics reveals that major changes have been
taking place in the industry. While the amount of wood treated with creosote and PCP
has remained relatively constant over the last 15 years, production of CCA treated
wood has increased about 800 percent (AWPA 1982). Over the same time period, many
new CCA-using facilities have been built and many oil-based using facilities have
converted or closed. As shown in Table 10-1, the percentage of plants using creosote
and PCP has dropped while the percentage of plants using CCA has increased. Also of
importance is the large drop in the percentage of facilities using two or more
preservatives. As a result of using a single preservative in the facility, the
wastewater treatment system is expected to be more efficient and reliable because no
cross contamination of materials can occur. It is also expected that the facilities'
overall waste generation rate should be lower due to the standardization of
preservative usage.
Table 10-1 Production Trends in the Wood Preserving Industry
Preservative
PCP
Creosote
CCA
Two or more
Percent of Plants
1974
9
53
50
41
67
1984
20
24
81
19
Source: Status of Pollution Control (Thompson 1975); Wood Preservation Statistics
(Micklewright 1985).
The alternatives to most widely used preservatives (PCP, creosote and arsenicals)
include naphtenates of zinc and copper. Zinc naphtenate appears to be safer than PCP
while exhibiting equivalent wood preservation characteristics.
An environmental issue related to the choice of preservative is the disposal of used or
spent treated wood. Disposal by incineration is an option likely to be considered as an
B18-28
-------
alternative to landfilling. While wood treated with creosote can be burned with
virtually complete destruction, wood treated with PCP produces hazardous hydrogen
chloride fumes which should be scrubbed and removed from the flue exhaust. Wood
treated with CCA also presents a problem, since the ash produced by burning will still
contain heavy metals and does require special disposal. Currently, all used treated
wood is disposed of in landfills and dumps. State and federal regulations on
incineration of treated wood have been promulgated (USEPA 1984, 1985).
11. CONCLUSIONS
While the wood preserving industry has achieved a large reduction in the amount of
waste it produces (a current reduction index of 3.0 or 75 percent), it appears that
further reductions are possible. Our estimates indicate that possible waste reductions
are characterized by a future reduction index of 0.5 to 1.6 (13 to 40 percent) which is
indicative of a moderate potential. Several major methods appear to be quite
effective, including covering the pad and treated wood storage area, use of non-
chromate cooling water treatment chemicals, and use of air-flotation on the oil/water
separator effluent. Waste reductions will also occur as more and more facilities
implement waste reduction programs and as older equipment wears out and is
Replaced. Since many of the proposed waste reduction measures are cost-effective,
implementation can best be achieved through an increase in awareness.
12. REFERENCES
AWPA. 1982. American Wood Preservers Association. Wood preservation statistics
for 1981. Proc. Am. Wood-Preservers' Assoc. Vol. 78.
Hunt, M.N., and Garratt, G.A. 1953. Wood preservation. 2nd ed. New York, N.Y.:
McGraw Hill Book Co.
Micklewright, J.T. 1985. Wood preservation statistics, 1983-84. A draft report to the
American Wood Preservers Institute. International Statistics Council, Inc. October
1985.
Morgan, J., and Burdell, 1978 C.A. Zero discharge of wood treating plant effluent and
specified rainwater runoff. Proc. Am. Wood-Preservers' Assoc. vol. 74.
Thompson, W.S. 1973a Pollution control. In Wood deterioration and its prevention.
2 vols. Syracuse, N.Y.: Syracuse University Press.
1973b Status of pollution control in the wood preserving
industry. Proc. Am. Wood-Preservers' Assoc. vol. 69.
B18-29
-------
1975. Status of pollution control in the wood preserving
industry in 1974. Proc. Am. Wood-Preservers' Assoc. vol. 71.
USDC 1983. U.S. Department of Commerce, Office of Technology Assessment.
Technologies and management strategies for hazardous waste control,' vol. 2;
alternatives for reducing hazardous waste generations using end-product substitution.
PB83-239962. Washington, D.C.: U.S. Department of Commerce.
1985. U.S. Department of Commerce, Bureau of the
Census. Wooden containers and miscellaneous wood products. In 1982 Census of
manufacturers. MC82-I-24C Washington, D.C.: U.S. Government Printing Office.
USEPA. 1973. U.S. Environmental Protection Agency, Office of Air and Water
Programs. Air pollution engineering manual. AP-40. 2nd ed. Washington, D.C.: U.S.
Environmental Protection Agency.
1979a. U.S. Environmental Protection Agency. Effluent
Guidelines Division, Office of Water and Waste Management Proposed development
document for effluent limitations; guidelines and standards for the timber products
processing; point source category. EPA-440/l-79/023b. Washington, D.C.:U.S.
Environmental Protection Agency.
1979b. U.S. Environmental Protection Agency, Office of
Water Planning and Standards. Economic impact analysis of alternative, pollution
control technologies. Wood preserving subcategories of the timber products industry.
EPA-440/2-79-018 Washington, D.C.: U.S. Environmental Protection Agency.
1980. U.S. Environmental Protection Agency, Office of
Research and Development. Treatability manual, vol. 2; industrial description. EPA-
600-8-80-042b. Washington, D.C.: U.S. Environmental Protection Agency.
1984. U.S. Environmental Protection Agency. Office of
Pesticides and Toxic Substances Wood preservative pesticides; creosote, penta-
chlorophenol, inorganic arsenicals. Position document 4. Washington, D.C.: U.S.
Environmental Protection Agency. Reprinted by: National Forest Products
Association.
1985. U.S. Environmental Protection Agency. Before
the administrator. In the matter of chapman chemical co. et. al. petitioners. FIFRA
docket nos. 529 et. al. settlement agreement; signed September 30, 1985.
Van Frank, A.J., and Eck, J.C. 1969. Water pollution control in the wood preserving
industry. Proc. Am. Wood-Preservers' Assoc. vol. 65.
Wilkinson, J.G. 1979. Industrial timber preservation. London: Associated Business
Press.
13. INDUSTRY CONTACTS
R.G. Smerko, President, and W.G. Talarek, General Counsel, American Wood
Preservers Institute, Vienna, VA.
J.A. Smigel, P.E., Chief Environmental Engineer, Atlantic Wood Industries, Savannah,
GA.
B18-30
-------
H.M. Rollins, P.E., H.M. Rollins Wood Preserving Service Co., Gulfport, MS.
H. Ervin Jr., Manager, Technical Services, International Paper Co., Dallas, TX.
C. Kempinska, Coordinator, Technical and Regulatory Affairs, and C.P. Markle,
Manager, Environmental Regulatory Programs, Koppers Co. Inc., Pittsburgh, PA.
L.D. Lonning, Manager, Technical Services, McFarland Cascade, Tacoma, WA.
C.A. Burdell, Director, Technical Services, Southern Wood Piedmont Co., Spartanburg,
SC.
P.M. Castle, Vice President/General Manager, William C. Meredith Co. Inc., East
Point, GA.
818-31
-------
-------
GOOD OPERATING PRACTICES
-------
-------
1. INTRODUCTION
1.1 Purpose
This study is part of an assessment of source reduction practices in U.S. industry and is
intended to complement the series of process studies in Appendix B to this volume. In
the process studies, good operating practices are evaluated according to their current
and potential waste reductions within the individual processes. This study seeks to
generalize over these separate treatments in order to develop a more global
perspective on sound waste-minimizing practices.
The purpose of this study is to identify and expand upon the elements of a set of
source reduction practices which can be implemented through an organizational
modification of a production operation (as opposed to technology modification or
changes to input materials). In numerous case studies of successful source reduction
efforts, a factor weighing heavily in the achievement of waste reduction has been a
procedural or policy change within the plant or firm. In the context of waste
minimization, the general objective of good operating practices or procedures is to
minimize material losses and the probability of their occurence. It may have been
something as simple as a change in the method of handling the hazardous materials, or
something as complex as a change in management outlook. Proper procedures and
policies leading to reductions in the amount of hazardous wastes generated should be
considered in the overall operating plan.
This discussion of good operating practices is, by necessity, a qualitative one. Ideally,
an investigation into the feasibility of making an operating change would express the
end result and potential implementation quantitatively in terms of cost and other
measurable quantities. This approach does not appear viable in this study for the
following reasons:
o When good operating measures are implemented, it is often not possible to
monitor or document the results.
o Good operating measures are often employed in conjunction with other
plant changes, which makes the specific waste-reducing/cost-saving impact
of the good operating measures difficult to separate from the total impact.
B19-1
-------
In sum, no solid data exist for the general role of good operating measures in source
reduction. A qualitative discussion may nevertheless point out the characteristics of
good operating practices. Also, the characteristics of operations which are amenable
to operating improvements may be explored qualitatively.
1.2 Definitions
In the context of this study, a good operating practice is defined as a procedure or
institutional policy within a service or manufacturing operation which results in a
reduction in hazardous waste generation. Good operating practice relates primarily to
the human aspect of production, i.e. organizational structure, initiatives, operations
planning and control, as opposed to changes in technology or materials. The following
is a representative list of areas which might easily lend themselves to changes in
operating practices along with a brief description:
1) Material handling improvements - any change in material handling
procedures which reduces the ratio of waste to material delivered.
2) Management initiatives - any revision of operational supervisory structure,
schedule, or set of managerial procedures and incentives which lead to a
reduction in waste per unit output.
3) Employee training - any additional training of employees or increase in
employee awareness of operating practices and their impacts on waste
generation which results in lower waste per unit output.
4) Scheduling improvements - tighter or more accurate operations scheduling
and plant area communications which result in lower waste per unit of
production.
5) Spill and leak prevention - a change in existing procedure which reduces
waste resulting from spills or leaks and thereby lowers waste per unit of
production.
B19-2
-------
6) Preventive maintenance - maintenance procedures designed to reduce
incidences of equipment breakdown and inefficiency or of process fluid
leakage, thereby lowering the amount of waste per unit output.
7) Corrective maintenance - efforts, such as resetting control valves or
adjusting process temperatures, can increase productivity and prevent
loss.*
8) Material/waste tracking or inventory control - any improvement in the
tracking of a material's location, quality, age, use or alteration of its
purchased lot sizes, which would result in lower waste per unit output.
9) Additional documentation - additional procedural guidelines or material
information which results in less waste per unit output.
10) Waste stream segregation - any measure taken to isolate waste streams by
1) toxicity, 2) type of contaminant, and/or 3) physical form, which reduces
the amount of waste produced per unit output.
The above elements of good operating practices are often referred to as "good
housekeeping".
1.3 Format
This study is presented in three parts. The first part (Section 2) presents the types of
plant operations which could minimize waste generation and describes some
representative good operating practices. The division of these practices into waste
stream segregation, institutional measures, procedural measures, and loss prevention
programs is made according to the factor playing the largest role in their
implementation: technology, personnel, procedure, or program structure. Good
operating practices are in fact difficult to separate from each other because they
constitute an integal part of sound overall plant operation. For example, the labels
that aid in material handling serve also to track materials and control inventory. Good
* Westinghouse Electric Corp. 1985: Personal communication.
B19-3
-------
material storage practices can be part of a spill or fire prevention program, as is
preventive maintenance. Management initiatives and employee training affect all
facets of plant operations and therefore all technological, procedural, and program
practices.
Section 3 records some observations on good operating practices ~ as they apply to
different types of processes — and draws conclusions from the foregoing exposition of
individual practices in Section 2. Generic batch and continuous processes are
described in terms of how the suggested good operating measures can be integrated
into the production process.
Section 4 lists the references used for this study. The list reveals a reliance on
material from pollution prevention programs and on specific case studies and topic
treatments drawn mostly from chemical engineering publications. This material was
chosen for its explanatory and illustrative value.
2. TYPES OF GOOD OPERATING PRACTICES
2.1 Waste Stream Segregation
Hazardous waste which is hauled to off-site disposal facilities is often a combination
of two or more waste stream types or of one type of waste and water. Waste stream
segregation as a good operating measure may involve:
o Isolation of hazardous materials from nonhazardous materials.
o Isolation of hazardous waste by contaminant.
o Isolation of liquid from solid waste.
Waste stream segregation can result in smaller waste haulage volumes and easier
disposal of the hazardous waste generated while producing the same amount of
product. In addition, segregation can often result in simplifying waste treatment or
make it easier to recover and recycle materials.
When a non-hazardous material is mixed with a hazardous material, there exists the
possibility that the entire mixture must be classified as hazardous. By separating
hazardous from non-hazardous materials or not allowing them to mix in the first place,
819-4
-------
the quantity of waste requiring regulated disposal is reduced. ICI Americas, Inc., of
Goldsboro, N.C., an agricultural chemical research and development firm, designates
separate containers for hazardous and non-hazardous waste, thus preventing cross-
contamination of the non-hazardous waste with the hazardous waste. In conjunction
with a procedure for segregating chlorinated solvents from non-chlorinated solvents
and a policy of returning unused chemicals to the distribution center, the overall
volume of waste exiting ICI was reduced from 100 drums in 1981 to 60 drums in 1984.
The product output of the firm nearly doubled in that period; observing that disposal
costs doubled as well (from $120/barrel to $266/barrel), the firm saved roughly $37,000
in 1984 alone (Huisingh et al. 1985).
The isolation of hazardous waste by contaminant often allows wastes to be recycled or
reused, thereby reducing the off-site disposal requirement. Rexham Corporation of
Matthews, N.C., does high-tech printing and coating, including film substrate, for the
photographic industry. Toluene is used to clean the ink from the press, and runoff
toluene is collected as waste. Rexham has nearly eliminated its toluene waste by
segregating cleanup toluene according to the color and type of ink cleaned and then
reusing the collected wastes to thin future batches of the same ink (Huisingh et al.
1985). The procedure has no effect on product quality and has resulted in almost 100%
reuse of the toluene solvent.
Martin Marietta Aluminum of Torrance, CA. reduced its cleaning and waste hauling
costs by $50,000/year through the simple operation of filtering aluminum particles
from soluble oils (Suzuki 1985). Prior to installing the filter apparatus, the company
was forced to transport its oil wastes to a Class A dump site, rather than taking it to a
municipal site, due to EPA regulations on effluent disposal. Disposal costs were too
high. Installation of the filter cost $1,000 and was performed by Martin Marietta
employees. The oil recovered from the filtering operation is reused, resulting in lower
oil purchase costs and oil waste transport costs.
In paints and coatings formulation the volume of hazardous waste can sometimes be
reduced by segregating empty paper bags that contained toxic pigments (such as lead
chromate) from non-hazardous discards.
B19-5
-------
For segregation to be cost effective, however, the savings resulting from reuse,
recycling, resale, or lower disposal costs must outweigh the costs of the additional
personnel allocations and of the on-site equipment and storage facilities needed.
2.2 Personnel Practices
2.2.1 Management Initiatives
Revisions in operational supervisory structures, schedules, or sets of procedures
contribute greatly to a company's efforts to reduce waste and/or come into regulatory
compliance. The need to head off rising disposal costs and to confront environmental
issues and responsibilities has encouraged many firms to institute self-standing
environmental programs and policies. These include in-house chemical waste manage-
ment programs, which can take the form of ongoing waste management/reduction
programs, environmental auditing programs, or periodic environmental reviews of
operations.
Waste Reduction Programs
In-house ongoing waste reduction programs can range from complex programs like 3M
Corporation's Pollution Prevention Pays program, which is independently staffed and
extends to 3M operations worldwide, to simple basic pollution awareness programs
wherein operational managers and employees are asked to identify ways of reducing
waste. Established in 1975, the 3M Corporation's 3P program has since achieved the
prevention of an estimated 103,000 tons of sludge and solid waste annually. The
company estimates its cumulative savings over the period at about $248 million. The
savings are in the form of retained sales on reformulated products, conserved
materials and energy, and the ability to delay or completely eliminate the purchase of
pollution control equipment.
According to information from 3M, the initiation of a "3P" type program begins with a
strong management commitment. The suggestion is to "work from the top down". The
plan can be implemented beginning with an environmental audit. The wastes which
emanate from the operation can be documented via an environmental accounting
schematic (see Figure 2-1) and quantified, after which all costs associated with these
B19-6
-------
FIGURE 2-1 ENVIRONMENTAL ACCOUNTING SCHEMATIC
ENVIRONMENTAL ACCOUNTING SCHEMATIC
TAYLOR INSTRUMENT - ROCHESTER NY
AIR DISCHARGES
R
E
C
Y
C
L
E
SCRAP
PAPER
BOILER/
UTILITY
BLDG
NK
ACID/CAUSTIC
RINSE
^
COOLING
HATER
SPENT
ACID/CAUSTIC
DRUMS
-)• TRASH
CLEAN SOLVENTS
<•
SILVER
SCRAP
1
PROCESS
DISCHARGE
CHLORINATED
\
CLEANING
SOLVENTS
/
HG
SCRAP
SANITARY
SEWAGE
WATER DISCHARGES
Source: 3M Corporation
L
D
A
S
T
E
B19-7
-------
discharges can be calculated. This procedure gives an overall picture of the types and
amounts of waste being generated and the costs associated with generating the waste.
Once the environmental picture is established, meetings are held with laboratory,
research and development, manufacturing, and process engineering personnel to
identify a strategy for reducing discharges and to choose the problems and solutions on
which to concentrate. At 3M, solutions generally fall into one or more of four
categories: 1) product (re)formulation, 2) process modification, 3) equipment redesign,
and 4) recovery/reuse. The choice of a problem and solution can be influenced by the
chance of success, the potential to apply the solution to other problems, and the age of
the process. At 3M, processes that have remained unchanged for 10 years are
targeted.
Following the implementation of a solution, further efforts are made to recognize
contributors, publicize success stories, and keep account of the quantities of waste
reduced and the costs saved. In addition, the 3P program scrutinizes all new projects
using 3P techniques both in the laboratory and in the initial engineering phase. In all,
more than 1,500 3P projects worldwide have been approved since the program began,
323 of which were approved in 1984 and are expected to contribute $42.5 million to
future cost savings in 3M's estimation. An example *of a publicized 3P "success story"
is given in Figure 2-2.
Other examples of in-house waste reduction programs are found at Borden Chemical
Company in Fremont, CA, and at Emerson Electric Company in Murphy, North
Carolina (Huisingh et al. 1985). Borden Chemical Company's management initiative
consisted of process reviews by management and the educating of employees to be
aware of hazardous waste reduction opportunities. As a result of this process, a 93%
reduction in the organics entering the company's wastewater treatment system was
realized. In Borden's case the residual phenol from the hose that conveys phenol from
delivery trucks to storage tank had previously been permitted to empty into the
treatment system. Currently the hose is flushed with water and the water-phenol
mixture directed to a storage tank for use in the process. This change in practice
came about because, 1) Borden had taught its personnel that any amount of chemical
waste is important in the total amount of waste produced and 2) Borden management
had "walked through" the movement of materials in the production process.
B19-8
-------
FIGURE 2-2 SAMPLE PUBLICITY FOR POLLUTION PREVENTION PROJECT
Pollution
Prevention
Pays
* *~* -r x on o o* •
A Compendium of 3P Success Stones
Keener Cleaner
Tank Laved, Solvent Saved, Toil Shaved
Problem
One hundred ten tons of tank
cleaning solvent were being tost
annually at 3M's plant in Hilden, West
Germany. The plant had no solvent
recovery capability so the used
solution was processed by an outside
contractor from whom 3M bought the
recovered solvent.
Another aspect of the problem was
the excessive manpower being spent
to clean Hilden's two 300-gaJlon
vessels.
Clean-up was necessary to avoid
contamination of an ensuing batch
when the color of a coating solution
was changed. The tank was flushed
with solvents and roughly cleaned by
agitation of the solution-mixing
propeller. Manual cleaning with
brushes followed, sometimes
requiring that an employee enter the
tank. Cleaning a tank took three hours
and the time of one employee.
Annually, cleaning the plant's two
tanks required 800 hours.
Solution
Hilden developed a fully-automated
cleaning system in which the solvent
mixture is pumped under high
Sotv«m flno«m«a By routing spray h««a
HiWcn'i land Maning preottms.
pressure through a rotating spray
head. The old mixing tanks were
replaced by new specially designed
stainless steel vessels.
The spray head is portable and can
be inserted in a tank by one worker in
five minutes. Because the new
system is dosed, an operator does
not have to enter the vessel to dean it
and has no direct contact with
solvent.
The system incorporates a
sedimentation tank to recover solvent
from the cleaning solution that used to
be given away and bought back.
L
• The new system cost $69,000 and
in its first year saved S6i ,500 in
solvents and labor costs.
• Clean-up time for a tank was
reduced from three hours and one
operator to 10 to 20 minutes plus
five minutes of an operator's time :c
insert the spray head. The new
system also does a more thorough
cleaning job than was oone
manually.
• Reduced down time for tank
cleaning meant improved
productivity.
• Employee safety and comfort were
improved by eliminating tanx entry.
me Idea Team
The technical employees who
concaved and implemented
the new tank cleaning method:
Manfred Hart
Manufactumg.
HlkJen, Wem Germany
. Neuaa, We*t Germany
Environmental Engineering and
Pollution Control Dept./3M
P O. Box 33331. Bldg. 21-2W
Saint Paul. Minnesota 55133
(612) 778-*79l
Source: 3" Corporation
B19-9
-------
The Emerson Electric Company developed a chemical waste management program on
the initiative of its Special Products Division and with the support of the president and
vice president. In addition to having addressed several immediate waste management
concerns, the program seeks to 1) eliminate unauthorized dumping of spent chemicals,
2) educate employees on the proper disposition of waste chemicals, 3) prevent
chemical spills, and 4) comply with environmental laws and regulations. Through its
waste management program, Emerson has been able to adjust its effluents to conform
to effluent guidelines, to reduce its downtime, and to implement various pollution
reduction measures with a maximum payback period of 2 years for a single project.
Environmental Auditing
Environmental audits are efforts to improve environmental quality management and to
check environmental performance (NCPPP 1985). Specifically, an environmental audit
is "a thorough examination and testing of a company's or facility's operating records
and environmental practices, to gather information about its compliance with federal,
state, and local environmental regulatory requirements" (Russell 1985). Environmental
auditing differs from monitoring in that auditing actively seeks to ensure the
effectiveness of a company's environmental activities, only one of which may be
effluent data collection through monitoring.
There are two types of environmental audits: a) those carried out by firms them-
selves, using their own employees or outside contractors, or b) those imposed by
regulatory agencies. Audits originating in-house fall within the context of good
operating measures. Firms typically conduct audits to determine their own regulatory
compliance and thereby avoid fines, but can also use the information to formulate
expenditure and operating budgets and to search for ways to reduce hazardous waste
and its associated costs.
The size and composition of an internal audit team depends upon the resources
available to the company, the size of the waste problem, the objective of the audit,
and other similar factors. Since auditing can be a legal procedure, the team ideally
could include an attorney and a Certified Public Accountant who can advise technical
and supervisory personnel on liabilities resulting from audits and on tax and financial
issues. The team should also include at least one engineer knowledgeable of plant
B19-10
-------
operations and another who is familiar with industry practices and environmental
regulations. If a component of the audit is problem rectification, then more engineers
may be included: an environmental department representative, a principal design
engineer, a facilities engineer, a process engineer, an instrument/control engineer,
etc.
Auditing procedures can vary according to what the firm wishes to accomplish with it.
A sample step-by-step procedure for audit preparation and performance might be the
following, (Conrad 1985):
1) Assemble pertinent documentation. Documentation of plant operations can
include current operating instructions, piping and instrumentation
diagrams, layout drawings, instrumentation logic, maintenance and
preventive maintenance records, environmental incident reports and
emergency procedures. Documents on audit procedures may include
outlines for interviews with operating personnel, check lists, manuals,
and/or other working papers.
2) Conduct an environmental process review. An environmental process
review is an analysis of the plant's design and operation for current or
potential points which can generate process wastes. The analysis can
include an evaluation of the process raw materials, the process equipment,
the way in which the equipment is integrated into the overall facility, and
other process-related concerns.
3) Conduct a procedure evaluation. Using the documentation on operating
procedures, a check is done to ensure that the plant is operated as
designed. The process procedures must also correspond to the process
instrumentation logic, which describes the action of valves, instruments,
and interlocks, and identifies switch points, alarm points, and fail-safe
points. Evaluation of material handling and warehousing procedures should
reveal practices which have waste-generating potential.
4) Carry out a site inspection. Questions that have arisen during the process
review and procedure evaluation can be answered during the site inspec-
tion. The inspection permits an assessment of the accidental spill/leak
B19-11
-------
potential in the physical dimensions of the storage, process, and effluent
collection areas of the site. Other practices can also be evaluated.
5) Report on and follow-up the findings. The report emanating from an
environmental audit lists the personnel interviewed, the probable impacts
of each problem, recommendations on solutions, and the cost
outlay/savings estimates for each solution. Follow-up meetings of the
team are held to ensure action on recommendations.
Some examples of problems frequently encountered in an audit are given in Table 2-1.
Through environmental audits, firms can rectify problems before being forced to do so
by regulatory agencies. This helps to avoid compliance costs and fines for violations.
Moreover, auditing can increase a firm's efficiency and profitability by:
1) Spotting opportunities to reduce waste and, therefore, disposal costs.
2) Lessening the likelihood of an environmental accident.
3) Compiling environmental cost information for financial purposes.
4) Teaching employees to be alert to environmental requirements and the cost
of materials lost through poor practices and carelessness.
These are some of the obvious benefits of implementing the practice of environmental
auditing. Audits can also lower the potential for lawsuit, improve public relations,
protect investments in pollution control equipment, and keep environmental insurance
rates low, all of which add up to substantial long-term cost savings. The official EPA
policy position encourages environmental audits, but also states that "...auditing should
remain a voluntary activity" and that "EPA will not routinely request environmental
audit reports." (Federal Register 1985).
Waste Reduction Audits
The concept of waste reduction audits is quite new. This type of audit differs from
the conventional environmental audit in one key aspect: the regulatory compliance is
B19-12
-------
TABLE 2-1 SOME FREQUENT PROBLEMS ENCOUNTERED BY AUDITORS
WATER
Process discharges to the storm sewer
Inoperative or poorly maintained sampling equipment
No effluent flow-measurement device
Process-area diking broken or nonexistent
Storm drains in liquid bulk-transfer areas
Oil-water separators improperly installed
No laboratory quality-control/quality-assurance program
Runoff from storage piles going to the storm sewer
Improper permit reporting
Outdated or poorly written Spill Control Plan
AIR
Vents and exhausts not registered
Lack of data on registered or unregistered emission points
Lack of Air Pollution Episode Alert Plan
"Guesstimates" on performance of control equipment - lack of test data
Improper worker protection in dusty environment
Neighborhood complaint file incomplete and/or lacking resolution of complaints
Dust from plant operations on building roof and pavement
SOLID WASTE
Sloppy housekeeping
Careless drum or container handling (leaks and/or dents)
Mixing process waste with office trash
Unreported in-plant dumpsites
Poor recordkeeping
Incomplete analysis of hazardous wastes being sent offsite
Incomplete files on past disposal practices
Unlabeled drums
Leaking containers
Source: Managing Your Environmental Audit (Russell 1985).
819-13
-------
not addressed. The main objective of the waste audit is to identify options which have
a potential of minimizing waste generation.
Preferably, the audits should be conducted by experienced engineering and operations
personnel not directly involved with the audited facility. The waste audit counsists of
nine steps (Fromm and Callahan 1985):
1. selection of the audit team
2. compilation of a waste stream list
3. generation of waste reduction options
4. ranking of options
5. preparation of documentation in support of options
6. joint review of options with plant personnel
7. analysis of revised rankings
8. final selection of options for further study and audit report preparation
The options selected through the audit activities may then undergo detailed technical
and economic feasibility evaluations, if required.
Environmental/Safety Reviews of Operations
Periodic environmental/safety reviews of operations are necessary to ensure that a
company's hazardous substances management and control system remains abreast of
modifications in plant operations and of developments in hazardous substances
management technology. This is important both for worker health and safety and for
environmental compliance and waste reduction.
The National Agricultural Chemicals Association recommends regular meetings of
supervisory personnel on the subject of safety and health rules (NACA 1981a, 1981b).
These meetings are held to evaluate current rules and to formulate new procedures
which reflect changes in operations. The review procedure itself is carried out in
conjunction with a program of employee drills and safety meetings covering a range of
hazardous substance handling topics.
B19-14
-------
An example of an environmental safety review program is the Toxic Chemicals Review
Committee of West Point Pepperell, a textile manufacturer with 22,000 employees and
40 manufacturing facilities in eight states (Huisingh et al 1985). On the impetus of a
suggestion by the Director of the Medical Department, the committee was formed in
1978 and contained the following personnel:
o An industrial hygienist.
o Three research chemists.
o A professional engineer.
o A corporate attorney.
o A corporate safety officer.
o A chemical production representative.
o A corporate information specialist.
The committee began reviews in 1976 of chemicals in current use and of chemicals
proposed for use. It based its analysis of each chemical on the following criteria:
o Personnel safety considerations.
o Fire hazard potential.
o Hazardous properties (ignitability, toxicity, corrosivity, reactivity).
o Availability of less hazardous alternatives.
o Biodegradability.
o Heavy metal content.
o Potential for accumulation in the facility.
o Potential for release to the environment.
o Hazard potential when reacted or mixed with other chemicals.
o Proposed manner of use.
o Ultimate destination of the chemical.
o Hazard potential to the consumer.
Since 1976, the company has rejected or substituted a number of chemicals deemed
inappropriate for its operations. Procedures were established for the purchasing of
chemicals so that a share of the burden of reviewing chemicals could be passed on to
the individual manufacturing operations. All chemical purchases may be made only on
approval by the Corporate Research Chemist, who follows committee guidelines.
B19-15
-------
A large part of the waste reduction or waste prevention at West Point Pepperell has
come about by replacing hazardous chemicals with safer ones and by preventing
hazardous chemicals from being brought into use in the first place. West Point
Pepperell now generates only a small amount of solvent waste which is sent off-site to
a refining company.
Environmental/safety reviews of operations can be conducted at low cost depending on
the size of a company's hazardous materials problem, the amount of in-house resources
that can be committed to the effort, and other similar considerations. A review
procedure can be used in conjunction with safety audits to preserve worker health,
and/or with environmental audits for ensuring compliance with environmental regula-
tions. The potential benefits from adopting this procedure are therefore significant.
In fact, "The best type of audit for assuring environmental compliance is not really an
audit at all, but a continual review of operations by plant management and personnel"
(Russell 1985). A formal audit is then conducted occasionally to verify progress.
2.2.2 Employee Training
An effective "good operating" campaign to reduce the amount of hazardous waste
generated in-plant must incorporate an employee training program. The personnel
responsible for operating and monitoring process equipment, for off- and on-loading
hazardous materials, for storing and transfering toxic substances during plant
operations, and for similar and related tasks, need to be trained to adhere to safe
operating procedures, proper equipment use recommendations, correct process control
specifications, and good industrial hygienic practices. A sound policy of keeping
employees informed of proper practices and of rectifying abuses contributes greatly to
the protection of worker health, plant facilities, and the environment.
According to recommendations made by the National Agricultural Chemicals
Association, the training of employees in safe hazardous materials handling should
take place in three stages: 1) prior to job assignment, 2) during on-the-job training,
and 3) ongoing during the employee's tenure with the company (NACA 1981b). Prior to
assuming a position, an employee should be given a review of the types of materials to
be handled, including their toxic properties, the health consequences if exposed, and
any byproducts resulting from fire or explosion. The care and fit of any protective
equipment required should also be covered.
B19-16
-------
When the employee begins working, the focus should turn to the safe operation of
equipment and proper materials handling and spill cleanup procedures. Techniques are
demonstrated for minimizing the amount of material allowed to escape and the human
exposure to it. Employees are also taught the signs and methods of detecting the
release of toxic material into the work environment.
Ongoing training should consist of regularly scheduled drills and safety meetings in
conjunction with supervisory reviews of industrial hygiene, materials handling, and
emergency practices. In this manner, the procedures in effect always correspond to
the chemicals and equipment in use.
2.3 Procedural Measures
2.3.1 Documentation
Documentation of process procedures in a manual or set of guidelines ensures that job
duties are precisely defined. This is especially important in an integrated production
process where unsatisfactory performance during one step can adversely affect the
remaining steps of the process. In general, a well-written operating manual or set of
instructions (Stus 1984):
o Informs the employee how his job fits into the overall process.
o Describes startup and shutdown procedures, normal operations procedures,
emergency procedures, and special procedures.
o Identifies important control parameters, including target emission or
effluent levels.
o Defines operator's job responsibilities.
o Describes potential personnel hazards.
This information is necessary if an employee is to operate equipment safely and
economically. Ideally, the topic of environmental control should be given separate
B19-17
-------
treatment. The employee should be made aware of controlled versus uncontrolled
waste generation rates, the target rates for the process, and the consequences to the
organization of not meeting the targeted rates. In addition, effluent sampling
procedures and control equipment failure procedures should be outlined. Adherence to
procedures and use of the information provided can contribute to reduced waste
generation.
Process documentation of this type also promotes product consistency, lessening the
likelihood of producing unacceptable product which must be discarded. Moreover,
procedural guidelines can be helpful in reducing waste generation during maintenance
or emergency shutdowns.
Material safety data sheets (Form OSHA 20) are prescribed for hazardous materials by
the Occupational Safety and Health Administration of the Department of Labor.
Although intended primarily as a safety precaution, the material safety data sheets
(MSDS) can aid in reducing hazardous waste generation. These sheets contain the
manufacturer's information on the chemical, physical, and toxicological properties of
the hazardous substances being used; they also contain recommendations on the proper
handling and storage procedures for these substances. Following these
recommendations reduces the likelihood of spills and of waste creation due to cross-
contamination with other chemicals. In the event of a spill, a response crew can refer
to the appropriate MSDS for instructions on how to clean up the chemical while
producing the least hazardous waste.
Consolidated Diesel of Whitakers, N.C., uses material safety data sheets to screen all
hazardous materials coming into the plant (Kohl et al. 1984). The company routes
MSDSs through hazardous material and medical personnel during the requisition
process. The approval of these individuals is required before a substance is allowed
onto the plant site. This ensures that each substance is properly documented and
evaluated in terms of its hazardous characteristics prior to its use. This can reduce
hazardous waste generation by preventing the use of some materials which would
require regulated disposal.
In addition to having material safety data sheets, each substance should be stored and
handled in containers that are clearly marked with information on contents, storage
B19-18
-------
and handling reommendations, spill procedures, and requisite first aid for exposure.
Some innovative products have recently been developed to aid in labeling. The
Legitronic 2060 electronic labeling system, manufactured by Weber Marking Systems
of Arlington Heights, Illinois, prints labels on an as-needed basis which contain
information on product identification, weight, lot number, health hazard warnings,
first aid and spill procedures, and storage and handling instructions. The labels can
contain Department of Transportation symbols when needed. The system is in use at
Stephan Company of Northfield, Illinois, where 400 separate products are produced at
four separate production facilities (Bobrowicz and Powers 1985).
Another labeling method rapidly coming into use is the automatic identification
system using bar codes (Powers 1985). Bar coded labels can link containers and
materials to a computer through the production, inventory, storage, and shipping
stages. This improves the accuracy of material tracking and inventory accounting,
thereby allowing work-in-process monitoring and preventing material from going
astray on the plant site and/or becoming dated. Hazardous waste in the form of
discarded chemicals and waste resulting from inefficient operations are reduced in this
manner.
2.3.2 Material Handling and Storage
Most processes involve the storage of raw materials, products, and wastes and the
transfer of these items from one area of the plant to another. Proper materials
handling and storage ensures that raw materials reach the production process without
spills, leaks or other types of losses which could result in waste generation. Similarly,
proper materials handling ensures that products and wastes which result from the
production process are transferred to off-site locations without additional waste
generation.
There are strong economic incentives for firms to handle and store materials properly.
Losses from improper handling and storage can often be curtailed without incurring
large capital costs (e.g. by changing procedures or organization). Thus, firms tend to
employ proper materials handling procedures because it is profitable to do so. The
proper storage of hazardous materials includes:
B19-19
-------
o Adequately spacing rows of drums to allow for a visual inspection of each
container for corrosion and leaks.
o Stacking containers no higher than recommended by the manufacturer and
in such a way as to minimize the chance of tipping, tearing, puncturing, or
breaking.
o Refraining from stacking equipment against containers.
o Maintaining distance between different types of chemicals to prevent
cross-contamination.
o Providing adequate lighting in all areas where hazardous substances are
stored.
o Insulating electrical circuitry and checking it frequently for corrosion and
potential sparking.
o Keeping aisles clear of obstructions.
»
o Maintaining a clean, even surface in areas trafficked by personnel and
equipment.
o Raising drums off the storage area floor to prevent corrosion through
concrete "sweating".
o Curbing or diking around process storage tanks/area.
In addition, using larger containers for chemical storage reduces the ratio of container
surface area to volume, implying less cleaning waste per unit of volume stored.
Larger containers also obviate the need to handle wastes in 55-gallon drums, which
themselves can become hazardous waste and are unsafe and inconvenient. An example
of an alternative to 55-gallon drums would be polyethylene containers enclosed in rigid
wire mesh. These can be constructed to hold up to 6 times the capacity of a 55-gallon
B19-20
-------
drum. The containers are portable, reusable, and can be outfitted for top or bottom
discharge, cleaning access, locking, and easy trucking.
2.3.3 Material Tracking and Inventory Control
Accurate material, product, and waste tracking provide operations supervisors with
information on the inventories of hazardous chemicals at each location on the plant
site. This enhances the development of a materials handling and storage procedure
which avoids hazards. Also, the availability of information on the quantities of waste
being generated from discarding unused chemicals may contribute to concerned
management setting a waste reduction program in motion.
Computers are increasingly being used in industry to monitor inventories and track
materials through processes. Material tracking computer software is available which
can list the names and locations of all hazardous materials on site in addition to
maintaining environmental monitoring, employee health, and chemical exposure
information (Giltenan 1984). Process control equipment allows process specifications
to be controlled using sensors; when undesired changes occur, corrections are sent to
•
control instruments. This reduces the number of discards by increasing the quantity of
product on-specificaton. Also, the electronic bar code labelling systems mentioned
above allow information on a container's contents to be read by lasers into computers.
The bar codes enable continuous monitoring of inventories.
2.3.4 Scheduling
Timing often plays an important role in the production of waste generating products.
It has a special significance in batch production operations, where equipment cleanup
waste volumes are directly related to cleaning frequency. To reduce cleaning
frequency, it is necessary to maximize batch size and equipment dedication to a single
product. This requires good scheduling and planning. The reader is further referred to
the study of process equipment cleaning in the appendix.
B19-21
-------
2.4 Loss Prevention Practices
2.4.1 Spill Prevention
The spillage of a hazardous chemical results in hazardous waste. Washdowns of spilled
toxic chemicals create liquid hazardous wastes which are subsequently transferred off-
site for disposal. Mopups using absorbent materials create hazardous solid waste
which must also be disposed of. To minimize the costs incurred in chemical spill
incidents — which can be a range of costs from disposal fees and employee care to
fines from enforcement actions ~ the best practice is to prevent spills from occurring
in the first place.
The likelihood of a spill can be lessened by 1) conducting hazard assessment studies at
appropriate points in the design and operation phases, 2) using properly designed
storage tanks and process vessels only for their intended purposes, 3) equipping all such
containers with overflow alarms, 4) testing the alarms periodically, 5) maintaining the
physical integrity of the containers over time, 6) setting up administrative controls for
all loading/unloading and transfer operations, 7) installing sufficient secondary
containment facilities, 8) having a good valve layout, 9) having interlock devices to
stop flow to leaking sections, 10) disallowing the operators to bypass the interlock or
to alter the set points, 11) isolating equipment or process lines that are not in service,
and 12) documenting the spillages and related dollar values (Kletz 1982, Sarokin 1985).
These measures can be implemented through a spill prevention program that provides
input into plant design as well as the administration of plant operations.
Hazard assessment studies are best performed during the conceptual design, the
design-freeze, the prestartup, and the operational phases of a plant (Ozog 1985).
Moving from the design to the operation phases, hazards become easier to detect, but
changes in the process become more costly since the financial commitments to a set
process design are in place. During the conceptual design phase, the focus is on
hazards which affect the general public and the environment, including the release of
chemicals, explosions, and fugitive emissions. The chemicals involved in the process
are identified by their physical, chemical, and toxicological properties, and it is
determined whether special handling, monitoring, and/or control is required. Also at
this time, a separate study team is put together to compile a list of events which could
B19-22
-------
credibly lead to hazards. A consequence analysis is done for each event, out of which
a set of recommendations for precautionary measures emerges.
The design-freeze phase is that interlude between the completion of the piping and
instrumentation diagrams and the determination of operating procedures. For new
facilities, a full hazard and risk assessment should be performed here. A
comprehensive effort could encompass (Ozog 1985):
1) Definition of scope.
2) Collection of data.
3) HAZOP study of entire facility.
4) Qualitative ranking of hazards.
5) Quantitative analysis of selected hazards.
6) Integration of hazard probability and impact estimates.
7) Risk comparison.
8) Evaluation of risk-mitigation measures.
At this point, alterations in process design still can be made relatively inexpensively
since equipment has not yet been ordered.
Prior to starting up the process, a final check should be done to verify the status of
recommended changes that emerged from the hazard and risk assessment. A site
inspection should be conducted to form an impression of spatial dimensions, and the
hazard and risk assessments should be brought up-to-date. Design changes at this
point are expensive.
Shortly after startup, operating procedures may be revised, and these revisions should
be reviewed in the context of the hazard and risk assessments. Also, changes in the
process may be made during the operational life of the plant; these changes should be
similarly reviewed under further hazard and risk assessments. Assessments should be
carried out regularly during the operational life of the plant.
Sound practice dictates that storage tanks and process vessels containing hazardous
chemicals be properly designed and used only for their intended purpose. They should
also have, as a minimum, overflow alarms. Once these conditions are met, it remains
B19-23
-------
to ensure the structural soundness of the containers, to enforce proper hazardous
materials handling procedures, and to construct and maintain secondary containment
and collection facilities (Shields 1980).
The structural soundness of the tanks and vessels can be assured through an inspection
and record-keeping initiative. Monthly visual inspection of tanks, including all
equipment connected to them, the tank's seams, and the supporting structures, should
be conducted. Physical inspection and testing should take place at the half-life of
each tank, or more frequently if the tank materials are subject to attack by the
chemicals in the tank. Discharges from internal heating and cooling coils should be
monitored and tested at least quarterly. All these efforts can be conducted in
coordination with other existing safety and operating checks.
Spills are most likely to occur during loading/unloading operations and in-plant
transfer activities. For this reason, written procedures should be kept current and
made easily accessible to all employees. In addition, a spill response plan should be
formulated, and employees should be trained both in the current operating procedures
and in the correct implementation of emergency spill measures. A policy of taking
disciplinary action on violations, if strictly enforced, can ensure employee adherence
to procedure.
An effective spill prevention program will carry over into the design phase of plant
operations. In structuring a production/storage facility, equipment should be designed
or ordered which, by its construction, does not allow hazardous substances to leak
from the process. The variations on equipment types and designs are numerous.
Successful applications have been double mechanical seals, rupture disk-relief valve
combinations, seal welded or all welded piping construction, flange guards, and bellow-
sealed valves, to name only a few. Some case examples of spill prevention measures
through equipment designs and facilities repairs are given below.
Sealless pumps - In modernizing its benzene facility in 1977, a major oil company
selected a sealless-type pump for its benzene transfer operations. The design prevents
leakage and subsequent exposure of workers and the environment to EPA-regulated
chemicals (Suzuki 1985).
B19-24
-------
Interlock devices - In some installations, filtration units are generally washed with hot
water and the wash liquid allowed to run into a drain through a drain valve. If by
mistake the drain valve is left open when the filter is put back in normal operation,
the filtrate liquor will drain out resulting in a spill. By installing an interlock device
that stops feed to the filter when the drain valve is open, such spills can be prevented
(Kletz 1982).
Bellow-seal valves - The design of these valves relies on the use of hermetically sealed
bellows (torsional or compression type) which connect the valve body to its moveable
stem. The design is preferable to a standard packing-type seal for highly hazardous,
flammable, or corrosive services.
2.4.2 Preventive and Corrective Maintenance
Maintenance programs, whether preventive, corrective, or a combination of both, can
prove instrumental in cutting production costs stemming from expensive repairs,
excessive waste disposal, and business interruptions. They can also prevent hazardous
waste releases due to equipment failure and facilities degradation. These programs
•
have difficulty, however, in gaining the acceptance and support of upper management.
Preventive maintenance planned to minimize equipment breakdown and malfunction is
a major part of an overall maintenance program, as losses for this reason can account
for three to four times the level of operating and maintenance costs (Rimberg 1980).
Despite this fact, preventive maintenance is often viewed or perceived, as an
expenditure which does little, if anything, to improve efficiency and profits. For
extensive maintenance programs to be adopted, it must be quantitatively
demonstrated that a particular program can result in 1) cost reductions, 2) cost
avoidance, and 3) a good return on the investment.
To demonstrate these points, measurements must be taken of maintenance labor
productivity and management, planning, and scheduling effectiveness. Indices can be
devised to measure how well a maintenance program is working. An index of labor
productivity is the ratio of work order process time to time needed to complete
maintenance tasks, for example. A maintenance control index, which is a measure of
management effectiveness, can include (Rimberg 1980):
B19-25
-------
1) The current backlog of work orders.
2) The backlog of maintenance manhours.
3) The number of open work orders.
4) The number of hours assigned for work in a specific area.
Indicators of the results of maintenance planning can be (Rimberg 1980):
1) Jobs finished on time.
2) Forecasting accuracy.
3) Maintenance hours planned vs. hours worked.
4) The amount of emergency work.
5) The amount of overtime.
6) The amount of equipment downtime.
The performance of a maintenance program can be determined by assessing (Rimberg
1980):
1) Productively employed maintenance manpower.
2) The effectiveness of maintenance while a plant is in operation.
3) The ratio of maintenance labor to maintenance materials costs.
4) The ratio of maintenance cost to unit of product.
5) The ratio of maintenance personnel to total plant personnel.
Careful recording of these measurements can reveal the value of a maintenance
program. The following example of air pollution control equipment maintenance
serves to illustrate the above:
"As an example of the need to measure maintenance performance, the following
situation was encountered: A plant had 12 dust collectors of assorted sizes with
50 to 250 bags of the pulse jet and shaker types. All collectors were usually
maintained on an emergency basis; that is when malfunction aborted production.
In addition, the source was fined $25,000 by the state for continuous failure to
comply with emissions regulations. Corporate management requested that plant
management compute the costs incurred from loss of production and excessive
baghouse component replacement (valves, bags, etc.). Loss of production
819-26
-------
accounted for approximately $500,000 and baghouse repairs $50,000. Labor costs
were extremely high, although not easily estimated, since the maintenance
department did not assign task and job costs. In addition to these costs,
excessive amounts of plant management time were being consumed handling
regulatory matters. The major underlying problem in this situation was the lack
of appropriate maintenance management procedures. Upon final evaluation of
the economics of the situation, a contract preventive maintenance program was
instituted. It should be noted that the most important step in initiating this
activity was to persuade operations personnel that the program would make the
process equipment more responsive to their needs. An expenditure of $36,000
was made to execute the program and after 1 year costs of malfunction and loss
of production were reduced to $128,000. In-plant labor costs were also clearly
defined and amounted to $32,000." (Rimberg 1980)
To illustrate a preventive maintenance program, the following is the required
information to develop a program directed at mechanical equipment (Hellhake 1981):
1. A list of all plant equipment.
2. A knowledge of operating time for each item or area.
3. A knowledge of which items are critical to the process(es).
4. A knowledge of problem equipment.
5. Vendor maintenance manuals.
6. Manual or computerized repair histories.
A separate card or instruction sheet can be typed for each equipment item in the
preventive maintenance program. These cards can be filed in a number of ways: 1)
based on maintenance schedule (most frequently used), 2) based on plant area (useful if
a plant section is down unexpectedly and one wants to do the maintenance in that
area) or 3) based on specific equipment number.
Like effective spill prevention programs, an effective maintenance program carries
back into the design phase. Designs should take into account the access of
maintenance equipment and personnel to process equipment components and
accessories. Design consideration should include larger access doors, wider internal
catwalks, accessible components, hopper access doors, and duct cleanout and
B19-27
-------
inspection hatches. Specification of easily maintained equipment at the outset serves
to reduce the cost of preventive and corrective maintenance during operation.
Preventive maintenance can be less expensive and more efficient in some cases if
contracted out to a maintenance firm. Cost and efficiency advantages can be found in
1) the increased time that plant personnel can spend on production operations, 2) the
easy handling of fluctuating workloads by contract maintenance firms, 3) the avoided
expenditure on crews, repairs, repair facilities, tools, and measurement instruments,
and 4) the greater operational experience of contract maintenance personnel, which
leads to more rapid identification and troubleshooting of equipment malfunctions and
thereby to product improvement. Also, the concern over environmental compliance
can be placed to some degree on the contracting firm (Rimberg 1980).
2.4.3 Emergency Preparedness
Part of overall good operations resides in maintaining a state of emergency
preparedness for chemical spills and/or fires. The exposure of both the plant's
employees and the immediate environment to releases of hazardous or toxic chemicals
can be minimized with a well-conceived emergency management procedure.
The Occupational Safety and Health Administration, by virtue of regulations published
in the Federal Register (39 CFR 1910.1200, 48 FR 53280, November 25, 1983), requires
chemical manufacturers and all plants using chemicals to implement a hazard
communication procedure. The procedure should provide for the identification of the
hazards of all chemicals produced and for the distribution of this information to
chemical users and their employees. As management and employee awareness of
hazards is the first step to preparedness, compliance with this regulation can form the
informational basis for an effective emergency program.
The essential requirement of the regulation is that by May 25, 1986, all hazardous
substances must be properly labeled with material safety data sheets (see Section
2.3.1) and all employees adequately trained in the handling of these substances. To
meet the requirement, the following should be accomplished (Brown 1985):
B19-28
-------
o The OSHA standard should be reviewed and the extent of its applicability
determined for the individual plant.
o Chemical users should list chemical substances used and obtain a material
safety data sheet for each.
o Chemical manufacturers or importers should list all substances and deter-
mine the hazards of each.
o Employee hazard information and education procedures should be
formulated and should include informing employees of the hazards of
nonroutine tasks and educating outside contractors about on-site hazards.
Figure 2-4 describes a sample management program to comply with the regulation. It
is important to note that as with any other successful organizational innovation, a
commitment from management is the primary motivation for the change.
Using materials safety data sheets and an in-place employee training program, an
emergency procedure can be developed. Emergency response preparation and
procedure can vary greatly in detail, depending on the size and location of the plant
(i.e. the potential size and spread of a spill), the number and toxicity of chemicals in
use, and like factors. One organizational suggestion is to maintain a separate
comprehensive dossier for each chemical (Berry 1980). Readily giving a complete
technical report can serve two functions: 1) it can provide a measure of legal
protection for the firm by being an indication of "good faith", and 2) it can prevent a
small spill incident from evolving into a larger one due to incorrect cleanup
procedures.
One portion of the chemical substance file can be a record of the data required for
briefings of the media, emergency teams and regulatory personnel. These data should
include (Berry 1980):
o Hazardous material classification (DOT).
B19-29
-------
FIGURE 2-4 SAMPLE PROCEDURE FOR OSHA COMPLIANCE
Corporate Management Commitment
Operating Unit Determines Policy and Establishes Position
Line Management Develops Program and Procedures and Directs
Implementation
Regulation Applicability Determined
Input, Output, and Intermediate Chemicals Listed
Chemical Hazard of On-Site Material Safety Data Sheets and Labels
Production Determined Prepared and Assembled
Workplace Hazards Assessed
Protective Measures Established and Labels Installed
Information Program Written
Affected Personnel Trained
Source: Understanding and Complying with OSHA's Hazard Communication
Regulation, (Brown 1985).
B19-30
-------
o Emergency service numbers (U.S. Coast Guard; state regulatory and
response agencies; police, fire, and health departments; health care
facilities; area water treatment plants; area health facilities; media
outlets).
o Location and quantity of chemical spill, the route of entry into the human
body, and the medium holding the chemical.
o Date and time of discovery, bodies of water affected, number of employees
involved and extent of injuries, if any.
o Weather conditions, containment/removal actions, other chemicals present,
and potential problems.
The remainder of the file can be comprised of specific information on the fugitive
chemical and the remedial action (Berry 1980):
o Name, address, and phone of vendor (plus information on other vendors and
consultants).
9
o Symptoms of exposure and recommended first aid.
o Fire protection measures.
o Environmental protection measures.
o Protective equipment required.
o Sampling and laboratory measurement methods.
o Details of training and education efforts for employees, health care
personnel, law enforcement personnel, and the media.
B19-31
-------
Additional considerations must include such elements as community awareness
programs, site emergency evacuation plans, site emergency services, and coordination
of site emergency plans with surrounding facilities and local authorities.
If carefully acquired and kept up-to-date, this information can support capable
hazardous spill management. Training employees in proper spill procedures and
informing those capable of rendering assistance can also help to minimize human and
environmental injury.
A good example of planning for agricultural chemical fire contingency is provided by
the form distributed by the National Agricultural Chemicals Association. The form
lists the information useful in the event of a chemical fire. For instance, the phone
numbers of plant authorities, state agencies, and emergency personnel are kept on file.
Sketches are made of the plant layout and vicinity areas as well as of the drainage
likely to occur during liquid spills and washdowns. The owners and uses of surrounding
areas are described, as are any local water supplies which might be affected or
utilized. Finally, a record is kept of scheduled reviews or updates of the plan in order
to keep the plan current.
3. OBSERVATIONS AND CONCLUSIONS
3.1 Process Characteristics and Good Operating Practices
The applicability of good operating practices is determined at the level of the
individual plant or company. Two plants producing the same product may have
different process or organizational structures which make a particular measure more
or less applicable. For example, two plants using bar coded labels may realize
different degrees of success in material tracking or inventory control depending on the
routing of the materials through production operations, the care taken to ensure error-
free data entry, etc. Nevertheless, certain process characteristics and economic
conditions do appear to enhance the effectiveness and/or implementation of good
operating measures.
In less automated plants, fewer automatic process controls are in place and the level
of human involvement is higher. There is, therefore, greater room for human error in
B19-32
-------
equipment operation, material handling, inventory control, production scheduling, etc.
Training programs and procedural reviews can thus have a greater effect in reducing
waste generation from those activities. In more automated processes, there is not as
much call for procedural refinements or training improvements.
At plants using highly toxic materials, the handling and storage of these materials
tends to be more closely regulated both by government and by the firms themselves.
Potential health hazards cause plant managers and personnel to track materials
carefully, maintain equipment regularly, and react quickly and decisively in the event
of a spill. Health and safety concerns at a plant producing epichlorohydrin, for
example, typically would result in more rigor in implementing operating measures than
do similar concerns at a paint-mixing plant.
Firms which use the same processing facilities for various products generally appear to
have more potential to implement good operating practices. Producing a product mix
implies a change of inputs and, perhaps, procedures for alternative product runs. In
the production of product mixes, scheduling, material handling, supervision and
training, and material tracking are not only more difficult but also more important in
terms of maintaining throughput, minimizing leaks and spills, and finally minimizing
waste generation. Moreover, there is generally more equipment cleaning as the
equipment is prepared for the next product run. Sometimes it is possible to segregate
the cleaning wastes by contaminant to allow for reuse and therefore produce a
subsequent reduction in waste generation. The reader is referred to the process
equipment cleaning study for additional information.
The foregoing observations suggest that greater potential for the implementation of
good operating practices and the associated reductions of waste generation exists in
batch processes rather than in continuous processes (except, perhaps, where batch
processes use highly toxic chemicals such as in pesticide formulation). This is not to
say that good operating practices are not important in continuous processes; on the
contrary, preventive maintenance and spill prevention, for example, are essential
considerations. Rather, batch processes are generally less automated, and have a
higher degree of human involvement, coupled with a lower degree of equipment
dedication to a single process medium in comparison to continuous processes.
B19-33
-------
Figures 3-1 and 3-2 present generic, block flow diagrams for batch and continuous
processes. The line segments to the right show the range of individual process
operations to which each good operating practice applies. Omitted are management
initiatives and employee training which, as personnel measures, apply to the entire
range of process activities. Documentation in the form of procedure manuals is
omitted for the same reason.
3.2 Economic Aspects
Economic considerations suggest that plants which 1) use raw materials with lower
unit costs, and 2) generate waste with lower unit disposal cost/remarketing value, are
likely to have more waste reduction potential. Processes using raw materials with
high unit costs, e.g. gold plating operations, will more readily have undergone
modifications to recover lost values and/or improve the efficiency of input materials
use. Similarly, plants generating wastes with high disposal costs relative to the
revenue from the sale of the product will more readily have implemented good
operating techniques. Good operating practices, such as waste stream segregation,
will also be employed to purify and preserve wastes which can be marketed. In this
case, disposal cost must meet or exceed the cost of containing the waste less the
revenue received for it.
At the level of the individual firm, the available resources and the target compliance
parameters (and their enforcement) determine the breadth and quality of the pollution
abatement effort. Large generators, which generally have the necessary financial and
personnel resources and which have been under regulation longer, have developed
comprehensive environmental programs in an effort to reduce disposal costs and avoid
fines stemming from non-compliance. While good operating practices are certainly in
evidence in these companies, their environmental efforts include examining and
implementing extensive process changes to reduce waste. Small generators, on the
other hand, do not have large amounts of financial and personnel resources and are
only now coming under regulation. Small firms can little afford the expense of
investigating various pollution abatement methods, may not have the expertise to do
so, and most likely could not withstand the losses from business interruption that the
implementation of these methods may entail. For small generators, then, good
operating practices offer more promise for waste reduction. Many of these practices
can be instituted quickly with little or no capital investment and require merely an
B19-34
-------
EBUIPNENT
~t
! B*« GUALITY |
j MATERIAL — *• CONTROL p~*i
| DELIVERY | |
I REJECTS
BACK TO
SUPPLIER
t $
i i
HATER/
CENTRAL i j— — \ [SEPARATION
STOHA8E f—**j JlTrBiIIf *~^ *MD
i RE*CTION [ C—
'
PRODUCT
PsTOR
t— t-| AN[
| SHIPI
SOLVENT i '
p-I i . »ASTE 1 j DISPOSAL
i RECOVERY 1 I ^ [ ! LANDFILL
_J U>J TREATMENT LcJ ...„.„
| OF VALUES | i- 1 i SURFACE
i J j IMPOUNDMENT
! ! DEEPHELL
1 PRODUCT j INJECTION
J! INCINERATION
1
r
'INS
RAH MATERIAL HANDLING IMPROVEMENTS
k H h>
PRODUCT HANDLING HAS
AND STORA8E IMPROVEMENTS
HASTE HANDLING AND
STORA6E IMPROVEMENTS
HASTE STREAM SE8REBATION / RECOVERY / REUSE
EBUIPMENT CLEANING AND PRODUCTION SCHEDULING
(J
INVENTORY CONTROL
INVENTORY
CONTROL
RAH MATERIAL TRACKING AND DOCUMENTATION
HASTE TRACKING AND DOCUMENTATION
SPILL AND LEAK PREVENTION/ PREVENTIVE MAINTENANCE/ PERSONNEL TRAIKIM6
Figure 3-1 Generic Satch Process
819-35
-------
INTERMEDIATE PRODUCT
1 BAM | BUALITY 1
i MATERIAL \-*t COMTHOL U
j DELIVERY !_
REJEC1
P
i
fcTN^H JSHH 1
i — H ES r
, . ,. ._J J REACTION 1
e
y
r»ACK TO
! SUPPLIER
|
SEPARATION SECOVE
*MD H OF VAL
PURIFICATION j " T*L
|
| PRODUCT
U STORAGE 1
AND f4—
SHIPPING i
1
. . M10TP DISPOSAL
UmATMENT^^"^
JES 1 ' , SURFACE
_j j IMPOUNDMENT
T DEEPHELL
1 PRODUCT ; INJECTION
| . INCINERATION
^
K—
RAN MATERIAL HANDLINC IMPROVEMENTS
PRODUCT HANDLING
AND STORA6E IMPROVEMENTS
NASTE HANDLING AND
STORAGE IMPROVEMENTS
k
SCHEDULING
H .
HASTE STREAM
SEGREGATION
INVENTORY CONTROL
INVENTORY
CONTROL
^
RAN MATERIAL TRACKING AND DOCUMENTATION
-H
HASTE TRACKING AND DOCUMENTATION
SPILL AND LEAK PREVENTION/ PREVENTIVE MAINTENANCE/ PERSONNEL TRAINING
Figure 3-2 Generic Continuous Process
B19-36
-------
examination of existing procedures. Therefore the benefit/cost ratio is potentially
very high.
4. REFERENCES
Anonymous. 1983. Double rupture disk assembly eliminates premature failures.
Chem. Process. 46(13):135.
Anonymous. 1984. Portable, reusable container features 6-drum capacity. Chem.
Process. 47(13):114.
Berry, R. 1980. First response procedures for hazardous materials. Poll. Enq.
12(10):37-41.
Bobrowicz, D., and Powers, J. 1985. Electronic labeling system keeps tab on 400-plus
products. Chem. Process. 48(2):90-91.
Brown, H. 1985. Understanding and complying with OSHA's hazard communication
regulation. Plant Eng.. 93(4):148-150.
Conrad, J. 1985. Total plant-safety audit. Chem. Eng. 91(10):83-86.
Drake, E.M., and Kalelkar, A.S. 1981. Handle with care: using risk analysis for
hazardous materials facilities. Risk Management. March 1981. pp.44-50.
Federal Register 1985. Environmental auditing policy statement, EPA, 50(217): 46504-
46508
Fromm, C.H., Callahan, M.S., 1985. Waste reduction audit procedure. Hazardous
Materials Control Research Institute Conference Proceedings, pp. 427-435, Atlanta,
GA.
Giltenan, E. 1984. Keeping chemical records on track. Chem. Business. 6(ll):47-49.
Gordon, C. 1984. Selling a maintenance program to top management. Plant Eng.
38(1):49-50.
Hellhake, F.J. 1981. Setting up a preventive maintenance program. Chemical
Engineering. 88(14): 145-150.
Huisingh, D., et al. 1985. "Proven profit from pollution prevention (Conference draft).
Washington, D.C.: The Institute for Local Self-Reliance.
Kletz, T.A. 1982. Minimize your product spillage. Hydrocarbon Processing, 61(3):
207-215.
Kohl, J., et al. 1984. Managing and recycling solvents; North Carolina practices,
facilities, and regulations. Raleigh, N.C.: North Carolina State University.
NACA. 1981a. National Agricultural Chemicals Association. Mimeo: Good
warehousing practices for agricultural chemicals.
B19-37
-------
1981b. National Agricultural Chemicals Association.
Mimeo: Good workplace practices for the handling and storage of pesticides.
NCPPP. 1985. North Carolina Pollution Prevention Pays Program. Publication:
Environmental auditing.
Ozog, H. 1985. Hazard identification, analysis, and control. Chem. Enq. 92(4):161-
170.
Powers, J. 1985. Automatic identification guides CPI plants to higher efficiency.
Chem. Process. 48(2):84-89.
Rimberg, D. 1980. Minimizing maintenance makes money. Poll. Eng. 12(3):46-48.
Russel, L. 1985. Managing your environmental audit. Chem. Eng. 92(134):37-43.
Sarokin, D. 1985. Reducing hazardous wastes at the source: Case studies of organic
chemical plant in New Jersey. In Source Reduction of Hazardous Waste: Seminar
Proceedings, at Douglas College, Rutgers University on August 22, 1985.
Shields, E.J. 1980. Prevention and control of chemical spill incidents. Poll. Enq.
12(4):52-55.
Stus, T. 1984. On writing operating instructions. Chem. Enq. 91(24):105-106.
Suzuki, R. 1985. Filter system for soluble oils saves $50,000/yr. disposal costs.
Chem. Process. 48(1):30-31.
Wichersham, C. 1985. Refinery selects canned pumps to reduce benzine transfer
costs. Chem. Process. 46(5):94-95.
3M. 1985. 3M Corp. Mimeo: How to initiate a '3P'-Type program in combustion
engineering.
5. INDUSTRY CONTACTS
Joseph P. Chu, Environmental Activities Staff, General Motors Technical Center,
Warren, MI.
Walter G. Vaux, Process and Chemical Engineering, Westinghouse Electric Corp.,
Pittsburgh, PA.
B19-38
-------
1. PROCESS: METAL PARTS CLEANING
2. SIC CODE: XXXX
3. INDUSTRY DESCRIPTION
Metal parts cleaning is an integral processing operation used by industries involved in
the manufacturing, repair, and maintenance of metal parts and equipment. The major
industries that practice metal parts cleaning are:
SIC SIC
Code Manufacturing Code Service
2500 - Metal Furniture 4010 - Railroad
3300 - Primary Metals 4580 - Air Transport
3400 - Fabricated Products 7530 - Auto Repair
3500 - Nonelectric Machinery
3600 - Electric Equipment
3700 - Transportation Equipment
3800 - Instruments and Clocks
3900 - Miscellaneous
It has been estimated that the above listed industries consist of approximately 336,000
plants (USEPA 1979). Since not all plants employ metal parts cleaning operations, and
since there is no direct connection between the size of the plant and the amount of
cleaning being performed, no additional industry information is available.
4. PRODUCTS AND THEIR USES
Due to the diversity of the industries utilizing metal parts cleaning, a comprehensive
listing of all products requiring cleaning would be beyond the scope of this report.
Virtually all fabricated metal objects require some form of cleaning.
5. RAW MATERIALS
There are basically five different types of cleaning media employed by industry. The
five types are nonchemical (or abrasives), solvents (halogenated and nonhalogenated),
B20-1
-------
alkalines, acids, and water. Alkaline and acid cleaners are usually referred to as
aqueous cleaners. A sixth type, emulsion cleaners, is a mixture of nonhalogenated
solvent and alkaline cleaner and is therefore not considered separately. Table 5-1
presents a breakdown of cleaning media usage (excluding water) by overall U.S. market
share.
Table 5-1 Raw Materials Used for Metal Parts Cleaning in 1981
Materials Market Share
Nonchemical
Aluminum oxide 3%
Silicon carbide
Silica
Binder (grease, oil, or water)
Halogenated Solvents 34%
Methylene chloride
Perchloroethylene
1,1,1-trichloroe thane
Trichloroethylene
Nonhalogenated Solvents 24%
Aliphatic hydrocarbons
Aromatic hydrocarbons
Ketones
Alcohols
Aqueous 39%
Alkaline (alkaline salts plus deflocculants,
inhibitors, and surfactants)
Acid (sulfuric, hydrochloric, nitric,
phosphoric, etc.)
Source: Estimation of Market Share of Metal Cleaning Techniques
(O'Brien 1981).
6. PROCESS DESCRIPTION
Detailed process descriptions of metal parts cleaning operations can be found in the
open literature (Durney 1984). The following discussion presents a general overview of
the operations involved. A process flow diagram of the major equipment items used
for metal parts cleaning is shown in Figure 6-1.
B20-2
-------
SOAK TANK
HEATING
COIL Jl
C OPTIONAL]
L A J
SOLVENT EMULSION OR AQUEOUS
SOLUTION
— > i I
ULTRASONIC
UNIT
C OPTIONAL 3
i
(~) f.
jsL
SPRAY UNIT
COPTIONAL3
DIPHASE SYSTEM
f
PARTS
j_
"1
SOLVENT
^_ TO
OR
c*n
"" I
| SPRAY
A~&AAA
MATER
X
DISPOSAL
RECYCLE
r
SPRAY
ATTA ["
1
PARTS
-*• OUT
pr
1
MATER TO
DISPOSAL
HAKE UP
MATER
VAPOR PHASE SYSTEM
COOLING ^
JACKET L
BOILIN6 ,_
SOLVENT r
HEATER
SPRAY
UNIT
FINNED
COIL
I CONDENSER
\-OOO-D
VAPOR
MATER TO
DISPOSAL
;> SOLVENT I SOLVENT
Figure 6-1 Metal Parts Cleaning Equifaent
B20-3
-------
The discussion below excludes paint stripping which is covered separately in the study
of paint application (B21).
Although most people think of metal parts cleaning as a very simple operation (such as
washing an oily part in a tank of solvent), many parts require quite complex
arrangements of several different cleaning operations. The sequencing of these
various cleaning operations usually depend on three main interrelated factors. These
factors are:
The nature of the contamination.
Not only is the composition of the contaminant important but also its
history. For example, subjecting the part to heat (or other cleaning
methods) can alter the physical properties of contaminants. Many times
parts are contaminated with several different materials so that careful
consideration must be made regarding the cleaning methods to be employed
and their sequencing.
The metal to be cleaned.
The selected process should perform the cleaning to the required degree
without adversely affecting the metal substrate. Since reactivity of metal
with alkalis and acids varies, cleaners for one type* of metal can be entirely
inappropriate for another type.
The required degree of cleanliness.
The required degree of surface cleaning varies with application. For
example, when parts are going to be cyanide zinc plated, critical cleaning
is not usually required. This is because cyanide based plating solutions
exhibit a strong cleaning action of their own and are very tolerant with
respect to the initial degree of surface cleanliness. On the other hand,
nickel plating requires a very clean substrate surface for nickel to adhere
resulting in more thorough and rigorous cleaning of the part.
Even after careful consideration of the above factors, no one particular cleaning
method can be readily specified. More commonly, several methods appear appropriate
and they all require experimental, evaluation.
R20-4
-------
The five major classifications of cleaners used by industry are abrasives, solvents,
alkalines, acids, and water. Cleaners (except abrasives) are normally held in large
open tanks, with the parts to be cleaned mounted on racks or held in perforated
horizontal barrels. The decision to use racks or barrels usually depends on the size and
shape of the part and the type of coating required.
Abrasives are designed to remove rust, oxides, and burrs and create a smooth surface.
Common abrasives are aluminum oxide or silicon carbide mixed with an oil or water-
based binder. This material is applied to a buffing wheel (usually made from an
absorbent material such as cloth), and then the metal part is held against the spinning
wheel. The speed of the wheel is determined by the type of metal being buffed and
whether the system is manual or automatic. Generally, manual systems run at lower
speeds and are considered to be less efficient. Automatic systems, while being more
efficient, generate more heat due to the higher speeds and can cause an oil-based
binder to burn. This, in turn, leads to more difficult cleaning in subsequent stages.
Another abrasive cleaning method, vibratory finishing, consists of a container that
vibrates an entire load of metal parts in a bed of abrasive material and water. Other
mechanical abrasive cleaning methods consist of tumbling barrels, centrifugal barrel
finishing, centrifugal disc finishing, and spin finishing.
Solvents, the most widely used class of cleaners, are used for removing oil-based
contaminants. Solvent usage is generally divided into three main categories or
operations. These are:
o cold cleaning
o diphase cleaning
o vapor phase cleaning.
Cold cleaning, which generally uses unheated or slightly heated nonhalogenated
solvents, is the most common cleaning method. It was estimated that over one million
cold cleaners are in operation, with more than 50 percent being used by the auto repair
industry (USEPA 1979). The second most common solvent cleaning method is vapor
degreasing. It was estimated that there were approximately 47,000 vapor degreasers
in operation (89 percent open top and 11 percent conveyorized) (USEPA 1979). Vapor
degreasers typically use a halogenated solvent vapor to dissolve oil-based
B20-5
-------
contaminants and the subsequent condensation of the vapor rinses the part clean. The
final system, diphase cleaning, is similar to the cold cleaning soak tank method except
that a water layer is floated on top of the solvent and provides for additional part
cleaning. All of the methods mentioned above are described in the following
paragraphs.
Cold cleaning operations can be divided into four distinct types of cleaning systems or
methods. The first method, wipe cleaning, consists of soaking a clean rag with solvent
and then wiping the metal part clean. This operation is strictly manual and the overall
efficiency is low. Rags must be replaced frequently or the parts will not be properly
cleaned. Another disadvantage of wipe cleaning is that the rags can leave lint on the
part.
The second cold cleaning method, soak cleaning, consists of soaking the parts in a tank
of solvent. Small parts are usually placed in a basket while larger parts are placed on
a rack. Sometimes a small pump and spray unit are installed on the tank so that the
parts can be more thoroughly cleaned. When soil contaminants are present (in addition
to oil-based contaminants), emulsion cleaners can be used. Emulsion cleaners typically
consist of five percent solvent, ninety-five percent water or alkaline cleaning solution,
and various surfactants. The main advantages of emulsion cleaners over solvents are
that they cost less and produce less air pollution. In addition, emulsion cleaners leave
a slight oil film on the cleaned parts. This can be an advantage if the part is going to
be stored for some time (the oil film provides rust protection), or a disadvantage if the
part is going to be immediately electroplated (additional cleaning steps would be
required to remove the oil film).
A third cold cleaning operation, ultrasonic cleaning, is identical to soak cleaning,
except for the addition of an ultrasonic unit in the tank. When ultrasonic waves pass
through a liquid media, cavitation occurs. Small vacuum bubbles are created which
almost immediately collapse. The rapid implosion of these bubbles provides a vigorous
cleaning action throughout the tank. Since this action occurs everywhere in the tank,
ultrasonic cleaning is useful for cleaning parts with difficult to reach surfaces or
crevices.
B20-6
-------
The fourth cold cleaning method, steam gun stripping, is used primarily for the
removal of paint from a metal part or object. Stripper (a mixture of nonhalogenated
solvents), is fed into the steam line through a valve and hose arrangement connected
to the stripper storage tank. After the part has been stripped of paint, the valve is
closed and pure steam is used to rinse the part clean.
Diphase cleaning systems make use of the fact that halogenated solvents are relatively
insoluble in water. Therefore, solvent cleaning and water washing can be performed in
the same tank quite easily. Parts to be cleaned first pass through a water bath, then a
solvent spray, and finally a water bath and spray (see Figure 6-1). Solvent used to
spray clean the part drops into the tank of water and sinks to the bottom. The solvent
runs to the deep end of the slopped tank bottom and can be pumped back up to the
spray unit. Since the unit is fully automated, and the solvent spray portion is fully
enclosed, very little solvent loss to the atmosphere occurs.
Vapor phase cleaning, or vapor degreasing systems consist of a tank of halogenated
solvent heated to the solvent's boiling point. Parts to be cleaned are placed in the
vapor zone so that the condensing solvent dissolves the oil-based contamination and
rinses the part clean. When the temperature of the part finally reaches the
temperature of the solvent, condensation ceases and the parts can be removed dry.
Since the potential exists for much greater air emissions than from cold cleaning dip
or soak tanks, special recovery equipment is installed on the tank. The first item, a
cooling jacket, is usually placed around the outside of the tank and is located at the
vapor/air interface. By cooling the air above the vapor, a cool air blanket is formed
which helps suppress vapor from escaping. A second unit, a finned coil condenser, is
installed inside the tank and is designed to fully condense any vapor that reaches the
coil. This condensate, which includes water evaporated from the parts and water
condensed from the air (this water forms frost on the coil and is removed during the
coils' defrost cycle), flows into a solvent/water separator. The separated water is sent
to the facility's wastewater system and the solvent flows into a sump to be used by the
cleaning tank spray unit. Vapor phase degreasers can be manual (open top) or
conveyorized (fully enclosed).
Alkaline and acid cleaners, more commonly called aqueous cleaners, both utilize the
same equipment and are therefore discussed together. Alkaline cleaners are used
B20-7
-------
primarily to remove soil from metal parts. Acid cleaners are used to remove rust,
scale, and smut (formed when parts are electrocleaned). Recently, very strong
alkaline cleaners (with cyanide and chelating agents added), have been formulated to
replace acid cleaners. The most common method for using aqueous cleaners is the
soak tank or soak cleaning method. Since the equipment used is almost identical to
the equipment used for solvent soak cleaning, the reader is referred to the previous
paragraph on soak cleaning. The only additional method that deserves special
discussion is the electrochemical cleaning method.
Electrochemical cleaning is similar to an electroplating operation in that the work
piece is connected to a source of current. Cleaning is achieved by the scrubbing
action of the gases generated at the electrodes. With direct current cleaning, the
metal part is attached to the cathode. Hydrogen gas is liberated at the part's surface
and provides the scrubbing action. Since the volume of hydrogen released at the
cathode is greater than the volume of oxygen released at the anode, scrubbing action
is greatest at the cathode. Offsetting this advantage is the potential for smut
formation.(metal contaminants in the solution plate onto the part) and the possibility
of hydrogen embrittlement of the metal. With reverse current cleaning, the part is
attached to the anode. While the cleaning effectiveness is less than with direct
current cleaning, smut formation and hydrogen embrittlement are prevented. The only
major disadvantage of reverse current cleaning is that many metals are subject to
attack (some of the metal will dissolve into the solution and plate onto the cathode).
Electrochemical cleaning is often employed prior to electroplating and after vapor
degreasing, soak cleaning, and/or acid cleaning.
Water, the final cleaning medium, is usually used in a soak tank and/or spray unit. The
importance of water as a cleaning medium is that almost every cleaning operation
mentioned requires that a primary water wash be performed before and after each
operation. By preventing cross contamination of various cleaning media (for example,
the dragout or carryover of alkaline solution into an acid bath), the life of a solution is
extended and the overall cleaning efficiency is maintained. Since proper cleaning
requires an exacting sequence of operations, water use is an integral part of any
cleaning operation.
B20-8
-------
7. WASTE DESCRIPTION
The primary specific wastes associated with metal parts cleaning are listed in Table
7-1, along with their sources. Generally, these wastes are produced whenever the
level of contamination in the cleaning media inhibits efficient cleaning. The
composition of the waste will depend on the cleaning media used and the type of soil
removed (oils, greases, waxes, metallic particles, oxides, etc.). The types of waste
produced are largely independent of the nature of the plant.
If a facility has a wastewater treatment system, alkaline and acid cleaning solutions,
along with primary rinse water waste are mixed together (one acts to neutralize the
other) and then treated. Secondary rinse water (if a secondary rinse is used) is usually
used to replace discarded primary rinse water or used as a makeup for cleaning
solutions. For facilities using small amounts of cleaner, the tendency is to drum the
material for disposal. Solvent waste is normally drummed for disposal or sold to an
off-site recycler. The only on-site recycling being performed is by companies using
vapor degreasers.
8. WASTE GENERATION RATES
Metal parts cleaning operations generate waste comprised of abrasives, solvents,
aqueous cleaning solutions, and spent rinse water. Except for the solvent waste, all of
the waste streams are very much diluted with water. Aqueous cleaners are usually
sold at various levels of concentration that must then be diluted before use. Based on
a review of available literature and information from industry contacts, a profile of
the wastes generated from metal parts cleaning operations in the U.S. was estimated
by project staff and is shown in Table 9-1. This profile shows a fractional share of
each waste stream in the total stream generated based on weight.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The list of individual primary waste streams and their sources along with a list of
source reduction methods is presented in Table 9-1. The following sections discuss
recommended waste reduction methods and identified procedures. The basis for
identification of those methods came from published accounts in the open literature.
B20-9
-------
Table 7-1 Metal Parts Cleaning Wastes
N5
O
No.
1.
2.
3.
4.
Waste
Description
Abrasives
Solvents
Alkaiines
Acids
Process Origin
Removal of rust;
scale polishing
of metal
Removal of oil-
based soils
Removal of
organic soils,
descaling
Removal of
scale, smut
Composition
Aluminum oxide, silica
metal, water, grease
Halogenated and non-
halogenated solvents,
oil-based contaminants
Alkaline salts, additives,
organic soils, water
Acids, additives, oxides,
metal, water
RCRA Code
-
F001
F002
F003
F005
F009
-
5.
Rinse water
Removal of
previous cleaning
material
Water with traces of
cleaners and additives
-------
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" have also been included. Good operating practices are defined as
procedural or institutional policies which result in a reduction of waste. The following
items highlight the scope of good operating practice:
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency prepare9ness
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Abrasives
Abrasive powders are usually mixed with an oil-based or water-based binder and are
then applied to a polishing or buffing wheel. Waste from this operation consists of
worn out cloth wheels saturated with abrasive, metal particles, binder, and various
oxides. Wastes from vibratory or mass finishing operations consist of abrasive, metal
particles, water, and oxides dispersed in a slurry. Alkaline or acid cleaners are
sometimes added to the slurry so that additional cleaning action is provided. Usually,
slurries are discarded when the abrasive has undergone a given amount of attrition or
breakdown. The following source reduction methods were noted for abrasive cleaners:
B2Q-1]
-------
o Use of greaseless or water-based binders for polishing or buffing.
When oil-based binders are used, the frictional heat generated during
buffing can cause the binders to burn. This, in turn, leads to the
requirement for additional cleaning e.g. using alkaline soaking. When
properly used, greaseless compounds produce an object that leaves the
wheel clean and dry (Durney 1984, Chap. 3B). Also, greaseless
compositions tend to adhere to the surface of the wheel so that wheel life
is extended.
o Use of liquid spray compositions.
Most abrasive compositions are formulated for use in bar form (the bar is
held against the wheel to apply abrasive). With a liquid spray system, a
spray gun applies the compound to the wheel automatically. Since the
optimum quantity of compound is always maintained on the wheel, wheel
wear due to compound deficiency, compound waste due to over-application,
and the requirement for subsequent cleaning (spray compounds are usually
water-based), are all significantly reduced or eliminated (Durney 1984,
Chap. 3B).
o Careful control of water level in mass finishing equipment.
If not enough water is used, work items leaving the equipment will be dirty.
In addition, too little water will increase the attrition rate of the abrasive
and increase replacement frequency (Durney 1984, Chap. 3C).
9.1.2 Solvent Cleaners
Solvents are used for various cleaning operations by a large segment of industry. By
far, the most common equipment item used is the soak tank, followed by the vapor
degreaser. While much more attention has been focused on vapor degreaser operation
because of its much greater potential for producing air pollutants, it should not be
overlooked that the main methods for reducing waste from both equipment items are
the same. Basically, the two most important source reduction goals are to minimize
solvent vapor loss and to maintain solvent quality. 3y reducing solvent vapor loss, the
composition of the solvent remains as close as possible to the solvent's original
composition. Solvents contain chemical stabilizers that help prevent acid formation
B20-]2
-------
and remove acid contaminants from the bulk of solution. As solvent is used, its ability
to neutralize or prevent acid formation lessens. Unless measures are taken to prevent
a solvent from going acid, the entire quantity of solvent has to be replaced
periodically. By maintaining the solvent's quality, the need for replacement is
reduced. The following source reduction measures were considered as ways of
reducing waste due to solvent cleaning operations (methods for reducing waste due to
solution drag-out are discussed in Section 9.1.5, Rinse Water):
o Installation of lids on tanks.
Lids should be placed on all tanks when they are not in use. By using a
cover during periods of non-use, vapor degreaser solvent loss can be
reduced by 24 to 50 percent. Additional reductions have been achieved by
using covers designed to allow for use even while the cleaning operation is
in progress (Durney 1984, Chap. 3E). Known as "silhouette entries", they
are covers with opening shapes that match the shapes of the parts being
degreased and therefore minimize the area for vapor loss*. Solid covers
should be designed to slide closed horizontally across the open surface.
This disturbs the vapor zone less than covers that are hinged.
o Increase in freeboard space.
Freeboard is the distance between the top of the vapor zone and the top of
the tank. Early vapor degreaser designs called for a freeboard equal to 50
percent of the tank's width. In the mid-1970's, EPA regulations
recommended a freeboard of 75 percent. This reduced emissions by 27 to
46 percent, based on experimentation. For shops where air turbulence is
present, increasing the freeboard to 100 percent can provide an additional
39 percent reduction (Durney 1984, Chap. 3E). As was previously stated
for installing lids, increasing the freeboard on all soak tanks should be
equally effective.
o Installation of freeboard chillers in addition to cooling jackets.
With this approach, a second set of refrigerated coils is installed above the
vapor degreaser's condenser coils. These coils chill the air above the vapor
zone and create a secondary barrier to vapor loss. Reductions in solvent
Westinghouse Electric Corp. 1985: Personal communication.
B20-13
-------
consumption of up to 60 percent have been measured in practice.
Unfortunately, water contamination of the solvent can occur due to frost
build-up on the coils. Therefore, special water collection equipment is also
required. For systems using azeotropic co-solvents that are water soluble
(such as alcohols), the potential for water contamination makes this
method impractical (Durney 1984, Chap. 3E).
Better operating practices.
Cross-contamination of solvent is a common occurence, especially when
the types of solvent being used have similar sounding names. As little as
one-tenth of one percent 1,1,1-trichloroethane mixed into a tank of
trichloroethylene can cause an acid condition (Smith 1981).
Just like solvent cross-contamination, water contamination can lead to
acid formation. In addition, water contamination can form minimum
boiling azeotropes (a mixture of solvent and water, inseparable through
simple distillation, with a lower boiling point than water or solvent alone).
To avoid water contamination, the water separator should be cleaned and
checked frequently for proper drainage. Next, the temperature "of the
water exiting the condenser coils shoufd be maintained at 90 to 100°F. If
the temperature is too low, moisture will condense out of the air. Finally,
parts should be checked to see that they do not enter the degreaser while
wet. This may call for using oil-based abrasives and cutting oils in
production steps prior to cleaning (Smith 1981).
Sludge that collects in the bottom of the tank should always be removed
promptly. Contaminants, such as paint, absorb solvent, dissolve into
solution, and reduce cleaning efficiency (Durney 1982). Zinc and aluminum
fines, which are particularly reactive in chlorinated solvent, can lead to
acid formation if allowed to collect. Organic soil contamination should not
be allowed to exceed 10 percent for cold cleaning operations and 25
percent for vapor degreasers. When these levels are exceeded, acid
formation can occur. While most solvents used in vapor degreasers are
cleaned by distillation and reused, waste solvent from cold cleaning
operations is normally drummed and disposed of (Payne 1961).
R20-14
-------
As solvents are used, their ability to neutralize acid lessens. Normally,
when an acid acceptance test indicates that a solvent is close to going
acid, fresh solvent will be added to boost the level of stabilizers in the
tank. This, however, is a poor solution, since the level of stabilizers in the
tank can never be made equal to the level of stabilizers in fresh solvent.
The proper solution is to analyze the solvent and add specific components
rather than fresh solvent. Usually, the expense of analysis will be offset by
the savings in solvent for tanks of 500 gallons or more (Durney 1982). This
method should also be useful for facilities that recycle their solvent, since
distillation usually removes most stabilizers from the solvent.
For facilities using a large number of cold cleaning tanks, standardizing the
solvent used would help by making the potential for recycling more
attractive. Standardizing, in this context, means using a minimum
(preferably one) type of solvent in all operations in the plant. Many times,
the decision to use a certain type of solvent as opposed to another is based
on an operator's preference, rather than a technical need or requirement.
In addition, the amount of waste generated from a single cold cleaning tank
is small, so that the potential for recycling is seldom considered.
Once standardization has been implemented, the next step would be for
companies to consider the possibility of consolidating cold cleaning opera-
tions into a centralized vapor degreasing operation. While cold cleaning
solvents must usually be discarded when the level of contamination exceeds
10 percent, vapor degreasers can operate up to a level of 25 to 30 percent.
In addition, vapor degreasers provide a much better level of cleaning and
the parts leave the unit dry. Another advantage of consolidation would be
that all solvent handling would occur at a centralized location.
Other waste reduction techniques based on better operating practices
would be to locate cold cleaning tanks away from heat sources, control the
amount of heat supplied to vapor degreasers, avoid spraying parts above
the vapor zone or cooling jacket, and avoid solvent vapor drag-out.
P20-15
-------
Solvent vapor drag-out, or solvent vapor removal from the tank, occurs
when a work piece is inserted or withdrawn from the tank too quickly. The
speed of the work should not exceed 11 feet per minute. In addition, the
geometry of the work piece can affect drag-out. If the space between the
wall of the tank and the work piece is too narrow, then a piston effect will
occur and solvent vapor will be forced out of the tank. As a general
measure, the cross-sectional area of the work load should not exceed 50
percent of the tank's open area (Smith 1981). Drag-out of the solution or
solvent is further discussed in Section 9.1.5.
9.1.3 Alkaline Cleaners
Alkaline cleaners are used to remove a wide variety of organic contaminants from
metal surfaces. The most common equipment used is a heated soak tank with spray
unit. Alkaline cleaners are viewed as viable substitutes for solvents. The following
source reduction methods for reducing wastes due to the use of alkaline cleaners
(except for solution drag-out) were noted:
o Remove sludge frequently.
Waterloo Industries, Inc. in Waterloo, Iowa manufactures steel cabinets
that are alkaline cleaned prior to phosphate coating. Waterloo installed a
100 gallon per minute Liquid-Life separator unit designed to remove sludge
and particulate matter continuously from the alkaline bath. The unit
consists of a pump, hydrocyclone, and sludge retention tank. Since the unit
was installed, replacement chemical costs have been reduced 20 percent,
the time interval between dumping and total cleanout of the system has
been lengthened from four to thirteen weeks, and maintenance associated
with the cleaning process has been reduced because a pump is the only
moving part in the system. Liquid-Life claims that the system can also be
used for solvent and acid cleaning operations (Anonymous 1982).
o Better operating practices.
All parts that enter the soak tank should be free of solvent or acid. The
best way to ensure this is to practice efficient rinsing of the part. For
parts that are first mechanically cleaned, water-based abrasives and
cutting oils should be used. This reduces the cleaning load (i.e., oil) that
enters the cleaning tank and hence extends solution life.
B20-16
-------
9.1.4 Acid Cleaners
Acid cleaners are used to remove oxidation, scale, and rust from metal surfaces. The
most common equipment used is a heated and lined soak tank. Since the basic
equipment and operating procedures used with acid cleaners are very similar to
alkaline cleaners, the source reduction methods are the same.
9.1.5 Rinse Water
Water is used to remove or dilute cleaning solutions that are dragged out with cleaned
parts. If these cleaning solutions are not removed, they can affect the quality of the
work and contaminate subsequent cleaning and processing operations. Since rinsing is
essentially a process of dilution, the general trend in the past was to use large volumes
of water. Today, however, efficient rinsing is required to achieve the proper level or
degree of dilution and conserve water. By conserving water, capital and operating
costs for waste treatment unit are minimized. Source reduction methods considered
for reducing the amount of cleaning solution drag-out and reducing the amount of
water required are:
o Proper design and operation of rack system.
Through proper design and operation of a rack system, solution drag-out
can be significantly reduced. Parts should be racked so that the surface is
nearly vertical and the longest dimension is horizontal. Also, the lower
edge should be tilted from the horizontal (this allows run-off to occur at a
corner rather than the entire edge). Withdrawal from the cleaning solution
should be made slowly and the part allowed to drain over the tank for a
minute or two. Additional drainage time can be provided by installing
sloped drain boards at the end of the tank. As shown below, proper racking
can reduce solution drag-out (Durney 1984, Chap. 32).
Drag out
Item gal/1000 ft2
Vertical parts, well drained 0.4
Vertical parts, poorly drained 2.0
Vertical parts, very poorly drained 4.0
Horizontal parts, well drained 0.8
Horizontal parts, very poorly drained 10.0
Cup-shaped parts, very poorly drained 8.0 to 24.0+
B20-17
-------
For items with cup-shaped recesses, drainage can be a difficult problem. If
the part cannot be positioned to allow for drainage, special measures must
be taken. Some of these measures include drilling or repositioning drain
holes in the part, tilting the rack as its removed from the bath, and/or
installing air-jets to blow off cleaner solution from the part.
As a final point, rack maintenance should not be overlooked. If rack
insulation is cracked, solution can be carried out in the gaps and fissures.
In addition, exposure of the rack metal during electroplating operations can
lead to contaminated bath solutions. Uninsulated racks used for cleaning
should be stripped regularly since the rough surface will hold solution by
capillary action. Rack maintenance should be viewed as a routine
operation and not something to perform when a problem develops.
o Proper design and operation of barrel system.
While barrels are normally fully immersed during electroplating operations,
maximum rinsing efficiency occurs when the barrel is only immersed
partially. The proper depth and rate of rotation depends on many factors
but normally occurs when the barrel is immersed to about 38 percent of its
diameter. In most plants, rinse tanks for barrel operations are designed
and operated the same way as electroplating baths (i.e., the barrel is fully
immersed). After immersing the barrel, it is raised over the tank while
rotating and allowed to drain. This operation should always be performed
twice over each rinse tank. Normally, however, the barrel is only rinsed
once or is not drained properly. In addition, many facilities employ only
one rinse tank. At a minimum, two counterflow oold rinses and a final hot
rinse should always be used (Durney 1984, Chap. 32).
o Proper installation and operation of rinse tanks.
Since the process of rinsing is one of dilution, good practice should assure
that the rinse tank is well mixed at all times. Tanks can be mixed by
agitating the water with oil-free air, introducing fresh water at the bottom
of the tank, and by other mechanical means. The water in the tank should
exhibit a rolling turbulence without undue splashing. For cleaners that are
not very easy to rinse off, heated rinse tanks are often employed.
B20-18
-------
Once agitation and complete mixing in the tank have been assured, the
next concern is the sequencing of rinses. Pinkerton and Graham (Durney
1984, Chap. 32) presented the mathematical derivation of rinsing equations
proposed by several authors for five different rinsing systems. For each
case, the drag-in and drag-out volumes and concentrations were held
constant. The five systems examined, along with the flow rates required to
maintain equilibrium, are listed below.
—— flow
System gpm
1. Single rinse 10.0
2. Two rinses in series, equal flows of fresh rinse 0.6
water to each tank.
3. Three rinses in series, same condition as in 2. 0.3
4. Two counterflow rinses, fresh water feed to second 0.3
tank only.
5. Three counterf low rinses, fresh water feed to 0.1
third tank only.
Based on these results, the conversion from a single rinse to a counterflow
rinse system can be well justified. For facilities with limited floor space,
single rinse tanks can often be converted by welding one or more dams
across the tank. For manual operations, the dams are designed to rely on
an overflow or cascade action, while automated systems would require
pumps to be installed. Most modern facilities are designed with multiple
rinse tanks after each cleaning operation. While theoretical wastewater
reductions of 90 percent have been reported (USEPA 1983), the actual
reduction expected would be closer to 66 percent*.
Installation of water sprays on rinse tanks.
For installations with single rinse tanks and limited space, rinsing efficien-
cy can easily be increased by installing a spray system. By spraying work
items with fresh water as they are raised above the rinse, the equivalent of
an extra one-half counterflow bath is obtained (Durney 1984, Chap. 32).
Sprays should be properly designed to provide uniform coverage or rinsing
of the part and not produce undue splashing. Spray rinsing is also
beneficial on multitank counterflow systems where each spray unit is fed
water from the succeeding rinse tank.
National Association of Metal Finishers 1985: Personal communication.
B20-19
-------
o Installation of fog nozzles on heated aqueous cleaner tanks.
A fog nozzle is a special high pressure water spray unit that produces a
finely atomized mist of water or fog. Since the water is so finely
dispersed, only a small amount of water is used compared to a normal spray
unit. Therefore, fog nozzles can be used over heated cleaner tanks for
rinsing work items without introducing a surplus of water. Two main
benefits of using fog nozzles are that they help cool the part so that the
cleaning solution has less chance of drying on the part and that they reduce
drag-out by diluting the solution retained on the part. The elimination of
rejects and the reduction of drag-out from cleaner and plating baths
through the use of fog nozzles has been reported (Durney 1984, Chap. 32).
o Chemical rinsing.
In some facilities, rinse water from an alkaline cleaning operation is reused
to rinse parts from an acid cleaning operation*. The basic premise is to
combine rinsing and waste treatment in one operation. While this
procedure would reduce the amount of waste rinse water generated and
reduce the degree of waste water treatment required, the potential for
contaminating the parts with metal hydroxide precipitates would be
increased. Therefore, this method would be limited to those parts not
requiring rigorous cleaning.
o Use of deionized water for rinsing.
Use of regular tap water is a major source of impurities in any closed loop
system. By using deionized water, many rinses can be reclaimed using a
simple evaporation system. In addition, use of deionized water can extend
plating bath life by reducing impurity drag-in and reduce the number of
rejects produced. Many packaged systems commercially available can
supply deionized water of adequate quality (most electroplaters do not
require extremely high purity water). (Durney 1984, Chap. 5).
9.2 Implementation Profile
The principal source control methods discussed for reducing waste from metal cleaning
operations are proper equipment operation, maintenance of cleaning solution integrity,
* National Association of Metal Finishers 1985: Personal communication.
B20-2n
-------
and installation of counterflow rinse tanks with spray or fog nozzles. These measures
are technically simple. A major constraint, however, on the implementation of
counterflow rinsing is in its large space requirement. Many older facilities would
have difficulty in finding the space needed for extra tanks.
As increased regulation raises the cost of waste disposal and treatment, efforts to
decrease waste volumes become economically justifiable. Spending to reduce cleaning
waste can produce treatment and disposal cost savings which more than offset the
expenditure. In addition, material costs can be cut since the life of the cleaning
solutions will be lengthened. Aside from the expense of having to make space for
extra equipment in some facilities, most of the measures discussed in this study would
involve minimal capital outlays (e.g., proper equipment operation requires only that
management thoroughly train the employees using the equipment and that the
equipment be correctly maintained). These low-cost measures generally have a fast
payback and are among the first a firm should implement. Furthermore, an increased
awareness on the part of production process designers that waste generation due to
cleaning operations is a significant (yet variable) operating cost, can result in a
substantial cost avoidance once production is underway.
9.3 Summary
Table 9-1 presents summary data on metal parts cleaning waste sources and control
methods. The rankings shown in the table were based on review of the available
literature. It is estimated that reductions in waste generation that have occurred as a
result of practicing the noted methods are significant, as evidenced by a current
reduction index of 2.0 (50 percent). By implementing additional waste reduction
measures or increasing the use of existing measures, the amount of waste currently
being generated can be reduced significantly, to a level characterized by a future
reduction index of 1.2 to 1.9 (30 to 48 percent). The most effective measures for
achieving this reduction include better operating practices, and frequent removal of
sludge from alkaline and acid cleaners.
10. WASTE REDUCTION THROUGH MEDIA SUBSTITUTION
Since the choice of which cleaning medium to use is seldom clear cut, many
opportunities exist for substituting one cleaning medium with another. Currently,
B20-21
-------
TABLE 9-1 SUMMARY OF SOURCE CONTROL METHODOLOGY FOR METAL PARTS CLEANING
O
I
NJ
1
Waste Stream
1
1 Abrasives
1
1
!
| Solvent
1 Cleaners (»)
1
1
Alkaline
| Cleaners (»)
1
1
| Acid Cleaners
1
1
1
I Rinse Water
1
1
1
i
1
1
I All Sources
I 1
1 Control Methodology
1 1
| 1 Use water-based binders |
12. Use liquid spray compositions 1
|3. Control water level in equipment i
| Overall |
|1. Install lids/silhouettes on tanks
|2. Increase freeboard space on tank |
|3. Install freeboard chillers on tanks |
14. Implement better operating practices |
I Overall |
|1. Remove sludge frequently/continuously!
|2. Install lids on tanks
|3. Implement better operating practices 1
| Overall |
|1. Remove sludge frequently/continuouslyl
|2. Install lids on tanks |
|3. Implement better operating practices 1
| Overall I
11. Operate rack system properly I
|2 Operate barrel system properly |
|3 Operate rinse tanks properly |
|4 Install water sprays on rinse tanks
|S Install fog nozzles on heated tanks |
|6 Use chemical rinsing |
17. Use deionized water for rinsing |
1 Overall !
| All Methods
Found Documentation
Quantity | Quality
1 i
t I
1 !
1.00 | 2
3 1
2 1
i 1
3 i
2 50 | 2
2 1
0 !
2 1
1.33 | 1
2 1
0 I
2 1
1 33 | 1
3 I
1 1
2 1
1 I
i !
i 1
1 I
1.57 | 2
i
1
2 1
2 1
2 1
00 |
2 1
2 1
2 1
2 1
00 |
3 1
0 1
2 1
67 I
3 1
0 1
2 1
67 |
3 I
3 1
3 1
3 1
3 1
1 I
2 I
57 |
Waste |
Reduction I
Effectiveness I
1 1
2 !
2 i
1 67 1
2 1
2 1
3 1
4 I
2.75 |
3 1
1 1
3 1
2 33 |
3 1
' 1
3 1
2.33 |
3 I
1 I
3 1
3 1
3 I
2 1
2 1
2 36 !
Extent of i
Current Use |
1
2 1
1 1
3 1
2.00 |
2 1
2 1
1 i
2 1
1.75 |
1 I
1 i
2 I
1.33 |
1 1
1 I
2 I
1.33 |
3 I
2 1
2 I
2 I
2 1
1 1
1 1
1 56 |
Future | Fraction of | Current
Application | Total Waste | Reduction
Future Reduction Index 1
1
Potential | | Index Probable I Maximum 1
2 1 | 0.5 0.3 | !
2 I 1 0.5
3 1 1.5
0.8 | 08|
0.4 | I
2 33 1 0.00 ! 1.5 | 0.5 | 0.8 1
2 I i 1.0 | 0.5 | |
2 I 1.0
2 | | 0.8
0.5 | |
1.1 I I
4 I | 2.0 2.0 | 2 0 |
2.50 | 0.04 | 2.0 1.0 | 2.0 |
3 | 0.8
3 | | 0.3
1.7 | 1.7 |
0.6 | . 1
4 | | 1.5 | 1.5 | |
3.33 | 0.33 | 1.5 | 1.3 | 1 7 |
3 | | 0.8
3 i I 0.3
1.7 | 1 7 |
0.6 I i
4 | | 1.5 1.5 | |
3 33 | 0.12 | 1.5 1.3 | 1.7 |
< 1 1 2.3 0.8 | I
4 i | 2.0
31 I 1.5
3 | | 1.5
3 1 1 '.5
1 | | 0.5
2.0 | 20 i
'. 1 1
'. 1 1
1. 1 1
0. 1 1
31 I 0.5 1. | |
3 00 | 0 51 | 2.3 111 20|
| 1.00 | 2.0 1 2 | 1 9 |
(*) These
streams include listed "F" and/or "K" RCRA wastes.
-------
there is a major effort to switch from cold tanks using non-halogenated solvents and
vapor degreasers using halogenated solvents to cold tank alkaline and emulsion
cleaners. This major change-over is, in large part, caused by the regulatory pressure
to reduce air emissions from degreasing operations.
Detailed cost estimates comparing vapor degreasing and cold cleaning operations with
abrasives, alkaline cleaning, and emulsion cleaning systems were presented in a recent
report (GCA 1983). For all substitutions except abrasives, solvent cleaning had lower
operating costs than either alkaline or emulsion systems. Major disadvantages of
cleaning media substitutions were that use of abrasives was limited to parts with
simple shapes, use of alkaline cleaners did not always leave the parts as water-free as
required, and use of emulsion cleaners left an oil film on the part that required
additional cleaning. Since no information was provided concerning the effect of
substitution on waste generation rates, no recommendations can be made except for
the substitution of solvent and/or alkaline cleaning solutions with abrasives. Extensive
application of this method can be made by facilities engaged in paint stripping from
metal parts which is covered in the study of paint application.
While stricter environmental air pollution laws are forcing a shift away from solvent
degreasing operations towards greater use of alkaline or emulsion cleaners, client
specifications often force many jobPiflo^ to perform certain operations in a specific
way. Whereas captive shops can usually perform any type of cleaning so long as a
certain level of cleanliness is maintained, job shops must clean parts according to the
standards and methods specified by the client. Many of the standards dictate not only
the type and cleaning sequence to be employed, but also the number of rinse stages to
be used between each step. Therefore, the revision of standards can be an effective
way of minimizing waste.
11. CONCLUSIONS
While industries engaged in metal parts cleaning operations have done much to reduce
the amount of waste they generate, it appears that further reductions are possible.
Our estimates indicate that possible waste reductions range from 30 to 48 percent.
Several methods that appear promising for the industry as a whole would be to
implement better operating practices on all systems, and frequently remove the sludge
from aqueous cleaners. Especially in light of rising disposal costs, many of the
B20-23
-------
proposed waste reduction measures are cost effective, and are justifiable on economic
as well as environmental grounds.
In general, it could be useful if the organizations that set cleaning criteria or
standards review them in the context of associated waste generation.
12. REFERENCES
Anonymous. 1982. Cyclonic separator saves pretreatment chemicals. Prod. Finish.
March, 1982.
Durney, L.J. 1982. How to improve your paint stripping. Prod. Finish. December
1982.
, ed. 1984. Electroplating engineering handbook. 4th ed. New York, N.Y.:
Van Nostrand Reinhold Co.
GCA Corp. 1983. Preliminary analysis of possible substitute for 1,1,1-trichloroethane.
Washington, D.C.: U. S. Environmental Protection Agency.
Hoogheem, T.J. et. al. 1979 Source assessment: solvent evaporation-degreasing
operations. EPA-600/2-79-019f (PB 80-128812). U.S. Environmental Protection
Agency.
Morrison, P.W., ed. 1973. Environmental control in electronic manufacturing. New
York, N.Y.-. Van Nostrand Reinhold Co.
O'Brien, T. 1981. Estimation of market share of metal cleaning technique. Chapel
Hill, N.C.: GCA Technology Division.
PACE. 1983. Pace Company Consultants and Engineers, Inc. Solvent recovery in the
United States 1980-1990. Houston, TX.: PACE CO.
Payne, H.F. 1961. Organic coating technology. 2 vols. New York, N.Y.; John Wiley &
Sons.
Smith, C. 1981. Troubleshooting vapor degreasers. Prod. Finish. November 1981.
USEPA. 1973. U.S. Environmental Protection Agency, Office of Air Quality and
Water Programs. Air pollution engineering manual: AP-4D. 2nd ed. Washington,
D.C.: U.S. Environmental Protection Agency.
. 1979. U.S. Environmental Protection Agency, Office of Air Quality Planning
and Standards. Organic solvent cleaners - background information for proposed
standards. EPA-450-2-78-045a. Research Triangle Park, N.C.: U.S. Environmental
Protection Agency.
. 1983. U.S. Environmental Protection Agency, Office of Water Regulations
and Standards. Development document for effluent limitation; guidelines and
820-24
-------
standards for the metal finishing industry. EPA-440-1-83-091A. Washington, D.C.:
U.S. Environmental Protection Agency.
Versar, Inc. 1985. National profiles report for recycling: a preliminary assessment.
Draft report for waste treatment branch. EPA Contract No. 68-01-7053. U.S.
Environmental Protection Agency.
13. INDUSTRY CONTACTS
Joseph P. Chu, Environmental Activities Staff, General Motors Technical Center,
Warren, MI.
David Anzures, National Association of Metal Finishers, San Fernando, CA.
Walter G. Vaux, Process and Chemical Engineering, Westinghouse Electric Corp.,
Pittsburgh, PA.
B20-25
-------
-------
1. PROCESS: PAINT APPLICATION
2. SIC CODE: XXXX
3. INDUSTRY DESCRIPTION
The application of paint is practiced in most industries, and thus does not have an
individual Standard Industrial Classification (SIC) code. Usually, industries engaged in
paint application can be segregated according to one of three paint or surface coating
classifications: architectural, product, or special purpose. Industries using
architectural paints would be engaged in construction activities; (this would include
building contractors, painters, and home repairers). Product coatings would be used by
industries engaged in manufacturing durable goods, such as automobiles, appliances,
furniture, etc. The last classification, special purpose coatings, would include
organizations involved in industrial, military, and highway maintenance. Also included
would be any person or industry using aerosol spray paints.
4. PRODUCTS AND THEIR USES
Due to the diversity of the industries utilizing paint application as part of their
operation, a comprehensive listing of all the products that are painted would be beyond
the scope of this report. Basically, surface coatings are used wherever it is desired to
provide decoration, protection, and/or safety marking to a product or item.
5. RAW MATERIALS
In 1983, the paint manufacturing industry sold $8.6 billion worth of paint. This
represented 463 million gallons of architectural coatings, 331 million gallons of
product coatings, and 130 million gallons of special purpose coatings (Webber 1984).
While most of the architectural coatings sold were water-based, most product and
special purpose coatings were solvent-based. For additional details concerning the
types of paint and the industries that utilize them, the reader is referred elsewhere in
this appendix to the process study on paint manufacturing.
B21-1
-------
6. PROCESS DESCRIPTION
Detailed process descriptions regarding paint application can be found in the open
literature (Payne 1961, Letsky 1960, Campbell and Glenn 1982). The following
discussion highlights the various types of equipment used for applying product coatings
and the supporting operations required for producing a finished product.
Before a product coating can be applied to a surface, the surface must be free from
contamination. Many different types of abrasives, alkalines, acids, and solvents are
used by industry to clean metal surfaces. Because of the complexity and diversity of
the contaminants, many different cleaning operations are usually required. In
addition, extensive use of water is required to rinse the parts free of cleaning solution
so that residual cleaner does not contribute to the contamination problem. Once a
part is cleaned, surface treatment such as phosphate coating can be applied if desired.
The purpose of surface treatment is to condition or prepare the surface so that the
paint forms a better bond with the metal surface. For detailed physical descriptions
characterizing cleaning and metal surface treatment operations, the reader is referred
to those studies in this appendix.
After the item has been cleaned and treated, paint can be applied. Depending on the
size, shape, complexity, and quantity of items to be painted, different application
methods can be employed. When it is desired to paint a large number of very small
items, the most commonly used methods are tumbling, barreling, or centrifuging. For
all three methods, the parts are placed inside a barrel, solvent-based paint is poured
onto the items, and the barrel is then rotated. After a short time and at the correct
point of tackiness, the parts are transferred to an oven in a wire basket. If the time in
the barrel is too short, the parts leaving the oven will be stuck together. If the time in
the barrel is too long, the finish will be damaged and marred. While paint consumption
using these methods is very small, the empirical nature of the operation requires that
the operator be highly experienced to achieve reliable results.
For cylindrical items, a commonly used method is dipping. Here the paint is held in a
large tank and the object to be painted is slowly lowered into the tank and then
withdrawn. Many complex items can be dip painted provided that the drainage points,
(the places where the excess paint drips off), can be located where they are not
B21-2
-------
noticeable. Items as large as refrigerators, washing machines, and automobiles have
been dip painted. A variation of dip coating or dipping is electrocoating. A water-
based paint is mixed into a large tank of water and given an electrical charge. The
object to be coated is connected to an electrical ground and the electrostatic potential
causes the paint particles to plate onto the object's surface. As the thickness of paint
builds up, most of the water is forced out of the paint so that the coating adheres
tightly to the surface. Following coating operation, water sprays are used to rinse the
object, with the rinse water being returned to the main dip tank.
Flow coating is often employed for items that would be difficult to dip because of
their size or shape, or as a means of avoiding the installation and operation of large
dip tanks. A flow coating system operates by using high pressure sprays to flood the
item with solvent-based paint. After spraying, the item is allowed to drain and the
excess paint is recirculated. Since a considerable amount of bubbling occurs due to
spraying, the item is then passed through a solvent chamber where the solvent vapors
allow the paint to reflow. Following this operation, the item is then oven-dried. The
main disadvantage of flow coating is high solvent loss, which can be three times as
large as for dipping and twice as great as for spraying.
For relatively flat items of large area, roller coating and curtain coating machines are
used. Roller coating is used extensively by the canning industry for painting flat metal
sheets that are then fabricated into cans. It is also used for spreading or applying glue
to wood in the manufacturing of plywood. A roller coating machine operates by
metering paint or coating material onto a roller and then transporting the item past
the roller by means of a conveyor belt. Curtain coating machines were originally used
for coating candies with chocolate and are presently being used in the furniture
industry (for table tops, doors, etc.) and the electronic manufacturing industry. A
curtain coating machine consists of a long, horizontal rectangular pressurized
container that is placed at a right angle to the direction of work flow. Along the
bottom of this container is an adjustable slit that allows the coating to flow and form
a vertical curtain. A conveyor belt is placed on each side of the curtain so that work
items are passed through the curtain and coated without the conveyor belts being
coated. When items are not being coated, the paint flows between the two conveyors,
is collected in a trough, and then pumped back up to the curtain head. Coating
thickness can be controlled by the speed of the conveyor belt.
821-3
-------
While all of the above-mentioned methods have found widespread acceptance by
industry, the most widely used method for applying paint is still the spray gun. A
spray gun operates by using compressed air, to atomize the paint and produce a fan or
circular cone spray pattern. Many installations are automated so that a fixed gun is
turned on when an object passes in front of it. In its simplest use, the gun is hand-held
and the object remains stationary. Some of the variations on spray gun painting are
airless spray guns and electrostatic spray guns. Airless spray guns force the paint out
at high pressure so that air is not required for atomization. By eliminating the use of
compressed air, operating costs are lower, spray mists are not produced, and expensive
exhaust systems are not required. Electrostatic spray units are designed so that the
atomized paint leaving the gun has a positive charge. This positive charge causes the
paint to be attracted to the object which is connected to ground. Since more of the
paint reaches its target (thereby reducing overspray), less waste is generated. Other
spray systems, besides hand-held guns, include revolving bells and discs that rely on
electrostatic and centrifugal forces to atomize the paint. Since neither of these last
two methods relies on air, their advantages are similar to those for the airless spray
gun.
Following the application of paint, the item is passed through a drying or curing oven.
The curing methods employed, infrared or ultraviolet, will depend on the type of paints
being used. Once dried, the items are sent to inspection and final packaging or
assembly. If a part fails inspection because of a bad finish, it is usually reworked by
stripping off the paint and returning it to the cleaning operation. A block flow
diagram of the operation used in product coating application is presented in
Figure 6-1.
7. WASTE DESCRIPTION
The primary specific wastes associated with product coating application are listed in
Table 7-1, along with their primary sources. Waste would consist of empty paint
containers, spent cleaning solutions, paint overspray (including paint collected by air
pollution control equipment), spent stripping solutions, and equipment cleaning wastes.
8. WASTE GENERATION RATES
Paint application operations generate waste comprised of empty containers, solvents,
aqueous (alkaline and/or acidic) cleaning solutions, rinse water, paint stripper, and
B21-4
-------
UNPAINTEO
PARTS
CLEANING AND
PAINT STRIPPIN6
SURFACE TREATMENT
OPTIONAL
ROLLER COATING
CURTAIN COATIN6
SPRAY COATIN6
A
B
C
D
AIR ATOMIZED
PRESSURE ATOMIZED
ELECTROSTATIC
ELECTROSTATIC /
CENTRIPUBAL
0©
PROCESS HASTE CATEGORIES! !
x-s !
(0 SPENT PARTS CLEANING SOLUTIONS j
I
(?) SPENT PAINT STRIPPER
(D PAINT OVER SPRAY
© EQUIPMENT CLEANING NASTES
j
Figure 6-1 Paint Application
B21-5
-------
Table 7-1 Paint Application Wastes
'JO
N3
i—»
I
OS
No.
1.
2.
3.
4.
5.
6.
7.
8.
Waste
Description
Leftover paint in
metal gallon cans
Spent solvent
Spent alkaline
cleaners
Spent acid
cleaners
Rinse water
Paint
Paint stripper
Water and/or
solvent
Process Origin
Raw materials
Cold cleaning
and/or vapor
degreasing
Removal of
organic soils,
descaling
Removal of
scale, smut
Removal of
previous cleaning
solutions
Overspray from
paint application
Stripping of bad
paint from parts
Cleaning of
equipment
Composition RCRA Codes
Metal can with a few
ounces of paint inside
Halogenated and non- F001
halogenated solvents, F002
oil-based contaminants F003
FOG5
Alkaline salts, additives,
organic soils, water
Acids, additives, oxides,
metal, water
Water with traces of
cleaner
Paint
Non-halogenated solvents, F003
paint
Halogenated and non- F002
halogenated solvent, F003
water, and paint F005
-------
equipment cleaning wastes. A survey of the available literature did not yield waste
generation rates for these various waste streams. In addition, volume estimates would
be difficult, due to the diverse types of operations being performed throughout the
industry. Therefore, only fractional waste generation data (the percentage of waste
each stream represents as compared to the total waste generated), could be estimated
by project staff, based on the available information and industry comments. These
values are shown in Table 9-1.
9. WASTE REDUCTION THROUGH SOURCE CONTROL
9.1 Description of Techniques
The list of individual primary waste streams and their sources, along with a list of
source reduction methods, is presented in Table 9-1. The following sections discuss
recommended waste reduction methods and identified procedures. The basis for
identification of these methods came from published accounts in the open literature.
In addition to the waste reduction measures classified as being process changes or
material/product substitutions, a variety of waste reducing measures labeled as "good
operating practices" have also been included. Good operating practices are defined as
procedural or institutional policies which result in a reduction of waste. The following
items highlight the scope of good operating practices:
o Waste stream segregation
o Personnel practices
Management initiatives
Employee training
o Procedural measures
Documentation
Material handling and storage
Material tracking and inventory control
Scheduling
o Loss prevention practices
Spill prevention
Preventive maintenance
Emergency preparedness
B21-7
-------
For each waste stream, good operating practice applies whether it is listed or not.
Separate listings have been provided whenever case studies were identified.
9.1.1 Empty Containers
Facilities that use small quantities of paint and a large variety of color often purchase
paint in small cans and containers (less than five gallons) for use in spray gun
equipment. After emptying the can of paint into the spray gun's holding cup, a coating
of paint remains inside the can. Since these cans are seldom cleaned, the entire can
must be discarded as waste. Also included in this waste stream would be outdated or
leftover paint removed from storage. The waste reduction methods are as follows:
o Better operating practices.
By preventing the contamination of non-hazardous materials such as trash,
with potentially hazardous materials such as solvent-based paint, the
volume of hazardous waste is minimized. Good examples of hazardous and
non-hazardous waste segregation were identified in the paint manu-
facturing process study.
Container waste is generated by the paint that remains inside a can after
the can is emptied and by the paint that is placed in storage, not used, and
becomes outdated. Facilities should strive to consolidate paint use so that
the purchase of paint in bulk is practical. Since one large bulk container
has much less surface area than an equivalent volume of small cans, the
amount of wasted paint is much less. Large bulk containers can often be
returned to the paint manufacturer for cleaning and reuse; (while cleaning
still generates some waste, the paint manufacturer is in a better position
to handle it).
When the purchase of paint in bulk containers is not practical, then paint
should be purchased in the smallest amount required so as to minimize the
amount of residual. Workers should not open a one gallon can of paint
when only a quart or two are required. It should be assumed that any paint
that is left over will require disposal as a waste. While the per unit cost of
paint when purchased in various small quantity containers is greater than
the cost of paint when purchased in larger quantities, the savings in
reduced paint wastage and disposal costs could still be substantial.
B21-8
-------
9.1.2 Parts Cleaning Waste
Wastes associated with parts cleaning would consist of spent solvent, alkaline and/or
acid baths, rinse water, and whatever contaminants were removed. Since the methods
used to minimize waste from the cleaning operation are highly specific to the types of
cleaning methods employed and because the subject of cleaning is so diverse, it is
recommended that the reader refers to the study elsewhere in this appendix on metal
parts cleaning for additional information. Note that paint stripping waste, however, is
covered in this study in Section 9.1.4.
9.1.3 Paint Application Waste
The wastes generated during paint application are primarily due to paint overspray or
the failure of all of the paint to reach the target. If overspray did not occur, then no
paint application waste would be generated and the need for equipment cleaning would
be greatly diminished. Since the degree of overspray is a function of system design
and operation, it can usually be reduced but never completely eliminated. The
following source reduction measures are noted:
o Use of paint application equipment with low overspray values.
When comparing different types of paint application equipment, the
transfer efficiency or degree of paint overspray produced can vary
considerably. Brewer (1980) lists the following efficiency ranges for
several different spray systems:
Spray Method Efficiency
Conventional air-atomized 30-60%
Conventional pressure-atomized 65-70%
Electrostatic air-atomized 65-85%
Electrostatic centrifugally-atomized 85-95%
One company, Flexsteel Industries in Dubuque, Iowa, changed from a
conventional spray system to an electrostatic system and reduced the
amount of overspray it was generating by 40 percent (Campbell and Glenn
B21-9
-------
1982). This saved the company $15,000 a year in paint cost and repaid the
investment within two years.
Other types of equipment that can yield even higher efficiency values are
roller and flow coating machines (90 to 98 percent) and electrocoating
systems (90 to 99 percent). Roller and flow coating machines, however,
would be limited in their applicability based on the shape of the parts.
Electrocoating systems would require a shift from solvent-based to water-
based paint and is discussed in Section 10, Product Substitution. Other high
efficiency systems, such as those that use powder coatings, are also
discussed in Section 10.
Better operating practices.
Since many spray systems are manually operated, the equipment operator
has a major impact on the amount of waste produced. D. Lenchkus (1982)
reports that spray operators generally believe that a large amount of air is
required to achieve proper atomization of the paint. When air pressures
are set too high, the paint has a tendency to bounce off the surface and
increase overspray. Overspray can run as high as 70 percent and achieving
a level of 50 percent is rare. Another reported factor affecting overspray
is the practice of arching the spray gun instead of keeping it perpendicular
to the surface. When the gun is arched 45 degrees away from the surface
at the end of each stroke, overspray can be as great as 65 percent and an
uneven coat of paint can result. By providing the proper training for spray
system operators, waste generation can be minimized. Manual operations
have been eliminated altogether in many facilities with the use of robots.
Preventive maintenance also plays a critical role in reducing the amount of
overspray, stripping waste, and equipment cleaning wastes produced.
Whenever a bad finish is produced, the paint is normally stripped off and
the entire paint application procedure is begun again. By maintaining all
application equipment in good working order, the likelihood of producing a
bad finish is lessened. Spray guns should be cleaned after use or whenever
there will be an appreciable interval between use. For hand-held units, a
solvent rinse with occasional blow-back (cover the fluid tip and operate the
B21-10
-------
trigger; this blows the paint back to its container) is adequate. All moving
parts should be lubricated frequently and properly adjusted. A spray gun
should always be able to provide a smooth change from solid fan to round
cone by adjustment of the controls.
Roller coating and flow coating machines must also be properly maintained
if they are to produce good finishes. Rollers on roller coating machines
should be cleaned regularly to remove dried paint and inspected for
swelling of the material. If swelling is evident, the rollers should be
replaced immediately. Curtain or flow coating machines have a curtain
head that must be kept clear at all times. If this aperture becomes
blocked, the curtain will break and give an uneven finish. To clean the
machine, solvent should first be circulated through the machine. After
this, the aperture should be fully opened and cleaned with a soft rag or
brush. Metal instruments should never be used for cleaning the curtain
head.
In addition, by fully inspecting parts before they are painted, the painting
of potential rejects can be avoided. Whenever a part is painted and then
rejected, additional stripping, cleaning, and painting wastes are generated.
If potential rejects are discovered before painting, they can probably be
corrected (cleaned if the problem is contamination) and then painted with
very little increase in waste generation.
9.1.4 Paint Stripping Waste
Paint stripping is required whenever a bad finish is produced or when the old coating
must be removed prior to re-painting. Reasons for bad finishes can include improper
cleaning, improper paint application, or both. Since the need for paint stripping is
based in parts, on the failure of other operations, some of the control methodologies
are aimed at reducing the number of rejects produced. Source reduction methods
considered for reducing wastes due to paint stripping are:
o Implementation of proper paint application techniques.
By properly applying paint to a given surface, the likelihood of producing a
bad finish is reduced. This, in turn, reduces the generation of waste
B21-11
-------
associated with paint stripping and repainting. Since the proper application
of paint is based on the type of equipment used and the ways in which it is
operated and maintained, the reader is referred to Section 9.1.3 for further
information.
o Implementation of proper cleaning techniques.
Proper cleaning requires the identification of the contamination, deter-
mination of the cleaning operations required to achieve a given degree of
cleanliness, and the correct practice of these operations. In addition,
proper design of the rinse systems is essential for achieving good cleaning
and reducing the chance of cleaning solutions becoming unwanted
contaminants. For an in-depth study of proper cleaning techniques, the
reader is referred to the process study on metal parts cleaning.
o Use of mechanical paint stripping methods.
Cryogenic paint stripping uses liquid nitrogen and plastic beads to
embrittle and then remove layers of paint from an object. The Cryogenic
Coating Removal (CCR) system consists of a feed system that propels non-
abrasive plastic shot at a cryogenically-embrittled rotating target. Since
the paint is under thermal tension, it debonds when impacted by the shot.
From this system, the loose paint and shot are separated in a second
system and the shot returned to the process. A full scale CCR system has
been installed by a major domestic auto manufacturer and the results were
impressive. Compared to hot caustic stripping, the operating costs were 44
percent lower; (this did not include accounting for increased parts damage
and disposal costs due to the use of caustic). Recent tests have been
successful on a wide variety of paints and thicknesses which makes CCR
appear to be an attractive substitute system for conventional paint
stripping operations.
For facilities handling items made of soft metals which are inappropriate for sand or
glass bead blasting, plastic bead blasting appears to be an attractive option. Hill Air
Force Base (AGB) in Ogden, Utah has used plastic bead blasting with great success
(Higgins et al 1985). Previously, paint removal from aircraft was a time consuming
and messy operation. Sprays or baths containing acidic methylene chloride, phenolic,
B21-12
-------
or hot alkaline sodium hydroxide solutions were used to soften the paint on the plane.
Then the paint was scraped off and large volumes of water were used to rinse off the
surface. The resulting rinsewater (tens of thousands of gallons per plane) had to be
treated and the wet paint sludge disposed. The new system utilizes high pressure air
to propel the plastic beads against the painted surface to dislodge the paint. Paint and
plastic are then recovered and separated with the plastic beads being returned to the
feed system. Plastic beads that have broken down due to attrition are removed and
disposed of along with the chipped paint. As reported by a source with the U.S
Department of Defense, plastic bead blasting for all DOD facilities would cost $13
million to install and result in an annual savings of $100 million. With regard to waste
generation, the currently used wet process annually produces 7 million gallons of paint
stripper waste and 100 million gallons of wash water nationwide in all DOD facilities.
The new system would eliminate these wastes and produce 500 thousand pounds of dry
waste (paint plus spent plastic beads) instead. Because of the favorable economics,
many airlines are looking into this process and many small scale systems are available
commercially (Forth 1985).
o Use of non-phenolic strippers.
Several 'non-phenolic/non-acid paint strippers have been developed in
recent years in response to the need to reduce toxicity associated with
phenol and acid additives (such as cresolic acid). Comparisons between
non-phenolic, non-acid brush-on strippers such as Sprazee (made by BASF)
and T-5873 (made by Turco) and phenolic/non-acid strippers such as T-5469
showed that the non-phenolic/non-acid strippers could provide the same
degree of removal power (Staebler, Jr. and Simpers 1983). The main
disadvantage was that much more stripper was needed for a given percent
removal and a longer contact time was required. Some of these
disadvantages were eliminated when non-phenolic/acid strippers were
tested.
o Better operating practices.
In addition to the methods discussed in the study on metal parts cleaning,
the following example of proper equipment layout and its effect on waste
generation was identified. Solvent soak tanks should be located away from
paint curing ovens to minimize the adverse effects of solvent on a painted
B21-13
-------
surface or item. At one facility, a halogenated solvent soak tank was
located near a paint curing oven. Heat from the oven caused solvent to
evaporate from the tank. Being heavier than air, the vapors settled to the
floor and entered the oven. Inside the oven, the vapors contacted the open
flame and broke down, forming hydrochloric acid. Since the hydrochloric
acid vapor altered the cure rate of the paint being applied, all of the
resulting finishes had to be stripped off and reapplied (Durney 1982).
9.1.5 Equipment Cleaning Waste
Regular cleaning of paint application equipment is required to ensure that the
production of rejects is minimized. Cleaning is also required whenever a different
batch of paint is to be applied, (so as to minimize paint contamination from a previous
batch of paint), or whenever the equipment undergoes a period of nonuse. Since the
need for equipment cleaning is directly tied to the paint application operation, all of
the methods listed in Section 9.1.3 are applicable.
o Better operating practices.
An aspect of good operating practice not discussed in Section 9.1.3 would
be the effect of scheduling improvements on equipment cleaning waste
generation. Since the need for equipment cleaning can be tied to the
frequency of paint color or type changes and the periods of inactivity,
proper scheduling of production runs could reduce the need for equipment
cleaning. By producing large batches of similarly painted items instead of
small batches of differently painted items, a large reduction in the need
for cleaning can be achieved. Some reductions would also be possible by
painting light-colored items first and then the dark-colored ones. The
reverse method is used in the paint manufacturing industry to reduce the
need for cleaning of mixing tanks.
Another good operating practice would be to segregate all solvents used for
equipment cleanup*. When a similar or compatible type of paint was going
''Westinghouse Electric Corp. 1985: Personal communication.
B21-14
-------
to be used in the equipment, the waste solvent then could be used as
thinner or makeup for the fresh paint. This method was identified in both
the paint manufacturing and printing operations process studies.
9.2 Implementation Profile
The two main methods discussed for use by industries performing paint application
would be to operate all equipment properly and to install high efficiency paint
application systems. Proper operation of the equipment would require that companies
provide thorough training for their employees and maintain the integrity of the
equipment. No additional equipment would be required and the only technical obstacle
would be lack of information. Installing high-efficiency paint application equipment
would require many companies to modernize their facilities. In addition, many high-
efficiency systems use water-based paints only, so that different paint formulations
would be required. Currently, extensive development work is being performed to find
water-based substitutes for solvent-based paints.
Very few of the proposed control methods would require large outlays of capital.
Many of the methods could result in net savings by reducing the cost of raw materials
and waste disposal. Even the .installation of new high-efficiency equipment has often
been cost-effective. Therefore, it is very unlikely that implementation of the methods
discussed here would entail economic hardships for industry.
9.3 Summary
Table 9-1 presents summary data on paint application waste sources and control
methods. The ratings shown in the table were based on review of the available
literature. Based on the measures already undertaken in the paint application field to
minimize waste, it is estimated that 48 percent (a current reduction index of 1.9) of
the potential waste that could be generated has been eliminated. By implementing
additional waste reduction measures or increasing the use of already implemented
measures, the amount of waste currently being generated could be reduced by 28 to 43
percent (a future reduction index of 1.1 to 1.7) . The most effective measures for
achieving this reduction would be to use proper parts cleaning techniques, provide
better operator training, and install high efficiency paint application equipment.
B21-15
-------
TABLE 9-1 SUWARY OF SOURCE CONTROL NETHODOLOGY FOR PAMT APPIICATIOK
03
ho
h-•
I
I—"
ON
Waste Stream
Control Methodology
Found Documentation i Waste ! Extent of 1 Future I Fraction of | Current | Future Reduction Index i
1 Reduction I Current Use I Application I Total Waste | Reduction | - — — I
I
1
Empty Containers
Parts Cleaning
Wastes (»)
Paint Apoi ication
Wastes
Paint Stripping
Wastes (')
Equipment Cleaning
Wastes (»)
1 Implement better
Overall
| Quantity | Quality | Effectiveness 1 | Potential | |
operating practices | 11 2| 3 I 21 4| |
1 1 GO
2 00 1 3 00 ! 2 00 ! 4 00 ! 0 01
1 See study on Metal Parts Cleaning | — i — | — | — | — I
Overal 1
| — | — I — ! — I — ! 0 68 I
?. Ose equipment with low overspray j 3 3 | 4 | 2 | 21 |
2 Implement better operating practices | 2 \ 3| 3 | 21 4 | |
Overall
I 2.50 I 3.00 I 3 50 | 2.00 | 3 00 | 0.06 |
) Implement proper painting techniques | 2
2. Implement proper cleaning techniques 1 2
3 Use mechanical paint stripping system) 2
4. Use non-phenolfc/non-acid stripper | 1
5 Implement better operating practices | 1
Overall
31 3 | 2| 31 |
31 3| 21 31
3 1 < 1 11 2 I |
1 1 H H 1 1
21 2 1 3 I 3 | |
I 1.SO | 2.40 | 2.60 | 1.80 | 2.40 1 0 01 |
1 Use equipment with low overspray 1 3
2 Implement better operating practices 1 2
Overall
All Sources 1 All Methods
3| 3 | 21 2 i
31 2| 21 4| 1
1 2.50 1 3 00 I 2.50 1 2.00 | 3.00 I 0 24 I
1.00 1
Index | Probable | Maximum
1.5 I 1.5 | 15
15 1 5 I 15
2.0 1 1.2 1.5
2.0 i 1 2 | 19
l.C 1 1.0 1
1.5 | 1.5 I 1.5
20| 1.3 1 15
1.5 I 1.1 |
15 1.1 |
1.0 I 1.5 I 1.5
0.3 | 0.2 |
151 0.4 |
1.5 | 0 9 | 15
1.5 ! 0 3 |
1.0 I 1.0 I 1.0
1.5 1 0.9 | 10
1.9 | ! 1 I 17
(*") These »aste streams include listed T' and/or "K" RCRA wastes
-------
10. WASTE REDUCTION THROUGH PRODUCT SUBSTITUTION
The bane of many finishing departments using solvent-based paint application systems
has been the increasingly high cost of meeting air and water quality regulations, as
well as problems in the disposal of hazardous and flammable wastes. The conventional
solvent-based systems for paint application often pose potential health hazards, due to
the emission of toxic solvents and the disposal of a large amount of waste paint sludge.
To minimize the quantity and toxicity of waste paint requiring disposal, new coating
types have been developed which eliminate the hazardous components of the paint and
also allow for maximum re-use of paint overspray, thus increasing the efficiency of
the paint application process. These alternatives include the use of water-based,
radiation-curable, or powder coating techniques.
10.1 Water-Based Coatings
Water-based coatings are paints containing a substantial amount of water (often as
much as 80 percent) in place of the volatile solvent. The polymers used include alkyd,
polyester, vinyl acetate, acrylic, and epoxies, and can be dissolved, dispersed or
emulsified. These coatings are supplied as baking finishes, as well as air-drying
formulations. One feature which makes water-based coatings attractive is that no
major equipment changes are necessary to apply water-based coatings with solvent-
based coating equipment. The main waste reduction advantage, however, lies in the
ease of recovering paint overspray. Overspray from a water-based coating can be
collected or captured with water in the spray booth. The solution can subsequently be
concentrated and reused as paint again. In addition to substantial reductions in
environmental hazards due to significantly lower air emission levels and smaller
amounts of waste paint sludge generated, water-based coatings can also provide
energy savings by making possible the recirculation of the hot air used to cure the
paint. Water-based coatings have increasingly been used by industry as an alternative
to solvent-based systems. For example, Emerson Electric Company has reported the
use of a water-based electrostatic paint system in place of a conventional organic
solvent paint system (Huisingh et al. 1985). This substitution has resulted in a number
of favorable changes. These include:
Improved quality of application.
Decrease of down time from 3% to 1%.
B2J-17
-------
Reduction in generation of aromatic waste solvent by 95%.
Reduction of paint sludge by 97%.
Increase in efficiency with up to 95% recovery and reuse of paint.
The new system also reduced hazardous waste disposal costs and decreased personnel
and maintenance costs by 40 percent.
10.2 Radiation-Curable Coatings
Another alternative to solvent-based coatings is radiation-curable coatings. These
coatings do not contain or use organic solvents. Reactive monomers are applied as a
liquid to a surface which is then exposed to high energy radiation such as UV or IR
light. Radiation-curable coatings account for 12 to 15 percent of the curable coating
market and were expected to capture 20 percent by 1985 (Campbell and Glenn 1982).
The advantages of using this coating technique include the reduction in wastage from
solvent loss and a substantial decrease in energy and maintenance requirements.
10.1.3 Powder Coatings
Powder coatings represent an attractive third alternative to solvent-based coatings.
Use of powder coatings has been referred to as a "dry painting process". The process
is simple in operation and can be done manually or by highly automated equipment.
With manual systems, the powder is sprayed on the object and the overspray is readily
retrieved and recycled. The unused powder is then collected and separated from an air
stream by various vacuum and filtering methods. It is returned to the feed hopper for
reuse. This ability to recycle the coating material provides a very high efficiency-use
ratio. Powder coatings can also be effectively applied in a fluidized bed*. Since the
system is entirely enclosed, there is no overspray to collect for the application area.
Hence, the coating powder remains clean at all times and the potential for waste
generation (contaminated powder) is nearly zero. Fluidized bed coating with powdered
paint is a well-proven technology*.
* Westinghouse Electric Corp. 1985: Personal communications.
B21-18
-------
Powder coatings fall into two broad categories - thermoplastic and thermosetting, the
choice of which depends on the application. Powder coating still represents a small
fraction (5 to 6 percent) of the U.S. industrial coating market, but its share is rapidly
increasing. Better coating materials such as epoxy-polyester hybrids have been
developed to improve film deposition characteristics and to allow for lower curing
temperatures. Improvements in spraying equipment have also contributed significantly
to the rapid spread of powder coating use for more than 200 different types of
consumer and industrial products.
The greatest non-product advantage of using powder coating lies in the elimination of
environmental problems inherently associated with most liquid-based finishing
systems. Powder coatings contain no solvent and thereby emit negligible volatile
organic compounds. Furthermore, since powder overspray can be readily recycled,
material utilization is high and solid waste generation is minimal. Recent case studies
are demonstrating that powder coating systems can be more efficient, cleaner, and
more environmentally acceptable, while producing a higher quality finish in
conventional solvent-based, water-based, or high solid compliance coating systems
(Letsky 1960). A model study developed in 1982 (Cole 1984) showed that powder
coating systems are economically feasible since the higher installation cost can' be
offset by lower maintenance, materials, labor/cleaning, energy, and waste disposal
costs. Compared to solvent-based coatings, powder coatings provide a 90 percent
waste reduction, reduce rejects by a factor of 4 or greater, and require 30 to 50
percent less energy and 30 to 40 percent less labor. However, the broader issue of the
quantity and toxicity of wastes generated during the production (as opposed to
application) of powder coatings was not addressed in this study.
Regulations dealing with control of volatile organic emissions have been a major
motivating factor behind implementation of the product substitution alternatives
described above. As stricter air quality standards and the costs for waste disposal
escalate, the use of these substitutes will increase. Additional implementation will
also occur as these products become proven in use and gain public acceptance. Since
they all offer attractive economic incentives even when the effects of waste
generation are ignored, their share of the product coatings market should increase
steadily.
B21-19
-------
11. CONCLUSIONS
While the product coating industry has done much to reduce the amount of potential
waste it could generate, it appears that further waste generation reductions are
possible in the range from 28 to 43 percent. Several methods that generally appear to
be most effective would be to emphasize proper cleaning techniques, provide better
operator training, and install paint application systems with low overspray.
Three practiced product substitution alternatives were identified: water based
coatings, radiation curable coatings and powder coatings.
12. REFERENCES
Anonymous. 1982. Cryogenic paint stripping. Prod. Finish. December 1982.
Brewer, G. 1980. Calculations of painting wasteloads associated with metal finishing.
Cincinnati, Ohio.: U.S. Environmental Protection Agency.
Campbell, M.E., and Glenn, W.M. 1982. Profit from pollution prevention. A guide to
industrial waste reduction and recycling. Toronto, Canada: The Pollution Probe
Foundation.
Church, F.L. ed. 1984. Powder coatings blitz keeps on rolling along. Modern Metals.
November 1984.
•
1984b. Powder coating sales reach near-boom levels. Modern Metals.
January 1984.
Cole, G.E. 1984. VOC emission reductions and other benefits achieved by major
powder coating operations. Paper No. 84-38.1 presented at the Air Pollution Control
Association Meeting on June 25, 1984.
Durney, L.J. 1982. How to improve your paint stripping. Prod. Finish. December
1982.
Forth, Karl, assoc. ed. 1985. Stripped clean and dry. Aviation Equipment Maintenence
Oct. 1985.
Higgins, T.E. et al 1985. CH2M Hill Evaluation of industrial process modifications to
reduce hazardous wastes in the armed services. Paper presented at the 40th Annual
Purdue Industrial Waste Conference at Purdue University on May 14-15, 1985.
Huisingh, D., Martin, L. et al. 1985. Proven profit from pollution prevention.
Washington, D.C. The Institute for Local Self-Reliance.
Lenchkus, D. 1982. Increasing productivity. Finishing Wood and Wood Products
Magazine. 87(4). Chicago, 111.: Vance Publishing Corp.
B2J-20
-------
Letsky, B.M. 1960. A practical manual of industrial finishes. New York, N.Y.:
Reinhold Publ. Corp.
Payne, H.F. 1961. Organic coating technology. 2 vols. New York, N.Y.: Wiley.
Staebler, Jr. C.J., and Simpers, B.F. 1983 U.S. Environmental Protection Agency.
Industrial Environmental Research Laboratory. Reduced-pollution corrosin-protection
systems. EPA-600/S2-83-003 Cincinnati, OH. U.S. Environmental Protection Agency.
Webber, D. 1984. Coating industry heading for record year. Chem. Enq. News.
62(40): 51.
13. INDUSTRY CONTACTS
Joseph P. Chu, Environmental Activities Staff, General Motors Technical Center,
Warren, MI.
Walter G. Vaux, Process and Chemical Engineering, Westinghouse Electric Corp.,
Pittsburgh, PA.
B2J-21
-------
-------
PROCESS EQUIPMENT CLEANING
-------
-------
1. INTRODUCTION
Equipment cleaning is a periodic preventive maintenance function performed in order to
remove an undesirable deposit (dirt, rust, scale, clingage, tars, etc.) from the internal
surfaces of process equipment. Generally, equipment cleaning is performed for the
following purposes:
Maintain or restore the operating efficiency of the equipment.
Extend the lifetime of the equipment by minimizing corrosion.
Remove deposits to allow for inspection or repair of the equipment.
Avoid or limit product contamination.
Equipment cleaning is widely practiced in all segments of the manufacturing industries.
However, from the waste generation viewpoint, chemical processing and compounding
operations are of special significance, due to the nature and amounts of cleaning wastes
generated annually.
2. DESCRIPTION OF CLEANING TECHNIQUES
Usually, a distinction is made between chemical and mechanical cleaning. Chemical
cleaning requires the use of chemicals, such as those listed in Table 2-1, which are
employed to chemically attack the deposits and render them either water-or solvent-
soluble. The basic removal mechanism is usually through oxidation, reduction, chelation,
or conversion of insoluble oxides into soluble salts. Physical or mechanical cleaning relies
on breaking the adhesion of the deposit to a surface through the use of mechanical devices
such as scrapers, squeegees, rags, drag-lines, pigs, or through the use of high pressure
liquid jetting or hydroblasting.
While other classifications of cleaning techniques are possible, the classification proposed
by Loucks (1973) was adopted, resulting in the following six categories of cleaning
techniques:
"Fill and empty" technique
"Circulation" technique
B22-]
-------
TABLE 2-1. SOME Cl IEMICAL CLEANING COMPOUNDS AND THEIR USAGE
Cleaning Compound
Chemical Action
Usage
Remarks
en
N)
NJ
Hydrochloric Acid
Sulfuric Acid
Nitric Acid
Hydrofluoric Acid
Sulfamic Acid
Citric Acid
Caustic Soda, Soda Ash
Ammonia
flthylene Diamine
Tetra-Acetate (EDTA)
Dissolves most water scales
and corrosion products
Dissolves most corrosion
products
Same as HCI
Dissolves silicate deposits
Dissolves calcium salts
Dissolves iron oxides
Dissolves oil and grease
Forms soluble complexes with
copper ions
Dissolves water scales at
alkaline pi Is
Used on boilers, heat
exchangers, pipelines, etc.
Limited use
Used for stainless steel
and aluminum
Used as an additive to HCI
(as ammonium bifluoride)
Used as an additive to HCI
Used mostly to clean boilers
Frequently with added ammonia
and oxidizers
Used to remove oil and grease
before acid cleaning and to
neutralize the acid after
cleaning
Used to remove copper from
large boilers
Corrosive to steel; tempera-
tures must be below 175°F
Cannot remove water
scales
Cannot be used for copper
and ferrous alloys
Very dangerous to handle
Easy to handle; soluble
calcium salts
Not good for water scale
removal
Dangerous to handle
Needs to be handled
carefully
Used for cleaning water systems Expensive
without shutdown
Source: Roosting Capacities with Chemicals (I oucks 1973), Handbook of Industrial Water Conditioning (Retz 1900).
-------
"Flow-over" technique
"Gas Propel" technique
"Process Simulation" technique
"Onstream Cleaning" technique.
These techniques are briefly described below.
2.1 "Fill and Empty" Technique
In this method, the process vessel is isolated from other equipment and filled with the
appropriate cleaning medium. The solution can be heated and agitated to facilitate scale
removal. After a period of time (usually four-to-eight hours for chemical cleaning), the
solution is drained. Rinse water or diluted alkali or acid solutions can then be used to
ensure complete removal of the cleaning chemical. The drained chemicals and the
subsequent rinses are either reused, treated, recycled, or land-filled, depending on their
composition and on the availability of options at the particular site. The drawbacks
include the necessity to maintain large quantities of cleaning solution and relatively high
equipment downtime.
2.2 "Circulation" Technique
Circulating or flow-through cleaning is usually performed on small vessels or equipment,
such as heat exchangers. The vessel is filled with cleaning solution to an overflow line
and allowed to stand for a short period of time. Fresh cleaning solution is then pumped in,
while the used solution is displaced and routed to waste-disposal or returned to the pump
to be recirculated. In large boilers, nitrogen gas is sometimes used to provide agitation
for more effective scale removal.
2.3 "Flow-over" Technique
For larger vessels and storage tanks, cleaning by filling or recirculating requires too
much cleaning solution. This equipment is cleaned by applying the cleaning liquid with
spray jets and allowing the liquid to flow off.
B22-3
-------
2.4 "Gas-Propel" Technique
For cleaning agents that are not very corrosive steam can be used to propel them through
the system. This technique is useful for pipelines, where fill-and-empty or circulating
methods are less efficient. Here, inhibited organic acids or chelants are put in a flow of
steam which carries the liquid drops and solid debris through the hydraulic obstacles of
the system. This method is quick, efficient, and requires a minimum quantity of
materials. At the end of the cleaning process, the system is dried by stopping the steam
flow and blowing through with dry air or gas. This method can be used to clean large
vessels when flow-over methods are not suitable. Sometimes, surfactants and air can be
used with cleaning agents to make a low-density foam. This method is used mostly for
cleaning the tube sides of large surface condensers.
2.5 "Process Simulation" Technique
For equipment that is easily fouled, two or more process units are usually provided in
parallel. Such equipment is often cleaned by a simulated operation. Here, the unit is
isolated from the process and the process stream is replaced by a cleaning fluid, with the
system's pumps and temperature control still in operation. An example of this operation
is the removal of iron oxides and copper from high-pressure steam generators using
ammoniated ethylene-diamine-tetra-acetate (EDTA).
2.6 "Onstream Cleaning" Technique
The ideal way to clean any equipment is to do so when it is in normal operation. Chelants
such as EDTA, when added to the process stream, can dissolve deposits. This technique is
used for cleaning reactor jackets, large gas-compressor-station engines, closed circuit
heating and air conditioning systems, etc. In-service cleaning of large circulating water
cooling systems is done through intermittent pH control - to the acid side of neutral and
back again.
On-stream cleaning is also accomplished with mechanical devices, e.g., rod-out of process
connections with ram valves, or heat exchanger tube cleaning using moveable tube insert
brushes propelled by process fluid.
B22-4
-------
3. WASTE GENERATION ASPECTS OF EQUIPMENT CLEANING
The composition and quantity of cleaning wastes differ widely depending on:
Type of deposit being cleaned.
Type of cleaning fluid.
Type of cleaning method.
Size of equipment being cleaned.
Cleaning frequency.
In general, batch operations generate more cleanup waste than continuous processes. This
is mainly because of the higher cleanup frequency associated with utilizing the same
equipment (typically a mix tank or an autoclave), for batch production of differing
materials. For example, in plasticizer manufacturing, the same jacketed kettle reactor is
often used for different esterification reactions to produce a large variety of esters of
phtallic, trimellitic, adipic, sebacic, stearic or oleic acids. The changeover from
production of one type of chemical to another requires cleaning of the reactor internals.
By contrast, equipment used in the continuous processes (e.g., vinyl chloride monomer
manufacture), is cleaned once every one or two years during a scheduled plant
maintenance shutdown period. »
In batch operations, such as in paint manufacturing, food processing, or fermentation,
where cleanup frequency is high and the cleaning solution inventory is large, it is a
standard practice to provide "clean-in-place" (CIP) systems. Such systems consist of
stationary spray head manifolds inside the production vessels, effluent piping, tankage,
heat exchangers, pumps, solids separation equipment, distribution systems for solution re-
use, and chemical dosing units. Minimization of cleaning solution use through counter-
current staging of rinse operations is especially attractive for large integrated CIP
systems.
Waste generation by equipment cleaning operations generally can be addressed by
examining cleaning operations performed on various process equipment. The following
sections deal with the cleaning of specific process equipment.
B22-5
-------
3.1 Batch Reactors and Mixers
Batch reactors or mixers are cleaned to remove material that adheres to the vessel walls
(clingage). This is done to ensure product purity of the subsequent batch and to prevent
clogging of the outlet connections. Clingage problems dominate in the industries where
batch polymerization is performed (plastic, resins, synthetic rubber), in the paint
manufacturing industry (where equipment cleanup waste associated with clingage removal
is a major waste stream), in paint application throughout the industrial sector, and in
specialty chemicals manufacturing, where batch processing is prevalent.
3.2 Tanks and Containers
Tanks and containers are used for storage of process inventories and must be periodically
cleaned to prevent accumulation of contaminants or to remove sedimented solids. Tank
cleaning is typically performed using the flow-over method, because of the large surface
areas involved. Small vessel cleaning is done using circulation or fill and empty
techniques.
3.3 Heat Transfer Equipment
Heat exchanger fouling is a commonly encountered problem caused by the formation of
deposits on heat transfer surfaces. These deposits decrease heat transfer efficiency and
increase pumping energy requirements. Deposits form through precipitation of insoluble
substances which adhere to the exposed heat exchange surfaces. The rate of deposit
formation depends on the nature of the fluids, film temperatures, turbulence, and the
smoothness of the metal surface. Fouling deposits are removed periodically to restore
heat transfer efficiency, decrease hydraulic resistance, and, in some cases, circumvent
corrosion. Cleaning is usually done by one of two methods - hydroblasting or chemical
cleaning (VanMatre, 1977).
Hydroblasting uses high pressure water jets to remove water scale deposits. This cleaning
procedure involves the dismantling of the heat exchanger. The shell side scales are
removed by using the "bundle blaster", which is usually equipped with cleaning heads that
rotate 90 degrees. The tubeside scales are removed by using the "lancing machine", which
employs stabbing devices that drive in and out of the tubes. The advantages of
B22-6
-------
hydroblasting are that the results are easily inspected and corrected, if necessary. The
disadvantages of this technique are that the equipment is out of service for a long period
of time, and that dismantling involves high labor costs.
Chemical cleaning involves the use of chemicals, as discussed in Section 2, mainly to
remove water scale deposits. In some instances, chemical cleaning can be done with the
equipment still on stream, as discussed in Section 2.6. Though on-stream cleaning is
generally not as effective as out-of-service cleaning, reduction in the equipment
downtime offers an economic incentive.
4. WASTE REDUCTION THROUGH SOURCE CONTROL
4.1 Description of Source Control Techniques
The problem of waste minimization associated with process equipment cleaning can be
addressed using two distinct approaches:
Reducing the frequency of cleanups
Reducing the quantity and/or toxicity of wastes.
«
t
The first approach relies on reducing the need for cleaning. The second approach deals
with how to reduce wasteloads associated with the cleaning process when it has to be
performed. In the following sections, both approaches are discussed.
4.1.1 Reduction of Cleanup Frequency
Reduction (or elimination) of the need for cleaning is accomplished by identifying the
causes of undesirable deposit formation, followed by identifying and implementing the
means to prevent or limit deposit formation. Discussion is provided below of the
following measures:
o Proper plant and equipment design
o Recognition of cleaning costs as a separate item
o Conversion from batch to continuous process
o Maximizing dedication of process equipment
o Avoidance of unnecessary cleaning
o Inhibition of fouling deposit formation rate.
R22-7
-------
Identified case examples are provided where applicable.
o Proper plant and equipment design.
Cleaning considerations are, (or should be), an integral part of the plant and
equipment design. Typically, these considerations encompass avoidance of un-
drainable pockets and minimization of equipment or piping surfaces exposed to
the process fluid.
o Recognition of cleaning as a separate item by management.
Often the costs of cleaning are lumped together with other maintenance costs,
thus losing their identity and visibility to management. Without a proper
identification of the overall magnitude of these costs and the various
individual cleaning cost elements, the analysis of alternatives to frequent
cleaning cannot be meaningfully performed.
o Conversion from batch to continuous process.
Such a conversion, where possible, will reduce cleaning wastes for the reasons
discussed in Section 3. The additional benefits of the continuous process
include lower labor requirements, ease of automation and control (which
minimizes human error leading to inferior quality products), and the
elimination of the need for manual material transfer operations characterized
by a high probability of spillage.
Historically, the changeover from batch (or cyclic) to continuous operations
has been very common in the chemical industry because of technological
progress and increased product demand (Peters and Timmerhaus 1980).
However, in certain cases, batch operations have proven advantageous over
continuous processing schemes. Here, the examples include the development
(but not the widespread use) of continuous processes for neoprene rubber
manufacture (Anon. 1965, Euleco 1975) and for phenolic resins manufacture
(Aho 1958, SIR 1968). It must be noted that the choice between batch and
continuous operations is governed primarily by other factors than by waste
generation. Common considerations include product diversity at the site,
production volume, capital equipment cost, and operating cost.
822-8
-------
Maximizing dedication of process equipment.
Making the process equipment more dedicated to a single product line or a
single process function will generally reduce the need for frequent cleaning.
This is typically accomplished through careful scheduling of production or by
providing more specialized equipment. Both routes are extensively practiced
in industries such as paint manufacturing and specialty chemicals
manufacturing. Producers of specialty chemicals and formulations generally
use batch processes with a number of similar products made in the same
equipment. By proper scheduling, entire year supplies of individual products
can be made in one campaign minimizing the numbers of changeovers and the
down-times involved. Generally, the volumes of the specialty products
produced are too low to justify continuous process equipment. Individual plant
production volumes for such specialty items may range from 10 to 300 tons per
year and volumes produced per batch may range from 1 to 10 tons.
Generally, equipment utilization strategies and the resulting production
schedules for batch processing are a product of optimization analysis, which
maximizes the production rate with due consideration of such constraints as
equipment cost (size) and availability, equipment downtime, labor cost and
availability, length of operation(s), ancillary equipment availability, cost of
cleanup, and cost of raw material losses. General descriptions of optimum
strategy formulation are available (Peters and Timmerhaus 1980). Based on
interviews with people who have had experience in the preparation of
production schedules, formal optimization methodologies are not invoked
often. Rather, a "common sense" approach based on trial-and-error is more
widely used.
Avoidance of unnecessary cleaning
At times, equipment cleaning is performed routinely with little consideration
for the rationale. In one actual case, a ball mill was used once every week to
wet-grind a certain powder. The ball mill, with corrosion-proof internals, was
totally dedicated to a single powder formulation, yet for no apparent reason, it
was cleaned after each batch. The only justification provided was that other
non-dedicated ball mills at the facility were cleaned after every use. In
another example from the food processing industry, cleaning of the equipment,
R22-9
-------
in some instances, can be avoided, if the equipment can be adequately
sanitized by heating.
Inhibition of fouling deposit buildup rate
Fouling rates have been extensively studied and are usually attributed to the
following processes (or their combination):
Crystallization.
Sedimentation.
Chemical reactions and polymerization.
High temperature coking.
Corrosion.
Bacterial growth.
Inhibition of fouling is of particular importance to heat transfer, where fouling
associated with tars and scale formation is frequently experienced. Inhibition
(slowing down) of buildup formation rates will reduce the cleaning frequency
while improving the overall operating efficiency (i.e., heat transfer rate and
hydraulic resistance). In general, the fouling deposit formation can be slowed
down by the following measures:
Use of smooth heat transfer surfaces.
Lower film temperatures or increased turbulence.
Control of stream composition.
Choice of heat exchanger type.
The use of smooth heat transfer surfaces relies on reducing the adhesion
between the deposit precursor and the heat transfer surface. The use of
electropolished stainless steel tubes in a black liquor forced-circulation
evaporator (used in the pulp and paper industry), resulted in a dramatic
reduction of cleaning frequency from a period of once-a-week to once-a-year
(Uddeholm Co., Tubec Tubes Brochure). Smooth non-stick surfaces can also be
provided by using Teflon*. Complete Teflon heat exchanger designs are
available (DuPont 1985). The additional advantage of Teflon tubes is their
ease of cleaning and corrosion resistance.
•^Registered trademark of E.I. Du Pont Company.
B22-10
-------
For the heat exchangers, the fouling rate can be inhibited by lower film
temperatures through the use of additional surface area and/or increased
turbulence (high velocities). Avoidance of excessive overdesign and
maintenance of sufficient velocity (e.g. through recirculation) during turndown
operation will lower fouling rate and resulting cleaning frequency of heat
exchangers (Garrett-Price 1985).
Inhibition of fouling rates can also be accomplished through prior removal of
deposit precursors. Such inhibition can be achieved by early .addition of
polymerization inhibitors in monomeric resin manufacture. Other cleanup
frequency reduction measures exist, such as application of less corrosive and
more thermally stable heat transfer fluids and better design and/or control of
fired heaters.
For closed cooling water systems, the fouling rate can be inhibited by proper
water treatment, which includes pH control, hardness removal and dosing with
biocide and corrosion inhibitors. Additionally, fouling rates can be controlled
by lowering the number of concentration cycles in the cooling tower and by
using make-up water with low total solids content. In certain applications,
e.g., heating or cooling of slurries where deposit buildup due to sedimentation
can be significant, the choice of heat exchanger type becomes important. For
example, in fermentation plants, spiral plate heat exchangers are often used in
preference to the shell-and-tube design because of the spiral plate exchanger's
lower pressure drop and more favorable solids re-suspension, or non-plugging
characteristics.
Slowing down of the deposit buildup rate is not only limited to heat transfer
equipment, but is also important in crude oil storage tanks and crude desalting
units. Crude oil's exposure to air can cause formation of gums and resins
during long exposure periods inside storage tanks. Use of nitrogen blanketing
or floating roof tanks has been suggested as a means of minimizing deposit
formation due to air oxidation (Jacobs 1979), along with the removal of
sludge-forming impurities prior to storage. Similarly, in the manufacturing
exposure to air causes the formation of solid films that strongly adhere to the
internal surface of the mixers. This can be avoided by using closed storage
B22-11
-------
and transfer systems . For example, Ford Motor Company modified their paint
storage and transfer system to be totally enclosed with full recirculation,
resulting in less frequent and easier cleanups and improvement in paint quality
(Colleta and Bowers 1981). Other applications include coating of the reactor
internals with special chemicals to prevent scale formation. These practices
were reported in the suspension polymerization process for polyvinyl chloride
manufacture (Cameron et. al 1980) .
4.1.2 Reducing the Quantity and Toxicity of Cleanup Waste
When equipment must be cleaned, the cleanup should be performed efficiently with a
minimum generation of hazardous waste. Generally, the following considerations should
be emphasized:
o The amount of clingage (or deposit) should be minimized.
o The amount of cleaning solution should be minimized, especially if it is
expensive and/or hazardous.
o The choice of the cleaning medium should favor possible re-use or recycle, or
result in the least toxic waste.
o Disposal of cleaning waste must be done in an environmentally acceptable
fashion.
The sections below address each of the above aspects, except the last item
(environmentally acceptable disposal), which is outside the scope of this study. Case
examples found in the reviewed literature are also provided.
o Minimizing clingage
The importance of minimizing clingage can be illustrated by waste generation
in the paint manufacturing industry. Following pigment dispersion, a batch of
paint is emptied from the mixer tank, leaving behind a residual layer (clingage)
which adheres to the interior surface. The tank can then be cleaned using a
caustic rinse and the resulting solution allowed to settle. The sludge is then
removed and sent to land disposal with or without additional dewatering. It is
clear that the ultimate quantity of sludge generated depends on how much
residual paint was left in the tank. To reduce clingage, the following measures
should be considered:
B22-12
-------
Provision of adequate drainage time.
Provision of non-stick surfaces, e.g., Teflon.
Use of mechanical wall wipers (dual shaft mixers).
Use of manual wall wipers (squeegees).
Choice of cylindrical tanks with height-to-diameter ratio close to 1 in
order to minimize wetted surface.
Maximizing batch size.
All of the above measures are self-explanatory and do not require elaboration.
All are practiced extensively throughout industry. The use of tank linings as
means of reducing adherence and preventing corrosion has been addressed
(Zolin 1984, Kays 1977). Use of dual shaft mixers with a slow scraper blade
wiping the walls and the bottom of the mixing tank is common in applications
involving viscous liquids*. Manual wipers are widely used in the cleaning of
smaller portable tanks in the paint industry (see the process study elsewhere in
this appendix on Paint Manufacturing).
Minimizing the amount of cleaning solution.
The use of chemical cleaning (e.g., with mineral and organic acids) results in
appreciable quantities of hazardous cleaning wastes which should be treated
prior to land disposal. As appropriate treatment facilities are not readily
available in every case, mechanical cleaning such as hydroblasting appears
preferable over chemical cleaning. When chemical cleaning has to be used (for
example, in cleaning of water scales from heat transfer surfaces), it is
important to give preference to "flow-over" methods which require less
solution than "fill-and-empty" methods (See section 2.)
When cleaning chemicals are used, there are generally four parameters that
must be controlled: time, temperature, concentration, and turbulence. By
using higher values of these parameters, the amount of cleaning chemicals can
be reduced.
Myers Mixing Company, Montebello, CA 1985: Personal communication.
B22-13
-------
Adding proper additives to the cleaning agents will also result in a more
efficient cleanup. By adding wetting agents, the cleaning chemicals can
penetrate the scales more easily. Defoamers can be added to prevent the
spillage of soapy bubbles, which form due to the presence of wetting
agents. Suspending agents and emulsifiers can be added to prevent the
settling of the soil and non diffusing liquids, respectively.
The use of clean-in-place (CIP) systems generally reduces the usage of the
cleaning medium. This is especially effective when coupled with high
velocity automated spray manifolds, and staged or counter-current rinsing.
According to one source (USEPA 1979), an 80-90 percent reduction in
aqueous solution use was achieved by installation of high pressure spray
nozzles for tank rinsing in the paint industry.
In another example, a considerable reduction of toxicity of the reactor
cleanup waste in phenolic resin manufacture was achieved by Borden
Chemical (Huisingh et. al. 1985), where a two-step rinse sequence was
employed. The first rinse was done with a small amount of water and
therefore generated a concentrated stream which could be recycled to the
process. The second, full volume rinse, generated wastewater with a much
lower content of toxic material then the original method.
The ultimate approach to cutting down on the use of cleaning solutions is
their elimination altogether. This is done by cleaning the equipment on-
stream using mechanical devices. One example is provided by the use of
"pigs" (contoured plugs propelled by the process liquid) to clean pipelines.
Another example is a commercial system which uses steel brushes fitted
inside heat exchanger tubes which are propelled by process fluid and
reversed periodically by a flow divertor (WSA 1985).
Choice of cleaning medium and cleaning solution reuse.
From the waste generation standpoint, the hierarchy of cleaning medium
choices appears to be process fluid over water over chemical solutions.
Process fluids can be and are used to clean deposits. This choice is favored
by the fact that following a simple filtration to separate out the solids, the
B22-14
-------
cleaning solution can be re-used as part of the formulation or process make-up
stream. Examples include re-use of rinsewater from latex tank cleaning as
part of latex formulation (Riley 1974), and the use of warm oil cleaning of
crude oil storage tanks (Barnett 1980). The last example deserves more
description. The sludge sediment in the crude oil storage tanks contains large
quantities of entrapped oil (60 to 90 percent oil by volume). In the application
described, a light gas oil, clean light crude stock, or other available low
viscosity straight run distillate is warmed, mixed with dispersant additive,
circulated through a tank to re-suspend the sludge and dissolve the oil, and is
then sent to the refinery's slop oil recovery system where the oil is recovered
and solids are efficiently separated. Previously used cleaning procedure relied
on direct mechanical removal of sludge from the tank. Whereby large
quantities of oil were lost and much higher waste quantities were produced.
4.2 Summary and Assessment of Source Control Techniques.
The techniques described in this report are summarized in Table 4-1. The compendium of
methods presented, of course, is not complete by any means. Apart from its possible use
as a guideline, it serves mainly to categorize various approaches and provide descriptions
of and references to their known applications. To assess the overall potential and extent
of possible future reductions of wastes generated in cleaning operations, along with the
current extent of waste reduction achieved, each of the methods listed in Table 4-1 was
rated on an integer scale of 0 (lowest) to 4 (highest) in the following three categories:
Effectiveness in reducing waste.
Extent of current use in the chemical processing industry.
Future application potential.
Based on the ratings, the current and future waste reduction extents were computed,
using the methodology described in the introduction to this appendix. The results are
presented in Table 4-2.
5. CONCLUSIONS
It appears that waste reductions on the order of 18 to 35 percent can be achieved in the
area of process equipment cleaning. This is shown by the future waste reduction index of
B22-15
-------
Table 4-1. Methods to Reduce Generation of
Waste from Process Equipment Cleaning Operations
Approach
Method
Reduction of Cleaning
Frequency
Proper plant and equipment design
Recognition by management of cleaning costs as a separate
item
Conversion of process from batch to continuous or semi-
continuous.
Maximizing dedication of process equipment.
Avoidance of unnecessary cleaning.
Inhibition of fouling deposit formation through:
Use of smooth heat transfer surfaces (electropolished
or Teflon tubes)
Lower film temperatures or increased turbulence.
Control of stream composition.
Choice of heat exchanger type.
Reduction of Quantity
and/or Toxicity of
Cleanup Wastes
Minimizing clingage through:
Provision of adequate batch drainage time.
Provision of non-stick surface.
Use of mechanical or manual wall wipers.
Choice of cylindrical tanks with low height-to-
diameter ratio.
Agitator rotation after batch dump.
Maximizing batch size.
Minimizing the amount of cleaning solution through:
Use of high pressure spray nozzles.
Use of "flow-over" technique.
Use of on-stream mechanical cleaning.
Use of clean-in-place systems with staged or counter-
current rinsing.
Choice of cleaning medium and cleaning solution reuse
- Process fluid over hydroblasting over chemical
cleaning.
Reuse of cleaning fluid.
R22-J6
-------
TABLE 4-2 SUBWRY OF SOURCE CONTROL HETHODOLOGY FOR PROCESS EQUIPHENT CLEANING
1 i Found Docufflentatiar ,' Waste 1 Extent of 1 Future I Fraction of | Current I Future Reduction Index
I AnDrna^h 1 rnntpnl MofhnHnlnnw 1 1 o A »• ' 1 r i 'i
i Application
T 1 M i n j • i
! 1 1 Quantity Quality effectiveness 1 Potential 1 Index | Probable
| Reduce Cleaning |1. Proper plant and equipment design | 1 | 21 2 i 2 1
1 Frequency |2 Cleaning cost as a separate item | 1 | 2 ( 3 | 1 | 2
I |3. Convert batch to continuous process j 2 1 3 j 3 ! 1
I (4 Maximize dedication of equipment ! 1 2 | 3 I
|5 Avoid unnecessary cleaning | 1 | 2 | 4 i
1 16 Use smooth heat transfer surfaces | 2 I 2 3 !
1 |7 Lower film tenp & increase turbulence| 2 I 3 I 2 i
|8 Control stream composition | 3 | 3 | 3 |
1 |9. Choice of heat exchanger type | 1 | 2 | 21
1 + + t
1 ! Overall | t 56 ( 2 11 | 2 73 ! 23
t 4 + + t j
| Reduce Quantity (f. Adequate batch drainage time I 0 2 1 I
and/or Toxicity 12 Use non-stick surfaces ( 1 | 1 | 2 (
I of Cleanup Kastes|3. Use mechanical/manual wall wipers | 2 | 2 | 3 |
14 Use tanks with low I/O ratios | 0 I 2 ( 2 (
2
3
1 2
1 2
3
I 3
1 2.11
! 2
I 2
i 3
1 1
|5 Agitator rotation | 1 1 1 | 21 2
16 Maximize batch size I 0 I 2 | 2 I 3 | 2
1 |7. Use high pressure spray nozzles | 21 2 ( 3 f 21 3
|8 Use "flow-over" techniques 1 1 | 2 | 3 | 21 1
|9 Use on-stream mechanical cleaning | 2 | 2 3 | 1 | 3
110 Use CIP systeu with staged/CCO rinses| 2 I 2 ( 3 | 2 | 2
(11. Proper cleaning medium choice 1 1 | 2 i 3 | 2 2
(12 Reuse of cleaning fluids I 1 2 4 | 1 | 2
). .f +
1.0 | 0.3
I 0 8 I 1.1
I 2.3 | 0.2
I 1.5 I 0.8
3.0 i 03
1 0.8 | 1.1
1 15| 0.3
| 2.3 I 0.6
15| 0.4
-_ — — — 4 ___— +_..- _--
0.50 | 3.0 | 0.6
~ 4 _- 4—- _---
1 0.3 | 0.1
| 0.5 | 0.8
I 1.5 | 1.1
( 1.0 | 0.3
I 0.5 | 0.3
I 1 5 | 0.3
1 '.5 | 11
1 1.5 | 0.4
| 0.8 I 1.7
I 1.5 | 0.8
Maximum
1.1
1 1
-•— __-- —
t.i
*-- __-
1 7
| 1.5 | 08
1.0 | 1.5 |
+ 4 4 4 4 4 1 1 1 1 *
1 Overall | 1.08 | 1 83 2 50 | 1.92 1 2.08 | 0.50 | 1 5 | 0.7 | 1.7
i 4 , 4 4. i A _ i I ± 1
T T - * f + 4 + _ 4-____ ____ 4 ». _ ^ -._— .-..-4 ____ _ _ — .
All Sources | All Methods | 1 00 | 2.6 | 0.7 | 1.4
jO
r-o
ro
i
-------
0.7 to 1.4 on scale of 0 to 4. However, the current reduction index of 2.6 is indicative
that significant waste reduction has already been achieved, due to partial implementation
of the techniques summarized in Table 4-2.
Of the 19 techniques listed, the use of smooth heat exchanger surfaces, on-stream
mechanical cleaning and cleaning fluid re-use appear to be most attractive for future
applications as evidenced by high future reduction indices.
6. REFERENCES
AESI. 1981. American Electroplaters Society, Inc. 3rd Conference on advanced pollution
control for the metal finishing industry. Held at Orlando Hyatt House, Kissimmee,
Florida on April 14 - 16, 1980. EPA-600-2-81-028. Cincinnati, Ohio; U.S. Environmental
Protection Agency.
Aho, C.E. 1958. U.S. Pat. 2,831,842 (Apr. 22, 1958) to E.I. du Pont de Nemours & Co., Inc.
Anonymous. 1965. Chemical week 97(19):35-40.
Betz Laboratories Inc. 1980. Handbook of industrial water conditioning, 8th ed., Trevose,
PA.
Brode, G.L., 1982. Phenolic resins. Kirk Othmer encyclopedia of chemical technology, 3rd
ed. Vol. 17, pp. 384-416
Cameron, J.B., Lundeen, A.J., and McCulley, Jr., J.H. 1980. Hydrocarbon processing
59(3):39-50
Campbell, M.E., and Glenn, W.M., 1983. Profit from pollution prevention: a guide to
industrial waste reduction and recycling. Toronto, Canada: The Pollution Probe
Foundation.
Colletta, V., and Powers, J. 1981. Chemical processing 44(4):20-1.
Du Pont, 1985. Chemical processing,48(6);42-3.
Euleco, 1975. Euleco continuous process: technical bulletin. Milano, Italy: Euleco,
S.P.A.
Edwards, V. 1982. Chemical processing, 45(13):137. '
Garrett-Price, B.A., Watts, R.L., Smith, S.A., et al. 1985. Fouling of heat exchangers,
characteristics, costs, prevention, control and removal. Park Ridge, N.J.: Noyes
Publications.
Huisingh, D., et al. 1985. Proven profit from pollution prevention. Washington, D.C.:
The Institute for Local Self-Reliance.
B22-18
-------
Hoblack, R., and Kawlor, L. 1985. Chemical processing, 48(8):84-6.
Jacobs Engineering, 1979. Alternatives for hazardous waste management in the
petroleum refining industry. EPA-530-SW-172C. Washington, D.C.: U.S. Environmental
Protection Agency.
J'Ardene, R. and Model, A.E. 1985. Chemical processing, 48(l):26-7
Kays, W.B. 1977. Construction of lining for reservoirs, tanks and pollution control
facilities. New York, N.Y.: Wiley.
Loucks, C.M., 1973, Boosting capacities with chemicals, Chemical engineering deskbook
issue, 80(5):79-84, McGraw Hill Co.
3M. 1985. 3M Corporation brochure: "How to initiate a '3P'-type program in combustion
engineering".
McGrath, J.E. et al. 1979. Elastomers, synthetic. In Kirk-Othmer encyclopedia of
chemical technology. 3rd ed. Vol. 8, pp. 446-640. New York, NY: Wiley.
Mrozineri, W., Short, A.J., and Bachetti, J.A. 1983. Chemical processing, 46(1):30-31.
Perry, R.JH., and Chilton, C.H., eds. 1973. Chemical engineers' handbook. 5th ed.
Chapter 19. New York, NY: McGraw Hill Book Co.
Peters, M.E. and Timmerhaus, K.D., 1980. Plant design and economics for chemical
engineers, 3rd edition, McGraw-Hill Book Co.
Requena, J., and Gaines, A. 1984. Chemical processing, 47(2):26-7.
Riley, J.E. 1974. Development document for efficient limitation guidelines and new
source performance standards for the tire and synthetic segment of the rubber processing,
point source category. EPA-440-l-74-013a. Washington, D.C.: U.S. Environmental
Protection Agency.
Roebuck, A.H. 1978. Chemical engineering 85(17):107-10.
Sandor, B. 1985. Chemical processing, 48(l):26-7.
SIR. 1968. Brit. Pat. 1,136,189 (Dec. 11, 1968) issued to Shell International Research.
Szostak, R.J., and Toy, D.A. 1985. Chemical processing, 48(10):44-5.
Talcott, R.M., Willis, C., and Freeman, M.P. 1980. Filtration in Kirk-dthmer
encyclopedia of chemical technology. 3rd ed. Vol. 10, pp. 284-337. New York, NY:
Wiley.
Uddeholm Corporation (Sweden), Technical Brochure on Tubec Tubes.
USEPA. 1976. U.S. Environmental Protection Agency, Office of Water and Waste
Management. Development document for final effluent limitation guidelines for the
pesticide chemicals manufacturing industry. EPA-440-l-73-060d. Washington D.C.: U.S.
Environmental Protection Agency.
B22-19
-------
. 1979. U. S. Environmental Protection Agency, Office of Water and Waste
Management. Development document for proposed effluent guidelines, new source
performance standards, and pretreatment standard for the paint formulating, point source
category. EPA-440-l-79-049b. Washington D.C.: U.S. Environmental Protection Agency.
VanMatre, J. 1977. Hydrocarbon processing, 56(7):115-117.
WAPORA, Inc. 1975. Assessment of industrial hazardous waste practice, paint and allied
products industry, contract solvent reclaiming operations, and factory application of
coatings. PB-251669. Washington, D.C.: U.S. Environmental Protection Agency.
Wilkes, T., and Hodel, A.E. 1984. Chemical processing, 47(9):38-9.
Wilkins, G.E. 1977. Industrial process profiles for environmental uses; chapter 10,
plastics and resin industry. Radian Corp. EPA-600-2-77-023. Cincinnati, Ohio: U.S.
Environmental Protection Agency.
WSA 1985. Water Services of America, Inc., Chemical engineering progress. 81(7):7.
Zolin, B.I. 1984. Chemical processing, 47(9):63-5.
B22-20
-------