-------
EPA/600/R-95/070
May 1995
POLLUTION PREVENTION POSSIBILITIES
FOR SMALL AND MEDIUM-SIZED INDUSTRIES
Results of the WRITE Projects
by
Ivars J. Licis
Pollution Prevention Research Branch
Risk Reduction Engineering Laboratory
Cincinnati, OH 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH 45224
Printed on Recycled Paper
-------
FOREWORD
This report is a compilation of 41 pollution prevention projects carried out under the WRITF
s Th±naded ^ P6rS°nS ^^ ^ idertifyinfl 3nd '^"emerrting ^iK^o^
ies. The reader is encouraged to contact the EPA project officers for more information.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
-------
CONTENTS
DISCLAIMER «
FOREWORD iii
ABSTRACT iv
ACKNOWLEDGEMENTS viii
EXECUTIVE SUMMARY 1
1. INTRODUCTION 11
2. WRITE TECHNOLOGY EVALUATION SUMMARIES 14
CALIFORNIA:
#01 Watts Nickel and Rinse Water Recovery via an Advanced Reverse Osmosis System 14
#02 Evaluation of Three Oil Filter Designs for Pollution Prevention Effectiveness 17
#03 Evaluation of Five Waste Minimization Technologies at the General
Dynamics Pomona Division; Electroplating Rinse Water Reduction 20
#04 Evaluation of Five Waste Minimization Technologies at the General Dynamics
Pomona Division; Sulfuric Acid Anodizing 25
#05 Evaluation of Five Waste Minimization Technologies at the General Dynamics
Pomona Division; Robotics Painting 30
#06 Evaluation of Five Waste Minimization Technologies at the General Dynamics
Pomona Division; Bead Blast Paint Stripping 34
#07 Evaluation of Five Waste Minimization Technologies at the General Dynamics
Pomona Division; Freon Recovery 37
CONNECTICUT:
#08 An Automated Aqueous Washer for the Metal Finishing Industry 40
#09 On-Site Newspaper Ink Recycling 44
#10 Cadmium and Chromium Recovery from Electroplating Rinsewaters 48
#11 Nickel Recovery from Electroplating Rinsewater by Electrodialysis 54
#12 Chromate Recovery from Chromating Rinsewater in the Metal Finishing Industry 58
-------
-------
#36 Onsite Solvent Recovery with Low Emission Vapor Degreasing 158
ERIE COUNTY, NY:
#37 Replacement of Hazardous Material in Wide Web Flexographic Printing Process 164
#38 Ultrasonic Cleaning as a Replacement for Chlorofluorocarbon-Based System 169
#39 Removal and Containment of Lead-Based Paint via Needle Sealers 173
#40 Low-VOC Wood Furniture Coatings 178
#41 Finishing Fabricated Metal Products with Powder Coatings 179
3. REFERENCES 180
4. INDUSTRY INDEX 181
5. EPA CONTACTS AND REPORT ORDERING INFORMATION 185
VII
-------
DISCLAIMER
The information in this document has been funded wholly or in part by the United States
Environmental Protection Agency. It has been subjected to peer and administrative review, and it has
been approved for publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
-------
NEW JERSEY: U.S. EPA project officer - Johnny Springer
New Jersey Department of Environmental Protection - project officers Mohamed
Elsaady, Anthony Tomljanovic, New Jersey Department of Environmental
Protection and Energy - Norine Binder; New Jersey Institute of Technoloav -
Daniel Watts
Hosts for technology evaluations- Cook's Industrial Lubricants Inc Linden NJ-
Newark Air Force Base, OH; Pioneer Metal Finishing Company, Franklinville NJ- '
Duffy Electric and Machine Company, Chillicothe, OH; Safety-kleen Corporation
Elgin, IL: Honeywell Space Systems Group, Clearwater, FL; Tooele Army Depot'
Consolidated Maintenance Facility, Tooele, UT '
WASHINGTON: U.S. EPA project officer - Ivars Licis
State of Washington Department of Ecology - project officer Robert Burmark.
Hosts for technology evaluations - Earle M. Jorgensen Steel Company, Seattle,
WA; Titus-Will Ford, Tacoma, WA; Municipality of Metropolitan Seattle, Atlantic
Base Garage; Ellington Field, National Aeronautics and Space Administration/Lyndon
B. Johnson Space Center, Houston ,TX; Navistar International Transportation
Corporation, Plastics Division, Columbus, OH; Cooper Industries, Belden Division,
Richmond, VA; Durr Automation Inc., Davidsburgh, Ml
ERIE COUNTY, NY: U.S. EPA project officer - Paul Randall
Erie County Department of Environmental Planning - project officer Paul Kranz.
Hosts for technology Evaluations - Lustreprint Company, Erie County; Conax
Buffalo Inc., Cheektowaga, NY; New York State Thruway Authority; Dinaire
Corporation, Buffalo, NY; Diversified Control Inc., Orchard Park, NY
EPA CONTRACTORS: Battelle, Columbus OH; PEI, Cincinnati OH; SAIC, San Diego, CA
(Respective project authors are appear in Section 3, under Reports)
EPA CONTRIBUTORS: Special thanks to Lynn Apel and Harry Freeman for the design of the
WRITE program and their participation in getting it started.
Thanks to Rada Olbina for collaborating in editing the final report. Thanks also to
Debbie Hettel for typing the final draft manuscript.
IX
-------
ABSTRACT
The Waste Reduction Innovative Technology Evaluation Program (WRITE) was a pilot program
with six (6) states and one (1) local government, to identify priority needs at the respective governmental
level, find promising waste minimization technologies and perform an evaluation to determine
performance, pollution prevention (P2) impact and costs. The research concentrated on environmental
problems and (P2) opportunities for small to medium-sized businesses, technology at pilot- or full-scale,
and use of voluntary business/EPA partnerships.
A total of 41 technologies were tested and evaluated. Industries included were coating,
depainting, electronics, metal plating and finishing, printing, surface cleaning, steel, and a number of
miscellaneous categories. Many of the evaluations were able to list cost savings coupled with reduced or
eliminated waste streams. It was found that technology benefits are largely application specific. However,
favorable (or even unfavorable) findings for one application can be useful as a beginning for estimating
other beneficial uses.
IV
-------
EXECUTIVE SUMMARY
The Risk Reduction Research Laboratory conducted a pilot program with six states and one local
government to evaluate promising P2 technologies. Included were a full-scale technology demonstration
P2 impact assessment, and appraisal of relative performance and economics. A total of 41 evaluations '
were conducted. The results of 39 are summarized in terms of major effects on wastes (Table 1) and
performance and economics (Table 2). The technology and test programs are described for two '
evaluations being completed.
Most of the work focuses on improvements to activities that have traditionally produced organic
solvent and toxic metal waste. Selection of specific waste reduction technologies is a product of EPA and
state environmental organization priorities, available technologies and the existance of sites willing to host
the technology evaluations
Overall findings indicated that:
. Technologies exist that reduce pollution, do not require large capital investment, produce
equal or better performance, and feature short term payback.
. Many processes, such as surface cleaning and solvent substitution, have potential for multi-
industry applications. However, most technology performance is sensitive to the
requirements of a specific application, and needs to be evaluated for that purpose prior to
use. K
. Compliance issues for specific applications should be investigated with local authorities prior
to making technology changes. Many states are starting to incorporate multi-media P2 in
their interpretation of environmental management. The objective is to allow flexibility in order
to produce less environmental hazard overall.
. The forces regulating the adoption of new technologies, even technologies with superior
performance, favorable economics and waste reduction, are governed by other equally
important criteria that were not specifically identified nor evaluated by this effort. These
criteria are non-technological and include categories such as requirements of future
regulation, worker/customer acceptance, tradition, social custom, etc.
Economic: figures presented here are conservative. Capital investment is estimated on the
basis of choosing between the cost of new technology or keeping the status quo at no
expense, rather than the difference in capital outlay between purchasing existing vs new
technology.
. The accounting system used for calculating economics did not include longer range expenses
and penalties of missing opportunities for adopting P2 technology (future disposal costs,
waste management overhead and impact of new regulations, worker exposure equipment
and training; productivity; health benefits; liability; etc.). While more difficult to quantify, these
expenses can be of equal significance.
-------
ILLINOIS:
#13 Ink and Cleaner Waste Reduction Evaluation for Flexographic Printers 62
#14 Alkaline Noncyanide Zinc Plating and Reuse of Recovered Chemicals 65
#15 Recycling Nickel Electroplating Rinse Waters by Low Temperature Evaporation
and Reverse Osmosis 70
#16 Evaluation of Ultrafiltration to Recover Aqueous Iron Phosphating/Degreasing Bath 74
#17 Waste Evaluation of Soy-Based Ink at a Sheet-Fed Offset Printer 78
MINNESOTA:
#18 Modifications to Reduce Drag Out at a Printed Circuit Board Manufacturer 82
#19 Sponge Rollers and Flow Controller for Rinse Water Reduction at a Printed Circuit Board
Manufacturer 35
#20 Carbon-Black Dispersion Preplating Technology for Printed Wire Board Manufacturing 90
#21 Evaluation of an Electroclialytic Process for Purification of Hexavalent Chromium 94
#22 Substituting Cadmium Cyanide Electroplating with Zinc Chloride 101
NEW JERSEY:
#23 Mobile Onsite Recycling of Metalworking Fluids 106
#24 A Fluid Sorbent Recycling Device for Industrial Fluid Users 110
#25 Electronic Component Cooling Alternatives: Compressed Air and Liquid Nitrogen 114
#26 Evaluation of ZERPOL (Zero Liquid Discharge System) at Pioneer Metal Finishing 119
#27 A Replacement Solvent Cleaner/Degreaser Study at Duffy Electric and Machine Company . 123
#28 A Supercritical Fluid Cleaning Study: Application to Instrument Bearings 127
#29 Replacement Non-Methylene Chloride Paint Remover 133
WASHINGTON:
#30 Recycling Electric Arc Furnace Dust: Jorgensen Steel Facility 135
#31 Low-Volatility Solvent and Filtration System for Mechanical Parts Washing 139
#32 Power Washer with Wastewater Recycling 144
#33 Bicarbonate of Soda Blasting Technology for Aircraft Wheel Depainting 148
#34 Onsite Solvent Recovery with an Atmospheric Still 152
#35 Onsite Solvent Recovery with Vacuum Heat-Pump Distillation 155
VI
-------
ID
08
09
10
11
12
13
14
15
16
17
STATE
CT
CT
CT
CT
CT
IL
IL
IL
IL
IL
SHORT TITLE
An Automated Aqueous
Rotary Washer for the
Metal Finishing Industry
(Compared to 1 . Vapor
degreaser2. Alkaline
tumbling 3. Aqueous
manual washing)
On-Site Newspaper Ink
Recycling
Cadmium and Chromium
Recovery from
Electroplating
Rinsewaters
Nickel Recovery From
Electroplating
Rinsewater by
Electrodialysis
Chromate Recovery from
Chromating Rinsewater
in the Metal Finishing
Industry
Ink and Cleaner Waste
Reduction Evaluation for
Flexographic Printers
Alkaline Noncyanide Zinc
Plating and Reuse of
Recovered Chemicals
Recycling Nickel
Electroplating Waste by
Low Temperature
Evaporation and Reverse
Osmosis
Evaluation of
Ultrafillration to Recover
Aqueous Iron
Phosphating/
Degreasing Bath
Waste Evaluation of Soy-
Based Ink at a Sheet-Fed
Offset Printer
USE8
S
Pr
M
M
M
Pr,S
M
M
S,M
Pr,S
OLD
1 . Organic
VOC's/wastes
2. Alkaline wastewater
3. Surfactant waste from
manual washing
Waste ink and solvent
Cadmium/cyanide
rinsewater, chromium
rinsewater, treatment
plant chemicals and
metal hydroxide sludges,
treatment plant
wastewater
Nickel bearing
wastewater
Process rinsewater,
treatment plant sludge
Organic solvent air
emissions, solvent waste
Zinc Cyanide plating
waste
Nickel electroplating
waste
Aqueous Phosphating
waste water
Air emissions from
petroleum-based inks
and cleaners, liquid
waste from washup
trays, inks and cleaners
on used raas
NEW
1 . Eliminate organic
wastes
2. Reduced waste volume
(-90%)
3. Reduced waste volume
(-80%)
Reduced ink/solvent
consumption and waste
generation (9,000 gal/yr
ink)
Cd: reduced wastewater
by approx 2 million gal
(w/ion exchange),
generated 660 gal
regenerant. Cr: potential
to reduce 0.5 million gal of
wastewater. Test run did
not recover chrome.
Recycle 29,964 Ibs/yr
nickel and over 1 ,000,000
gal/yr water
Reduce wastewater by
450,000 gal.
Generates 4 drums of still
bottoms
Reduced solvent air
emissions by 80%
Reduced hazardous
wastes. Increased
aqueous waste
Eliminate Zinc Cyanide
waste. Generate alkaline
noncyanide (ANC)
rinsewater with zinc
hydroxide recovery, water
recycling
Reduced plating waste,
nickel plating bath reuse.
Permeate from RO,
increased electricity use,
cooling water
99.8% reduction in H/W
(1,500 gal/yr)
Approx 1 7% reduction in
ink use. Inconclusive
differences for air
emissions, liquid and solid
wastes. Same cleaner for
both.
-------
ACKNOWLEDGMENTS
The U.S. EPA would like to acknowledge the cooperation and effort contributed by the WRITE
Program participants:
CALIFORNIA: U.S. EPA project officer - Lisa M. Brown
California Department of Toxic Substances, project officer- Robert Ludwig
Hosts for technology evaluations- General Dynamics, Pomona
Division, Pomona, CA; Hewlett Packard, Sunnyvale, CA; Orange County Transit
Authority, Orange County, CA.
CONNECTICUT: U.S. EPA project officer - Lisa M. Brown
Connecticut Hazardous Waste Management Service, Connecticut Technical
Assistance Program, project officer - Rita Lomasney. ESSAR Environmental
Services The Connecticut Hazardous Waste Management Service consultant- Sumner
Kaufman
Hosts for technology evaluations - Quality Rolling and Deburring, Thomaston;
The Hartford Courant; the Torrington Company, Torrington, CT; Automatic Platinq
of Bridgeport, Bridgeport, CT
ILLINOIS: U.S. EPA project officer - Paul M. Randall
Illinois Hazardous Waste Research and Information Center, project officer- Gary
Miller
Hosts for technology evaluations - MPI Label System Inc., University Park, IL;
P&H Plating Company, Cook County, IL; Graham Plating Company Chicago IL-
R.B. White, Bloomington IL; Office of Printing Services, University of Illinois
MINNESOTA: U.S. EPA project officer - Teresa Harten
Minnesota Environmental Control Agency and Minnesota Technical Assistance
Program Aency.andMinnesotaTechnicalAssistanceProgram, University of
Minnesota - project officers, Cindy McComas and Paul Pagel.
Hosts for technology evaluations - Micom Inc, Brighton MN; Hutchinson
Technology Inc., Hutchinson, MN; McCurdy Circuits, Orange County, CA;
Pararnax Inc., St. Paul, MN; SL Modern Hard Chrome, Camden, NJ-' Aeroquip Inc
Industrial Connectors Div., Van Wert, OH
VIII
-------
ID
29
30
31
32
33
34
35
36
37
38
39
40
41
STATE
NJ
WA
WA
WA
WA
WA
WA
WA
EC
EC
EC
EC
EC
SHORT TITLE
Replacement for Non-
Methylene Chloride Paint
Remover
Recycling Electric Arc
Furnace Dust: Jorgensen
Steel Facility
Low-Volatility Solvent and
Filtration System for
Mechanical Parts
Washing
Power Washer With
Wastewater Recycling
Bicarbonate of Soda
Blasting Technology for
Aircraft Wheel Depainting
Onsite Solvent Recovery
with an Atmospheric Still
Onsite Solvent Recovery
with Vacuum Heat- Pump
Distillation
Onsite Solvent Recovery
with Low Emission Vapor
Degreasing
Replacement of
Hazardous Material in
Wide Web Flexographic
Printing Process
Ultrasonic Cleaning as a
Replacement for a
Chlorofluorocarbon-Based
System
Removal and
Containment of Lead-
Based Paint via Needle
Sealers
Low-VOC Wood Furniture
Coatings
Finishing Fabricated
Metal Products with
Powder Coatings
USE'
D
St
S
S
S
04
S
S
Pr
S
D
C
C
OLD
Spent methylene chloride
K061 Waste (EAF Dust),
Steel making residues
Spent organic solvent,
volatile releases, or
mineral spirits replaced
monthly
Spent organic solvent, or
steam/spray cleaning &
wastewater
Spent organic solvent,
dissolved paint
Spent methyl ethyl ketone
(MEK)
Spent methylene chloride
(MC)
Organic solvent air
emissions, spent organic
solvent
Volatile organic
constituents (VOC's)
from ink and make-up
solvent
Chlorofluorocarbon
(CFC) air emissions,
spent CFC's
Lead contaminated dust
emissions and blasting
grit
Organic
solvent air emissions and
wastes
Organic solvent air
emissions and wastes
NEW
Spent N-methyl-2-
pyrolidone (NMP).
Eliminates 49,000 Ibs of
methylene chloride waste.
Reuses 240,000 Ibs
activated carbon
On-site recycling of EAF,
residues into ceramic
products
Closed loop system, low
volatility solvent, solvent
disposal reduced 94% by
weight"
Eliminated solvent releases
& spent solvent. Reduced
wastes by 80%, waste
segregation. Electrical
consumption up
Eliminated solvent.
Bicarbonate slurry/water,
paint chips, loud noise,
Rotoclone wastewater
Reduced solvent waste by
70%, generated still
bottoms, still cooling water
Reduced waste solvent
volume 92%,
generated still bottoms,
Air emissions reduced
>99%.
72.5% reduction in VOC's
(at. 80% water, 20%
solvent)
Reduced volatile emissions
by 86%
Reduced dust emissions by
99%, reduced H/W by
97.5%
Not available (NA), project
being completed
NA, project being
completed 5/95
-------
-------
n Converting from hard to sponge rollers.
0 Converting from continuous to conductivity sensor controlled rinse.
p Based on limited data. Findings were favorable but not conclusive.
q Depends on work load. Payback reached after 38 units tested using air and 96 using N2
r Presented economics excludes recycling of semi-aqueous solvent/rinse because supplier does not
have a recycling system. Recycling plus impact of Clean Air Act Amendments, etc. could favorably
impact this system.
s Based on cleaning 150 bearings/yr.
1 Projected savings based on 1000 parts/yr.
u Projected for formulations at Jorgensen. Actual values will depend on product type and amount sold
at $2 to $650/ton.
v The $24.6 K/yr is based on the actual formulation tested at 80% water.
w The $117 K/yr is projected on using a 100% water-based formulation.
x Project not completed.
ID
01
02
03
04
05
06
07
STATE
CA
CA
CA
CA
CA
CA
CA
SHORT TITLE
Watts Nickel and Rinse Water
Recovery via an Advanced
Reverse Osmosis System
Evaluations of Three Oil Filter
Designs for Pollution Prevention
Effectiveness
Electroplating Rinsewater
Reduction
Sulphuric Acid Anodizing
Robotic Painting System:
. Robotics Arm Spraying
. Proportional paint mixing
. Electrostatic spray guns
. Automatic paint line cleaning
and solvent collection
Solvent Recycling Stills
Plastic Bead Blast Paint
Stripping
Freon Recovery from
Degreasinc
PERFa
2
3'
3
2
2-
2
2
2-
ECONOMICS
$SAVEDb
17.1s
0.359
157h
496
15
38.5
4
5
490
CAPITAL0
75
.369
164h
4,100
995'
1,400
14
18
270
YRSd
4.4e
~1
>8
67
38
3.5
3.6
0.6
-------
TABLE 1. RESULTS OF WRITE TECHNOLOGY EVALUATIONS
MAJOR IMPACTS ON WASTES
LEGEND
a Type of industry using technology. Industries are: C - Improved Coating Technology (low or no
solvent paints and varnishes, improved application); D - Depainting (paint, varnish surface
coating removal); E - Electronics; M - metal plating or finishing; P - Paper; Pr - Printing; S -
Surface cleaning; St - Steel; O - Other, explained below.
Q! Reduction of waste oil, internal combustion engines.
O2 Metal working oils/fluids recycling.
O3 Recycling of spilled oils/fluids, recycling of sorbent pads.
O4 Spray painting line cleaning solvent (MEK) recycling.
b Solvent life projected from 16 month run. Product acceptance problems, et al., cut solvent life
test short. Potential for significantly longer solvent life seems likely.
ID
01
02
03
04
05
06
07
STATE
CA
CA
CA
CA
CA
CA
CA
SHORT TITLE
Watts Nickel and Rinse
Water Recovery via an
Advanced Reverse
Osmosis System
Evaluations of Three Oil
Filter Designs for
Pollution Prevention
Effectiveness
Electroplating
Rinsewater Reduction
Sulphuric Acid Anodizing
ROBOTIC PAINTING:
. Proportional paint
mixing
. Electrostatic spray
guns
. Automatic paint line
cleaning and collection .
Solvent recycling stills
Plastic Bead Blast Paint
Stripping
Freon Recovery from
Degreasing
USE3
M
o,
M
M
C
D
S
OLD
Rinse water from nickel
plating, waste water
treatment sludge
Used motor oil, used oil
filters
Rinsewater, hazardous
waste (H/W) sludge from
treatment, waste copper
anodes
Chrome rinsewater,
chrome air emissions,
Paint waste, overspray,
spent solvent
Spent Methylene
Chloride
Spent solvent
NEW
Recovery of nickel.
Reduced water
consumption and waste
water by 98%. Reduced
plating chemical use and
treatment sludge.
Reduced waste oil by 66%
(157gal/bus/yr). Reusable
filters
Reduced rinsewater
discharge by 50 gpm.
Copper recovery
Eliminate chrome waste.
Reduce rinsewater volume
by 60%
Decrease electric
consumption by 70%
Generate waste H3SO4
Paint waste reduced
approx. 50%,
Recycled cleaning solvent.
Still bottoms generated
.Eliminate Methylene
Chloride
.Plastic bead residue
Reduced solvent
consumption by 35%
(229,000 Ibs of freon)
-------
ID
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
STATE
NJ
NJ
NJ
NJ
NJ
NJ
NJ
WA
WA
WA
WA
WA
WA
WA
EC
EC
EC
EC
SHORT TITLE
Mobile Onsite Recycling of
Metalworking Fluids
A Fluid Sorbent Recycling
Device for Industrial Fluid Users
Electronic Component Cooling
Alternatives: Compressed Air
and Liquid Nitrogen
Evaluation at Zerpol (Zerc Liquid
Discharge System) at Pioneer
Metal Finishing
A Replacement Solvent
Cleaner/Degreaser Study At
Duffy Elec-ric and Machine
Company
A Supercritical Fluid Cleaning
Study: Application to Instrument
Bearings
Replacement for Non-Methylene
Chloride Paint Remover
Recycling Electric Arc Furnace
Dust: Jorgensen Steel Facility
Low-Volatility Solvent and
Filtration System for Mechanical
Parts Washing
Power Washer With Wasiewater
Recycling
Bicarbonate of Soda Blasting
Technology for Aircraft
Wheel Depainting
Onsite Solvent Recovery with an
Atmospheric Still
Onsite Solvent Recovery with
Vacuum Heat-Pump Distillation
Onsite Solvent Recovery with
Low Emission Vapor Deg'easing
Replacement of Hazardous
Materials in Wide Web
Flexographic Printing Process
Ultrasonic Cleaning as a
Replacement for a
Chloroflucrocarbon-Base'l
System
Removal and Containment of
Lead-Based Paint via Needle
Sealers
Low-VOC Wood Furniture
Coatinas
PERF3
2
2
2p
2-
2-
2-
1
2
2
2
2+
2
2
2+
2
2+
1
NAX
ECONOMICS
$SAVEDb
1.6 to 7.8
52% to
75%
~$5.25 per
unit tested
w/air or N7
22.5
-0.3r
m:
315
10.6001'
0.5
22
80
10
18
25
24.6V
117W
27
-7.2
-
CAPITAL0
0
0.7
Air 0.2
N20.5
120
(for 2000 gpd)
NA
75 to 1 00
80
10,500"
1.5
55
227
13
24
210
63
63
44.4
NA
-
YRSd
NA
~0
~0«
~0q
5.3
NA
-71s
-11'
<1
1U
4
3
<4
<2
<2
10
2.56V
0.5W
1.6
NA
--
-------
ID
18
19
20
21
STATE
MN
MN
MN
MN
22
MN
23
24
25
26
NJ
NJ
NJ
NJ
27
28
NJ
NJ
SHORT TITLE
^^^—^—^— _
Modifications to Reduce
Drag Out at a Printed
Circuit Board
Manufacturer
Sponge Rollers and Flow
Controller for Printed
Circuit Board
Manufacturing
Carbon-Black Dispersion
Preplaling Technology
for Printed Wire Board
Manufacturing
Evaluation of an
Electrodialytic Process
for Purification of
Hexavalent Chromium
Substituting Cadmium
Cyanide Electroplating
with Zinc Chloride
Mobile Onsite Recycling
of Metalworking Fluids
A Fluid Sorbent
Recycling Device for
Industrial Fluid Users
Electronic Component
Cooling Alternatives:
Compressed Air and
Liquid Nitrogen
Evaluation at Zerpol
(Zero Liquid Discharge
System) at Pioneer Metal
Finishing
A Replacement Solvent
Cleaner/Degreaser Study
At Duffy Electric and
Machine Company
A Supercritical Fluid
Cleaning Study:
Application to Instrument
M,E
M,E
M
M
O2,St
O,
M
OLD
-^—^—.
Plating waste and
waste water
Dragout, rinse water,
and treatment process
wastes
Formaldehyde 200
gal/yr, rinsewater,
Spent: chromic acid
etchant, strip solution
Cadmium cyanide
wastes, grease and oil,
chromium, treated water,
wastewater treatment
sludge
Spent fluids sent to
disposal
Spent sorbent pads,
spilled fluids
Freon (R12) releases
Plating wastewater,
sludge, hydrogen
cyanide at 2.69
(compliance 1.2 ppm).
360 gal of spent
petroleum distillate
(w/toluene and
ethylbenzene) Approx
50% can be recycled.
Freon-113 releases,
toluene, hexane,
isopropyl alcohol,
acetone
NEW
Reduced dragout by
-50%, reduced rinse water
volume
Reduced drag out,
reduced waste water,
reduced water use
Eliminated formaldehyde
Reduced: rinsewater
-88%, copper -23%
Reduce: chrome 7,000
Ib/yr, etchant 21,000 gal/yr
Generate; Sodium chloride
10,000 Ib/yr, sodium
sulfate 5,000 Ib/yr,
increase water
consumption 5.000 qal/yr
Decreases (Ibs): Cd-
12.100.CN-835, oil&
grease-9480. Increase:
Zn- 22,300, Cr-3740,
sludge-101,000, treated
water- 4.9 million gal.
Reduced fluids waste by
-1,250 gal for small shops
via on-site recycling
Spent sorbent pad waste
reduced by up to 87%.
Eliminate R12 releases.
Requires N2 or clean
compressed air.
——————_—
Reduced water use by
66%; -80% process water
recycled; hydrogen
cyanide to .058 mg/m3;
potential for metals
recover
•
78 gal spent liquids; 39 gal
semi-aqueous cleaner,
39.1 gal isopropyl alcohol
Eliminate all releases
under the old system.
-------
SECTION 1
INTRODUCTION
This report is a summary of work done under the WRITE Pilot Program with state and local
governments. Between 1989 and 1994, forty-one (41) candidate, pollution prevention technologies were
evaluated. The states of California, Connecticut, Illinois, Minnesota, New Jersey, and Washington, and
Erie County New York, participated in the pilot program. Specific objectives of the WRITE program were
to:
. Identify priority problem areas within each jurisdiction
. Find new technologies that have good potential to reduce or eliminate EPA priority or problem
wastes
. Identify companies using or planning to use the technology in the near future, and willing to
host a demonstration,
. Evaluate the technology via demonstration and determination of P2 impact and economics.
The demonstrations included designing a test to quantify performance and associated
environmental impact and gather data regarding previous or traditional technology (historical in most
instances) that was replaced by the new.
Where complete performance or environmental impact of the new technology was not directly
available from test data, such as confidentiality of certain aspects, cost of obtaining full data sets was
beyond the scope of the project (such as getting full air emissions data, for example) these factors were
calculated or estimated on the basis of worst case scenario. The methods of obtaining these values is
explained in the final project reports.
In some instances, evaluation of the performance of a product or process was based on
qualitative data. Where this was the accepted and traditional method of determining product or output
quality, experts or experienced personnel in performing this determination were used in arriving at the
qualitative determinations.
To the extent practical, comparisons were made between the traditional methods and the new
technologies to estimate the net effects, assessing the pollution prevention (P2) impact, and estimating
the economics for both old and new systems. For assessing economics, the typical assumption used for
the old system was that it was already in place and no capital outlays were required. Exceptions would be
if modifications were required to get or stay in compliance using the old system.
TECHNOLOGY EVALUATIONS
The evaluations consisted of performance and P2 impact data from the new technology
demonstrations compared to new data, historical records or estimates of the old technology. The
economic comparisons were based on historical records and estimates of the old system compared to
estimates for the new. Energy consumption comparisons between the two systems was included in the
evaluations where applicable.
11
-------
TABLE 2. RESULTS OF WRITE TECHNOLOGY EVALUATIONS-
PERFORMANCE AND ECONOMICS
LEGEND
Annual saving in $000.
c Capital costs in $000.
d Payback period in years.
• Unit operated a, ,ess than 50% capacity. Protons near fu,, capac«y give a 2-year payback
' e^Tar"'6™' inCr6aSed ""' 2* "ased °n «<*« — »• oil con*. Need data on long term
9 per bus.
h per facility (fleet of 450 busses).
' Quired repiacemen, of exis«ng tanKs, adding cooiing equipment and modi^ng water treatment
J Compared to vapor degreasing.
Compared to alkaline tumbling.
Compared to hand-aqueous washing.
- * not ge, a successful
-------
P2. Waste Minimization. Source Reduction and Recvclina-
During the start of the WRITE program, and prior to the Pollution Prevention Act of 1990, the
focus of the WRITE Program was equally on source reduction and recycling. After its passage, the
Agency placed more priority on source reduction. The chronological progression of the WRITE
technology evaluations reflect this change of emphasis, with fewer recycling projects being selected during
year two and three. From a practical standpoint the full hierarchy of waste management; source
reduction, recycling, treatment and disposal should be considered in arriving at the smarter way of doing
business.
Federal/State EPA Partnerships--
State and local government entities were very helpful in identifying pollution problems and
priorities in their jurisdictions. In many instances they were able to make valuable contacts in finding host
sites for the demonstrations and in over a quarter of the technology evaluations, they were able supply
people and equipment to test and evaluate the technologies and draft reports.
Additionally, there was significant benefit in working with the multiple partners in terms of
promoting the P2 paradigm shift which is an important part of pollution prevention.
13
-------
08
CT
An Automated Aqueous Rotary
Washer for the Metal Finishing
Industry (Compared to: 1. Vapor
degreaser2. Alkaline tumbling 3
Aqueous, manual washing)
2-
1.-15
2. 15k
3. 4
09
CT
On-Site Newspaper Ink
Recycling
2-
NA
318
10
10
CT
Cadmium arid Chromium
Recovery from Electroplating
Rinsewaters
Cd:2
Cr: Om
Cd:41
Cr: 5.7 m
Cd: 16
Cr: 16
Cd:<1
11
CT
Nickel Recovery From
Electroplating Rinsewater by
Electrodialysis
1
188
110
12
CT
Chromate Recovery from
Chromating Rinsewater in the
Metal Finishing Industry
22
87
13
IL
Ink and Cleaner Waste
Reduction Evaluation for
Flexographic Printers
16.5
14
IL
Alkaline Noncyanide Zinc Plating
and Reuse of Recovered
Chemicals
2-
62
88
1.5
15
IL
Recycling Nickel Electroplating
Rinse Waters by Low
Temperature Evaporation and
Reverse Osmosis
1-
NA
115
2.8
16
IL
Evaluation of Ultrafiltration to
Recover Aqueous Iron
Phosphating/
Degreasing Bath
NA
NA
0.6
17
IL
Waste Evaluation of Soy-Based
Ink at a Sheet-Fed Offset Printer
NA
NA
NA
18
MN
Modifications to Reduce Drag
Out at a Printed Circuit Board
Manufacturer
3.4
~0
~0
19
MN
Sponge Rollers and Flow
Controller for Printed Circuit
Board Manufacturing
2"
2°
16.5n
1.5°
0.2n
0.6°
~0n
0.4°
20
MN
arbon-Black Dispersion
Preplating Technology for
rinted Wire Board
Manufacturing
87
221
<4
21
MN
Evaluation of an Electrodialytic
~ rocess for Purification of
Hexavalent Chromium
2+
126
563
4.5
22
MN
Substituting Cadmium Cyanide
Electroplating with Zinc Chloric-la
17
1973
115
-------
verify results reported by HP.
Cost information was provided by HP and the AROS manufacturer. Where possible the costs
were checked against other sources.
Results
The AROS unit achieved good separation of contaminants from the dirty rinse water Removals
of contaminants usually ranged from 90% to 97%. Overall the HP staff regard the AROS unit as havinq
shown good performance during the test period. Rinse water quality was maintained at a low level of
nickel contamination. It was reported that no printed circuit boards were rejected because of Watts nickel
plating deficiencies.
The recycling of the rinse water resulted in a 98% reduction in the use of new deionized makeup
water for this plating process, equivalent to about 425,000 gallons annually per shift, per plating line.
The AROS unit also successfully produced concentrated Watts nickel solution of adequate qualitv
to return to the plating bath solution. Fresh Watts nickel solution costs about $5.00/gal so recovery and
recycling presented a significant direct savings. It was also calculated that a reduction of approximately
three (3) tons of category F006 waste (waste water treatment sludge from electroplating operations) at the
industrial waste water treatment system could be attributed to the AROS unit.
Cost information provided by HP (see Table 1) indicated that the AROS unit would produce an
estimated annual cost savings of $26,250 at the HP facility when the unit was operated at less than half its
rf SS ^drfullc caPacity- Th's savings is reduced by an estimated annual operating and maintenance cost
of $9,419 for a net annual savings of approximately $17,100/yr.
Capital investment is approximately $75,000, which represents $63,000 for the unit and $12 000
for making the installation permanent and for training of operational personnel. Dividing $75 000 by'
$17,100 results in a payback period of 4.4 yrs and a 23% return on investment.
The AROS unit at HP was operated at less than 50% of its capacity, partly due to the reduced
operation of the Watts plating line. Operating the AROS unit at a level closer to design capacity would
produce significantly better economics.
In terms of the overall facility, the AROS unit treated only about 3% of the total wastewater flow
Therefore, in its cost analysis, HP made no allowance for reduced labor cost at its main wastewater
pretreatment plant. If, however, the AROS unit treated a larger percentage of the total wastewater flow a
labor reduction credit could be included in the cost analysis.
The HP facility has a fully amortized wastewater treatment facility in place. At a new facility under
design, sizing the unit to operate closer to design capacity, reducing the capacity for a water treatment
plant, reducing the size of Dl water production units and gaining some benefit in reduced operating costs
could all lead to more attractive economics.
at 2 yrs.
The projected payback period for the unit operating near capacity at the HP facility was estimated
15
-------
ID
STATE
SHORT TITLE
PERF"
ECONOMICS
41 EC Finishing Fabricated Metal
Products with Powder Coatings
NAX
10
-------
#02 EVALUATIONS OF THREE OIL FILTER DESIGNS FOR POLLUTION PREVENTION
EFFECTIVENESS
Participants
The state of California Department of Toxic Substances Control, Sacramento, CA assisted in
implementing the evaluations and participated in test design and review of the project report. The host for
this evaluation was the Orange County Transit Authority (OCTA), Garden Grove, CA. OCTA personnel
operated the busses equipped with the test oil filters as well as performing required maintenance.
Science Applications International Corp, on contract to EPA, helped to design the test program, supplied
test personnel and equipment, and drafted the final report.
Technology /Testing
Used oil and oil filters are a large source of waste in the U.S. Two approaches to minimizing this
waste, as well as minimizing consumption of non-renewable energy, would be to design reusable filters
and filters that extend oil life. Filters that can be cleaned instead of discarded and improved filtration
function to permit more mileage between oil changes have a large potential due to the large numbers of
internal combustion engines in use and vast quantities of waste generated.
Three types of oil filters were included under this project for use on diesel engine busses. The
designs included a reusable wire rnesh, a disposable fiber and a disposable paper filter. The test program
started in January, 1991 and underwent four months of testing plus engineering and economic
evaluations.
Twelve OCTA fleet busses with Detroit 6V92T diesel engines were used for the test. The twelve
were selected from the same manufacturer. The busses were as similar as possible in type, age,
mileage, type of service route, previous service history and oil analysis.
The busses were grouped in three sets of four. Each was fitted with a primary filter and
secondary-partial flow-centrifugal filter. One group of the four served as the baseline and was equipped
with regular screw-on pleated paper, primary filters. The second set was equipped with a composite,
synthetic media, primary filter. The third set used primary filters made of metal screens that could be
cleaned and replaced.
A fitting was installed on the test vehicles that permitted samples to be withdrawn immediately
downstream of the primary filter with the engine running. During the 4-month test period, weekly oil
samples were taken to analyze physical and chemical properties, additives, and contaminants.
Additionally, a particle count was conducted every other week (see Table 1.).
The test parameters examined fell into 2, basic groups, those that reflected oil quality or condition
and those that indicated potential engine problems. Parameters monitored were based on guidelines
provided by the Detroit Engine Company.
17
-------
nriman « focused on changes in EPA identified priority toxic substances and the
primary effects due to the changes made. At the same time secondary implications were considered to
the extent possible within the scope of the evaluations.
While significant efforts are being carried out in the area of comparability, risk and true costs of
tokina fTr pnn8,;nhG nT ^Tl ^ ^ °f infOrmed Jud9ement in determining what constitutes
'
Part of this challenge is also due to the fact that the answers for determining common
denominators do not lie solely within the technical or technology sectors, but rely on the other facets of
«mnnn r ^9hS^i0"TOTn°mlC!,and ^^ and Varied PercePtions °f what are environmental priorities
among the stakeholders. To enable consensus definitions, all of these sectors need to come into play
Recognition of the need for better consensus definitions and priorities is reflected by the recent EPA
adoption of the Common Sense Initiative (CSI) which includes participation of all stakeholders in the
development of regulations and implementation of P2 in technology development and improvement
OVERVIEW AND RETROSPECTIVE
The central purpose of the WRITE Program was to enhance and speed up the process by which
th!f heH , 9'f! 96t imP'emented- ln the Purs"it of this objective a number of issues were identified
technoloS'Tn eanerarnn9 ^ ^^^ °f implementin9 P2 technology and on the adoption of new
Identifying New Technology-
The states and local governments were to identify problem areas in their jurisdictions and
participate in finding host companies to test, demonstrate and evaluate technologies that solve these
nfr0bmnf ' Sa{lS^ ^ ab°Y? °riteria was a difficult P^ of the program. There was an overall shortage
rlPpnThTI f9 Td ^ TK d t0 particiPate- ln this sense- qualified pertains to companies using a
dean technology of mterest. There was also an apparent reluctance of companies entering into a joint
effort with a regulatory agency such as the EPA, even within a non-adversarial or partnering arrangement.
Adoption of New Technology-
Generally, expectations regarding the rate at which new technology gets implemented were
overly optimistic. While there is a vary large time gradient from simple changes of procedure to very
complex and interlinked supply, process and product changes, it is even further complicated by a variety
of non-technolog.cal issues. Therefore, even for win-win-win technology (i.e. technology with superior
performance reduced or eliminated waste, a small amount of capital investment and a very short payback
period), significant roadblocks need to be overcome. A more rigorous examination of cause and effects of
these dynamics may be desirable, but both outside the scope and mission of this program
12
-------
Based on 48,000 miles driven per bus, per year, the savings in changing from 6,000 to 18 000
miles would be approximately 170 gal of waste oil and $350. For OCTA this would mean savings of
76,450 gal and $157,000. More effective filtration of engine oil has good potential for pollution prevention
if engine life is not adversely impacted.
The full report, entitled "Evaluation of Three Oil Filter Designs for Pollution Prevention
Effectiveness", by Lisa M. Brown and Robert Ludwig will be available as an EPA/600 series report.
19
-------
SECTION 2
WRITE TECHNOLOGY EVALUATIONS
alP^«oa,,y by state
CALIFORNIA
A total of seven (7) technologies were evaluated in California, five (5) for the General
i "* 0) ** *" Oiane C°Unt T ^
ROH ,- Th| evaluationj> wuere done under a Memorandum of Understanding between the EPA Risk
2£?T fhngmeTing Laboratory and the California Department of Toxic Substance Control Vhe project
officers for the evaluations were Lisa M. Brown and Robert Ludwig, respectively. P J
SYSTEM71"3 N'CKEL AN° R'NSE WATER RECOVERY VIA AN ADVANCED REVERSE OSMOSIS
Participants
The host for the evaluation was the Hewlett Packard (HP) plant in Sunnyvale California whn
also ran the reverse osmosis system during the test. The state of California!S?e^n^Sc
Substances Control.Sacramento, CA, assisted in implementing the evaluations Science
lnternat,onal Corporation (SAIC), Santa Ana, California, on contract to EPA helped to de-
program, supplied test personnel and equipment, and drafted the final report.
Technology/Testing
.P}a^ ind"strV produces large quantities of metal contaminated wastewater requiring
HF ^Ti ? AdHVanC6d RfJe rSe °Sm°SiS SyStem (AROS>' manufactured by WT was
ant an
n HF H '
onni Piant ^ and reCOVer Watts nickel P'atin9 bath solut'on and rinse water The
technology approaches zero discharge capability. An 8 - month test program was conducted to as
the effectiveness of the AROS and estimate the incremental cost savings ^^g^2?u^
The test included continuous monitoring of flow volume, conductivity, and pH at various
monitoring points in the system. Streams monitored include the deionized rinse water makeuD line the
concentrate return tae and the permeate return line. The plating bath was sampled and a llyzed weekly
Analyses were conducted for nickel, pH, Nickel PC-3 (Saccharin), boric acid, chloride, and dudUNy.
Independent sampling and analysis were performed by the EPA contractor over a 1 -day period to
14
-------
Copper spheres In anode baskets are used In the new system Instead of the conventional anode
bar-and-hook system that was utilized in the old plating system. This allows a 1 to 1 ratio of anode to
cathode for a very even plating across the panel and through the holes.
In conjunction with the installation of the new production equipment, Chemcut Corporation was
required to provide a non-sludge-producing treatment system for all waste streams generated by the
process. This resulted in the installation of a new copper-recovery system in which short-bed ion-
exchange columns and electrowinning technologies are used. This system now produces salable scrap
copper metal, eliminating a major waste stream to the conventional sludge-producing waste treatment
system.
Figure 2 shows a flow diagram of the printed circuit board copper recovery system. The copper
recovered with this process represents the equivalent of approximately 365 tons of sludge that would
have gone to a landfill.
Short Bed
Copper
Ion Exchange
Ion Exchange
Regeneranl
Copper
Plating Bath
Dump
Collection
Copper
Electrowinner
Ion Exchange
Regenerant
Short Bed
Copper
Ion Exchange
Scrap
- Copper
for Resale
Eleclrowinner
Dumps
To
Final pH
• Adjustment
and Discharge
to Sewer
Figure 2. Flow diagram of circuit board copper recovery system
The objectives of the computerized printed circuit board plating operation evaluation included
the following:
1)
2)
Verification of reductions in wastewater rinse flow by comparing records of present
wastewater flows vs. previous wastewater flow data.
Establishment of a material balance around the printed circuit board process, if possible
using available plant data. Copper-bearing waste streams are to be identified, sampled '
for metal concentrations and documented for appropriate metal recovery processes.
21
-------
TABLE 1. ESTIMATED ANNUAL SAVINGS FROM USE OF THE
AROS UNIT AS REPORTED BY HEWLETT-PACKARD COMPANY. 1900 COSTS
Description of Costs
Sewer discharge fees and water costs
Deionized (Dl) water production costs*
Plating Wastewater treatment costs**
Purchase of new plating chemicals at 85%
reduction
Total estimated annual savinqs
Estimate
Savings
($/aal)
0.004
0.0064
0.0062
5.00
Quantity
(pal)
1,275,000
1,275,000
1 ,275,000
1260x
0.85
Total
Annual
Savings ($)
5,100
8,160
7,905
5,355
Dl water production cost is for chemicals, electricity and resin replacement only. No labor, depreciation, or other costs are
included on the assumption that they would remain essentially the same, whether the AROS unit was in use or not.
Plating wastewater treatment cost includes sludge disposal, chemicals, and electricity. As in the note above no labor
depreciation or other costs were included. «w vo, ,u iauur,
The full report, entitled "Watts Nickel and Rinse Water Recovery via an Advanced Reverse
Osmosis System", by C. Schmidt, et al., is available as EPA/600/R-93/150.
16
-------
Under the new system, the copper is recovered using ion-exchange and electrowinning
equipment. Sludge disposal from this process has been eliminated. Anode basket plating has the
following features:
b Constant anode surface area (promotes uniform plating on parts), less product waste
ID Improved handling of anode material (anode material can be added without shutting down
plating process)
(3 No wasted anode material (anodes can be allowed to dissolve, completely eliminating the
need to recycle unused anode material)
ID Easier to maintain proper current density (presence of the basket and the constant anode
surface area aids in maintaining current density)
As presented in Table 2, cost savings in labor and waste treatment were found to be the major
cost parameters, resulting in a favorable net annual operating cost savings between the original and newly
installed printed circuit board plating systems. The payback period for the printed board plating system
was estimated to be 8.3 years.
Due to the lack of computer automation, operation of the original system required six employees
(i.e., four hourly and two engineers), versus two employees (i.e., one hourly and one engineer), for the
new system. This resulted in a cost savings in labor of $324,000 based on the plant operating two shifts
per day, 8 hours per shift, and 6 days per week.
The cost savings in waste treatment of $130,000 represents the savings resulting from the
recovery of copper from rinse water and process tank solutions, which were previously treated and
disposed of as hazardous sludge. Copper is now recovered as a salable by-product using ion-exchange
and electrowinning equipment. The recovery process, estimated to be approximately 15 percent of the
total capital installed cost of the printed circuit board system, produces solid slabs of copper which weigh
approximately 30 pounds. One copper slab represents an equivalent of 0.5 tons of sludge that previously
required landfill disposal at a cost of $1200. Now, the recovered copper has a by-product value ranaina
from $15 to $20 per slab.
Cost savings in labor, waste treatment and water usage resulted in a net operating savings when
compared to an estimate of the operating cost of the old plating system. These savings may increase with
time as a result of startup problem resolution and reductions in chemical costs. Although a payback of 8.3
years was calculated on the capital investment of $4.1 million, this does not consider the fact that a retrofit
of the old system to the present operating capacity and waste reduction goals would have resulted in even
higher capital costs. Looking at the expenditure from this standpoint, it could be reasoned that the new
system is providing an instant payback because of lower capital costs and operating costs savings.
Transfer of this type of computerized plating technology to smaller companies, however, would be difficult
because of the high capital costs.
23
-------
Results
Analyses taken during the 4-month test period revealed little differences in oil sampled from the
three types of primary filters or among the 12 busses with the exception of the wear metals and particle
counts below or equal to the 25 urn range.
TABLE 1. OIL ANALYSES
Parameter Group Test Parameters
Physical/Chemical Flash point, Fuel, Viscosity, Water content,
Percent solids, Glycol, Soot, TBN
Additives Magnesium, Calcium, Barium,
Phosphorous, Zinc, Molybdenum
Contaminants Silicon, Boron, Sodium
Wear Elements Iron, Chromium, Lead, Copper, Tin,
Aluminum, Nickel, Silver, Manganese,
Antimony, Cadmium, Titanium,
Particle Counts >5, >1 Q, >15, >25, and >50 um
For the wear elements, the synthetic fiber filtered oil contained slightly higher metal concentration
This difference was observed for one bus that produced consistently higher concentrations throughout the
test. Busses with the synthetic fiber filter also had the lowest concentrations of particles equal or less than
25 um. Busses with the reusable filters were in the middle, leaving the regular filters with the lowest metal
concentrations and highest concentrations for particles less or equal to 25 um.
All of the busses tested were able to go beyond the 6,000 mile oil change interval regularly
practiced by OCTA. Ten of the twelve busses operated for 4 months without needing an oil or filter
change. Miles traveled for the two were 14,429 and 21,571.
Little differences were revealed among the performance of the three filters for the relatively short
test of 4-months. The major differences were the particle size cut below 25 um. This particle size range
was removed more effectively with the synthetic fibers. A longer test period and a larger sample size
would be needed to correlate engine life with fine particle removal.
The cost of an oil change at OCTA is estimated at $30 (Pleated paper filter), and $40 (Synthetic
fiber filter) without the cost of oil. The reusable filter is $364, installed, with an estimated 10-year life With
labor and gaskets and 8 cleanings per year it equals $27.50 per cleaning cycle. While the cost per filter
per use is similar for the three, with the reusable being slightly less, the latter reduces the amount of waste
generated by spent filters at 100 drums per year (8x450 busses).
Significant source reduction of used oil could be achieved with an extended oil change interval
For fleets with routine oil quality monitoring such as OCTA, the interval could be safely increased by a
considerable amount.
18
-------
TECHNOLOGY AT THE GENERAL DYNAM,CS
Participants
Technology/Testing
•*
The new automated anodizing system, which eliminated both water and air
of r ^a ,u"wted,o: 1 ^ ^ ^ 3 *^ °W Chromic *
-------
#03 EVALUATION OF FIVE WASTE MINIMIZATION TECHNOLOGIES AT THE GENERAL DYNAMICS
POMONA DIVISION: ELECTROPLATING RINSE WATER REDUCTION UTNAMlUb
Participants
The state of California Department of Toxic Substances Control, Sacramento CA assisted in
implementing the evaluations. The host for the evaluation was the General Dynamics, Pomona Division
Plant. General Dynamics personnel operated the computerized electroplating system. PEI Inc on
contract to EPA helped to design the test program, supplied test personnel and equipment, and drafted
the final report.
Technology/Testing
Technology Description--
The Chemcut Corporation installed a new computerized printed circuit board plating system at
General Dynamics at a cost of $4,100,000. Figure 1 shows the tank configuration layout of the printed
circuit board system, and Table 1 presents the standard operating sequence for the process The new
plating system eliminated! rinse tanks from the process by use of a spray-rinse system contained in the
transporter hoist system, allowing the circuit boards to be rinsed for only a short duration after their
immersion in a process solution.
Storage 19-45
Flow Rinse 18
Rack Strip \l
1 6
Rack Strip ]J
Transfer 13
Load/Unload 11/12
Eless CU f ?
67
Eless CU 66
65
Eless CU 64
| Reducer 1 63
1 Activator 62
1 PrA-frin K1
Etch Clean 60
Cln. Cond. 59
| Cln. Cond. 58
1 Acid Dip 57
1 Copper 56
1 Copper 55
Transfer 54
|Dry 53
Dry 52
1 Anodfi Main. 51
*
J
Copper 84 1
Copper 83
Copper 82 1
Copper 81 1
1
Copper 79 J
Copper 78
Copper 77
Acid Dip 76 1
Etch Clean 75]
Acid Clean 74 1
Transfer 73
Acid Dip 72 J
Aux. Plate 71 j
Eless CU = Eledrole$s Copper
Cln Cond = Cleaner/Conditioner
Figure 1. Tank configuration of printed circuit board system
20
-------
C°StSl SUCh aS adding comPuter'zed hoists and on-demand rinse operations must
'f
n h t the 6XiSting process" For GD' jt was more cost-effective to
replace the existing chromic acid process than to make all of the modifications.
The costs associated with testing in order to obtain military approval to modify design
sensitive a\To "*"* "* addr6SSed in the economic evaluation. This information was considered company
O&M costs are typically much lower for sulfuric acid anodizing than for chromic acid because it is
less energy-intensive, it has a smaller plating interval, and wastewater treatment costs are less.
ran, • ?6CaUSe SU'lUriC audd JS mUCh m°re conductive as an electrolyte, the anodizing process usually
requires less power than that required by the chromic acid process. Specific electrical energy
requirements for aluminum anodizing are shown in Table 2. As shown in Table 2, a cost savings for
etectncity was estimated to be $10,900. This was based on a conventional sulfuric acid usage rate of 130
Wh/m2 and a chromic acid anodizing usage rate of 400 Wh/m2. The amount of material anodized was
based on the maximum design capacity of the new system (i.e., 625 ft2/h) and full-time operation (i e
4 992 hours per year). The cost for electricity (i.e., $0.14/kWh) for a typical industrial consumer was "
obtained from Southern California Edison Utility Company
TABLE 2. SPECIFIC ENERGY REQUIREMENTS FOR
ALUMINUM ANODIZING
Type of anodizing Voltage Electrical
. (Y) energy, Wh/m2
Sulfuric acid
Conventional 12-22 70-130
Hard 25-80 150-480
Chromic acid 30-50 240-400
Wastewater treatment costs are less for sulfuric acid than for chromic acid because of a decrease
in metals removal requirements. The chromic acid process requires the reduction of chrome whereas the
sulfuric acid process requires only copper removal. Chrome reduction usually requires additional tanks
and chemicals for treatment and settling. The sludge from the chromic acid anodizing process is also
considered hazardous and must be sent to hazardous waste landfill at a cost greater than that required for
the disposal of the sludge produced by the sulfuric acid process. The annual cost savings resulting from
decreased sludge disposal requirements was estimated to be $1000, based on an annual usage of
approximately 2000 pounds of chromic acid anhydride by General Dynamics. The resulting sludge
produced from the use of this chemical, assuming a 25 percent solids composition would be
approximately 4 tons. GD is presently charged $250 per ton for sludge disposal. The company has a
contract with a smelter in Arizona which recycles their plant sludge for its intrinsic copper value Because
other hazardous metals from other plant processes are still contained in the resulting sludge which is sent
to the copper smelter, no cost break for landfill disposal could be given for the sludge from the new
sulfuric acid anodizing system.
27
-------
plating operation is to be obtained for comparison with the new process. Air emissions
above the baths were expected to be negligible and no air sampling was planned.
3) The use of copper spheres from the anode baskets for other electroplating operations is
to be investigated. Savings in sampling costs due to the new system vs. the old system
of sampling anode thickness to determine anode replacement is to be documented.
TABLE 1. OPERATION SEQUENCE FOR PRINTED CIRCUIT BOARD SYSTEM
Position Step
11 Load
58 Cleaner/conditioner
59 Cleaner/conditioner
60 Etch cleaner
61 Predip
62 Activator
63 Reducer
65-66 Electroless copper
57 Acid dip
76 Acid dip
77-83 Acid copper plating
53 Dry
51 Hold
11 Unload
Results
Reductions in rinse-water discharges were verified during the plant visit by checking rinse-water
flow rates via General Dynamics operating logs. Establishing a material balance around the printed circuit
board process was not possible from information available from plant personnel, operating data, or system
suppliers. This was due to the use of the proprietary process chemistry, developed overseas by Schering
AG, West Germany, the parent company of the system supplier, Chemcut Corporation.
Under the previous plating operation, rinse-water wastes containing copper were sent to the on-
site waste treatment plant for metals precipitation. The sludge was then sent off site for landfill disposal.
22
-------
Although operating costs for the sulfuric acid systems were determined to be about $15,000 lower
than the old chromic acid process, payback on the capital cost of $995,000 was calculated at 67 years.
The payback period for this process may be misleading since GDPD did not believe that they
could meet the 99.8 percent removal efficiency with existing technology and were concerned about
discharge of chromium in view of the 1986 California Safe Drinking Water Act. Given that compliance
with both air and water regulations would not be possible with available control technology even at higher
costs than that for the sulfuric acid system, the capital cost expenditure becomes less important.
The full report, entitled "Evaluation of Five Waste Minimization Technologies at the General
Dynamics Pomona Division Plant", by Lisa M. Brown and Robert Ludwig, is available as EPA/600/S2-91-
067.
29
-------
TABLE 3. ECONOMIC EVOLUTION OF PRINTED CIRCUIT BOARD
PLATING OPERATION WITH ON-DEMAND RINSING3
Economic data
Annual operating costs of old
process
Anode replacement $65,000
Chemical replacement $138,000
Labor $499,000
Water $12,000
Total $714,000
Annual operating costs of new
process
Anode replacement $46,000
Chemical replacement $125,000
Labor $175,000
Water $2,000
Total $348,000
Waste treatment savings $130,000
Net annual operating savings $496,000
Capital investment $4,100,000
Payback period 8.3 years
a Based on plant operation of 2 shifts per day, 8 hours per shift, 6 days per week.
The full report, entitled, "Evaluation of Five Waste Minimization Technologies at the General
Dynamics Pomona Division Plant", by Lisa M. Brown and Robert Ludwig, is available as EPA/600/S2-
91/067.
24
-------
1) Determination of the amount of batch paint waste saved by the use of the proportional
paint mixing by reviewing plant records on paint waste disposal in past years.
2) Verification of increased transfer efficiency of the electrostatic spray nozzles, which was
estimated to have increased from 35 to 65 percent. This was to be done by reviewing
plant records on paint use before and after installation of the electrostatic spray guns.
3) Determination of the amount of solvent reclaimed through the use of the solvent stills by
reviewing plant records on solvent purchases and dirty solvent sent off site for recycling
before and after installation of the solvent stills. Costs for disposal of the dried paint
sludge were also to be determined based on plant records on volume of sludge
incinerated.
4) Verification of the concentrations of VOC in the general area around the paint shop by
taking air samples.
5) An economic analysis of the proportional paint mixing electrostatic spray and paint shop
solvent still portions of the robotic paint system.
Results
The installation cost of the robotic painting system was $1,400,000. This system included a parts
conveyor, computer-controlled robotic spray arms, electrostatic spray guns, proportional paint mixing, and
cleaning solvent collection equipment. The disposal of 42 tons of waste in 1987 would have cost about
$73,000 at a disposal rate of $420/drum plus $7,000 per truckload for transportation (80 drums). The
disposal of 21 tons in 1989 would be about $36,500, for a savings of $36,500. The payback period from a
waste-disposal standpoint alone would be almost over 40 years. This substantially overstates the
payback period, however, because the savings in labor costs from painting and waste disposal and any
decrease in reject parts are not included. The information is considered company sensitive and was not
available.
The two solvent stills were installed at a cost of $14,000. These units are currently being used to
recycle 1,000 gallons per year of cleaning solvent (polyurethane thinner), resulting in $1,500 in purchase
savings and $2,500 in disposal savings, for a total savings of $4,000. The estimated payback period for
this equipment, therefore, is 3.5 years. This estimate does not include increased labor costs to operate
the equipment or decreased hazardous waste handling costs.
Air samples were collected in the general area around the paint shop to verify concentrations of
VOCs. No previous data were available to compare the results.
Air samples were collected using charcoal tubes on a table near two paint spray booths, on the
floor just outside the robotic paint booth, and on a lower shelf next to a small spray booth just inside a
drive-through door leading to the outside. The sampling results are summarized in Table 1. Only 1,1,1-
trichloroethane (TCA) was detected at all three sampling locations. Concentration levels ranged from 0.2
to 1.6 mg/m3, or 1.7 to 12 mg/m3 for an 8-hour time-weighted average (TWA). The OSHA TWA limits for
1,1,1 -trichloroethane are 1,900 mg/m3, well above the levels detected in the paint shop.
Trichlorotrifluoroethane was detected on the two charcoal tubes nearest the floor. Measured levels were 2
and 6 mg/m3, or 17 and 48 mg/m3 for the 8-hour TWA. This can be compared with the OSHA TWA limit
of 7,600 mg/m3 for trichlorotrifluoroethane. It is interesting to note however, that TCA and
trichlorotrifluoroethane are not used in the painting shops. Thus samples measured fugitive emissions
from other work areas.
31
-------
Results
The capital cost of the new sulfuric acid anodizing system installed by NAPCO, Inc., for GD in
December 1988 was $955,000. This cost included the computerized hoist and on-demand anodizing
rinse systems. Actual operating and maintenance (O&M) cost data were considered company sensitive
information and were not available. However, by understanding the differences between the sulfuric and
chromic acid anodizing processes, O&M costs were derived by using information obtained from outside
equipment vendors. Using this information an economic analysis of the sulfuric acid aluminum anodizing
process was performed. The results of the economic evaluation are presented in Table 1. The payback
period for the sulfuric acid anodizing system was estimated to be 67 years.
Conversion to sulfuric acid anodizing from chromic acid is not just a simple chemical substitution
according to system suppliers. The conversion requires a complete change over of anodizing equipment
and partial modifications to downstream waste-treatment facilities. Capital costs would be realized for
such major production equipment items as the anodizing tank, rectifier, and cooling equipment.
TABLE 1. ECONOMIC EVALUATION OF SULFURIC ACID ANODIZING PROCESS
Item $
Annual operating costs of chromic acid process
Electric (116,000 kWh/yr) 16,200
Water (6,000,000 gal/yr) 3,900
Total 20,100
Annual operating costs of sulfuric acid process
Electric (38,000 kWh/yr 5,300
Water (2,400,000 gal/yr) 1,600
Total 6,900
Operating Cost Savings 13,200
Waste reduction savings 1,000
Capital investment 955,000
Payback period 67 years
Replacement of the anodizing tank is required due to the differences in acidity between sulfuric
acid and chromic acid. Because sulfuric acid is much more corrosive than chromic aid, the materials of
construction of the anodizing tank must be upgraded. Because the chromic and sulfuric acid anodizing
processes have different voltage and amperage requirements, the rectifier must also be replaced. The
operating temperature of the electrolytic bath is also different for the two processes. The chromic process
is usually maintained by steam heat at an operating temperature of 90° to 100°F, whereas the sulfuric acid
process must be chilled using cooling water to an operating temperature range of 45° to 70°F.
26
-------
TABLE 2 SURROGATE RECOVERIES FOR CHARCOAL TUBE BLANKS
Surrogate recovery. %
Blank No.
Laboratory blank
GDP-C1
GDP-C2
GDP-C3
GDP-CB2
1,2-
Dichloroethane
91
79
92
88
88
Surrogate
recovery, %
Toluene
92
82
93
94
92
p-Bromo-
flourobenzene
92
83
95
96
94
Payback for the solvent stills is about 4 years. An even shorter payback will be realized when all
the cleaning solvent can be recycled.
While some of the painting technology discussed here could be too large an investment, many
smaller companies could incorporate parts of the system, such as the electrostatic sprayer and solvent
stills into their operations.
The full report, entitled "Evaluation of Five Waste Minimization Technologies at the General
Dynamics Pomona Division Plant", by Lisa M. Brown and Robert Ludwig.is available as EPA/600/S2-
91/067.
33
-------
Additional cost savings can be realized with the addition of computerized hoists and on-demand
rinse systems. Both of these systems help to reduce water consumption. According to GD, rinse-water
consumption was reduced from 20 to 8 gpm. The cost savings in reduced water consumption was
estimated to be $2300, assuming full-time operation (i.e., 4992 hours per year) and the Pomona City
Water Works rate of $0.65 per 1000 gallons. GD also expects that the computerized hoist system will
lower costs associated with rejects and rework.
Effluent from the detoxification/desmut rinse-tank was chosen for sampling, since it has the
greatest potential of all of the rinse tanks to contain metals. The sample taken represents a composite of
the rinses performed in the deoxidation/desmut tank. The rinse water from this and other rinse tanks in
the anodizing process undergoes pH adjustments and is sent to the wastewater treatment plant to remove
any metals remaining in the rinse water.
The results of the deoxidation/desmut rinse tank sampling are shown in Table 3.
TABLE 3. RESULTS OF METAL SAMPLES TAKEN FROM THE
DEOXIDATION/DESMUT TANK OF THE H2SO4 ANODIZING PROCESS
(mg/liter)
Metal
Aluminum
Cadmium
Chromium
Copper
Lead
Nickel
Silver
Zinc
Deoxidation/desmut
Rinse tank
14.0
<0.003
0.034
<0.020
0.058
0.045
0.008
<0.006
Method detection limit
<0.024
<0.003
<0.004
<0.020
<0.052
<0.031
<0.006
<0.006
The only metal of any consequence in the deoxidation/desmut tank rinse water was aluminum at
14 mg/liter. Subsequent analysis of the wastewater treatment plant effluent shows reduction to 0.35
mg/liter.
Material balance calculations were not possible from the data provided by GDPD and system
suppliers. The automated system has resulted in reduced rinse water, downstream waste treatment
requirements, and improved product quality through reduced part rejections. Computerization of the
system can be extended to the chemical and process parameters, thus further reducing labor costs,
chemical usage, waste products, and further improving product quality. An analysis of the cost of
computerization vs. waste reductions and cost savings would provide more insights on its transferability to
other processes.
28
-------
Painted
Part
Recovered
Plastic Media
Bead
Blaster
Stripped
"*"" Part
Paint Chips
To Incineration
Figure 1. Schematic of production paint shop bead blast paint stripper
A material balance around the bead-blasting unit was not possible from plant operating records
and information made available by plant personnel. It was found that operating records were not kept
with respect to when the unit was used, how many pieces were stripped or the length of time required to
perform the work. The unit is used on an as needed basis and reportedly only a few hours per week.
Air sampling was not conducted because fugitive dust emissions around the unit were reported
to be negligible. This assumption was considered reasonable in that the parts subjected to bead
blasting are enclosed in a glove box apparatus during the paint stripping operation. In addition, the area
around the bead-blast unit seemed free of debris and paint chips. Some fugitive emissions can occur,
however. Failure to have the access door tightly closed during paint stripping, or during loading or
unloading activity or during the cleaning of the stripping chamber and cleanout drawer all have potential
of allowing some dust to escape.
The bead-blast paint stripper was installed at a cost of $18,000. This system eliminated the
disposal of about 10,000 Ib/yr of spent methylene chloride and paint sludge. Although the waste from
the bead-blasting unit has not been quantified, plant personnel report that there is noticeably less waste
when stripping with the bead-blast unit. A recent study* indicated that waste generated from plastic
bead blasting was about one-half the volume as that generated by methylene chloride stripping. Both
the methylene chloride and bead-blast waste are disposed by incineration resulting in waste disposal
costs of $10,000 for methylene chloride and $5,000 for bead-based waste.
Based on the cost savings of $5,000 and the installed capital cost of $18,000 of the bead-blast
unit, the payback period for the cost of waste disposal alone, would be about 3.6 years. Cost
calculations did not include differences in labor hours and system maintenance requirements between
the two systems. Significant cost savings could be realized if the paint being removed by bead blasting
did not produce hazardous waste.
* Balasco A. A., et al. Demonstration Testing of Plastic Media Blasting (PMB) at Letterkenny Army depot. Prepared by
Arthur D. Little, Inc., for U.S. Army Toxic and Hazardous Materials Agency under Contract No. DAAK11-85-0008.
Aberdeen Proving Ground, MD. January 1989.
35
-------
#05 EVALUATION OF FIVE WASTE MINIMIZATION TECHNOLOGIES AT THE GENERAL DYNAMICS
POMONA DIVISION: ROBOTIC PAINTING
Participants
The state of California Department of Toxic Substances Control, Sacramento, CA assisted in
implementing the evaluations. The host for the evaluations was the General Dynamics, Pomona Division
(GDPD) Plant. GDPD personnel operated the painting equipment during the test. PEI Inc., on contract
to EPA helped to design the test program, supplied test personnel and equipment, and drafted the final
report.
Technology/Testing
The GDPD paint production operations facility was completed in December 1988, to replace
manual mixing and hand spraying of metal parts in naval weapon systems. It includes computer-
controlled robots (a GRI OM 5000 unit), which allows quick, automated precision painting, a proportional
paint mixer which feeds preselected quantities of individual paint components directly to the paint spray
nozzle, to avoid batch makeup operations, electrostatic spray guns and automatic paint line, waste
cleaning solvent collection systems to allow for recycle and reuse of waste paint. Spray paint booths
are also available for touchups. Stills are used for recycling paint cleaning solvents. Figure 1 is a
schematic of the robotic paint facility.
The painting facility uses both water-based primer and oil paints. For oil-based paints,
polyurethane thinner is used for paint thinning and equipment cleaning. A thinner containing isopropyl
alcohol and xylene is used with water-based primer.
The objectives of the paint facility evaluation included the following:
Atmosphere
Paints'
Thinners
Cleaning
So (vents
Recycled
Cleaning Solent
Waste Paint
To Ottsrte Incineration
Still Bottoms
To OffsHe Incineration
Figure 1. Schematic of production paint shop painting/recycling operations
30
-------
#07 EVALUATION OF FIVE WASTE MINIMIZATION TECHNOLOGIES AT THE GENERAL DYNAMICS
POMONA DIVISION: FREON RECOVERY FROM DECREASING
Participants
The state of California Department of Toxic Substances Control, Sacramento, California assisted
in implementing the evaluations. The host for the evaluation was the General Dynamics, Pomona Division
(GDPD) Plant. PEI Inc., on contract to EPA helped to design the test program, supplied test personnel
and equipment, and drafted the final report.
Technology/Testing
Three Freon recovery stills, manufactured by Recyclene, were installed in December 1988 to
collect and distill waste from solvent degreasing operations throughout the GDPD facility at a cost of
$240,000 plus $40,000 for add-on equipment to address operating problems. Recovered solvent from
these stills is tested and reformulated under a quality assurance program to ensure that it meets
manufacturer's original product specifications before being returned to material stores for reissue for
production operations. The still bottoms are sent out for incineration.
Prior to the installation of these stills, a single, on-line Freon recovery still was installed in
November 1985 to extend the life of Freon used in conveyorized cleaners. This extended the solvent
changeout period to once per year and saved 35,000 Ib annually in Freon purchases. This still is not
included in the analysis.
Other changes in operational procedures to reduce degreaser solvent changeouts have resulted
in a substantial reduction in Freon purchases. These operating changes were instituted beginning in April
1987 after GDPD completed a survey of all degreaser operations. The program objectives were to
standardize all solvent cleaning operations through employee education, improved operating procedures
and a solvent quality assurance program using vendor recommendations and operator input. The
degreaser survey revealed that the solvent was changed every 1 to 7 days based solely on a visual
analysis by the user. A quality assurance laboratory analysis program was implemented which extended
the time between changeouts to 7 days, then 14 days, and finally 30 days. At the same time, the number
of solvent types used was reduced from five to one over a 6-month period ending in February 1988.
Manufacturing Process Specifications were changed to restrict changeout authorization to the Process
Quality Assurance Laboratory to assure that proper justification for changeouts was provided. To further
reduce the changeout frequency, a moisture removing desiccant was added to each degreaser which
resulted in an extended use period of 30 days. A standardized operating procedure was developed to
reduce evaporative losses based on manufacturer recommendations and was included in all operating
instructions as well as posted on signs at each unit. Operator training was completed by all user
departments, equipment operation checked, and preventive maintenance procedures updated.
Standardized equipment specifications were also developed to further reduce evaporative emissions on
new degreasers and ensure compliance with California South Coast Air Quality Management District
regulations.
The 1988 level of 421,000 Ib. of Freon purchases provides the baseline for the technical and
economic evaluations that follow.
The objectives of the Freon recovery stills evaluation were to:
37
-------
TABLE 1. SUMMARY OF AIR SAMPLING PERFORMED IN THE
GENERAL DYNAMICS PAINT SHOP
Charcoal Tube 1 Charcoal Tube 2 Charcoal Tub
Location
Distance from floor, ft
Quantity in sampled air,
1,1,1-Trichloroethane
Trichlorotrifluoroethane
Concentration, mg/m3
1,1,1-Trichloroethane
Trichlorotrifluoroethane
8-hour TWA, mg/m3
1,1,1-Trichloroethane
Trichlorotrifluoroethane
TWA limits, mg/m3
1,1,1-Trichloroethane
Trichlorotrifluoroethane
On table near two paint On lower table
booths and exterior door near small booth
3.5
6
0
0.3111
0.0000
2.488
0.000
1900
7600
1
10
100
0.2145
2.1447
1.716
17.158
On floor near
robotic paint booth
0
26
100
1.5647
6.0180
12.518
48.144
«• ..j f°r< lue a'r samP'es' surrogate recoveries from compounds spiked on charcoal tubes from the
field and a laboratory blank were all within the target range of 60-145 percent as shown in Table 2.
The paint facility with proportional paint mixing, robotic controlled, electrostatic spray quns and
cleaning solvent collection equipment appears to be operating well. The main problem area at the time of
testing was the waste solvent collection process. The issue to be worked out is one of separatinq water
from solvent in the case of the water based primer cleaning solvent recycling. This problem had led to a
reluctance to utilize the solvent recycled from the water based primer cleaning because of its poor quality
Only polyurethane thinner solvent from cleaning paint equipment is currently recycled.
^on Aua'nt WaStG was reduced from about 42 tons in 1987 to 31 tons in 1988, and further to 17 tons in
1989. About 1000 gallons of polyurethane cleaning solvent per year was being recycled through the paint
shop solvent stills, during the time of the test, resulting in approximately 60 to 100 pounds of still bottoms
per week, or about 5,000 pounds per year. The still bottoms and waste paint are sent off site for
incineration.
Detailed costs were not available to develop a comprehensive economic evaluation of the waste
minimization methods applied to the production paint shop. The payback period of almost 40 years
calculated on the overall investment of $1.4 million is substantially overstated because labor costs savings
from paint and waste disposal and decreases in reject paints or batches of leftover pre-mixed paints were
not available. The very significant decrease of worker exposure to paint and solvents is also not included
in the calculation.
32
-------
This is a minimuim payback number that is based on recycling all of the Freon that is not lost
through evaporation and dragout. If less Freon is recycled, the payback period would increase but
increase to a maximum of 1 year. Notwithstanding the technical problems that have been encountered in
recycling the Freon, this appears to be an economical waste minimization technique.
The full report, entitled "Evaluation of Five Waste Minimization Technologies at the General
Dynamics Pomona Division Plant," by Lisa M. Brown and Robert Ludwig, is available as EPA/600/S2-
91/067.
39
-------
Participants
The state of California Department of Toxic Substances Control, Sacramento, California assisted
n^nting^evaluati°ns- The host for the evaluation was the General Dynamics, Pomona Division
(GDPD Plant. GDPD personnel operated the bead blast equipment during the test. PEI Incorporated on
contract to EPA, helped to design the test program, supplied test personnel and equipment, and drafted
the final report.
Technology/Testing
The plastic bead-blast paint stripper at GDPD was installed in June 1 988 to replace methylene
chloride stripping. Reusable plastic beads are used in this mechanical stripping operation, which is similar
to sand blasting. Paint is stripped from the hangers used to hold parts being painted in the paint shop and
from parts having paint defects.
The plastic bead-blasting booth is a Pauli and Griffin Pram Machine approximately 3 ft by 3 ft bv
3 ft., glove box and uses size 20- to 30-mesh Poly Plus beads. The unit is used only on an as-needed
basis, generally a few hours per week.
Waste generated during the operation of the plastic bead blasting unit consists primarily of paint
chips and a small amount of spent plastic beads which are sent off site for incineration Stripping bv
methylene chloride resulted in about 10,000 pounds per year of toxic solvent contaminated with paint
sludge, which also was sent off site for incineration. Figure 1 is a schematic of the bead-blast paint
stripper. K
The objectives of the plastic bead blasting evaluation included the following:
1) Establishment of a material balance around the bead blasting operation, if possible using
plant-supplied data on input and output of plastic beads and paint. The paint chips 'are mixed
in with the other wastes at the plant before incineration, and preliminary indications from
General Dynamics were that the amount of paint chips sent out for incineration would be
difficult to ascertain. If so, it was decided that an estimate of paint chips discarded annually
would be made after interviewing plant personnel.
2) Fugitive dust emissions around the bead blasting operation were reported to be negligible
during operation, as the system is completely closed. Thus, no air sampling was planned.
3) An economic analysis of the bead blaster was planned. Annual operating costs, which ar
primarily the cost of incinerating the paint chips vs. the cost of disposing of methylene
chloride, were to be calculated.
34
-------
Pollution Prevention Potential--
Pollution prevention was measured in terms of waste volume reduction (Table 1) and pollutant
reduction (Table 2). The total waste volume generated by the automated washer is much lower than
either the alkaline tumbler or hand-aqueous washer.
The processing energy requirement of the automated washer is higher than the energy
requirement of any of the three older processes. The moderately higher processing energy requirement
of the automated washer should, however, be weighted against the potentially higher energy requirements
of the older processes in such other areas as waste treatment. Secondary pollution resulting from energy
generation was not a part of this evaluation.
Although the waste volume generated by the vapor degreaser is lower than that of the automated
washer, it is considered more hazardous.
Perchloroethylene is a high risk pollutant and a health problem, with inhalation and skin the main
entry routes.
TABLE 1. COMPARISON OF ANNUAL WASTE VOLUME FROM
THE THREE CLEANING PROCESSES
Wastestream
Vapor Degreasing3
Wastewater in separator
Still bottom sludge
Air emissions
Alkaline Tumblingb
Wastewater
Volume Generated
Per Year
(gai)
200
1,440
see Table 2
1,010,880
Wastestream
Automated Washing3
Wastewater
Oily Liquid
Automated Washingb
Wastewater
Oily Liquid
Volume Generated
Per Year
(gai)
143,000
962
85,800
577
Hand-Aqueous Washing0 Automated Washing0
Wastewater 296,400 Wastewater 57,200
Oily Liquid 385
a Based on 5,200 bbl/yr run on automated washer instead of vapor degreaser.
b Based on 3,120 bbl/yr run on automated washer instead of alkaline tumbler.
0 Based on 2,080 bbl/yr run on automated washer instead of hand-aqueous washer.
Spent perchloroethylene is listed under Resource Conservation and Recovery Act (RCRA) as a
hazardous waste (EPA hazardous waste number F001). Other commonly used degreasing solvents are
methylene chloride, 1,1, 1-trichloroethane and trichloroethylene, all of which are considered highly
hazardous and on EPA priority lists for waste reduction. The use of the automated washer overall
reduces the use of these solvents.
The automated washer generates a wastewater containing surfactants, which are a lower hazard
both in terms of occupational safety and the environment. Surfactants are not RCRA hazardous wastes.
They can, however, cause environmental problems.
41
-------
Variables due to shape and size of parts being stripped, type of paint, hazard characteristics of
paint ingredients, operator skills needed, worker safety, are all factors that can have a significant impact
on economics. Because of the unavailability of operating data and the infrequent use of the bead blasting
unit, a comprehensive economic evaluation could not be developed. Overall, the technology is relatively
inexpensive and is easily transferred to other industries and small companies.
From a pollution prevention standpoint, elimination of the generation and disposal of a toxic and
volatile solvent waste and associated air releases and worker exposure appears to have a significant net
benefit. While the reduced volume of waste from the bead blasting operation is incinerated as a
hazardous waste for this application, other applications for which the paint being removed is not
considered hazardous would benefit by totally eliminating the generation of hazardous waste.
The full report, entitled "Evaluation of Five Waste Minimization Technologies at the General
Dynamics Pomona Division Plant", by Lisa M. Brown and Robert Ludwig, is available as EPA/600/S2-
91/067.
36
-------
Cost Element
Labor
Energy
Chemicals
Water
Onsite Waste Treatment
Offsite Waste Disposal
Total
Alkaline Tumbling Cost3
($/Yr)
18,720
2,847
2434
4,700
7,299
0
36,000
Automated Washing Cost3
($/Yr)
10,380
6,427
1,626
399
619
1,574
21 ,025
aBasedon3,120bbl/yr
Reducing the amount of solvent used can also reduce possible liability resulting from health
claims or pollution fines, but these savings were not quantified by this study and were not in the economic
calculations.
Automated washing reduces the volume of wastewater that has to be treated (either onsite or at
the publicly owned treatment works) and discharged downstream. This is done without compromising the
cleaned product quality, and no additional skill is required. Parts cleaned in the automated washer can be
either electroplated or sent out as finished products.
One current limitation is that the automated washer cannot yet totally substitute for the three older
processes. Certain delicate parts have to be cleaned by more gentle means. Some difficult-to-clean parts
have to be processed through the hand-aqueous washer. Most jobs that can be run on the older
processes can, however, be routed through the automated washer. Thus, the automated washer is a
good technology for metal finishers who are considering an expansion in capacity.
TABLE 5. OPERATING COSTS OF HAND-AQUEOUS WASHING VS.AUTOMATIC WASHING
Cost Element Hand-Aqueous Washing
Cost3
($/Yr)
Labor
Energy
Chemicals
Water
Onsite Waste Treatment
Offsite Waste Disposal
Total
16,640
3,256
33,134
1,213
2,140
0
56,383
Automated
($/Yr)
6,920
4,285
1,084
266
413
1,050
14,018
Based on 2,080 bbl/yr
43
-------
1) Determine the amount of Freon solvent reclaimed by reviewing plant records on Freon
purchases and solvent sent off site before and after installation of the Freon stills. The
volume of still bottoms sent out for incineration and associated costs were also to be
determined, based on plant operating records.
2) Perform economic analysis of the Freon still recovery system capital, annual operating
costs/credits, and payback.
Results
Testing of the Freon recovery stills indicated that the distillation process was working but that the
Freon being recycled was contaminated with alcohol and water, which degraded the quality of the
recovered product. GDPD installed separators and molecular sieves to dry and remove water and alcohol
from the Freon after its distillation and recovery. These additions to each still cost $13,000 plus
installation but allowed the Freon recovery to proceed as planned and produce a quality recycled solvent.
As a result of the control measures implemented, GDPD experienced a 35 percent reduction in
1988 cleaning solvent purchases going from 650,000 pounds to 421,000 pounds.
Reduction in Freon purchases through 1988 have primarily come through improved operation
procedures such as extended changeout times for Freon, and reduction in evaporative losses. A baseline
of 421,000 Ib of Freon for 1988 is used to calculate reductions in Freon usage attributable to the use of the
three Freon stills, which are assumed to be utilized at full capacity for the first time in 1989. The following
calculations were performed to calculate simple payback for the installation of the three Freon stills:
TABLE 1. ESTIMATED PAYBACK FOR FREON STILLS
Item Lbs.
$
Capital cost: $270,000
Annual operating costs:
Estimated 1989 Freon production adjustment @15%: 357,850 Ib
Evaporative and dragout Freon losses @37.5%: 134,194 Ib
Amount of Freon processed for recycling: 223,656 Ib
Still bottoms from Freon recycling @5% 11,183 Ib
Avoided purchases (amount recycled) of Freon: 212,473 Ib
Cost savings from avoided Freon purchases @$1.64/lb
$348,456
Cost of still bottoms incineration @$500/drum & 715 Ib/drum:
$7820
Cost savings from avoided Freon disposal: $148,583
Total cost savings: $489,219
Payback: $270,000/$489,100 = 0.55 years
38
-------
The ink in the distillation still is sent through 100- and 325-mesh filters to remove paper dust and
transferred to a blending tank. At this point, a grind test and a drawdown test are performed, and the
amount of virgin black ink, required for blending,is determined (typically three to four times the amount of
the processed ink). The virgin ink is added to improve the color, consistency, and other functional
properties of the processed ink to an acceptable range. The processed ink, blended ink, is called the final
"recycled" ink.
Approximately 175 gal. of waste ink is generated each week. Previously it was sent to a vendor
location where it was blended with other solvents to create a supplemental fuel. Since October of 1990,
waste ink has been recycled on-site and reused for printing.
For the test, two batches of waste ink were processed through the recycling unit and samples of
the waste and the recycled ink were collected for analysis. Samples of the virgin ink used at The Courant
were also collected and analyzed. A comparison of the analytical results of the waste and recycled inks
indicates the improvement achieved by the recycling process, and a comparison of the recycled and virgin
inks indicates how closely the recycled product approximates the virgin product.
Results
The results of the product quality analyses are shown in Table 1. The recycled ink fared well in
most of the analyses. The viscosity, as measured by ASTM D 4040-89, of the recycled ink was within ±1
Poise and in the normal range for newspaper inks. The grind (ASTM D 1316-87) and residue (U.S.
Printing Ink Method #12) analyses indicated that some very fine particulates were retained in the recycled
ink, although this did not cause any problems in the printing process at The Courant. Tack (ASTM D 131-
89) was measured at speeds specific to The Courant (1200 rpm at 1 min for web-fed inks). One sample
was slightly above industry recommendation and the other was within this standard. Press operators at
The Courant did not think that the sample that was slightly out of range was of significant concern.
Relative tinting strength was measured by a method similar to ASTM D 387, D 2745, and D 4838.
One sample was slightly out of range. Since the recycled ink is blended with virgin ink, the ratio of virgin-
to-processed ink could be increased to improve the tinting quality of the ink. Water content (ASTM D
1744-83) analyses showed that most of the water is removed in the recycling process. No industry
standards are indicated for this because it depends on the individual printing process. Operators at The
Courant observed no problems resulting from water. Water pickup (ASTM D 4942-89) analyses
determines the emulsifying capability of the ink. This parameter also varies with the printing process, nd
the recycled ink results posed no problems.
The visual effect and behavior of the recycled ink, once it is printed on a newspaper was
evaluated by (a) densitometer readings of black image areas of newspapers printed with virgin and
recycled inks and (b) analysis by 11 experienced viewers of newspaper pages printed with recycled or
virgin inks. The densitometer measurements show that the recycled ink was much denser than the virgin
ink on the wrapper, or exterior pages, of the newspaper. The virgin ink was only slightly denser than the
recycled ink on the core, or interior pages. The results of the visual judging (Table 2) showed that the
newspapers printed with recycled ink were of comparable quality to those printed with virgin ink.
45
-------
CONNECTICUT
The five (5) technologies evaluated for Connecticut included four dealing with metal cleaning and
plating industries and one (1) with the recycling of newspaper ink. Rita Lomasney of the Connecticut
Technical Assistance Program was in charge of coordinating the WRITE program for the state. Summer
Kaufman of ESSAR Environmental Services was the Connecticut Hazardous Waste Management Service
consultant and project officer. Lisa M. Brown was the project officer for EPA.
#08 AN AUTOMATED AQUEOUS ROTARY WASHER FOR THE METAL FINISHING INDUSTRY
Participants
The Connecticut Hazardous Waste Management Service assisted in implementing the
evaluations. The host for the evaluation was the Quality Rolling and Deburring (QRD) Company of
Thomaston, CT. Battelle, Columbus Laboratories, on contract to EPA, helped design the test program,
supplied test personnel and equipment, and wrote the draft report.
Technology/Testing
One of the major steps in metal finishing is cleaning metal parts to remove oil and grease, dirt,
and metal chips. Cleaning may involve washing with a detergent or degreasing with a solvent. Before
installing the automated aqueous washer, QRD routed metal cleaning jobs through one of three cleaning
processes: vapor degreasing, alkaline tumbling, or hand-aqueous washing.
The product quality evaluation was based on (a) an examination of the cleaned metal parts from
some of QRD's normally scheduled cleaning jobs, (b) an examination of cleaned test panels from the
alkaline tumbler and automated washer.
Three cleaning jobs (each containing several thousand small metal parts) were selected from the
several that QRD receives every day. Job-A consisted of steel caplets that were suitable for cleaning on
the vapor degreaser. Job-B consisted of aluminum rivets that were suitable for cleaning on the hand-
aqueous washer. Job-C consisted of steel cylinders that were suitable for cleaning on the alkaline
tumbler. Job-C involved cleaning as well as electroplating (nickel plating).
Each job was split into halves. Half was cleaned by the automated washer, and the other half by
one of the older processes to provide a one-to-one comparison between the automated washer and each
of the three older processes.
After each cleaning run, a predetermined number (150) of randomly selected, cleaned parts were
examined for product quality.
Results
Performance--
Visual examination revealed no notable contamination on any of the parts for all three jobs nor on
the cleaned test panels. The waterbreak test indicated that the parts designed for electroplating had been
adequately cleaned.
40
-------
The Courant generates approximately 175 gal/wk, or 9,100 gal/r of waste ink. This waste ink
consists of 5,460 gal of ink, 546 gal of solvent and 3.049 gal of water. Recycling reduces 6,006 gal of ink
and solvent waste. The Courant is also considering installing an activated carbon filter for polishinq off
organics in the wastewater from the separator so that the water can also be reused.
The waste ink at The Courant has been tested and is not considered a hazardous waste per
Resource Conservation and Recovery Act (RCRA) regulations and can be disposed of according to state
regulations for oily wastes. Solvent washes for other inks that contain lead or chromium in their
formulation are, however, listed as hazardous wastes (EPA Waste Number K086) under RCRA In
addition, other waste inks could contain constituents that render them flammable or toxic.
The economic evaluation took into account the capital and operating cots of the recycling
equipment, as well as the savings resulting from reduced amounts of raw materials (virgin ink and solvent)
and disposal costs. A return on investment of about 9% is obtained in the tenth year of recvclina With a
payback period of about 10 years for the $318,000 capital requirement, the recycling equipment tested
here is a large investment, even for a medium-to large-size newspaper such as The Courant.
The waste ink recycling evaluation demonstrated that the potential for waste reduction with ink
recycling is promising The Hartford Courant reduced waste volume from over 9 000 gal of waste ink to
approximately 46 gal of paper dust and 3,049 gal of wastewater per year. The recycled product fared well
in both product quality testing of the recycled ink and quality of the actual printed material The slight
deviation of some recycled ink test results from the industry standard did not cause any noticeable
reduction in print quality. The blanket wash solvent in the waste was also recovered and reused
> ent'tled "°n~Site Waste lnk Recyclin9" by Ar"n Gavaskar, et al., is available as
47
-------
Different surfactants vary widely in terms of aquatic toxicity and ease of biodegradation.
Surfactants accumulate within aquatic organisms and impair their functions. When compared with
alkaline tumbling or hand-aqueous washing, the automated washer generates lower amounts of these
surfactant wastes.
The results of the study indicate that measurable pollution prevention accrues from using the
automated washer instead of any of the three older cleaning processes.
Tables 3, 4, and 5 compare the operating costs of the older cleaning processes and those of the
automated washer. The results of the economic calculations showed that, based on a capital requirement
of $207,260, the payback period for QRD (where the ROI exceeds 15%) was about 7 yrs.
TABLE 2. POLLUTANTS GENERATED BY THE THREE CLEANING PROCESSES
Pollutant
Vapor Degreasing'
Perchloroethylene
Perchloroethylene
Perchloroethylene
Alkaline Tumbling"
Anionic surfactant
Hand-Aqueous Washing"
Non-ionic surfactant
Medium
Sludge
Water
Air
Water
Water
Amount
Generated
Per Year
(b)
45
negligible
6,145
43
105
Pollutant
Automated Washing*
Anionic surfactant
Non-ionic surfactant
Automated Washing"
Anionic surfactant
Non-ionic surfactant
Automated Washing0
Anionic surfactant
Non-ionic surfactant
Medium
Water
Water
Water
Water
Water
Water
Am- unt
Generated
Per Year
(b)
2
22
1
13
1
9
8 Based on 5,200 bbl/yr run on automated washer instead of vapor degreaser.
b Based on 3,120 bbl/yr run on automated washer instead of alkaline tumbler.
c Based on 2,080 bbl/yr run on automated washer instead of hand-aqueus washer.
TABLE 3. OPERATING COSTS OF VAPOR DEGREASING VS. AUTOMATED WASHING
Cost Element Vapor Degreasing Cost3 Automated Washing
($/YR) Cost3
($/Yr)
Labor (base rate)
Energy
Chemicals
Water
Onsite Waste Treatment
Offsite Waste Disposal
Total
13,866
2,943
1,795
• 0
Negligible
1,440
20,044
17,300
10,712
2,711
655
1,032
2,614
35,044
Based on 2,080 bbl/yr
42
-------
Figure 2 shows the chromium system configuration. The primary ion exchange resin is anionic
to remove hexavalent chrome. In the future, a cationic resin component will be added to the primary
resin to remove any trivalent chrome that may be present in the rinsewater. The anionic resin is a also
regenerated with a 15 to 20% NaOH solution. The resulting solution (sodium chromate) is run through a
secondary (cationic) exchange unit that is designed to convert the regenerant back to chromic acid and
return it to the plating tank.
Workpie
'*••-. s'*'Pump Make-up
k-'VXS^^fc^^^
j— M Chromium
i \ Plating Tank
t
Recovered }
Chromium *
Secondary Cation
Exchange
1
ro Chrome
Reduction +
Wastestream
IV s^ *'i
— — !i^^
41
^^i 1
ft/hsa / 1 Rinse 2 \
n—r*-
i
i •
-* fl
w
/v/rw
Cartndge
^
Primary
Anion
Exchange
1 " 1 1
4
Regonermnt
Water
Legend
—•— Rinse Water
....*..... Workpieco
-_>.__ Regenerant
Figure 2. Ion exchange recovery of chromium from plating risnewater
The test included collecting three samples each of the rinsewater, before and after passing
through the ion exchange system. These samples were analyzed in the laboratory to evaluate the
removal of contaminants.
Results
The results of the laboratory analysis of the cadmium rinsewater samples showed that most of
the cadmium and cyanide were removed by ion exchange - in some cases, to below detection levels.
The Ph of the rinsewater remained steady at alkaline levels throughout the testing.
A statistical t-test (95% significance level) was performed based on the averages and standard
deviations of the 1-min "before" and "after1' (CD-X1) data. Suspended solids levels were very low in both
"before" and "after" samples. After ion exchange, concentrations of cadmium, cyanide, and iron in the
rinsewater decreased significantly. Overall dissolved solids levels also showed a significant decrease
after ion exchange. This indicated a decline in dissolved mass levels.
Interestingly, conductivity did not show any significant change after ion exchange indicating that
the current-carrying capacity of the rinsewater did not change. During ion exchange, heavier ions
(cadmium, iron, etc.) transfer to the resin and lighter sodium ions are transferred to the water. Thus,
dissolved mass in the water decreases but conductivity remains relatively constant. Small amounts of
fresh makeup water were added to the rinsewater loop from time to time to compensate for the water
lost to evaporate and dragout with the parts; this also helped maintain conductivity.
49
-------
The full report, entitled, "An Automated Aqueous Rotary Washer for the Metal Finishing Industry",
by Arun Gavaskar, et al., is available as EPA/600/R-92/188.
#09 ON-SITE NEWSPAPER INK RECYCLING
Participants
The Connecticut Hazardous Waste Management Service assisted in implementing the
evaluations. The host for the evaluation was the Hartford Courant. Courant personnel operated their
equipment during the test period. Battelle, Columbus Laboratories, on contract to EPA, helped design the
test program, supplied test personnel and equipment, and wrote the draft report.
Technology/Testing
The recycling process is shown in Figure 1. The major components of the recycling unit were
purchased from Separations Technologies Inc. Other equipment was added as required. Trays
containing waste ink (consisting of 75% black and 25% colored ink) from the press room are emptied on a
1/4-in. wire mesh to remove nuts, bolts, and other gross contaminants. The waste ink was then placed in
a large waste ink storage tank. Once enough ink was collected, the batch was processed. Processing
involved vacuum distillation, filtration, and blending.
Waste ink from the storage tank was transferred to the still and distilled at 140°C under vacuum.
Solvent and water from the waste ink are vaporized, condensed (by a chiller), and collected in a separator
tank where where the condensate forms water and solvent phases under gravity. The water is drained off
and discharged to the municipal sewer under permit, and the solvent is reused in cleaning the presses.
Irt Urgor P**** — — — —
1 V4 in.\Wn Uesh Vacuum
1 I X
X
Wu»h* I
StOf^M I ,
3bc*: Recycled Ttnk \ ~ "]
Ink to — DittfflttKKi
Pn&t Room \SHI
^1
Jl "
rota
«•*»
rnm .
™ 1 '
1 ' ** ~L ^ "" 1
i i Btenang 1 I
r*>* *-• ' —
^ (RtcycMW \ Fjr»
Hfteft
u — — — .
I
1
,
Condenser i
1 1
_. Jo A0UM •* — . . _ _ ""
To
*~l
1
1
|
> flB*i±» |
1
J*~~**
Figure 1. Waste Ink Recycling Process
44
-------
Prior to ion exchange, approximately 69 Ib. of cadmium and 281 Ib. of the cyanide were
discharged annually. Now, because most of the cadmium can be recovered and reused, this pollutant is
virtually eliminated from the waste stream. Some cyanide is also destroyed in the cadmium recovery
process.
On the chromium line, without ion exchange, approximately 80 Ib. of chromium is discharged
annually. With ion exchange, most of the chromium will be captured on the resin, which will be
regenerated with NaOH. The regenerant then will pass through a cation exchange resin from conversion
of sodium chromate to chromic acid. However, when this recovery was performed during the pilot unit
testing, the final regenerant liquid still showed a pH of 13.08. This indicates that sodium chromate had not
been converted to chromic acid. However, when this recovery was performed during the pilot unit testing,
the final regenerant liquid still showed a pH of 13.08. This indicates that sodium chromate had not been
converted to chromic acid; if it had been, the pH would have been much lower. This may be because (a)
an excess of NaOH was used to regenerate the resin and/or (b) insufficient resin was available to
exchange all the sodium in the regenerant. Further testing is needed to determine the feasibility of the
chromic acid recovery process.
The economic evaluation, comparing the costs of the ion exchange operation with those of the
former practice (counterflow rinse*) are summarized in Tables 2 and 3. The cost of cadmium anodes is
approximately $15/lb, making the value of the 69 Ib/yr of recovered cadmium approximately $1,036/yr.
The chromium deposited on the ion exchange resin also has value if it can be successfully
recovered as chromic acid. The cost of chromic acid is $2.50/lb. Approximately 80 Ib/yr of chromium
metal is deposited on the ion exchange resin. This corresponds to about 154 Ib. of chromic acid (CrOg).
However, further testing is needed to establish the feasibility of chromic acid recovery from the chromium
in the regenerant.
The purchase price of the cadmium ion exchange system was $8,100. The EMR equipment price
was $4,125. Installation costs at Torrington, including materials and labor, was approximately $3,500, to
which $5,000 was added to approximate the cost of in-house pilot testing to determine specifications for
the individual plant.
The purchase price of the chromium ion exchange system is estimated to be $8,200. Installation
cost at Torrington is expected to be $3,500, including materials and labor. The approximate cost of
$5,000 for in-house testing was also added for this unit.
For the cadmium ion exchange system, the return on investment (with cost of capital equal to
15%) was less than 1 year. For the chromium process, the return on investment (with capital at 15%) was
over 5 years. Because chromic acid recovery from the regenerant is yet to be established, no recycled
chromium value is assumed. These calculations take into account taxes, depreciation, inflation, etc. and
is based on the worksheets provided in the Facility Pollution Prevention Guide (EPA 600/R-92/088. The
costs include engineering and installation as well as increased overhead.
51
-------
TABLE 1. ANALYTICAL TESTS
Batch Sample Type
No.
1 ,2 Waste Ink
1 Recycled lnkd
2 Recycled lnkd
Virgin Ink
Industry
Viscosity
(Poise)
NAC
19
21
20
--
Grind (mil)
4/1 Oa
NA
0.4/0.3
0.6/0.3
0.3/0.0
<0.4/<0.2
Residue
%
NA
0.0817
0.0735
0.0019
<0.01
Tack (gram-
meter)
3.4
4.4
3.9
4.0
3.7-4.3
Tinting
Strength
o/ b
69
96
92
100
>93
Water
Content %
23.6
0.102
0.049
0.057
-
Water
Pickup %
NA
86
80
50
-
4/10 refers to 4 or 10 scratches at reported endpoints.
Strength of recycled ink was compared to the virgin ink and given as a percentage of the virgin ink strength.
NA=Not analyzed. Tests could not be performed because of the large amount o water in the sample.
Processed ink blended with virgin ink in the ratio of 1:3.
TABLb 2. RESULTS OF VISUAL JUDGING3 FOR PRODUCT QUALITY
Parameter
Glossiness
Smoothness
Opacity
Rub Resistance
Blackness
Absorption/
Bleed-Through
Sharpness
#Viewers
Preferring
Virgin Ink
0
0
0
3
0
2
1
Wrapper Page
(outer)
#Viewers With
No Preference
or Preferring
Recycled lnkb
11
11
11
8
11
9
10
Upper 95%
Confidence
Bound on the
Proportion
Preferring
Virgin Ink
0.238
0.238
0.238
0.564
0.238
0.470
0.364
# Viewers
Preferring
Virgin Ink
2
4
4
1
4
1
3
Core Page
(inner)
# Viewers With
No Preference
or Preferring
Recycled lnkb
9
7
7
10
7
10
8
Upper 95%
Confidence
Bound on the
Proportion
P referring
Virgin Ink
0.470
0.650
0.650
0.364
0.650
0.364
Eleven experienced viewers of newspapers.
Processed ink blended with virgin ink in the ratio 1:3.
46
-------
TABLE 3. OPERATING COSTS COMPARISON FOR CHROMIUM SYSTEM
Item
Without Ion Exchange
Fresh waster
Wastewater treatment
Total
With Ion Exchange
Freshwater
Chemicals (50% NaOH)
Energy
Labor
Routine maintenance
-filters
-labor
Waste Disposal
-regenerant
-filters
Total
Amount Used
per Year
480,000 gal
1 ,920,000 gal
12,500 gal
240 gal
1,492 kWhr
149hr
12
24
840 gal
6
Unit Cost
$0.70/1 ,000 gal
$15/1 ,000 gal
$0.70/1 ,000 gal
$1 .50/gal
$0.075/kW hr
$7/hr
$5
$7/hr
$15/1 ,000 gal
$400/36 units
Total Annual
Cost
$336
$7,200
$7,536
$9
$360
$112
$1,043
$60
$168
$13
$67
$1,832
The full report, entitled "Cadmium and Chromium Recovery from Electroplating Rinsewaters" by
Arun Gavaskar, et al., is available as EPA/600/R-94/050.
53
-------
#10 CADMIUM AND CHROMIUM RECOVERY FROM ELECTROPLATING RINSEWATERS
Participants
The Connecticut Hazardous Waste Management Service assisted in implementing the project.
The host for the evaluation was the Torrington Company of Torrington, CT who also operated the ion
recovery unit. The ion exchange system was manufactured by CTEO Tek, Inc. and supplied by Plating
Services Inc. Battelle, Columbus Laboratories, on contract to EPA, helped design the test program,
supplied test personnel and equipment, and wrote the draft report.
Technology/Testing
The objectives of this study were to evaluate (1) the effectiveness of the ion exchange unit in
cleaning the rinsewater for reuse in the rinse tanks, (2) the pollution prevention potential of this
technology, and (3) the cost of ion exchange versus the cost of the former practice (disposal).
Figure 1 shows the cadmium ion exchange system configuration. The water from Rinse 1 tank
is first passed through a filter to prevent suspended solids from contacting the resin in the ion exchange
column. The anionic resin captures the cadmium-cyanide complex, and the water is returned to the
Rinse 2 tank. An emergency bypass valve allows this water to be discharged to waste in case cadmium
or cyanide levels are found to be too high.
The resin is periodically regenerated with a 15 to 20 % NaOH solution, and the regenerant is
taken to the electrolytic metal recovery (EMR) unit, where cadmium is recovered on the cathode and
returned to the plating tank. Some cyanide is destroyed by decomposition during the EMR process.
Workpiecs
•••..t s''Metering '\
\ /Pump Make-up \ /
M •*-
Make-up Water
'f
L,
.-•• H Cadmium
\ | Plating Tank
i.
Recovered
Cadmium
\
L
— •— s
1
iSi
muni
1 '
Still Rinse
(Dragout)
Emergency Bypass to
Cyanide Wastestream
\ L
r
EMR Unit
V
Rinse
*
tit
n
Can
1
CA
/ 1 Rinse
i
-i
A«
to
r~
\2
[--r*—
]
r
**
"I! t
Ion
Excbtng*
r •« _j
1
1
Ffog*n*r*nt \
' T»nk \
WJ
Legend
—>— Rinse Water
...>.... Workpiecs
—».-. Regenerant
Figure 1. Ion exchange recovery of cadium from plating risnewater
48
-------
The ions present in the rinsewater are transferred across the membranes into the concentrate
loop. This concentrate is periodically blended into the concentrate tank. Steam coils present in this tank
raise the temperature of the concentrate, which is then sent to an atmospheric evaporator (fan). The fan
drives off some of the water. The concentrated nickel stream from the atmospheric evaporator is sent to
the plating baths where the nickel is reused.
There are two nickel-plating lines at APB. Each line has its own plating bath and rinse tank. The
two lines and the electrodialysis unit are operated for two shifts (10 to 16hrs) per day. Some makeup
water is added daily to the system to make up for evaporative and other system losses. Testing for this
study was conducted over one complete shift. Samples of the water were collected from the various
locations, and water meters were installed as shown in Figure 1 to estimate water usage.
To evaluate the performance of the overall recovery system, periodic water samples were
collected from the two nickel rinse tanks and the rinse recycle tank. To evaluate the internal profile of the
recovery unit, periodic samples were collected from the dialysis feed tank, stack outlet water, and the
stack outlet concentrate. To evaluate the suitability of returning the concentrate to the plating baths
periodic samples were collected from the concentrate tank and the two nickel plating baths.
Results
Nickel concentration was reduced from nearly 3,000 mg/L in the rinse tanks to less than 100 mg/L
in the stack outlet and rinse recycle tank water. Other contaminants are reduced similarly. Both cation
and anion levels (total dissolved solids) were reduced by electrodialysis. The nickel concentrations in the
rinse tanks were contained below 3,000 mg/L by the balance between the contaminated dragout and the
relatively clean recycled water entering the tanks. Additional contaminant control is brought about by the
addition of approximately 60 gal/day of fresh water to make up for evaporative losses.
As much as 90% of the nickel is removed from the water loop as the water makes a pass from the
dialysis feed tank to the stack outlet water stream. The rest of the nickel remains in the rinsewater
returned to the rinse tank. Thus, all the nickel is recovered and reused. The nickel removed from the
water loop accumulates in the concentration loop.
On the test day, the evaporator was not functioning properly, and the nickel level in the
concentrate could not be brought up to that in the plating baths. Therefore, no concentrate was sent to
the plating baths. After the test day, however, APB did manage to adjust the evaporator and obtain an
evaporator outlet stream that was sent directly to the plating baths. When the nickel level in the
concentrate was increased, there was a concomitant increase in the level of contaminants (iron chloride,
etc.). On the test day; it appeared that with further concentration in the evaporator, the contaminant levels
in the concentrate would be of the same order of magnitude as the levels in the plating baths.
However.indicators of contamination, such as iron or total dissolved solids, needed to be continuously
monitored in the evaporator outlet stream and the plating baths to ensure good bath quality. The
existence of fouling substances that can be inadvertently introduced makes the system vulnerable to
contaminants.
Pollution Prevention Evaluation--
Based on readings taken at water meters installed in the water lines shown in Figure 1, it was
determined that 4,350 gal of water overflowed from the rinse tanks during one shift on the test day and
4,310 gal was returned by the recovery unit. Extrapolating to an annual basis, 2,154,000 qal of water
flows through the recovery unit.
55
-------
The results of the laboratory analysis of the chromium rinsewater samples showed that after ion
exchange, the rinsewater pH levels were slightly alkaline (9.31 to 9.45) because chromate ions (and any
other contaminant anions) had been substituted with hydroxide ions. The alkaline pH was neutralized in
the rinse tanks by the chromic acid residue on the parts (workpiece).
Similar statistical analyses were performed on the chromium data as have been described for the
cadmium. Suspended solids levels were significantly reduced by the cartridge filter ahead of the resin.
Chromium (total chromium) and iron levels decreased significantly after ion change. Iron removal may be
due either to removal of ferrous suspended particles on the cartridge filter or to deposition of complexed
iron on the resin. As in the cadmium tests, dissolved solids mass decreased significantly, but conductivity
(current-carrying strength) remained constant after ion exchange. This is because heavier chromates in
the rinsewater were replaced with lighter hydroxide ions.
Table 1 summarizes the waste volume reduction. By using ion exchange, large volumes of water
are saved from going to waste. This water can be reused as a rinse on the cadmium and chromium lines.
Without ion exchange, Torrington must maintain high rinsewater flow rates (8 gpm for the cadmium line
and 2 gpm for the chromium line). These continuous flows generate large amounts of wastewater that
have to be treated on site. With the ion exchange system on the cadmium line, Torrington requires only
50 gal/day to make up for dragout losses.
TABLE 1. WASTE VOLUME REDUCTION
Without Ion
Waste Description
Cadmium System
Wastewater
Chromium System
Wastewater
Exchange
Amount
Generated
per
Year3
1,920,000
gal
480,000 gal
With Ion Exchange
Waste
Description
Wastewater
Regenerant
Filter cartridges
Wastewater
Regenerant
Filter cartridges
Amount
Generated per
Year15
Ogal
660 gal
6
Ogal
840 gal
12
(a)Based on values of 16 hr/day, 5 days/wk, 50 wk/yr.
(b)Based on pilot tests conducted by the Torrington Company and resin capacity.
In terms of pollutant reduction on the cadmium line, the pollutants of interest are cadmium and
cyanide. Before ion exchange, cadmium in the rinsewater was lost to wastewater, which was sent to an
on-site wastewater treatment plant. The wastewater was treated in a steel cyanide treatment tank using
chlorine gas, sodium hypochlorite, calcium hypochlorite, and NaOH to oxidize the cyanide. The cadmium
and other metals formed hydroxides that settled in the clarifier as sludge, which was then hauled off site
for disposal. The treated water was discharged to the municipal sewer under a permit.
50
-------
The electrodialysis recovery system maintains a continuous supply of acceptably clean recycled
water. Nickel, a hazardous but valuable chemical, is recovered and reused. The maintenance and
energy requirements of the system are relatively high, but the elimination of wastewater treatment result in
annual savings leading to a payback period of 1 year at APB. There are other cost benefits that are not
easily quantifiable (e.g., eliminating nickel from treatment plant sludge and reducing permitting costs) that
may help offset the higher operating costs.
The long-term effects of closing the nickel recovery loop still are a concern, and future research
into the operation of this system is needed to ensure good product quality and to reduce maintenance
requirements.
The full report, entitled "Nickel Recovery From Electroplating Rinsewater by Electrodialysis" by
Arun R. Gavaskar, et al., will be available as an EPA\600 series report.
57
-------
TABLE 2. OPERATING
Item
Without Ion
Exchange
Freshwaster
Waste water treatment
Total
With Ion Exchange
Freshwater
Chemicals (50%
NaOH)
Energy
Labor
Routine maintenance
-filter cartridges
-EMR anode plates
-EMR cathode plates
-labor
Waste Disposal
-regenerant
-filters
Total
=========
Amount Used
per Year
1 ,920,000 gal
1, 920,000 aal
2,500 gal
96 gal
1,564kWhr
173hr
6
1
12
24 hr
660 gal
6
=========^===^^
Unit Cost
$0.70/1 ,000 gal
$22/1 ,000 aal
$0.70/1 ,000 gal
$1.50/gal
$0.075/kW hr
$7/hr
$5
$30
$30
$7/hr
$22/1 ,000 gal
$400/36 units
Total Annual
Cost
$1,344
$42.240
$43.584
$9
$144
$117
$1,211
$30
$30
$360
$168
$15
$67
$2,151
rnmnnn 6*fluatlon «h°wed *at rmsewater on both cadmium and chromium lines at Torrington
Company can be reused after subjecting it to filtration and ion exchange to remove impurities Larae
volumes of water are thus saved, and large amounts of hazardous metal sludge aretept om the
rrrT J^ Sldfhstreams from ™ exchan9e ™ WW* compared wilh the wastewaTer and
sludge wastestreams that are generated under the old method.
52
-------
QRD has operated the chromating line with the recovery system for about 1 year. The recovery
system configuration at QRD is shown in Figure 1. QRD uses three different chromate formulations, blue,
clear, and yellow. The chromating line receives a variety of parts that first are zinc-plated and then
chromated.
Contamination accumulates mainly in the Rinse 1 Tank, which functions as a dragout rinse. The
contaminated water is processed through the recovery unit at an average rate of 1 gal/min.
The recovery system is fully automated and integrated into the chromating line, requiring little
operator attention.
The chromating line operates during 1 shift/day, but the recovery unit remains on 24 hr/day so that
the rinsewater gets additional cleaning after contaminant contribution from the chromating line stops.
Approximately 50 gal/day of fresh tap water is added into Rinse 2 Tank daily to make up for evaporative
losses. Once every 3 or 4 weeks, the still bottoms are evacuated into the sump tank. The sump tank also
stores the rinsate generated when the chromate lines are purged before a formulation switch. When the
sump tank is full, QRD plans to process its contents through the recovery unit and concentrate it down to a
smaller volume. Approximately 200 gal of final concentrate is generated in one year of operation at QRD.
The 200 gal of concentrate, containing chrornates, can be either reused or disposed of as hazardous
waste. QRD has not reused the concentrate because it contains a mixture of three different chromate
formulations. Other plans that use only a single formulation may be able to reuse the concentrate,
provided that contaminants (such as zinc and iron) do not accumulate to the extent that chromating quality
is affected.
Testing for this study was conducted over one shift (6 hours of continuous chromating line
operation), and the following morning before the next day's shift. Samples were collected periodically from
the locations shown by asterisks in Figure 1.
Results
Rinsewater quality was monitored throughout the shift and the results are shown in Table 1.
Water quality in Rinse 1 continues to deteriorate as the contribution from the dragout increases.
Contamination in the Rinse 2 Tank is maintained at very low levels by the periodic influx of clean
processed water from the recovery unit.
Table 1 shows the characterization of the still bottoms and the chromate solutions. The
contaminants that are removed from the rinsewater can be seen accumulating in the still bottoms. From
this table, it can be seen that the still bottoms concentration of chromium is below that in the blue
chromate tank. Additional chromate would have to be added to the still bottoms solution to raise the
concentrate level. If only one formulation were being used, the replenished concentrate could possibly
have been returned to the chromate tank. Plants that reuse the chromate concentrate would have to
carefully monitor dissolved solids levels to ensure that contaminants such as zinc and iron do not cause
the chromating quality to deteriorate.
59
-------
#11 NICKEL RECOVERY FROM ELECTROPLATING RINSEWATER BY ELECTRODIALYSIS
Participants
TK u T,he Cuonnecticut Hazardous Waste Management Service assisted in implementing the evaluation
The host for the evaluation was Automatic Plating of Bridgeport (APB), Inc. Bridgeport CT APB
personnel also operated the plating and recovery equipment. Battelle, Columbus Laboratories on
contract to EPA, helped design the test program, supplied test personnel and equipment, and wrote the
draft report.
Technology and Testing
The goal of this study was to evaluate the technical, pollution prevention, and economic issues
involved in using an electrodialysis recovery system to recover rinsewater and nickel from the nickel-
plating line in a metal-finishing plant. The recovery system configuration is shown in Figure 1
Rinsewater overflow from the rinse tanks is sent to the dialysis feed tank by a level controller' After
electrodialysis, the recycled water is stored in the rinse recycle tank and sent back to the rise tanks A
small amount of sulfuric acid is added to the rinse tanks to remove flash rust after nickel plating From
the dialysis feed tank, water is circulated through the electrodialysis membrane stacks and back to the
feed tank. Part of the cleaned water coming out of the stack outlet is channeled into the rinse recycle
tank for storage. In this water loop, there is a carbon filter to remove organics and a cartridge filter to
remove and carryover carbon particles.
u
ulfurfc
Acid
f
i
-» i
Sour
\ '
Walar
Malar- 1
9
r
Nickel PUUng T*nk
r—
BBM.MM
tmtmmmm
^
L
\
Hlnaa
Racycla
Tank
Concanlrala Looo
,
M
Slack
i
-C\ Walar
Ma'l^-:
f
If
.-I — L
Almoa
i«p.
t
Air
Nlckal PUlIng Tank
Slaam
WaUr
M«lar-2
W.lar Loop
Figure 1. Nickel recovery system
54
-------
The amount of savings realized is highly dependent on the wastewater treatment costs that a
plant would incurs without the recovery unit. Because QRD performs a multitude of finishing operations
its unit cost for wastewater treated ($88/1000 gal) is relatively high.
The recovery system at QRD prevents the generation of large volumes of wastewater and
hazardous sludge thai: otherwise would result from the chromating operation. Further pollution prevention
would be possible at plants that use only one formulation, through potential reuse of the chromium
contained in the still bottoms. Zinc and iron levels would have to be closely monitored if such reuse was
practiced.
This recovery system has potential for use in many applications that generate wastewater. A
negative effect is the relatively high energy consumption of the recovery unit. Future versions of this
system are aiming to reduce this energy requirement.
The full report, entitled "Chromate Recovery from Chromating Rinsewater in the Metal Finishing
Industry" by Arun R. Gavaskar, et al., will be; available as an EPA/600 series report.
61
-------
In order to allow sufficient flow through the process, the rinsewater flow rate at APB when the
recovery unrt ,s ,n operation generally is kept twice as high as it is without the recovery unit The"e ore if
the recovery unrt were absent, the flow would be half as much (or 1,077,000 gal/yr) aN of which would be
Wa "0 S6 th - W3SteWater 'reatment P^nt with9theynickel ending up°n the
on
By using the electrodialysis system, the nickel and water are recovered and reused Based on th
concentrations and flows registered during testing, it is estimated that approximately 2 964 I > of nfckel is
recovered and reused each year when operating 2 shifts per day, 50 weeks per year.
Economic Evaluation-
Table 1 shows a cost comparison for operating with and without the recovery unit Consumotion
and unit cost data were obtained from APB's plant records. The electrodialysis unit has relatively hTah
energy and maintenance costs. The electricity required to maintain the voltage across the s acts is the
mam energy component. Most major maintenance is performed under a service agreement with the
manufacturer. Because wastewater treatment and its associated costs are eliminated thereTs a payback
™ ^^ $ 1 aooo^r the
TABLE 1. OPERATING COST COMPARISON
Items
Without Recoveiv Unit
Water
Chemicals
Wastewater Treatment
Total
With Recovery Unit
Water (make-up)
Energy
- Steam
- Electricity
Maintenance
- Service Contract
- Carbon Change
- Evaporator Cleaning
- Miscellaneous
Total
Annual
Usage
1 ,077,000 gal
188,41 8 lbNiSO4
1,077,500 gal
• ,
15,000
4.9x1 08Btu
78,600 Kwh
4 visits
6 changes
12 cleanings
Upkeep on electrical
units
Unit
Cost
$4. 19/1 000 gal
$1.97/lbNiSO4
$14/1 000 gal
• — • — .
$4. 19/1 000 gal
$6/106Btu
$0.118/Kwh
$3,000/visits
$1 ,020/change
$120/cleaning
========
Total
Annual
Costs
$4,515
$201 ,607
$15,085
$221,207
— — ^— . — . — _^_ __ ___ ^__
$63
$2,940
$9,275
$12,000
$6,120
$1,440
max $1,000
56
-------
Most printing processes begin with a photographic negative. Developing the neaative aeneratP
-------
#12 CHROMATE RECOVERY FROM CHROMATING RINSEWATER IN THE METAL FINISHING
INDUSTRY
Participants
The Connecticut Hazardous Waste Management Service assisted In implementing the
evaluations. The host for the evaluation was the Quality Rolling and Deburring Company (QRD) of
Thomastown, CT. QRD personnel operated the metal finishing and recovery processes during the test.
Battelle, Columbus Laboratories, on contract to EPA, helped design the test program, supplied test
personnel and equipment, and wrote the draft report.
Technology and Testing
The recovery unit tested was manufactured and provided by Cellini Purification Systems, Inc.*.
Similar units with varying capabilities may be available from other vendors.
From
In Electroplating
From Mechanical
Zn Plntlng
Still
Bottom*
Clean
•Clw" \Wattr
Chromate
Tank
* sampling locations
Figure 1. Chromating rinsewater recovery system
58
-------
#14 ALKALINE NONCYANIDE ZINC PLATING AND REUSE OF RECOVERED CHEMICALS
Participants
The P&H Plating Company of Cook County, Illinois provided the test site. The Hazardous Waste
Research and Information Center (HWRIC), a division of the Illinois Department of Energy and Natural
Resources, Champaign, Illinois provided test personnel, designed the test and drafted the final report.
Technology/Testing
This work was undertaken to evaluate the feasibility of using an innovative, closed-loop rinsewater
treatment system to precipitate plating chemicals for recovery and reuse and to produce purified water for
recirculation to the rinsing tanks and sprayers. The goal of this study was to achieve zero discharge of the
wastewater and total recycle of the recovered precipitate, thereby reducing the amount and toxicity of the
waste from a zinc plating operation. The zinc cyanide (CN) plating line was converted to one that uses an
alkaline, noncyanide zinc (ANC) plating bath. A recovery/recycle (R/R) unit was then designed, installed
and tested to determine how completely the goals of the project were being met. The effectiveness of the
R/R system in reducing the process wastes was evaluated by: quantifying the effectiveness of the
removal of the zinc through precipitation by pH adjustment, the basis of the recovery system; determining
the quality of the precipitate and the treated water that were recovered; comparing the plating quality of the
CN-based operation to that of the ANC-based processes that used both the recycled chemicals and the
treated rinsewater; and analyzing the costs associated with the change in the process and the installation
and use of the R/R system. The R/R unit was designed, installed, and tested by engineers from the
Center for Neighborhood Technology (CNT).
The CNT Recovery/Recycle-
The relatively simple principles involved in the design of the CNT R/R system are similar to those
for a standard wastewater flocculation treatment to remove metals before disposal of water and sludge
wastes. Ideally, 100% of the zinc in the CNT R/R system's rinsewater would be recovered and returned to
the plating bath. Additionally, all rinsewater would be recycled. The projected result would be substantial
savings for the company in plating chemicals and water from this rinsewater purification that both recovers
and reuses as well as treats.
In the plating process at P&H, there is the usual parts pretreatment or cleaning, followed by the
plating process, a spray rinse, and finally, submersion into two counterflow rinse tanks. Although cleaning
requirements for ANC plating are generally more stringent than those for CN plating, no change was
required to the pretreatment portion of the line at P&H. The company had already installed a very
stringent cleaning component to their plating lines to ensure good parts cleaning and, presumably, better
plate quality. The CNT R/R system was plumbed from the spray rinse tank into which the counterflow
tanks ultimately overflowed. The rinsewater flows into a tank where the Ph is measured and automatically
adjusted to a Ph between 10 and 10.5.
This Ph monitoring and control tank is a continuous flow stirred reactor (CFSR). It is designed to
stimulate precipitation by sparging with compressed air entering from the bottom of the tank. The flow
rate through the CFSR is set at 10 gallons/minute. The precipitate/water slurry next flows into a flat-
bottomed clarifying tank. To facilitate the settling process, the tank is baffled. A recirculating pump is
used to pull water from the clarifying tank through the dual filtering system to remove suspended
hydroxide. The filtered water than either flows to a storage tank for reuse or to the waste treatment area
for disposal. The precipitate that has settled in the clarifying tank is removed and combined with that
collected on the filters. This composite hydroxide is placed in a filter press to remove as much water as
possible. The water that is removed is returned to the precipitation reactor and the dewatered hydroxide is
analyzed and stored for future use or disposed if not needed. Figure 1 shows the system components
65
-------
TABLE 1. CHARACTERIZATION OF DISTILLATION STILL BOTTOMS AND
CHROMATE TANKS
Sample
No.
Time
Cr
mg/L
Still Bottoms
C-SB-1
C-SB-2
C-SB-3
C-SB-4
Chromate
Blue
Yellow
Clear
9:30 a.m.
12:30 p.m.
3:30 p.m.
8:00 a.m.3
Tanks
2:00 p.m.
2:00 p.m.
2:00 p.m.
162
168
216
224
1,254
1,471
3,860
Zn
, rn.9/k
258
277
385
420
1 ,739
15
0.743
TDS
mg/L
2,280
2,240
3,170
3,610
16,740
3,590
25,110
Conductivity
mhos/cm
2,700
2,500
3,200
4,000
12,000
15,000
28,000
PH
3.52
3.59
3.80
3.95
2.57
1.55
2.87
a Next day.
Table 2 lists the major operating costs with and without the recovery system. At QRD, the
recovery unit generates a savings in annual operating costs of $22,159. The recovery unit was purchased
by QRD for $78,000, incurring an additional $9,000 in installation and auxiliary equipment costs(such as
cooling tower, piping, etc.) A payback was estimated at 4 years.
TABLE 2. MAJOR OPERATING COSTS COMPARISON AT ORD
Item
Without Recovery System
Water (rinse)
Wastewater treatment
With Recover/ System
Water (rinse, makeup)
Water (cooling, makeup)
Water disposal
Energy
- electricity
- steam
Maintenance
Labor
Annual Amount
450,000 gal
450,000 gal
12, 500 gal
12, 500 gal
200 gal
60,950 kW/hr
3,000,000 Ib
-
100hr
Unit Cost, $
4.65/1 ,000 gal
88.00/1 ,000 gal
Total
4.65/1, 000 gal
4.65/1 ,000 gal
350.00/55 gal
0.068/kW hr
0.004/lb
-
8.00/hr
Total
Annual Cost, $
2,093
39.600a
41,693
58
58
1,273
4,145
12,000
1,200
800
19,534
60
-------
Plating Quality Comparison-
Alth°U. gh the r??s.0"s for P|atin9 are sometimes purely ornamental, more frequently it is for
pro 1 , ON
Toxicity Comparison--
oiim- t The fhan9.^to,ANC plating resulted in the reduction in health and environmental risk due to the
el.mmat.on of cyan.de from the process. This is a very substantial risk reduction CN Zing reau res
sbsontn ah n ea
subs Mutton to achieve source reduct.on, as was done for this project, not only reduced pro-ess costs but
also the company's hab.l.ty because of the reduced toxicity of the chemicals being
The exposure of shop workers to toxic chemicals presents the most serious health and safetv
cons,dera,on for the electroplating industry. Although no occupational illness has been"menS for
elec roplat.ng operators, they are routinely exposed to hazardous substances known to ca^se serous
health problems. The use of cyanide, generally considered the most potentially dangerous of the
elec roplatmg chem.cals, ,s carefully monitored and employees are trained to use it properly Combinina
employee education with substitution of less toxic chemicals may provide the lest co'stfy and mos™ 9
^
is
Economic Comparison-
honof t Althou9h redLJCed .jsks maV be sufficient to justify the change from CN to ANC the economic
benefits denved were significant. An economic comparison was calculated to include cost ofmaZ the
change as well as the capital investment required. The economics involved are compiled nT^e 1 An
of son/ ^.akir]9 the SW,ltCh fr°m CN t0 ANC requires disposal of the exitin9 1'800 9a"on CN bath at a cost
of $20/gallon for a total cost of $36,000. This was included as part of the capital investment fo this
project Add.ng the R/R system increases the capital investment to a total of $87,822 The ana ys?s in
Table 2 was prepared usmg 1 992 cost data, an inflation rate of 4%, a discount rate of 7 7% depred^on
schedule of 7 years, and a project life of 10 year, and process operation 8 hours/day 5 da^s/weeTfo 50
ZS yn7h te analHyS'S S,hri that the recyding °ption Provides the 9reater economic bS the
shorter payback period, and the larger return on investment.
67
-------
ILLINOIS
Five technologies were identified and tested with the help of the Illinois Hazardous Waste
Research and Information Center (HWRIC), Dr. Gary Miller, project officier. Two technologies focused on
waterbased inks and printing, three dealt with metal cleaning and plating. The EPA project officer was
Paul Randall.
#13 INK AND CLEANER WASTE REDUCTION EVALUATION FOR FLEXOGRAPHIC PRINTERS
Participants
The MPI Label System Incorporated of University Park, Illinois provided the test site. The
Hazardous Waste Research and Information Center (HWRIC), a division of the Illinois Department of
Energy and Natural Resources, Champaign, Illinois provided test personnel, designed the test and drafted
the final report.
Technology/Testing
MPI wanted to eliminate employee exposure to any liquid or gaseous hazards resulting from
handling of hazardous materials and an equal desire to minimize the possibility of future litigation resulting
from use of such substances. This decision forced each plant to substitute water-based inks for alcohol-
based inks, and at the same time, change from alcohol solvent cleaning agents to aqueous cleaning
agents.
The water-based inks were already available on the commercial market. Implementing this
change, however, required extensive cooperation between the ink manufacturer and the printing plant
because of different printing conditions, changing customer requirements, and changing paper stock to be
compatible with the water-based inks.
Because water-based inks are easier to remove or clean with aqueous agents when wet, a
terpene-type (d-limonene) cleaner was initially tried. Later, a dilute aqueous solution of detergent proved
to be even easier to handle, odor free, and less expensive. MPI reported that it is their standard practice
to try new cleaning agents as they are introduced to the commercial market by testing them to determine if
they can improve their process and be safer to use.
Printing wastes are generated at most stages of the printing process. Ink wastes result when the
reservoir, the various rollers, and the printing plate are cleaned at the end of a run. Excess ink in the
reservoir can be collected for reuse, but the other ink quantities removed during cleaning generally remain
as waste. MPI Label Systems keeps a 50 gallon barrel of water adjacent to each press. During cleanup
when water-based inks are used, soiled parts and cleaning towels are rinsed in this water to remove ink
residues. At the end of the week, barrels of rinse water are transferred to an ink splitting device that
absorbs the various ink pigments on a cellulose-based porous material. The nearly clear filtrate passes
through for disposal in the sanitary sewer, with permission from the local water treatment plant. The
pigment-colored cellulose is accepted at the local landfill along with paper wastes from the print line.
Some paper wastes are also accumulated during setup operations, although an experienced printer is
usually able to minimize them. Exceptional amounts of waste labels are, however, occasionally generated
during the production of multicolor labels because of color registration difficulties. The adhesive based,
print-free trimmings, part of the label stock, are collected on a separate waste roll.
62
-------
TABLE 2. COMPARISON OF ECONOMIC INDICES FOR THE ALKALINE NONCYANIDE PLATING
PROCESS WITH AND WITHOUT THE RECOVERY/RECYCLE SYSTEM
Index
Capital Investment
Payback Period
Net Present Value
Implied Rate of Return
Option
ANC
$36,000
3 years
$57,500
27.0%
ANC+R/R
$87,822
1 .5 years
$281,122
71 .9%
The company has recently converted all of its plating lines from CN-based to ANC-based With this
change the company may consider the delisting process.
The full report, entitled "Alkaline Noncyanide Zinc Plating and Reuse of Recovered Chemicals" bv
Jacqueline M. Peden, is available as EPA/600/R-94/148.
69
-------
Solvent emissions to the plant air have been reduced by at lest 80%. Toxicity of these emissions
has gone from hazardous to non-hazardous, based on EPA regulations. Solid waste generated and
destined for local landfills has been reduced in volume and is no longer classified as hazardous. This
waste is mostly bulk paper for which a recycling system is being sought.
The approximate annual savings at MPI Label Systems resulting from the ink and cleaner change
is estimated by the plant manager to total $16,500 (Table 1). There is essentially no difference in raw
material costs for the inks and cleaners. MPI feels that the overall productivity of the plant has increased,
but the specific economic value of this increase was not determined. Annual waste disposal and handling
account for at lest a savings of $15,000. The facility saves about $500 each year because of a lowered
insurance premium based on improved working conditions. Savings because of new wiping materials
(nondisposable) equals about $1,000 annually. Additionally, these changes did not require capital
investment nor increased operational expenses.
TABLE 1. ECONOMIC SUMMARY OF SAVINGS
USING WATER-BASED INKS AND CLEANERS
Parameter Savings
Water-based inks
Printing speed Approximately 10% faster
Raw Materials None
Waste disposal and handling Minimum annual savings = $10,000
Aqueous cleaners
Disposal Minimum annual savings = $5,000
Raw materials None
Overall Savings
Insurance liability Approximately $500/yr
Inventory None
Wiping materials Annually at least $1,000
Annual total At least $16,500
The full report, titled "Ink and Cleaner Waste Reduction Evaluation for Flexographic Printers'" by
Gary D. Miller, et al., is available as report no. EPA/600/R-93/086.
64
-------
This project was performed to evaluate the effectiveness of low temperature evaooration anri
on «d 80%
Results
C^SronTnTr^S,?,0"818'6"',Pr°dUC!iVity thr°U9hOUt the
-------
and the path of the water and the solids as they flow through the system.
Product
Pretreatment Rinse Four part plating process
Can be returned to plating 6atf)
rneter Counter How rinses
Precipitate collection
Filter Press
Figure 1 . P & H plating alkaline noncyanide plating line with CNT designed recovry/recycle system
Results
R/R System Performance--
Although system operation generated problems, it proved to be a highly efficient and economically
advantageous addition to the plating operation. The quality of the zinc hydroxide precipitate and the
purified water were generally of adequate purity for recycling. Careful and regular analysis of the essential
bath chemicals is recommended to ensure contaminants are not introduced through these recycling
efforts.
66
-------
Advantages of the reverse osmosis system include its relatively high production rates with respect to
ow concentration feed solutions. Additionally, it would require lower capital investment (about $50 000)
than a comparably sized low temperature evaporation system. Energy costs required to operate a 'reverse
osmosis system would be only about $2.50/1,000 gallons processed P
Disadvantages associated with a reverse osmosis system include its inability to concentrate the feed
invent i°seeoVfelSSnd, ^ 12>56° * 18'2°° m9/L 'eVe'S r6Vea'ed in this Studv' This factor alone would
prevent use of a stand-alone reverse osmosis system at the Graham Plating facility Another
be^usPd^hfn1^ 'T'>qUa'lyK^^^ Pr°dUCed by the System" This solution would P^ably have to
oe reused within the plant or further processed through the reverse osmosis svstem beforp " •
the POTW. ei»e ubmubis system oerore
Conclusions-
Both the low temperature evaporation and reverse osmosis systems appear to offer advantages
under specific operating conditions. auvdruages
The reverse osmosis system is best adapted to conditions where the feed solution has a relatively
low nickel concentration. It can process the low concentration feed solution with relatively hiqh
tf™y a e' °! 4'?°0,t0 ,5'°°0 mg/L At this P°int'the solution could be transferred tol low
temperature evaporator for further concentration.
J) The low temperature evaporation system appears to be best adapted to processing solutions with
relatively h.gh nickel concentrations. It can process these solutions so that a concentrate solution
composed of 8% or more nickel is produced along with a very high-quality distillate solution
whirh? 3M h Gral31m ?'atio9 fadlity> the Units would require a caPital investment of $115,000
which would be paid back in 2.8 years through a 27.6% implied rate of return.
ID Electrical conductivity measurements could be used as good indicators for correlating nickel &
TOC concentrations & membrane flux characteristics for this plant.
The full report, entitled "Recycling Nickel Electroplating Rinse Waters by Low Temperature
Evaporation and Reverse Osmosis", by Timothy C. Lindsey, is available as report no. EPA/600/R-93/160.
73
-------
TABLE 1. COMPARISON OF ANNUAL OPERATIONAL COSTS FOR CN PROCESS,
ANC PROCESS WITHOUT R/R UNIT, AND ANC PROCESS WITH R/R UNIT AT P&H PLATING CO.
Process Operation CN Costs ANC ANC + R/R*
($} Costs ($) Costs ($)
Bath makeup 1,771 1,860 1,860
Bath maintenance 22,325 21,225 19,425
Water usage
1. Use @ $7.56/7480 gallons 1,213 1,213 364
2. Sewering @ $5.59/7480 gallons 897 897 269
Wastewater treatment
1. Cyanide oxidation 14,000 0 0
2. Metal precipitation 69,000 69,000 20,700
3. Labor @ $15/hour 7,500 7,500 2,250
Sludge disposal @ $209/cubic 2,600 2,600 1,820
yard
Total 119,306 104,295 46,688
Assumes 70% water and 30% zinc hydroxide recycled.
The recovery efficiency for zinc hydroxide from the CNT system averages 84%. However, due to the
variability in the numbers and types of plating jobs and the general fluctuations in business (all of which
play a role in the amount of zinc hydroxide needed for the line), the company does not recycle all of the
precipitate that is covered. Additionally, the recovered zinc hydroxide is not a totally suitable substitute for
the zinc ingots traditionally used to add zinc to the plating bath. All of the precipitate produced is passed
through the filter press, which greatly reduces its volume. Approximately 30%, depending on production
needs, is returned to the plating bath. The remaining 70% is stored for later use or disposed as a
hazardous waste. While it would be possible to petition to delist this waste, the amount being produced is
less than 5% of the metal waste that the company produces and must routinely dispose. Since much of
the other metal waste was from cyanide-based plating lines, it is sent to disposal as a hazardous waste.
Currently, the company is finding its more economically advantageous to add the zinc hydroxide to the
hazardous metal waste rather than separating, storing it, and applying for delisting.
68
-------
g
the bath. Although the aqueous degreasers do not carry all the risks and S^
or9anic solvent cleaners' periodic
to extend the life of the aqueous cleaner baths. Rather than wasSng raw mater alsthT
aqueous cleaners have the potential to be recycled. Depending on the physical character sties oHhe bath
frac tons o", hl'hnnn ^nf" be eX'ended by Skimming ^taminants'off the top e~eaver
tractions to the bottom, or filtering out suspended species.
Conventional filtration techniques rely on depth or screen filters to remove oil and dirt from a
membranes with pore diameters ranging from 10 » to 10 6 m. The ultrafiltration process works by
producing two separate streams: concentrate and permeate. The permeate stream conta ns onlv th*
components in the feed solution small enough to pass through the membrane po"es(wa^so2ized
chs
The recent development of more durable membranes, such as PVDF has expanded the
app cation of ultrafiltration beyond its origins in the food industry to successfully handtelndustia? process
solutions wrth extreme pH, high temperatures, and high oil concentrations. Because d its un que
n?n± !n? tohcol?cent;a'e oilV wastewater and produce a clear filtrate, ultrafiltration has emerged as
promising techno ogy for extending the life of aqueous cleaner baths. Ultrafiltration of oil water emuki"",
is a more straightforward method for removing and concentrating oil than are ofter physca? chemical I or
^,T ,hTnS- .Ultraflltrati°n does not «»"!«» a stockpile of chemicals and does not produce a chemical
sludge that requires special treatment or disposal. Instead, ultrafiltration produces a water phase thaT
requires no further treatment and a concentrated phase only a fraction of the original volume that can
ene^y, S
HQto » H°ne,^f ^ 9feateSt limitations of ultrafiltration membranes is their tendency to foul Foulinq is
nf n± ? the deCreaSe '" Permeate flux over time' where the flux is defined as the volumetric flow rate
of permeate per cross sectional area per unit of time. Fouling is mainly due to the accumulation o
particles on the membrane surface and/or within the pores of the membrane itself. In indusWal
applications, where ultrafitration could be used to filter aqueous cleaning baths, fouling wiN typically be
due to oils, suspended solids, free surfactant, and metal precipitates. When a membrane shows stons of
u°nU™ hTXhCan Iar9e',y be reSt°red bV C'eanin9 the mem"rane, but a portion ofteTx may be 9
unrecoverable because of irreversible fouling. y
75
-------
#15 RECYCLING NICKEL ELECTROPLATING RINSE WATERS
Participants
The Graham Plating Company In Chicago, Illinois provided the test site and operated the
wastewater recycling system. The Hazardous Waste Research and Information Center (HWRIC), a
division of the Illinois Department of Energy and Natural Resources, Champaign, Illinois provided test
personnel, designed the test and drafted the final report.
Technology/Testing
Low temperature evaporators (Licon, Inc., Pensacola, FL) heat water under a vacuum to produce
steam at relatively low temperatures (150 to 160°F). The unit tested was a model C-3, single effect, pilot-
scale evaporator especially designed for conducting pilot-scale tests on a variety of feed solutions.
Figure 1 provides a schematic of material flow through a low temperature evaporation system.
Cooling
Water *^..-..-..-.-.
Outlet
Cooling
(4\Water '.
Steam or
©Hot *«oC
Water ^^
Supply
Feed
1
\
c
1
once
T&
ntra
nk
Feed/CorKentrate \1
Steam or Hot Water (£
s~*
Distillate (^
s~*
Cooling Water *£,
Feed
(T)
Concentrate
Figure 1. Basic flow diagram for single effect evaporator
70
-------
300
200
O
O
100
Increased surfactant
due to increased
Dura-Gard additions
Surfactant
Oil
0 100 200
Operating Time (hours)
Figure 2. Oil and surfactant in bath vs. time
The costs and benefits associated with installing an ultrafiltration system were analyzed to
determine the economic feasibility of this technology. Based on the estimated expenditures and savings
the payback period associated with this technology was only 6.9 mo. The net present value and Interest
rate of return Indices were $152,143 and 178%, respectively. Therefore, investment in an ultrafiltration
system represented a very attractive economic alternative.
In summary the findings of this evaluation indicated that:
o The application of ultrafiltration produced significant reductions in hazardous waste.
This is especially significant in comparison to earlier methods using separate degreasing
and phosphating tanks and organic solvent.
o For metal fabricating, and potentially similar applicants, use of this technology can be
economically attractive although capital investment is required.
o Other applictions for recycling/reusing wastewater via ultrafiltration have good potential
for P2 improvements as well as good economics. However, they should first carefully
investigate the highly sensitive parameters such as fouling, at small scale and identify
variability in their operation and its effect on the process.
Report
The full report, entitled "Evaluation of Ultrafiltration to Recover Aqueous Iron
Phosphating/Degreasing Bath" by Gary D. Miller, et al., is available from the National Technical
Information Service. The project Summary is distributed by EPA as report number EPA/600/SR-93/144
77
-------
TABLE 1. COMPARISON OF NICKEL CONCENTRATIONS IN CONCENTRATE,
DISTILLATE, AND PERMEATE
Osmosis
Product
Low Temp. Evap. Reverse
Drum A Drum B Drum C Drum D
Concentrations at beginning of test (mq/L):
Concentrate 4,140
Distillate 2.5
Permeate
Concentrations at end of test:
Concentrate 17,000
Distillate 1
Permeate
Ratio of distillate permeate to concentrate:
Distillate 0.02%
Permeate
2,540
2.2
128,000
0.3
0.01%
2,580
44.5
12,560
210
1.4%
1,425
14.5
18,200
790
1.54%
The feed solution processed through the reverse osmosis system contained initial nickel
concentrations of 1,425 to 2,580 mg/L (Table 1). Nickel concentrations increased steadily until
approximately 60% of the water volume was processed and nickel concentrations were up to 4,000 to
5,OOOmg/L. After this point productivity started to drop due to solids precipitating and fouling the
membrane. The final concentrations achieved with the reverse osmosis process ranged from 12,560
mg/L to 18,200 mg/L, well below the 8% nickel concentration required for the plating bath. Some of this
solution could be used to replace water losses in the electroplating process. The reverse osmosis
system, however, would probably produce excess volumes of concentrated rinse water composed of 1.2%
to 1.8% nickel. This material would have to be further processed with the use of an alternative technology
such as low temperature evaporation or be shipped to a facility that could exact the nickel for use in other
industrial processes.
The reverse osmosis system concentrated the organic constituents present in the rinse water feed
solution from initial TOC levels of 340 to 540 mg/L to levels of 2,800 to 3,500 mg/L. These concentrations
suggest that the organic bath constituents are concentrated by the reverse osmosis equipment at rates
that parallel the nickel concentration rates.
The quality of the cleaned rinse water permeate produced by the reverse osmosis equipment
averaged 89 to 134 mg/L nickel. These levels are about 98.5% lower than the nickel concentrations
present in the concentrated solution, but would not, however, be acceptable for discharge to publicly
owned treatment works (POTW). The nickel levels present in this solution could be further reduced by
passing this solution through the reverse osmosis equipment again.
72
-------
TABLE 1. QUANTITY OF INKS USED
Ink Color
Black
Blue
Fled
Yellow
Ink Type
Petroleum
Soy
Petroleum
Soy
Petroleum
Soy
Petroleum
Soy
Ink Used
(g)
241.8
162.7
124.8
79.5
109.7
87.1
85.8
53.6
Ink Used
(9)/100
Sheets
3.8
3.2
2.0
1.6
1.7
1.7
1.4
1.1
Differenc
e
19
25
0
27
The amount of blanket cleaners used is shown in Table 2. Almost 46% more blanket cleaners
were used during the petroleum ink run (3,455.4 g compared with 2,368.0 g). Some of this difference can
be attributed to the larger amount of blanket cleaners used during the longer make-ready that occurred
during this run.
Ink Type
Petroleum
Soy
Cleaner Name
V120
Clean Quick
V120
Clean Quick
Cleaner Used (a)
1,768.7
1,686.7
1,141.2
Both types of inks appeared to require approximately the same amount of cleaners and effort to
remove from the presses. On average in typical practice, the amount of cleaner used for the two inks
would be expected to be about the same.
Based on laboratory results and on the amount of inks and cleaners used (Tables 1 and 2) the
amounts of solids and volatiles in the inks and cleaners were calculated (Table 3). The amounts of
volatiles were used to estimate the total mass of air emissions. Over 90% of the liquid wastes originated
from the inks. As for the volatiles, over 99% originated from the cleaners in both cases
79
-------
#16 EVALUATION OF ULTRAFILTRATION TO RECOVER AQUEOUS IRON
PHOSPHATING/DEGREASING BATH
Participants
R.B. White, Incorporated of Bloomington, Illinois, provided the test site and operated their process
equipment during the test. The Hazardous Waste Research and Information Center (HWRIC), a division
of the Illinois Department of Energy and Natural Resources, Champaign Illinois, provided test personnel,
designed the test and drafted the final report.
Technology/Testing
R.B. White, Incorporated, of Bloomington, Illinois operates a sheet metal fabrication facility that
manufactures painted steel shelving units.
Cold-rolled steel arrives at the plant from the steel mill coated with mill oils to protect the bare
metal from corroding or staining during storage and fabrication operations. During fabrication, coolants
and lubricants are also applied to the metal working surface. Before being painted, the metal surfaces are
cleaned to remove the mill oils and metal working fluids and then preconditioned to bond well with the
paint coating.
Fabricated parts are cleaned and phosphated in a 5000-gal heated, aqueous immersion tank and
rinsed with a fresh water spray. The company previously operated separate degreasing and phosphating
tanks using trichloroethylene for degreasing. In 1985, it switched to a single-stage, aqueous iron,
phosphating/degreasing system to improve worker safety and reduce the generation of organic solvent
emissions and hazardous waste.
Although the switch eliminated the risks and liabilities associated with organic solvents, it
introduced a new waste disposal problem. Simultaneous degreasing and phosphating in the same bath
formed an oil-water emulsion. With extended use, the buildup of oil in the bath reduced cleaning and
phosphating efficiency, and product quality was compromised. Additionally, dragout of oil from the bath
into the rinse water eventually pushed oil and grease levels in the discharge over the allowable limit.
In the past, oil skimmers were used to control oil slicks on the surface and prolong the life of the
bath, but the skimmers were only partially effective. When oil in the bath began to sacrifice product quality
and the discharge levels edged closer to the maximum allowable limit, the bath had to be replaced.
Depending on production rates, the bath typically lasted 3 to 4 months. Replacing the bath required a full
day of lost production time to take the process off-line, make arrangements with a waste transporter to
drain and dispose of the contents, and recharge the tank with 5000 gallons of fresh water and raw
materials.
The spent bath was classified as RCRA hazardous waste because it failed Toxicity Characteristic
Leaching Procedure (TCLP) tests for xylene. Since land disposal of liquid wastes is prohibited, the bath,
sludge, and skimmed oil were incinerated in a cement kiln. Disposal costs, including transportation and'
incineration, ran about $1 /gallon v/hich came to $5000/bath, or about $15,000/yr. This amount was in
addition to the costs associated with lost production time and replacement of water and raw materials.
Aqueous cleaners have already replaced solvent degreasers in many industrial surface
preparation operations. The water-based cleaners effectively remove protective oils, cutting oils, hydraulic
fluids, silicone oils, water soluble coolants, shop dirts, finger prints, and other contaminants. Special
additives also make the aqueous cleaners versatile coating solutions. Some aqueous cleaners have
made it possible to eliminate some separate degreasing and coating processes as well as reduce waste
74
-------
Cost-
The main cost factor considered was that of raw materials. No equipment expenditures were
required by OPS to switch from using petroleum to soy inks. The operating conditions of the press such
as temperature and speed, were the same for each ink so there was no difference in overall rate of
production or utilities used because of the type of ink used. Insurance, monitoring requirements reportinq
and record keeping, and permit requirements are the same for each type of ink. Any differences in future
costs for remediation or property damage would be minimal. Employee health costs may slightly favor the
use of soy-based inks because of reduced employee exposure to breathing released petroleum based
chemicals.
The purchase prices of raw materials (inks, cleaners, and paper) for this print job are
approximately the same for both types of ink. Generally soy inks cost about 10% more. The average
purchase price for both the soy inks and petroleum inks is about $8.00/lb (or $0 018 g) Actual costs
depending on the color, range from about $4.00 to $12.00/lb. Costs do vary, however; when the amount
of materials used for petroleum inks is compared with the amount of materials for soy inks.
Overall, soy inks have some environmental and other advantages for sheet-fed offset printers
The mam environmental advantage is that they release less than 1/5th of the mass of volatile organic '
chemicals compared to petroleum inks. The soy inks also spread about 15% further which offsets the
small difference in cost that currently exists. In this study all other factors including make-ready time
appearance of printed product, and clean-up effort were essentially equivalent between the two tvpes of
inks.
The full report, entitled "Waste Evaluation of Soy-Based Ink at a Sheet-Fed Offset Printer" bv
Gary Miller, et al., is available as EPA/600/R-94/144.
81
-------
Results from the bench- and pilot-scale studies were used to develop a full-scale modified-batch
test conducted on site at the facility. Figure 1 shows how the full-scale test applied ultrafiltration directly
to the 5000-gallon iron phosphating/degreasing bath.
The objective was to directly measure the effect of ultrafiltration on the process solution under
actual plant conditions. The full-scale test took Into account the constant input of oil from the
production line and the daily addition of bath chemicals. The full-scale test also helped identify
problems with the ultrafiltration equipment and anticipate changes that should be made on a permanent
unit.
The full-scale in-plant testing featured an ultrafiltration system provided by Koch Membrane
Systems (Model UF-4) equipped with four 1-in tabular PVDF membranes (1000,000 MWCO, 4.4 tf total
area). Data obtained from the full-scale modified-batch test was used to determine whether ultrafiltration
would be a viable option for waste reduction at the plant. Technical, operational, and economic aspects
associated with the ultrafiltration equipment were examined to evaluate the feasibility of this technology
to improve the company's metal fabrication operation.
Pump
^
Iron Phosphating/
Degreasing Bath
(5000 gal)
T Pump—
Process
Tank
(SSgal)
V
a—
Concentrate
i rx ^
•
1 Permeate
Membrane
Figure 1. Modified-batch scheme ultrafiltration.
When field testing began, the iron phosphating/degreasing bath had not been replaced in over 3
months. The aqueous solution was murky with dirt and oil, and large patches of free oil floated on the
surface. The changes that took place over the next 11 days of ultrafiltration testing produced a dramatic
effect. Surface oil slicks disappeared and were replaced by a clean, light foam. The bath solution was
visibly clearer, and plant personnel testified that it looked like a freshly recharged bath. Results of total
organic carbon (TOC) analyses for the full-scale testing showed the change in oil and surfactant
concentrations during the test (Figure 2).
76
-------
period.
The total surface area of the boards plated for each rack was also tracked for each sampling
Two modifications to reduce drag out were made and independently tested after taking baseline
samples. The first modification was to slow the withdrawal rate of the racks as they were pulled from
the process tanks.
A second modification was tested using a withdrawal rate between the baseline and first
modification rates and increasing the drain time over the process tank before transfer to the rinse tank.
Softened Water
Electroless
Copper
A
A
Sampling
Location
Counter
Current
Rinse
A
1 3.3 gpm
f
to treatment (Ion Exchange)
i
Counter
Current
Rinse
A
Figure 2. Electroless Copper Process Diagram
Results
Results for the micro-etch bath are summarized in Table 1, and results for the electroless copper
bath are summarized in Table 2. The reduction in drag out for the micro-etch bath was 45% as a result
of the first modification and 41% as a result of the second modification. For the electroless bath, drag
out was reduced by 50% after the first modification and 52% after the second modification.
For modification 1, the reduction in drag out was calculated to prevent 194 g/day of copper
from the micro-etch bath and 9 g/day of copper from the electroless bath from entering the rinse-water
waste stream. The figures for the second modification were 180 and 9 g/day, respectively. These
figures assumed a copper concentration of 30 g/L in the electroless bath and a production level of 1200
fr of printed circuit board per day.
Economic calculations were based on the cost of existing treatment and disposal methods for
the rinse-water waste streams.
83
-------
#17 WASTE EVALUATION OF SOY-BASED INK AT A SHEET-FED OFFSET PRINTER
Participants
The Office of Printing Sen/ices at the University of Illinois was the host site for the project. The
Hazardous Waste Research and Information Center (HWRIC), a division of the Illinois Department of
Energy and Natural Resources, Champaign, Illinois, provided test personnel, designed the test and
drafted the final report.
Technology/Testing
Soy inks for sheet-fed offset printing are defined as those that have a minimum of 20% soy oil by
volume. The soy oil replaces petroleum oils in the ink vehicle and varnish components. Use of the soy oil
affects color, drying, and other operating characteristics of the inks. Soy-based inks used in sheet-fed
presses still contain at least 10% petroleum oils. Research is continuing to increase the proportions while
maintaining satisfactory printing characteristics. The objectives of this study were to determine
environmental impacts and associated economics.
Data for this study were collected during a full-scale print run on a Miller TP104 Plus six-color
press. For this evaluation a 4400 sheet, six-color, work-and-turn print job was selected. Only four of the
six colors were included in the study because the type of ink used for two of the colors was not changed
between the runs. All inks included in the study were manufactured by Handschy, Incorporated of
Bellwood, Illinois.
In-plant measurements consisted of weighing the containers of inks, blanket cleaners, and roller
cleaners used during make-ready and printing at each press unit before and after each print run. Cleanup
rags were weighed before and after each print run and wastes in the wash-up trays were weighed at each
press unit at the end of each run. Samples of each type and color of ink and each cleaner used were
analyzed, at the same temperature as the press, for total solids and volatile content, at the Hazardous
Waste Research and Information Center's (HWRIC) Hazardous Materials Laboratory.
The operators at the university Office of Printing Operations (OPS) used the same cleaners for
both petroleum and soy-based inks because no satisfactory substitutes had been identified.
Results
The study results showed that the petroleum-based ink run required larger quantities of three of
the four colors of ink than the soy-based ink print run (Table 1).
78
-------
#19 SPONGE ROLLERS AND FLOW CONTROLLER FOR RINSE WATER REDUCTION AT A
PRINTED CIRCUIT BOARD MANUFACTURER
Participants
The host for the evaluation was Hutchinson Technology, Incorporated (HTI) plant of Hutchinson,
MN. Hutchinson personnel also operated the plating equipment during the test. Battelle, Columbus
Laboratories, on contract to EPA, helped design the test program, provided test equipment and personnel
and drafted the test report.
The process lines used for evaluating the technologies in this study produce flexible circuits (FC's)
and printed wire boards (PWB's). The lines evaluated were typical for the industry. Two studies were
conducted for reducing waste water. Absorbent sponge rollers were added to reduce dragout on the
horizontal FC cleaning line. A flow controller valve was added to reduce rinse water on the PWB line.
Absorbent Rollers for Drag Out Reduction-
The FC manufacturing process at HTI uses an automated horizontal wet process line that
employs conveyors to transport the flexible circuits and liquid sprayers to apply the process solutions. The
production line consists of the following steps:
(1) Acid cleaner
(2) First rinse (water from second rinse)
(3) Second rinse (fresh reverse osmosis water)
(4) Micro etch (sodium persulfate solution)
(5) Third rinse (water from fourth rinse)
(6) Acid rinse (5% sulfuric acid)
(7) Fourth rinse (water from final rinse)
(8) Fifth rinse (water from final rinse)
(9) Final rinse (fresh deionized water).
At the start of the study, HTI was using pairs of hard roller squeegees (standard equipment on the
horizontal process line) to remove excess solution from flexible circuits as they exited each bath.
Absorbent rollers were tested as replacements for the upper squeegees between the acid cleaner and the
first rinse, Steps 1 and 2 above. As each piece leaves this bath, it carries solution containing cleaner and
copper salts. The acid cleaner, the squeegees, and the first rinse are shown in Figure 1. The soft,
absorbent rollers that replaced hard rubber squeegees in the wet process line were made of PVA
(polyvinyl alcohol), a porous material that has high absorption and retention capabilities and is pliable and
elastic. PVA's resistance to abrasion and chemicals makes it appropriate for use in this type of process.
The PVA rollers were evaluated to determine the reduction of drag out from the acid bath to the
subsequent rinse bath. Any reduction of drag out would mean that the rinse could maintain its dilution
capability with a lower flow of fresh water into the tank, thereby reducing the volume of wastewater exiting
the tank. It was expected that any reduction of drag out would necessitate more frequent replacement of
the acid bath due to accelerated buildup of the impurities that previously were transferred to the rinse
baths.
85
-------
TABLE 3. SOLID AND VOLATILE CONTENTS OF INKS
AND CLEANERS USED (g)
Parameter
Ink Solids
Cleaner Solids
Total Solids
Ink Volatiles
Cleaner Volatiles
Total Volatiles
Petroleum
Print Run
537.17
64.78
601.95
24.93
3,905.22
3,930.15
Soy Print
Run
371.36
48.64
420.00
11.54
2,992.76
3,004.30
As shown in Table 3, the estimated mass of volatiles emitted from the inks is less than 1 0% of
the total mass emitted from the inks and cleaners. Since most of the air emissions were from the
cleaners, less than 1% overall reduction in air emissions resulted from using the soy-based inks The
longer the print run the greater the reduction in volatile emissions from using soy inks.
Liquid Components-
The two main liquid wastes from the printing press were from washup trays, and inks and
cleaners on used rags.
The amount of liquids in the washup trays from this study was measured directly at the end of the
print runs. Almost 10% more liquids were generated in the washup trays after the petroleum ink run This
liquid was a mixture of inks and various cleaners. The amount of each ink and each cleaner that ended
up in these trays and the rags (less the amount of evaporated and on printed product) was not
determined.
Differences resulted from operator induced variability, extended make-ready during the petroleum
ink run, and more thorough cleaning conducted at the end of the soy ink run. From this it cannot be
determined which ink generated more liquid wastes.
Solid waste that may be generated in printing includes the waste sheets in starting (make-ready)
trimmings and excess number of pages printed to compensate for losses that may occur during folding
binding and any other final preparation steps. Because the soy inks generally spread further (by an
average of about 17%), less used ink containers may be generated with those inks. The amount of the
other solid wastes would be unaffected by the type of ink used.
Based on observations of manually operated presses at this and other facilities, automation or
efficient press set-up, rather than the choice of ink, results in less solid waste being generated during
printing. Once operators are familiar with the use of either the petroleum or soy inks, the amount of solid
waste generated during make-ready will not be noticeably different. In both cases, more waste paper
might be generated on some jobs because of difficulties in obtaining acceptable colors or other print
quality factors. In this regard, neither ink appears to have a clear advantage over the other.
80
-------
Results
Absorbent Rollers-
The average drag out rate decreased from 27.6 ml/ft2 with hard rollers, to 9.8 ml/ft with the
sponge rollers, or a reduction of 65%. At the 1.0 gpm flow rate used with the hard rollers, about 120,000
gallons of wastewater were discharged per year, assuming 250 eight-hour work days. If drag out is
reduced by 65%, the flow of water into and out of the rinse tank could also be reduced by 65%. This is a
projected savings of 78,000 gallons/year, with an equal decrease in wastewater generation.
With reduced drag out, the acid bath will require less frequent replenishment with concentrate and
water to maintain sufficient volume. However, it will require more frequent replacement to preserve
cleaner effectiveness.
The amount of cleaner concentrate and water needed to replenish the acid bath solution will be
reduced by the same amount as the drag out. At the hard roller drag out rate of 0.175 gpm for the typical
circuit production rate of 24 ft2/min, about 2100 gallons of cleaner concentrate and 18,900 gallons reverse-
osmosis water are required annually to maintain production. (0.175 gpm x 60 min/hr x 8 hrs/dy x 250
days/yr = 21,000 gal/yr solution; concentrate:water = 1:10). With the PVA rollers and the resulting 65%
reduction in drag out, the annual requirements for concentrate are an estimated 744 gallons, while water
requirements drop to 6696 gallons. Total drag out becomes 7440 gallons of cleaner pe year. This
produces an annual savings of 1356 gallons of cleaner concentrate and 12,204 gallons of water.
Controlled Flow-
Data were collected on the volume of rinse water used in the final counter-current cascade rinse
of the tin-lead plating line during the one week period before the conductivity sensor and automatic valve
were installed. At the 0.4 gpm flow rate, a total of 1810 gallons of water were used in the rinse when
continuous flow was maintained for the duration of each work shift. A total of 320 PWBs, each with a 4 ft2
surface area (12 in x 24 in per side), were processed during this week. The total PWB area produced was
1280 ft2, and 1.41 gallons rinse water were used per square foot of PWB. The water used was constant at
1810 gallons per week, regardless of the production rate, when the line was operated without the use of
the control system.
With the control system installed and set at 30 micro Siemens (mS), 430 gallons of water were
used to process 167 PWBs, or 668 ft2, during the week following installation. This is a rate of 0.64 gal/ft2
of PWB, an apparent 55% reduction when compared with the 1.41 gal/ft2 measured the week before.
However, in order to compare the actual water savings, it is necessary to note that, with the previous
method of preset continuous inflow, 1810 gallons of water would have been used regardless of
production. The second week required only 430 gallons, a 76% reduction from the continuous flow
practice of 1810 gallons per week. At an average production rate of 300 PWB per week, annual water
use would be 90,500 gallons without the flow controller (based on 1810 gal/wk and 50 weeks/yr.) and
38,400 with the flow controller (based on 300 PWB, or 1200 ft2/week,50 weeks/yr, at 0.64 gal/ft2).
Economic Evaluation for Absorbent Rollers-
Table 1 compares operating costs for the acid cleaning step and subsequent rinse with the hard
roller squeegees and with the soft, absorbent rollers. The costs are based on operating one, 8-shift per
day, 250 days per year.
87
-------
MINNESOTA
Of the five technologies evaluated one was performed under the auspices of the Minnesota
Technical Assistance Program (MnTap), Cindy McComas, project officer. Two more sites Hutchinson and
/M^3? W!.re ;dentlfied with the assistance of MnTap (Paul Pagel), project officer) while the remainder
(McCurdy, SL Modern Hardchrome and Aeroquip) were located and tested with the assistance of Battelle
Columbus Laboratories. All five are in the area of electroplating process improvements The EPA project'
officer was Teresa Marten. M J
#18 MODIFICATIONS TO REDUCE DRAG OUT AT A PRINTED CIRCUIT BOARD MANUFACTURER
Participants
The host for the evaluation was Micom Incorporated of Brighton, MN. Micom personnel also
operated the equipment during the test. The design of the test program was a collaboration with EPA and
the Minnesota Technical Assistance Program (MNTAP) of the University of MN on a cooperative
agreement with the Minnesota Pollution Control Agency. MNTAP supplied test personnel and helped to
write the draft report. K
Technology/Testing
The project evaluated modifications that reduce drag out at a single plating line of a printed circuit
board manufacturer. It was hoped that by demonstrating the success of the modification in a fully
operational setting, the technology would be transferred to other plating/rinsing systems within the
company as well as to other companies in the metal finishing industry.
Two interrelated modifications that effectively help prevent wastes from entering the rinsing
processes are (1) reducing drag out, which is the carryover of concentrated solutions from platinq baths
and (2) reducing rinse water flows. When drag out is reduced, rinse water can be conserved because
less will be needed to achieve effective rinsing.
Procedure-
The evaluation took place at the sensitize line where a number of process baths including etchant
(micro-etch), activator, accelerator, electroless copper, and rinse tanks, first etch and then chemically
deposit copper onto the insides of the circuit board through holes. Drag out from two of the line's process
baths, the micro-etch and the electroless copper, was a significant source of copper discharged into the
nne-water waste stream leaving the line. Figure 1 shows a schematic of the micro-etch process and
Figure 2 provides the same for the electroless copper process. Rinse water from the two processes had
to be treated by an onsite, ion-exchange unit for copper removal before it could be discharged to public
sewer. ^
Treatment of spent micro-etch and electroless baths included copper recovery in both cases and
regeneration of etchant in the case of the micro-etch solution.
To determine baseline drag out, samples were taken at the process tanks and at the two rinse
tanks following each process tank - both before and after a rack of circuit boards was moved through the
three-tank system. Over a 2-week period, 12 sample sets were taken to calculate 12 baseline drag out
values. Additional measurements taken to calculate drag out included rinse-tank water volumes for each
of the rinses. These volumes were measured during each of the 12 sampling periods.
82
-------
TABLE 2. OPERATING COST COMPARISON OF CONTINUOUS AND CONTROLLED FLOW
___ CONTINUOUS FLOW _ SENSOR-CONTROLLED FLOW
__ Unit Cost _ Units/yr _ Cost/yr _ Units/yr _ Cost/yr
MATERIALS AND EQUIPMENT
Controller Maintenance $25/hr N/At N/A 2 hrs $50
Deionized water for $17/1000 gal 90,500 gal $1,538 38,400 gal $653
rinsef
WASTE MANAGEMENT
Water Treatment __ $13/1 OOP gal _ 90,500 gal _ $1,176 _ 38,400 gal _ $499
TOTAL COST $2,714 $1,203
NET SAVINGS _ _ _ $1,511
* All Unit cost data were provided by HTI.
t Cost includes ion exchange regeneration and waste disposal costs.
t N/A = not applicable because maintenance not required.
Note: The average production rate of HTI of 300 PWB/week on the tin-lead line was used to estimate water
consumotion.
The cost of the rinse water control equipment was $250, and installation required about 10 hours
at $25 per hour, for an installed cost of $500. If the maintenance cost is included, the total capital cost for
the first year's use of the rinse water controller is $550. With a total savings of $1511, the payback is less
than 5 months (payback=$550/$1511 per year = 0.36 yrs).
Using absorbent rollers reduced the quality of acid cleaner concentrate and water required to
maintain the bath. It allowed the amount of water flowing into the rinse to be decreased in proportion to
the reduction in drag out and reduced the total copper loss from the acid bath and the subsequent rinse
by more than 50%. This reduces the heavy metal content of the sludge resulting from wastewater
treatment.
The acid cleaner was found to degrade the soft rollers over time, requiring that they be replaced
every five to six months. The materials compatibility of the rollers and the solution must be considered for
all applications. The replacement cost and labor to replace the rollers were low enough that they had little
impact on the annual savings. The payback period for the equipment change was less than one week.
Overall, the use of soft, absorbent rollers proved very effective in minimizing drag out even
through they were used only on the top surface of the flexible circuits due to the processing equipment
and items tested at HTI. It is expected that the rollers would be even more effective if used both above
and below the flexible circuits, as recommended by the manufacturer. The rollers can be applied to other
types of processing lines in which the items can be passed between a set of the rollers.
The full report, entitled "Sponge Rollers for Drag Out Reducation and Flow Controller for Rinse
Water Reduction At a Printed Circuit Board Manufacturer" will be published as an EPA series 600 report.
89
-------
Sampling
Baseline
Mod 1
Mod 2
Withdrawal
Rate ,
(ft/min)
100
11
40
:
Time of
Withdrawal
(seconds)
1.7
14.9
4.3
======
Drain
Time
(seconds)
3.4
2.5
12.1
=I===T
Total
Time
(seconds)
5.1
17.4
16.4
— — — _^_^_^_
Drag Out
(ml/ft2)
12.1
6.7
7.1
Sampling
Baseline
Mod 1
Mod 2
= ==;
Withdrawal
Rate
(ft/min)
94
12
40
=
^
Time of
Withdrawal
(seconds)
1.8
13.9
4.3
====i
Drain
Time
(seconds)
5.2
3.2
11.9
===
:^===
Total
Time
(seconds)
7.0
17.1
16.3
=====
Drag Out
(ml/ft2)
6.0
3.0
2.9
The economic evaluation showed that the company could save $2640 per year in rinse water
treatment and disposal costs by implementing modification 1 or $2460 per year by implementing
modification 2. An additional savings of $710 per year for modification 1, or $660 for modification 2 could
be realized in avoided water and sewer charges if the company reduced rinse-water flow rates in '
proportion to the reduced copper contamination resulting from the modifications.
The company decided to adopt the second modification as its normal operating procedure The
increase in drain time did not slow production at the company.
The full report, titled "Modifications to Reduce Drag Out at a Printed Circuit Board Manufacturer"
by Teresa M. Marten et al., is available as report no. EPA/600/SR-92/114.
84
-------
The BH™ bath (Step 5} is an aqueous, carbon-black dispersion which eliminates the need for EC
metalization before electrolytic plating. The steps before and following Step 5 are similar to those used in
the EC process.
Unlike the EC process, with BH™ technology the rinse after the microetch process step is the only
rinse water stream that goes to the ion-exchange system. The rest goes to the discharge line. The
carbon-black dispersion process uses only two rinse water flows, and the process solutions contain
nonhazardous material.
Waste Reduction Evaluation-
The amount of waste resulting from the EC operation, run at full production, was evaluated to
represent baseline data. The amount of waste from the carbon-black dispersion process using BH™
technology, run at 1/2 capacity, was then compared with this baseline. The rinse water characterization
included analyses for copper, pH, and total solids content.
Results
The production rate on the carbon-black dispersion process line using BH™ technology (i.e., 3.3
ft2/min) was found to be 2.1 times as fast as the production rate on the electroless copper (EC) process
line (i.e., I.6 ft2/min). Production rates were timed during field testing and compared with production
schedules maintained by McCurdy Circuits. McCurdy Circuits operates the electroless copper process at
approximately the full capacity of 1.6 ft2/min. which yields 200,000 ft^yr.
Tables 1, 2, and 3 show annual projections of the chemicals used, costs incurred, and wastes
generated, with the process operating at capacity.
Conclusions-
Because the carbon-black dispersion process reduces wastes, avoids many hazardous chemicals
and metals, is cost effective, and yields an acceptable product, it should be considered a viable alternative
to the EC process. If the shop involved is a job shop, client input and requirements would be important in
determining the feasibility of incorporating the carbon-black dispersion process. Although this study
provides generalizations for companies considering carbon-black dispersion, it is recommended that each
company examine its specific requirements to determine the suitability of this alternative technology for
specific applications.
Formaldehyde (a suspect human carcinogen that poses a significant health hazard when inhaled
or ingested or through direct physical contact) is completely eliminated in the BH process, whereas
approximately 200 gal/yr are used in the EC bath. Palladium and trace amounts of cyanide, also used in
the EC process, are not present in the carbon-black dispersion process.
Economic Evaluation--
The economic evaluation was based on data obtained from McCurdy Circuits, including the unit
costs and amounts used of chemicals and water, and from suppliers. The current capital cost of carbon-
black dispersion equipment was obtained from an equipment vendor. The calculations are based on the
production rate of 200,000 ft2 of PWB per year, approximately the rate of the current EC system. The BH
process cost basis is half a year, running at capacity, i.e., the time it would take to process approximately
200,000 ft2 of PWB.
91
-------
Conductivlty-Sensor-Controlled Flow Valve-
The tin-lead plating line for printed wire boards uses plating baths and racks to hold and move
items through the line. It consists of the following process steps:
(1) Acid cleaner (weak)
(2) Rinse (water from Step 3)
(3) Rinse (water from Step 5)
(4) Preposi-etch
(5) Rinse (water from Step 6)
(6) Rinse (fresh tapwater)
(7) Sulfuric acid dip (10%)
(8) Acid copper plate
(9) Rinse (water from Step 10)
(10) Rinse (fresh tapwater)
(11) Tin-lead plate
(12) Rinse (water from Step 13)
(13) Rinse (water from Step 14)
(14) Rinse (fresh deionized water)
PCs Hill
i
Rolen or
Squeegees
TTTTT
Add Cleaner
11111
TTTTT
ftnse
3) Ruse Wafer
To Wostetater Treatment
Sampkng Pouts
I - Add Cleaner
2 - Rmse Water
3 - Rinse Water Wet
Figure 1. Drag out reduction
A conductivity controller was installed in the final rinse tank (Step 14). When the study began
HTI was maintaining a continuous 0.4 gpm fresh water flow into this tank when the line was running
This rate had been established by HTI to provide adequate rinsing. The shortcoming of this approach
was that the inflow is not reduced during periods of light production. This resulted in a steady stream of
wastewater that must be treated, even though it may be only slightly contaminated.
HTI engineers adapted a flow valve so that it was controlled by a conductivity sensor This was
a simple adaption of standard equipment, at a cost of $250. Such systems can also be purchased from
equipment suppliers, at about the same cost.
When rinse water conductivity, as a measure of metal content exceeds a specified point the
valve opens, adding fresh deionized water to the rinse. Once conductivity drops below the set point the
inflow of water is stopped. As compared with the practice of maintaining a uniform flow this approach
should conserve water, particularly during periods of reduced plating line throughput. Decreased water
use would reduce the wastewater volume and treatment costs.
86
-------
In tests conducted by McCurdy Circuits, product quality of the carbon dispersion processed
boards was equivalent to that of the electroless copper processed boards. In addition, PWBs using the
BH carbon dispersion technology have passed MIL-P-55110D qualification and performance standards for
plated through-holes.
TABLE 2. COMPARISON OF ANNUALLY ADJUSTED MAJOR OPERATING COSTS*
Description
Chemicals
Tap water
D.I. water
Energy(b)
Labor
Waste disposal
Waste treatment labor
Totals
Electroless
Copper
$89,600
3,200
503
N/A
50,000
N/A
10,000
$153,000
Blackhole™
$37,500
403
38.3
N/A
25,000
N/A
3,330
$66,300
Blackhole™
Savings, %
58
87
92
0
50
0
67
57
Because the BLACKHOLE™ production rate is approximately twice as fast as that of the electroless copper process,
the costs are adjusted to take this into account. The BLACKHOLE™ costs reflect a half year of processing, where as
the electroless copper costs represent a full year.
TABLE 3. SUMMARY OF WASTE REDUCTION
Waste Types
Rinse water
Chemical use
Copper waste (in
rinse water)
Total solids
Electroless Copper
Process
13.8 gal/ft2
1 1 ,755 gal + 38 Ib
324 mg/ft2
23,800 mg/ft2
Blackhole™ Process
1.7 gal/ft2
90 gal + 61 1 Ib
248 mg/ft2
4,510 mg/ft2
Net Change in
Waste
12.2 gal/ft2
not calculable
76 mg/ft2
19,300 mg/ft2
The full report, titled "Carbon Black Dispersion Preplating Technology for Printed Wire Board
Manufacturing" by Dale Folsorrr et al., is available as report no. EPA/600/R-93/201.
93
-------
There is an overall savings of over 50% in annual operating costs. Although costs for roller
replacement and bath treatment increase when the PVA rollers are used, these increases are more than
offset by net savings in water and acid cleaner concentrate and in waste rinse water treatment The cost
of using the hard rollers is an estimated $29,865 per year. The estimated cost of using the soft rollers is
$13,322 per year, a reduction of $16,543.
Payback is defined as the minimum time to recover the capital cost of the equipment (payback in
years = capital cost/savings per year). The capital cost of are $200 for two sets of rollers yielding a
payback of less than a week ($2007$ 16,537). '
TABLE 1. OPERATING COST COMPARISON OF HARD AND SOFT ROLLERS
Hard Rollers
Soft Rollers
Unit Cost*
Materials & Equipment
Replacement rollers
Reverse-osmosis water
for rinse
Reverse-osmosis water
for bath replenishment
Acid cleaner concentrate
for bath replenishment
Bath replacement
(cleaner+water+labor)
Waste Management
Rinse water treatment
Bath batch treatment
Sludge disposal
Total cost
Net Savings
$50/ea
$10/100
gal
$10/1,000
gal
$11.75/gal
$396/bath
$13/1,000
$0.50/gal
$245/ton
t
120,000 gal
18,900 gal
2, 100 gal
4 baths
120,000 gal
1 ,240 gal
306 Ib
$1,200
$189
$24,675
$1 ,584
$1,560
$620
$37
$29,865
4
42,000 gal
6,696 gal
744 gal
6 baths
42,000 gal
1,860 gal
336 Ib
$220
$420
$67
$8,742
$2,376
$546
$930
$41
$13,322
* All unit cost data were provided by HTI.
t Hard rollers last indefinitely.
Economic Evaluation for the Sensor-Controlled Valve--
Table 2 summarizes the estimated annual operating costs for the tin-lead line, final rinse before
and after the rinse water controller was installed. Significant savings are realized from the reduced use of
water. These are slightly offset by maintenance costs for the controller. The effect on sludqe disposal
costs is negligible.
With the flow controller, a savings of 52,100 gallons per year could be realized, assuming he line
operates for 50 weeks. At a cost of $17 per 1000 gallons deionized water and $13 per 1000 gallons for
wastewater treatment this amounts to a savings of $1563 per year for water purchase and waste
treatment.
88
-------
Materials and Methods--
One test site uses a chromium plating solution and one a chromic acid for etching copper. At
each site, metal analyses were performed on the chromium solution and the catholyte solution to
determine contaminant levels in the process solution and the rate of metals buildup in the catholyte
solution. From the analyses, the rate of metals buildup in the catholyte solution was determined. This
corresponded with metals removal from the process solution, which in turn was used to determine the
bath life-extending capabilities of the lonsep™ process and the waste reduction potential.
Operating costs with and without the lonsep™ process were used along with installation costs to
determine the economics of the process.
Results
The results from the two sites are discussed separately for each site because the bath process
and the lonsep™ process used at the two sites were different. For each site, the discussion is divided into
the three project objectives: waste reduction potential, economic evaluation, and product quality
evaluation.
Chromium Plating at SL Modern Hard Chrome-
At SL Modern Hard Chrome's 1,400-gal hard chromium plating solution was recovered
continuously by placing the lonsep™ unit (cell) directly into the bath. The catholyte solution is contained in
a plastic 55-gal drum outside the bath and is circulated through the lonsep™ cell.
The waste generated by the process is sludge from the lonsep™ catholyte solution. This
catholyte sludge contains levels of chromium at 2000 to 3200 mg/L total chromium and 760 to 1050 mg/L
hexavalent chromium and therefore must be handled as a hazardous waste. With the catholyte solution
replaced monthly, the annual discharge of total chromium is 6.5 kg (14.2 Ib).
The bath life is projected to be over 40 years, which results in approximately 35 gal of solution
saved/yr (that would otherwise be disposed of), based on an operation schedule of 250 days/yr (1,400-gal
bath/40 yr).
Waste hexavalent chromium is significantly reduced with the use of the lonsep™ unit - from 165 Ib
to 14.2 Ib. The 14.2 Ib of chromium is the small amount lost to the catholyte solution. The reduction is the
result of not having to replace the bath periodically and of having all rinse water returned to the plating
bath to make up evaporative water losses.
Chromium is saved not only through reduced replacement of plating solution, but also through the
reduced number of rejects. For every reject, the company must strip and replate the part. The reject data
are based on estimates of plant personnel, because no records were maintained. This decrease in
rejects is estimated at 5% and corresponds to approximately 237 Ib chromium/yr that is not disposed of
with the stripper and that need not be purchased for addition to the plating solution. The 237 Ib of
chromium oxide is calculated from the company's plating rate from this bath of 42,900 ft2/yr, at a plating
amount of 1.77 oz CrO3 per ft2, and a 5% savings (O.OSx 42,900 ft2/yr x 1.77 oz/ft2 x 1/16 oz/lb.)
95
-------
#20 CARBON-BLACK DISPERSION PREPLATING TECHNOLOGY FOR PRINTED WIRE BOARD
MANUFACTURING
Participants
The host company for the evaluation was McCurdy Circuits of Orange County, CA. McCurdy
personnel also operated the plating process during the test. Battelle, Columbus Laboratories helped
design the test program, supplied test personnel and equipment, and wrote the draft report.
McCurdy operates two process lines for the through-hole plating of Printed Wire Boards (PWBs):
one uses electroless copper (EC) and the other uses the carbon-black dispersion process. The EC
process at McCurdy Circuits consists of the following 18 operational steps:
1. Acid cleaner
2. Rinse (to discharge line)
3. Microetch (sodium persulfate solution)
4. Rinse (to ion exchange line)
5. Activator-pre-dip
6. Catalyst
7. Rinse (to discharge line)
8. Rinse (to discharge line)
9. Accelerator
10. Rinse (to discharge line)
11. EC
12. Rinse (to separate ion exchange system)
13. Su If uric acid 10%
14. Rinse (to ion exchange system)
15. Anti-ox
16. Rinse (to discharge line)
17. Deionized (D.I.) water rinse (to discharge line)
18. Forced air dry
In the first 17 steps, racks of PWBs are moved from tank to tank with an automated hoist. All the
rinses are single flow through, which generates more wastewater than does cascading or multiple-use
rinses. Because of drag out, the rinse following the EC bath (Step 11) contains complexed copper, which
is hard to treat by typical metal hydroxide precipitation. Rinse water from the EC process is collected in
one of three drain lines: one to a discharge line, another to the first ion-exchange collection system, and
the third to an ion-exchange system for the EC rinse.
Whereas the EC process is essentially a batch process, the carbon-black process is a continuous
system in which parts are placed on a roller conveyor. This carbon-black dispersion technology, termed
BLACKHOLE™ (BH) technology by the vendor/inventor, consists of fewer baths and a simplified process.
It has only the following 11 process steps:
1. BH™ cleaner
2. Rinse (water from step 4, to discharge line)
3. BH™ conditioner
4. Rinse (fresh tap water, to rinse #2)
5. BH™bath
6. Dryer
7. Microetch
8. Rinse (water from step 10, to ion exchange system)
9. Anti-tarnish
10. Rinse (fresh tap water)
11. Dry
90
-------
L-'CO^I IfJUUI 1
Without lonsep™
Cr bath
Bath disposal
CrO3 due to rejects
Labor due to rejects
Strip solution reDlacement and disposal
Total without lonsep™
With lonsep™
Catholyte solution
Sodium sulfate
Sodium carbonate
Water
Barium carbonate
Sludge disposal
Labor
Maintenance
Power
Total with lonsep™
Annual Savings
Annual Use
35 gal
35 gal
237 Ib
140hr
100aal
120 Ib
288 Ib
660 gal
36 Ib
4 drums
41 hr
8,100Kwh
==^ — — — — — _
Rate ($)
$11.30/gal
2/gal
1.13/lb
20/hr
1 .50/gal
18/1 00 Ib
18/1 00 Ib
2.66/1000
gal
1.22/lb
205/drum
20/hr
.0902/Kwh
Annual Cost
($)
$ 396
70
268
2,800
1 SO
• \j\j
— ^— — — — — — _ _ __
$3,684
22
52
2
44
820
827
1,080
731
$3,578
$ 106
Wastewater discharge costs are not applicable. Analytical costs are assumed to be the same for both cases.
Chemical analyses of the chromium plating solution were conducted to verify that the solution met
operational specifications for hard chromium plating solutions. Analysis of the bath showed rt was
level was below the maximum °f 52 9/L and the
The chromium-plated parts, are inspected for pits, blisters, other deformities and chromium
hckness. S,nce mstal.ation of the lonsep™ unit, the number of rejects has decreased oy abouT^T The
improved chromium plating quality has produced more uniform chromium deposit and fewer pits
Paramax--
At Paramax, chromium-based etchant was tested as a batch process. Eight of the six cells in the
process ,ne were used, and over a 2-day period, 200-gal etchant baths were removed from he oocess
line to a treatment tank and pumped through the lonsep™ unit. The unit ran continuously Sr 3 days
97
-------
Annual chemical usage and cost for both processes are shown in Table 2 As seen in Table
BH has lower operating costs than does EC in all cost categories that could be obta^ed
-------
Economic Evaluation--
Because the lonsep™ unit at Paramax is still undergoing testing, most of the cost analysis was
based on Paramax estimated derived from economic information in plant records and experience of
Paramax personnel. These estimates indicated that the unit can prevent disposal of approximately seven
etchant baths/wk and, thus, save disposal and replacement chemical costs.
Operating cost factors involved in the economic analysis for Paramax include labor, maintenance,
chemicals, utilities (water and electricity), and waste treatment/disposal. Table 4 compares costs with and
without the lonsep™ unit.
TABLE 4. ECONOMICS OF THE ETCHING LINE
Description
Without lonseo™
Etchant (concentrate)
Etchant disposal
Labor for disposal
Water*
Total
With lonsep™
Catholyte solution:
Sodium chloride
Sodium sulfate
Soda ash
Water*
Labor
Maintenance
Power
Total
Annual Use
20,625 gal
41 ,250 gal
150 gal
20,625 gal
1 0,000 Ib
5,000 Ib
1 ,000 Ib
25,000 gal
2,000 hr
1 ,430 Kwh
Rate ($)
4.85/gal
2.31 /gal
20.00/hr
2.66/1 000/gal
3.50/50 Ib
17.50/1 00 Ib
0.23/lb
2.66/1 000 gal
20.00/hr
0.045/Kwh
Annual Cost ($)
$100,031
95,288
3,000
55
$198,374
700
875
230
66
40,000
30,000
65
$71,936
Water costs include sewage fee.
The capital cost of the unit (specific to Paramax), including installation and modifications, was
$563,000. Dividing this by the estimated annual savings results in a payback period of
4.5 yr.
99
-------
°F AN ELECTRODIALYTIC PROCESS FOR PURIFICATION OF HEXAVALENT
Participants
The host companies for this evaluation were Paramax Incorporated of St Paul MN and SL
t^t R tV3,? rhr,°m! iP ? Tden> NJ" B°th comPanies °Perated their electrodialytic equipment during the
test. Battelle, Columbus Laboratories helped design the test program, supplied test personnel and
equipment, and wrote the draft report. H«">u'»iei «*na
Technology and Testing
SL Modern Hard Chrome has specialized in industrial hard chrome plating for over 35 years and
alb SSto zLincSPeCtmm mater'alSl ranging from aluminum tnrough the copper, ferrous, and nickel base
u •• ^ Paramax uses a chromic acid solution to etch copper from printed wire boards (PWB) As coooer
builds up in the etching solution, the etching rate becomes unacceptably slow. Prior to installing lonseb™
he chromic acid solution was replaced with fresh solution at the operator's discretion Replacement '
frequency ranged from once a day to once a week. «^«nieru
lonsep™ Electrodialytic Process-
The lonsep™ process extends the life of process solutions by removing metals other than
chromium from the process solution. This improves plating and etching product quality and extends
process solution life. For this application, the process removed copper and converted some of the
tnvalent chromium to hexavalent.
Th* innco™ ,1 stno;fa two-compartment cell used for the purification of a chromium plating solution
The lonsep electrodialytic process uses a voltage gradient to separate salt in a solution into cations and
anions.
ANODE
MEMBRANE
CATHODE
Figure 1. lonsep™ eectrodialytic process
(Source: lonsep™ Corporation, 1989)
Chromium is present in the anionic chromate form in the plating and etching solutions Metal
contaminants (cations) migrate across a semipermeable membrane, under the influence of the electric
field. Conversion of the electroplatable metal cations to insoluble hydroxides occurs when the cations
migrate through the membrane; this migration eliminates the buildup of metals on the cathode
Membranes in the electrodialytic cells serve to physically separate the acidic, basic, and other process
94
-------
#22 SUBSTITUTING CADMIUM CYANIDE ELECTROPLATING WITH ZINC CHLORIDE
Participants
The host for the evaluation was Aeroquip, Incorporated, Industrial Connectors Division, Van Wert,
Ohio. Industrial Connectors personnel also operated the electroplating system during the test. Battelle,
Columbus Laboratories helped design the test program, supplied test personnel and equipment, and
wrote the draft report.
Technology/Testing
The cadmium cyanide and the zinc chloride plating processes for the rack plating line at Aeroquip
are compared in Table 1. Hydrochloric acid is used to condition parts (shown as step 12 in Table 1)
before plating in the zinc chloride process whereas sodium cyanide is used in the cadmium cyanide
process.
The cadmium cyanide plating line had separate tanks to apply either clear chromate or yellow
chromate coatings (steps 18 and 20). Previously, Aeroquip used clear chromate coating on most (90% to
95%) of the cadmium-plated parts.
Currently, Aeroquip uses yellow chromate coating on all zinc-plated parts because (1) Aeroquip
has adopted a worldwide standardization and (2) yellow chromate coating vastly improves the corrosion
protection of the zinc-plated fittings.
The yellow chromate solution used by Aeroquip contained approximately a five times greater
chromium concentration than did the clear chromate solution. In the water-soluble oil application step
(step 22) the concentration of the oil was reduced by a factor of approximately ten in the zinc chloride
plating process from the level used in the cadmium cyanide plating process. The change was necessary
to obtain improved adhesion of chromate coating during the subsequent heat-curing step.
All wastes from the plating operations end up in three waste streams -- treated wastewater,
dewatered sludge, and waste oil - that are discharged or sent for disposal from the wastewater treatment
plant. The treated wastewater is discharged to a waste sewer. The dewatered sludge is collected in a 20-
yd3 hopper and sent to an off-site hazardous landfill once a month. The waste oil is collected in drums
and sent to an off-site hazardous waste incinerator every 3 months.
Product Quality Evaluation-
Product quality was measured by the corrosion resistance of plated parts determined by salt-spray
tests carried out in accordance with the ASTM Method B117-90 (Standard Test Method of Salt Spray
[Fog] Testing). As part of their quality acceptance criteria for zinc-plated parts, Aeroquip's engineering
process specification has adopted the ASTM Method B633-85 (Standard Specification for
Electrodeposited Coatings of Zinc on Iron and Steel) requirement of 96 hr of freedom from white corrosion
products in salt-spray testing. An additional internal acceptance test by Aeroquip was performed. The
acceptance criterion is freedom from red rust after 360 hr exposure to salt spray.
101
-------
r * *K Table 1 summarizes the waste reduction potential of the new technology. The top part of Table 1
.
nT rT'aCement and the estimated amount of chromiumed from
P H fh Ibt°tt0^Part °f Table 1 lists the chemicals used to make the 'onsep™ catholytic
and the additional barmm carbonate used to adjust the sulfate levels of the chromium plating
TABLE 1. WASTE REDUCTION OF THE CHROMIUM PLATING LINE
Without lonsep™
Cr03 in bath solution 80 ,b (41 6|b Cr)
CrO, due to rejects 237 Ib (123lb Cr)
Total Cr03 317 Ib (165 Ib Cr)
With lonsep™
Catholyte solution
Sodium sulfate 120lb
Sodium carbonate ' 288 Ib
Additional barium sulfate from sulfate 43 |b
reduction - —
the amounl of ™tal
The sludge for 1 yr would contain 14.2lb Cr+6.
The annual operating costs with and without the lonsep™ units are itemized in Table 2 The
nfi,°±l,a t°"amP IOre,P Un™n the fa" °f 1991 was $20'000- Because there is a savings of
$106 in annual costs using the lonsep™ unit, there is a payback on the unit of 189 yr.
Although economic considerations are important, they are not the only justification for installina
waste reduction equipment. SL Modern Hard Chrome has zero discharge from its plating ope rations to
the city water treatment system because all rinsewater is returned to the plating baths.
96
-------
A second comparison test was conducted by Aeroquip to directly compare performance of
cadmium and zinc plated parts. Seven groups of parts of each zinc and cadmium plating, and
representing rack and barrel plating products were subjected to the 96hr white corrosion and 360 red rust
salt spray tests.
Results
The initial test zinc plated parts from the rack plating line passed the 96 hr test but a small
number of parts (3 of 48) in each laboratory did not pass the 360 hr. red rust test. The barrel plating line,
zinc plated parts passed both the 96 and 360 hr tests. White corrosion started to appear after 168 hrs.
Red rust was first noticed at 432 hrs.
The comparison test series resulted in all the zinc coated parts passing the 96 hr, white corrosion
and 360 hr, red rust tests. Red rust first appeared in some types of parts after 504 hrs.
All cadmium plated parts passed the 96 hr test for white corrosion. The appearance of white
corrosion was delayed to 336 hrs and beyond. No red rust was observed on any of the cadmium-plated
specimens after 504 hr of exposure, at which point the tests were ended.
These results demonstrated that while both types of plating passed the comparison tests, the
cadmium-plated parts exhibit a superior corrosion resistance to zinc-plated parts with regard to the
appearance of white corrosion products and red rust in salt-spray tests.
Waste Reduction Potential-
Waste reduction potential of the process substitution was determined on the basis of waste
volume reduction and reduction of toxic pollutants. Waste volume reduction was estimated for the treated
wastewater and the dewatered sludge. Toxic pollutants were cadmium, cyanide, chromium, and chlorine.
Tables 2 and 3 show the changes in the total waste and toxic pollutant reduction, respectively.
TABLE 2 ANNUAL GENERATION OF RELATED WASTEWATER AND SLUDGE FROM
CADMIUM- AND ZINC-PLATING PROCESSES (AEROQUIP DATA)
Year
1989
1991(a>
Plating
Process
Cd
Zn
Treated
Wastewater,
gal
40,000,000
44,900,000
Sludge
Ib
282,000
383,000
(a) Adjusted to the 1989 production rate of the electroplating process.
The increases in wastewater and sludge were due to an increase in plating bath concentration
from approximately 3 oz/gal of cadmium in the cadmium-plating baths to approximately 3.5 oz/gal zinc in
the zinc-plating baths. The decrease in oil and grease was due to an approximately ten-fold decrease in
the concentration of oil used in the water-soluble oil dip tank. The increase in chromium was due to an
approximately fivefold increase in the chromate bath concentration. The chromium, which also is a priority
pollutant, was effectively converted from the toxic hexavalent form to a much less toxic trivalent form in
the wastewater treatment plant, resulting in a reduced health risk relative to cadmium.
103
-------
Paramax has since established that the recovered etchant meets its requirements for etching solutions.
The company is currently using 12 cells to keep up with the buildup of copper and has successfully
recycled the etchant during production.
Calculation of the waste reduction potential was based on the difference between the amount of
chromic acid etch solution disposed without lonsep™ treatment and the amount of chromic acid etch that
can be reused after treatment with the lonsep™ unit.
Use of the lonsep™ unit prevented disposal of approximately 7.5 baths/wk. Paramax currently
disposes of 8.5 baths/v/k and estimates that, with the lonsep™ unit, disposal will be reduced to one
bath/wk. Each bath contains 110 gallons (the lonsep™ unit, can treat up to 500 gallons or about four
baths in 4 days), which means that 41,250 gallons of etchant bath solution is saved per year (7.5 bath/wk
x 110 gal/bath x 50 wk/yr). The etchant concentrate is diluted 50% to make up the bath. Therefore, the
lonsep™ unit could reduce 20,625 gallons of etchant concentrate per year needing disposal. This amount
of etchant contains approximately 7,154 Ib of chromium that would otherwise have gone to waste (80 g/L
Cro3 in etchant x 52 g Cr/100 g CrO3 x 20,625 gal x 3.785 L/gal). Table 3 summarizes the items evaluated
in the waste reduction analysis.
TABLE 3. WASTE REDUCTION OF THE ETCHING LINE
Amount Discarded
Description Per Year
Without lonseo™
Etchant
Water
Chromium
With lonsep™
Catholyte solution:
Sodium chloride
Sodium sulfate
Soda Ash
Water
Chromium
20,625 gal
20,625 gal
7,100lb
1 0,000 Ib
5,000 Ib
1 ,000 Ib
25,000 gal
42 Ib*
'Estimated
98
-------
TABLE 4. CAPITAL COST TO CONVERT (1992)
Parameter
Expense (cleanup of old equipment
and waste disposal)
New equipment
Subtotal
Total
Barrel
Plating Lines
$428,000
$424,000
$852,000
Rack
Plating Line
$ 999,000
$ 122,000
$1,121,000
Subtotal
$ 1 ,427,000
546,000
$1,973,000
Report
The full report, titled "Substituting Cadmium Cyanide Electroplating with Zinc Chloride
Electroplating" by B.C. Kim, et al., is available as report no. EPA/600/R-94/074.
105
-------
The etchant was sampled and analyzed for both chromium and contaminants at the end of the 3-
day treatment/recovery process. These analyses were used to determine whether the renovated bath was
within specifications and whether bath quality was an indication of product quality Althouqh total
chromium in the etchant remained constant, hexavalent chromium increased. The hexavalent chromium
started at approximately 74% of the total and increased to about 99%. It is believed that oxidation of the
tnvalent chromium back to the hexavalent form caused the increase in hexavalent chromium
concentration. The resulting hexavalent chromium concentration of 30.3 g/L in the etchant over the 3-dav
sampling period approached the minimum specification level of 31 g/L. This could be increased further by
longer treatment or by adding etchant concentrate. At the time this study was conducted, the etchant had
no been reused and the effect of recovered solution on in-house printed wire board product quality had
not been evaluated.
The cationic contaminants were within specification. The 10.8 g/L (mostly copper) was below the
maximum level of 25 g/L. Because the chromium and the contaminant levels are both near specification
levels, the etchant should be acceptable.
The full report, titled "Evaluation of an Electrodialytic Process for Purification of Hexavalent
Chromium Solutions" by Dale Folsom, et al., is available as report no. EPA/600/R-94/071
100
-------
Vacuum
MOM
ISOWcron
• Rern6t*FlMef '
Cfcfcrtoir* CfcW
1 Moon? ^- SO Micron
Filter
Initial
Fluid
Holding
Tank
Reoirculata
Recirculate;
Fail
Pass
Clean
Fluid
Holding
Tank
\
>
I Pump |
Sustained
Pasteurization
Heater
Additive
Injections
Fluid Test
Pass
Fail
To
Centrifuge
Bypass
Centrifuge
Centrifugal
Figure 1. Metalworking fluids recycling system flowchart
The main purpose of metalworking fluids in machining operations is to provide lubricity and
cooling without causing corrosion or other problems.
Results:
Degree of removal of non-dissolved and dissolved particulates during recycling is shown in
Table 1. High concentrations of these particulates affect tool life, surface finish, and chemical
breakdown. Particulates also provide substrates for microbial growth. At all three sites, the results
showed considerably lower concentrations of nondissolved particulates in the recycled fluids (E1-R, E2-
R, and S1-R) as compared with concentrations in the spent fluids (E1-S, E2-S, S1-SO).
Dissolved solids levels remained approximately the same after recycling, which indicated the
effect of contaminant precipitation and fresh additive introduction. At the three sites tested, the pH of
the recycled fluids was returned to a range between 8.5 and 9.5.
107
-------
Process
Step
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
=====
Tank
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
=
Operation
Zinc Chloride Cadmium Cyanide
Plating Line Platina Line
Soak clean
Rinse
Electroclean
Rinse
Rinse
Hydrochloric acid
pickle
Rinse
Rinse
Electroclean
Rinse
Rinse
Hydrochloric acid
pre-dip
Zinc plating
Rinse
Rinse
Nitric acid dip
Yellow chromate dip
Rinse
Chromate seal
Rinse
Drip tank dip
Water-soluble oil dip
^^ —
Soak clean
Rinse
Electroclean
Rinse
Rinse
Hydrochloric acid
pickle
Rinse
Rinse
Electroclean
Rinse
Rinse
Sodium cyanide pre-
dip
Cadmium plating
Rinse
Rinse
Rinse
Nitric acid dip
Rinse
Rinse
Yellow chromate dip
Rinse
Water-soluble oil dip
An initial series of tests were performed by both Aeroquip and an independent laboratory f Detroit
n T0ry (DTL)' 'TV*' TeStS induded f°Ur types of zinc P|ated P^s from *e rackS l?ne
Additionally, Aeroqu.p tested four types of zinc plated parts from the barrel plating line.
102
-------
Lubricity and wear preventive characteristics of a metalworking fluid affect workpiece quality and
tool life. Lubricity and wear characteristic were measured by the standard "four-ball test" (ASTM D 445).
For Site E1, the recycled sample caused a much lower average scar diameter than did the spent sample,
but not as low as the virgin sample. This indicated that the recycled and virgin samples performed about
the same. The presence of some emulsified tramp oil could have improved the lubricity results of the
spent sample E2-S.
A major factor in metalworking fluid spoilage (rancidity) is microbial growth. In the recycling
process, existing microbes are killed during the pasteurization step, the dead biomass is removed during
the centrifugation step, and a measured quality of biocide is added to control future microbial growth.
ASTM E 686-85 evaluates the effectiveness of biocides at use concentrations. No microbial growth was
observed in the samples up to 6 weeks after recycling.
Currently, there are no published standards for recycled fluids. Each user establishes
requirements based on the same factors used in selecting a virgin fluid. At the three test sites evaluated
in this study, recycled fluids appeared to satisfy the functional requirements of the users.
On an average, Safety-Kleen visits each user once every 10 weeks and recycles 250 gallons of
spent fluid per visit, thereby yielding a potential annual reduction of 1250 gallons for a typical small user.
Approximately 4 gallons of tramp oil per visit are generated during recycling. The tramp oil is hauled away
at a competitive fee by Safety-Kleen for use as supplemental fuel. Residue generated on the filters
(mostly metal chips) is transferred to the user's waste metal bin and later reclaimed for its metal value.
According to a 1991 study by the Independent Lubricant Manufacturer's Association, the volume
of metalworking fluids (concentrate) manufactured in the United States, has increased from 67 million
gallons in 1985 to 92 million gallons in 1990. By extending the life of metalworking fluids through onsite
recovery, considerable amounts of fluid can be prevented from going to waste. The total volume of fluids
going to waste, may be significantly higher than the manufacturer volumes (as much as 20 times higher.in
some cases) since many types of fluids are diluted into 3% to 5% solutions with water.
The economic evaluation compared costs for recycling versus costs for disposal. Recycling costs
included the onsite service charge for the customer and tramp oil disposal cost. Disposal costs included
spent fluid disposal cost and hazard analysis costs. The annual savings for a typical small user, who
recycles 1,250 gallons/yr of metalworking fluid was approximately $1,600 if the spent fluid was
nonhazardous, and $7,800, if the spent fluid was hazardous (by the Toxicity Characteristic Leaching
Procedure).
This evaluation found that recycling of metalworking fluids is a good option for small-to medium-
sized plants with machining operations. In the absence of published standards for recycled fluids quality
and performance, the user has to evaluate the recycled product by the same criteria used to select a virgin
brand. Direct, extended time testing of tool life and work piece quality vs. recycled fluid characteristics
may be desirable to establish recycled fluid standards.
The full report, titled "Mobile Onsite Recycling of Metal Working Fluids" by Arun Gavaskar, et al.,
is available as report no. EPA/600/SR-93/114.
109
-------
TABLE 3. TOXIC POLLUTANTS FROM CADMIUM- AND ZINC- PLATING
PROCESSES
(Ib/yr based on production rate of 3.29 million ft2)
rarfm- a .re 'J e ^^ hazard level of the Wa8te was substantially reduced by eliminatina
cadmium and cyan.de. Consequently, the process substitution has reduced the compaTs w tentlal
Sh^6 > acf'de"tal worker exposure and environmental release of these , he^Wmffii An
added benefrt .s the el.rn.nat.on of chlorine for destruction of cyanide in the wastewater treatment plam
Economic Evaluation-
«
Approximately 72% ($1,427,000) of the total cost was for expenses related to oleanina UD the
cadmium process equipment and for disposal of the waste generated from the deanuo oSnn rL
rema.nmg 28% ($546,000) was for installing new equipmen? The operZg cos s for t'he^Tatino
200 ™ *" *° Cadmium-Platin9 P^ess'resulting in an annual saJ?ngs9
15 vr The
Based on these costs, the estimated payback period is
cannot be justified on economic grounds alone. Justification
safety and reduced environmental pollution plus the market's requirements for zinc patedratheT°han
?hT "? f ,? '" """V W"6"**"- ln comparing the two processes for a new insta lation the
chloride plating process offers obvious advantages over the cadmium cyanide plating process
104
-------
Results
The average adsorbency ratio and extraction efficiency for low viscosity fluids is plotted against
the number of extraction cycles in Figures 1 and 2. The average adsorbency ratio 13.99 g to to 14.79 g
of fluid per g of dry weight of pad.
The results of the rate-of-release tests are given in Table 1. The MPP and MEP of the fresh pads
for the low-viscosity fluid were 6.19 and 5.21 g/g, respectively. The decrease in MPP was 23.6% and
28.9% for pads reused for four and eight times, respectively, and the decrease in MEP was 24.8% and
31.1%, respectively.
Although the pad performance was degraded by approximately 25% after four uses, the
degradtion in performance was relatively insignificant for 4 additional uses. For the medium and high-
viscosity fluids, the MPP and MEP were measured only for the fresh sorbent pads.
The results of the fluid pickup tests are presented in Table 2. Regardless of fluid types, the
sorbent pads effectively removed fluids from the floor. Only 2.4% to 5.2% of the spilled fluids were left
on the floor. Moreover, the sorbent pads effectively removed low and medium-viscosity fluids even after
they were reused four or eight times.
The objective of comparing costs of pad disposal versus reuse was met by using fluid capacities
and process time measured during the study and supplemented by literature and company hitorical
data. Fow low-viscosity fluid, substantial savings occurred as a result of pad recycling. Savings of up to
51.4% and 75.3% weire possible with as few as two and as many as eight reuse cycles, respectively.
Additional savings were also possible, but much less significant, as reuse cycles increased to more than
eight times. Similarly, the cost per use was greatly reduced, from $4.80 for single use to $1.19 for eight
uses (see Figure 3). For medium viscosity fluid, the annual pad recycling savings were 50.5% and the
per use cost was $2.38 for two uses. Additional uses and savings are very unlikely because the sorbent
pads became severely separated and deformed as a result of the extraction process.
16
IS —
14 —
13 -
12 -
11
10 -
9 -
8
Four extractor? cycles
Eight extraction cycles
1 23 45678
Extraction cycle
Figure 1. Absorbency ratio for low-viscosity fluid
111
-------
NEW JERSEY
technologies were evaluated under the New Jersey contingent of the WRITE oroaram
fNJDFPF P±!?™With M6 helpDof !he,New Jersey Apartment of Environmental Protection and
(NJDEPE, Project Officer, Norme Binder) and the New Jersey Institute of Technology (NJIT
Officer, Daniel Watts). The EPA Project Officer was Johnny Springer Jr. CnnOI°9y (NJI''
#23 MOBILE ONSITE RECYCLING OF METALWORKING FLUIDS
Participants
Phii.H ,The ^ f°r ^f evaluation were three, small-to-medium sized machine shops in the
Philadelphia PA, area (known to EPA only as E1, E2, and S1). The Safety-kleen Corporation of Elqin
B±LWarSn^H*±r;JPrTding *? r^0^9 flujds and operating themobi.e, on'site recyc inguni
LjHTTPII^ I -fill imni 1C I arirtroTyM-i^r* l*s>l»i,«tj-J*J._.«:_._.Li ._. .. _ j •*+••• t\* \*t it
_ units.
and wrote the draft report.
Technology/Testing
thn thStafety-Kleen Prides metalworking fluid recovery services to a variety of businesses
those that generate relatively small quantities of fluid waste. The mobile service performs^the r^vc.ma or
r.tfm6''3?? ^^'thUS elim'natin9 tne need to transport potentially hazardous waste Each mobile
gal/hr °P6 9 On ** ™" P°Wer> JS Capable °f Processin9 fl^d at a maximum rate of 300
The recycling process (Figure 1) consists of filtering, pasteurizing, and centrifuging the spent fluid
11f!uS J°U?h a, 1°°"t) fllter to remove anV Iar9e Particulates. It is then pumped thrc
ger to kill bacteria and fungi, as well as to reduce fluid viscosity
and other debris from the usable fluid, is next. Additives a
performance. In the final step, the fluid flows through a
'p0!"10'09? was evaluated at three small-to-medium sized machine shops (sites) in the
, PA, vicinity. The three sites were chosen from among Safety-Kleen's cus terror base Twn
ida E1 and E2> USSd emulsion-'VPe metalworking flu'ids. The th'd ^ite S s" used a
At each site, one sample each of the spent, recycled, and virgin fluids (at their normal use
concentratons) was collected and subjected to a series of tests. ThecompariLn^e then made
between the virgin, spent and recycled fluids.
f cu™latj°nuof verY sma» Particulates over time and use could limit the number of times a
as an indicat of^
A oh.n USerSu°f metalworkin9 fluidl often monitor the PH as an easily measured indicator of fluid quality
A change ,n pH may mdicate chemical degradation or degradation due to microbial growth The recvcL
process seeks to restore pH to a range of 8.5 to 9.5. This alkaline pH improves emulsion stabil^ an
corrosion resistance characteristics of the fluid. "UK.IUM biaointy ana
106
-------
TAE3LE 2. FLUID PICKUP BY SORBENT PADS
Fluid pickup (%)
Replicate No. /pad No.
Fluid
viscosity
Low
Medium0
High
Pad.
Condition
Fresh
4Xa
8Xb
Fresh
4X
8X
Fresh
4X
8X
1/28
964
932
942
1/31
97.1
975
95.8
1/34
100
N/A
N/A
2/29
98.2
97.2
95.8
3/32
96.2
94.1
93.8
2/35
94.2
N/A
N/A
3/30
98.2
96.2
95.8
3/33
97.5
94.2
99.5
3/36
100
N/A
N/A
Average
97.6
95.5
95.3
96.9
95.3
94.8d
98.1
N/A
N/A
a Pad extracted four times.
° F??airmeCd?um^scSySfiuid tests, pads were soaked at 50% pad sorbing capacity before extractions.
d Based on the performance of Pads No. 31 and 32 only.
N/A = Data noT available because pad could not pass through Extractor™
Because the capital cost for the Extractor™ was relatively insignificant ($699) and the annual
savings would be substantial, the payback period of the investment would be only 2.8 to 5 weeks.
The sorbent pad recycling evaluation demonstrated that roller compression technology can be
effectively used to extract low and medium-visocsity fluids from meltblown polypropylene sorbet pads.
The Extractor™ is particularly useful for low-viscosity fluid applications; the sorbent pads can be reused at
least eight times. For medium-viscosity fluids, no more than two to three reuse cycles are possible^The
potential to reduce waste by recycling sorbent pads can be substantial. For example, for a 1,858-m
(20 000-ft2) plant, annual sorbent pad consumption can be reduced from 3,600 pads to 1,800 or 450 if the
pad's can be reused for two or eight times, respectively. Correspondingly, the number of drums for
disposal of pads would be reduced from 24 drums (assuming 150 oil-saturated pads per drum) to 6.5 or
1 6 drums (assuming 275 desaturated pads per drum). The 14 to 16 drums of waste fluids extracted from
the sorbent pads would be processed for reuse or hauled away for disposal at a waste-to-energy facility.
The economic benefits of the roller compression technology were substantial. The use of the
Extractor™ by shops and plants that handle and/or use various oils and fluids would result in annual
savings of 51 % to 75% The savings come primarily from the lower disposal costs for spent pads.
Further savings may be possible if exracted fluids can be recycled. The per use cost of sorbent pads can
be significantly reduced from $4.80 for a single use to $1.19 or less for eight or more reuse cycles.
GD° The full report, titled "A Fluid Sorbent RecyclingDevice for Industrial Fluid Users" by Abraham S.C.
Chen, et al., is available as EPA/600/SR-93/154.
113
-------
concentrate (approximately 5% solution of the concentrate in tap water) 9enerated n° mst at the use
^
separation was not.cec, in any of the recycled samples, indicating the trampThad I beer "removed.
TABLE 1. ANALYSIS OF NON-DISSni VFn SOLIDS
Non-Dissolved Solids Concentration
(mg/100mL)E
Sample No.
E1-Sb
E1-R
E1-V
E2-Sb
E2-R
S1-S
S1-R
S1-V
Total
79.10
22.55
3.55
12.55
5.60
33.80
17.00
5.18
Inorganic
27.25
1.45
2.50
0.50C
3.00
14.50
1.95
0.78
' By ASTM D 2276. Particulates smaller than 8 microns
Analyzed skimming off and discarding the floating tramp oil. E1-S=spent emulsion site 1 •
E1-R=recycled emulsion, site 1; E1-V=virgin emulsion, site V etc emu's'°n, site 1,
Possible inhomogeneity giving a low value.
Dissolved Solids
(Conductivity)
jmhos/cm2
~—- i—
2,400
1,810
700
1,820
1,750
1,450
1,460
1,930
At all three sites, the recycled and virgin fluid viscosities were very close This indicated that thP
ecyc ng process had restored this parameter. The viscosity measurements also indicated fhauhe
recychng process succeeded in returning the fluids to the required use concentration Swateirafo).
108
-------
A higher level of accuracy or component identification confidence (CIC) is needed to avoid the
cost of erroneously replacing non-defective components, potential damage created during component
replacement, and multiple iterations of testing and repair. An experiment was performed to compare the
capability of each cooling method to identify components with thermally intermittent failure modes.
Thirteen test articles were evaluated with the use of each of the three cooling methods, r-12, air and
nitrogen.
Three technicians, working independently, evaluated the test articles following a randomized
sequence of cooling methods. For each evaluation, the technicians assigned a CIC level which reflected
their confidence that they had been able to isolate the cause of the circuit failure using the assigned
cooling method.
Electrostatic Discharge-
The amount of electrostatic charge buildup generated by the cooling material as it is dispensed is
a concern because of the vulnerability of electronic components. Two experiments were designed to
compare the electrostatic charge generated by the various cooling method/nozzle combination for R-12,
compressed air and liquid nitrogen.
The first experiment measured the electrostatic charge generated on the nozzle during release of
cooling material. During a 10-to 12-sec material release, the nozzle was held parallel to and
approximately 1 inch from the platen of an Ion Systems, Incorporated, Model 200 Charged Platen
Monitor*, which measured charge buildup. Two measurements were taken for each cooling
method/nozzle combination. The second experiment measured electrostatic charge buildup when cooling
material was dispensed toward circuit boards placed on the platen of an Ion Systems Model 200 Charged
Platen Monitor. Six circuit boards were evaluated, with two measurements taken for each cooling
method/nozzle combination. The six circuit boards were selected to provide component and density
variety.
Cooling Rate and Absolute Temperature Drop--
The cooling rate and absolute temperature drop were measured for each method. An experiment
was designed to estimate the rate of change of component temperature. Two test boards were
fabricated, one having integrated circuits and the other having wound-film capacitors. Each test board
contained three components with thermocouple buried inside and one exposed thermocouple. During
tests, all four thermocouples on a test board were connected to a four-channel data logger, which
simultaneously recorded temperatures of all four thermocouple as cooling material was directed at the
target component. Two measurements were taken for each combination of test board, cooling method,
direction, and distances. Before each measurement, the cooling material was dispensed directly at the
exposed thermocouple to determine the absolute lowest temperature that could be achieved given the test
distance, direction, and cooling method.
Operator Safety-
Exposure to sound created by the operation of the compressed-air tool was a safety concern. To
assess the potential safety hazard, personnel from the Newark AFB Bioenvironmental Engineering Office
took sound-level measurements during operation of the compressed-air tool.
115
-------
#24 A FLUID SORBENT RECYCLING DEVICE FOR INDUSTRIAL FLUID USERS
Participants
equipment, and wrote the draft report.
Technology/Testing
'
ion method extracts the sorbed fluid and permits reused 'the Dads
The extraction efficiency test (ASTM Standard Method F726-81 ) was used to determinp th*
srss.
he correlation of Performance of the sorbent pads vs. the number of cycles through the
.
110
-------
Distance from the target component affects the component cooling capabilities of both
compressed air and liquid nitrogen. As the distance from the component to the nozzle increased from
0.25 in. to 1 in., the minimum component temperature decreased for both alternative methods. A
comparison of component minimum temperature data for two different directions of application indicated
that R-12 is not sensitive to application direction. In contrast, compressed air provided lower component
temperatures for integrated circuits, but liquid nitrogen yielded lower component temperatures for wound-
film capacitors. The most likely explanation of this difference is the variability resulting from manual
positioning of the nozzles.
Results
Accuracy--
The accuracy testing identified that the compressed air system was correct 11 out of 13 times,
liquid nitrogen 9 out of 13 times,, and R12, 11 out of 13 times. Due to the number of criteria that impact on
variability and the limited scope of the project, the results of comparing accuracy among the three
techniques was considered inconclusive. Additional testing in this area could provide better information.
BSD Risk--
Averages of each pair of measurements indicate that both the compressed air and the liquid
nitrogen alternatives generated lower electrostatic charge buildup is not increased by using either of the
alternative component cooling technologies. If aerosol cans of R-12 have been used successfully, either
compressed air or liquid nitrogen should be acceptable alternatives.
Technician Safety-
A sound level of 81 DBA was recorded at the operator work position. Because the sound levels
did not exceed 84 DBA, additional measurements were not required by the Air Force and, in accordance
with Air Force Regulation 161-35, hearing conservation precautions were deemed unnecessary.
Pollution Prevention Potential-
The average R-12 release per article was 232.65 g (0.51 Ib). With the adoption of either
alternative technology, release of R-12 would be eliminated along with the wastestream of empty aerosol
cans. Neither usage nor production information for the United States was available when this report was
written; quantities consumed vary by user, ranging from a few cans per month in repair shops to over a
thousand cans per year in production operations. In light of the Montreal Protocol of 1987, substitutes for
R-12 will be required & subsequently R-12 is not an option for the future.
Economic Evaluation--
Data presented in the project report indicate that a material cost savings of $5.28 per circuit board
can be projected if testing is done with liquid nitrogen instead of R-12. This would result in payback of
$500 dispenser investment after 95 circuit boards have been tested. For a shop that has an existing
adequate air supply, the average operating cost savings for compressed air is $5.26 per board. This
would pay back a $200 air-tool investment after 38 circuit boards have been tested. The payback period
would be extended if additional investment were required to compress and deliver air to the work stations.
Table 2 summarizes investment and payback figures for each alternative technology.
117
-------
84
-I
• Four extraction cycles
• Eight extraction cycles
82 —
80 ~*
Extraction cycle
Figure 2. Extraction efficiency for low-viscosity fluid
TABLE 1. MAXIMUM PRACTICAL PICKUP AND MAXIMUM EFFECTIVE PICKUP
Pad
condition
Fresh
Extracted
four
times
Extracted
eight
times
Fresh
Fresh
Fresh
Ruid
viscosity
Low
Low
Low
Medium
Medium
High
Pad
texture
Unpleated
Unpleated
Unpleated
Unpleated
Pleated
Unpleated
__SB^^=!~-gB-!i
Pad no.
1
2
3
Average
4
5
6
Average
7
8
9
Average
10
11
12
Average
10B
118
12B
Average
19
20
21
Average
Ruid
sorbed at
saturation
(g)
346.54
360.28
350.83
325.55
255.95
203.96
195.71
218.54
194.06
195.57
197.65
195.76
445.65
447.36
452.59
448.53
306.25
292.09
303.41
300.58
444.54
417.91
392.16
418.20
Time to
"stop"
dripping8
(min)
120
120
120
120
120
120
120
120
>120
>120
>120
>120
>120
>120
>120
>120
>120
>120
>120
>120
120
120
120
120
Maximum
effective
PW
5.55
6.57
6.45
6.19
4.51
4.62
5.07
4.73
4.42
4.34
4.45
4.40
11.82
11.18
11.75
11.58
7.78
7.80
7.81
7.80
13.67
13.54
13.68
13.63
Time to
"stop"
dripping0 with
fan on (min)
61.0
61.5
62.0
61.5
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.0
60.9
60.9
60.9
fino
Maximum
pickupr
(fl/0)
4.69
5.57
5.37
5.21
3.90
3.75
4.13
3.92
3.58
3.47
3.58
3.54
9.19
8.58
9.36
9.04
6.86
6.96
6.95
6.92
12.14
12.19
12.38
19 94
b Maximum ta
Max mum Practical Pickup
Maximum Effective Pickup
drippfog oontlnutd at a rate of more than 5 to 15 drops/min
Ruid sorbed at the end of 2 hr/sorbent oad drv weioht
Ruid sorbed at the end of 1 h/with fan on/a%M pad dry weight
112
-------
#26 EVALUATION OF ZERPOL (ZERO LIQUID DISCHARGE SYSTEM) AT PIONEER METAL
FINISHING
Participants
The host for the evaluation was the Pioneer Metal Finishing Company of Franklinville, New
Jersey. The test was performed in cooperation with the New Jersey Department of Environmental
Protection and Energy and the New Jersey Institute of Technology (NJIT). NJIT prepared the draft report.
Technology/Testing
The Zerpol system has been in place at the Pioneer Metal Finishing facility since 1981. In this
system, aqueous effluent is accumulated in processing tanks to allow conditioning in a batch mode. In the
metal finishing application, conditioning is a stepwise process. Initially, an active oxygen compound is
added, as needed, to oxidize cyanide ions. This step is followed by the addition of sodium hydrosulfite to
reduce chromium, if present. Sodium hydroxide is added for pH adjustment to induce precipitation of
metals.
After a two to three day settling period, the clarified water is transferred to a storage tank and
used in the shop process as needed. Approximately two-thirds of the recovered water is used in the
process for noncritical rinsing and about one-third of the total flow is directed to the boiler. The net effect
of the operation is the near total reuse of the effluent stream, thus attaining a zero discharge condition at
the location. The residual solids streams generated from the precipitation process and from the boiler
blowdown are sent off-site for reuse or disposal. The metal contents are reclaimed by smelting. Figure 1
provides a diagram of the process.
Test Description-
A previous EPA supported study by researchers at the University of Central Florida evaluated the
effectiveness of the Zerpol process as a waste management technique for the metal finishing industry.
The study concluded that there were potential applications for the process in certain segments of the
metal finishing industry. However, the report raised some questions that could not be answered based on
the limits of the evaluation that was carried out. These questions included uncertainties regarding rinse
water quality, boiler economics and operation, product quality, and safety.
The purpose of this project was to confirm and extend the previous study of the Zerpol process to
provide information about its waste reduction potential, operation implications, safety issues, and
economics. As a result, this study focused on 3 main objectives. The first objective was to evaluate the
quality of the recovered and recirculated water to determine its ability to produce good quality plated
product effectively and efficiently. The second objective was to evaluate the quality and quantity of the
condensate produced by the boiler to determine its suitability for process critical rinsing requirements.
The third objective was to evaluate the capacity of the water storage tanks in relation to system
operational requirements and flow rate.
Three types of data were gathered during six days of testing, conducted over a 32 day period.
Composite stream samples were obtained for chemical analysis. Samples were taken from ten sampling
points and analyzed for cyanide ions, calcium, magnesium, copper, nickel, chromium, iron, zinc, cadmium,
phosphorus, total solids, total suspended solids, total dissolved solids (TDS), and pH. Actual flow rate
measurements of the generated effluent, recirculated water, steam condensate and boiler fuel gas were
also taken.
119
-------
COMPONENT COOLING ALTERNATIVES: COMPRESSED AIR AND LIQUID
Participant?
i ,hora, Th8 h°St ^IJI8 evalualion "** Newark Air Force Base (NAFB). Ohio. Battelle, Columbus
Laboratories supplied test personnel and equipment, and wrote the draft report.
Technology/Testing
Tire electronic circuit boards that are tested and repaired dally at NAFB come from a variety of
A,r Force Systems, such as inertial guidance systems used in KC-135, C-5, and C-141 arcTaft a^d af°L
saver advisory system sued in the KC-135. " ^ ••" aircran ana a tuei-
Aerosol cans of refrigerant, such as R-12 and R-22, are commonly used in the electronics
manufacturing and repair industries for trouble-shooting circuit boards that have known or suspected
thermally intermittent failure modes. Thermally Intermittent failures occur when tempTratu re ch^geland
material expansion or contraction aggravate the mechanical failure to create an electrical dlconXy
condition. Due to the elimination of CFC's, users are seeking technologies that will replace them!
Two alternative technologies were evaluated, one process utilizing compressed air and the other
using liquid nitrogen. In the air system (Figure 1), compressed air enters a tangential* drilled s^ona*
generator which forces the air to spin down the long tube's inner walls toward the hot air control valveat
sonic velocity A percentage of the air, now at atmospheric pressure, exits through the neediest
2S « ^ ^ The/ema'n'"9 air is forced back through the center of the sonic-velocity airstream
still spinning. Because it moves at a slower speed, a heat exchange to take place, with the slower
moving inner air stream giving up heat to the outer, faster-moving air column The slower inner air
column exits through the center of the stationary generator and out the cold exhaust To obtain the
0 to Jc' the to* requires
Compressed
Air In (70
(21.
F)
Control
Valve
Cold Air
Out(-46«F) Hot Air
(-43°C) Vortex-Generation Chamber Out (212°F)
(100°C)
Figure 1. Compressed air spray gun
The second alternative technology evaluated uses liquid nitrogen. A 1/2-L Dewar flask
of nitro9en
was
114
-------
recirculation rate is about 1,700 gallons per day.
Pollution Prevention Impact--
It should be noted that Pioneer Metal Finishing reported discharge violations frequently while the
former continuous discharge treatment system was in operation from 1975 to 1981. These violations
were a primary reason the previous treatment system was replaced. There have been no reported
violations since the Zerpol system became operational. While the previous system discouraged efforts to
reduce or minimize water use, the installation of the Zerpol system reduced water use and, reportedly,
stimulated additional pollution prevention efforts.
Table 1 provides data on the contents of the water sludge from the plating process. This material
is suitable for metal recovery by smelter. Measured levels of hydrogen cyanide at the effluent mixing tank
and at the boiler blowdown sump were low (0.058mg/m3 and 0.101mg/m3, respectively) and within the
OSHA standards.
TABLE 1 WASTE SLUDGE CONSTITUENTS
Material Analyzed lbs/1000 sq. ft.
CN
TDS
Ca 0.058
Mg 0.0047
Cd 0.000099
Cr 0.23
Cu 0.19
Fe 0.30
Ni 1.53
Zn 0.60
P
The operation of the Zerpol system and the boiler were fine tuned to match the requirements of
the plating process. The 25,000 gallon effluent processing tank used to hold the effluent for 5 work days
(old system) now holds effluent for about 18-20 work days. This results in a reduction in the amount of
chemicals used to precipitate the metal hydroxides in the effluent processing tanks. The Zerpol system
enables 80-83% of the process water to be recirculated. About 7-8% goes into the boiler blowdown and is
transported for treatment. The remaining 1-2% is lost to evaporation.
Economics-
The cost for a 2000 gal/day Zerpol system is $120,000. Table 2 provides information on the
annual cost savings achieved. The payback period ($120,000/$22,500) is 5.3 years.
121
-------
Pollution Prevention Potential--
Economics Evaluation-
used duSSa^^ cooling material
Cooling Rate and Absolute Temperature Drop--
TABLE 1. M,N,MUM TEMPERATURE: ACHiEVED (AT ,4-IN DISTANCE, AND ELAPSED T,ME FOR
THREE COOLING POINTS
Temperature Elapsed Temperature Elapsed Temperatur Elapsed
Integrated Circuit
Target
Component
Exposed
Thermocouple
Wound-Film
Capacitor
Target
Component
-45.0
-54.5
18.0
-27.5
-35.5
29.0
-175.0
-175.0
31.0
31.0
51.0
* Minimum thermocouple temperature assumed to be -175 °C based on wound-film capacitor tests.
116
-------
#27 A REPLACEMENT SOLVENT CLEANER/DEGREASER STUDY AT DUFFY ELECTRIC AND
MACHINE COMPANY
Participants
The host for this test was the Duffy Electric and Machine Company of Chillicothe, Ohio. Duffy
Electric also operated the cleaning unit during the test. Battelle, Columbus Laboratories on contract to
EPA, provided test personnel and equipment and drafted the final report.
Technology and Testing
Duffy Electric & Machine Company repairs and rebuilds electric motors. The company
overhauls large electric motors (AC and DC with greater than 15 hp output). The company also
overhauls small electric motors. The process involves gross cleaning of electromechanical devices to
achieve a level of cleanliness that facilitates inspection, repair, and testing.
The cleaning system used in this study is comprised of a cleaning unit and a rinse unit. The
cleaning unit is the IBR Series 400 parts washer, made by Inter Basic Resources, Inc., of Grass Lake,
Michigan.
The unit features an 11 -inch-diameter impeller blade, mounted at the inside bottom of an
immersion tank, that rotates at 100 rpm. Cleaning fluid is circulated through a 50-^ m filter at 5 gallons
per minute, at 1 foot of head pressure (Figure 1) to create turbulence in the cleaning fluid. A 100-VAC
electric motor powers the unit which is 16 in x 19 in x 11.5 in and it holds approximately 14.5 gallons.
An insertion heater is used to bring the ester bath to approximately 130°F.
The IBR alcohol rinser consists of a chamber that encloses a manifold sprayer, fitted with an
array of nozzles to spray isopropyl alcohol (I PA) onto the parts from a variety of angles. The unit uses
pneumatic pumps driven by air at 110 to 120 psig. Alcohol is drawn from a 5-gallon drum by an internal
pump. The manifold moves up and down at about 4 cycles per minute while rinsing the parts for a
typical 5 minute cycle. The runoff, caught by a pan and drain at the bottom of the unit, is pumped by a
smaller air motor back into the 5-gallon drum.
UGkUtt> LEVEL
10 yroKA4S"
CONTAINER.
Figure 1. Illustration of IBR parts washer and circulation system
123
-------
Cooling Method
Investment
Payback/
Compressed Air
Liquid Nitrogen
^a=a^^^s^== —
$200
$500
38
95
The cost of equipment to deliver compressed air that is clean, dry, and near room temperature in
the volume and pressure required to achieve maximum cooling capability will depend o
equipment and the number of tools to be used. U«H«"U o
Report
M-* I u6 o" ref°n' entitled "Electronic Component Cooling Alternatives: Compressed Air and Liauid
Nitrogen" by Stephen C. Schmidt, et al. is available as EPA/600/SR-94/1 70. Q
118
-------
The check of appearance and color was used to assess how soiled the cleaner and alcohol had
become. By the 7th week of the study, or after washing 30 parts, the cleaner and rinse were very darkly
colored. After 100 parts were cleaned, they appeared highly contaminated.
Specific gravity is useful to track how much soil loading the cleaner and alcohol experience. Both
solutions showed a small, monotonic increase in specific gravity overtime. The main contaminants
expected to be present in the cleaner are oils and suspended solids; in the alcohol, the main contaminant
is dragout cleaner.
The pH of the cleaning solution was measured for solution acidity or alkalinity changes over time.
This test was done to determine whether exposure to moisture over time caused any acid increase in the
ester cleaner. Acidity was determined by extracting samples of the cleaner with water (ASTM D 2110).
The pH dropped rapidly, reaching a steady state of about 5.04 in the seventh week. In all cases, the
materials in the compatibility tests experienced a net decrease in weight and thickness, which is probably
caused by removal of process oils, colorants, stabilizers, and other additives. The loss of these
constituents could have a negative effect on the performance of the elastomers.
TABLE 1. TOTAL SOLIDS RESIDUE ON PARTS
Sample
ID
H2
H3
H4
H5
H6
H7
H8
H9
H10
H11
H12
H13
Motor
Part
A
B
A
B
A
B
A
B
A
B
A
B
Sample
Date
6/17/93
6/17/93
7/16/93
7/16/93
7/27/93
7/27/93
8/6/93
8/6/93
8/19/93
8/19/93
8/27/93
8/27/93
Days of
Study
8
8
37
37
48
48
58
58
71
71
79
79
Weight of
Primary
3.22
3.05
1.99
1.44
1.83
1.33
1.24
1.62
2.60
1.50
2.79
2.73
Solids
Duplicate
3.27
3.29
1.76
1.24
2.06
1.55
1.01
1.66
1.98
1.30
3.42
2.81
(g/parts set)
Average
3.25
3.17
1.87
1.34
1.94
1.44
1.13
1.64
2.29
1.40
3.11
2.77
P2 Impact--
Contaminants in the ester cleaner primarily are oil, grease, and shop dirt. Therefore, the cleaner
itself is assumed to present little environmental or health hazard during use. Annual solvent usages were
calculated to be 51.4 gal of cleaner and 55.2 gal of IPA. These values represent a worst-case estimate for
the cleaner because it was not fully spent at the time the study was concluded. The annual volumes of
waste liquids were calculated to be 39.0 gal of cleaner and 39.1 gal of IPA.
The petroleum solvent formerly used by Duffy Electric Company had been supplied at a rate of
360 gal per year. Industry estimates indicate that about half of the amount of petroleum solvent supplied is
recovered. The remainder is lost due to dragout, evaporation, and spillage during transfers. Therefore,
125
-------
*
\
PLATING LINE
, C^densate from Tank Heater ^ . ^^^
^ Clarified water to Rinses
t Steam to Tank Heaters
f* \
CONOENSATE §>
TANK *5
1.500 gal K
i
< \r
Dfidensate to CONDENSATE
inses /<•>> TANK
"
© v , j
1 * ' W
V_ TANK
^K^ 1 — ZT_J
WATER TUBE ^^
BOILER
60' HP
_J '
i
BOILER
SLOWDOWN
1 @
SALT
CONCENTRA7OR
2,000 gal
Off-she ID sposal
1
1
MIXING
700 gal *
V V © V M
EFFLUENT fFCiiiojr
"SSK?" 4o8aSSa
25,000 gal © 25.000 aal
1 "~^
Sludge
\ v
METALS
CONCENTRATOR
11, 000 gal
Off-Site
Treatment and
Metal Reclamation
CLARIFIED WATER STORAGE ScSf EfiS
50,000 gal 75 ga| ^
^ - ivictAu-u^ waiei
Q Sampling Points
F:igure 1. Pioneer Metal Finishing Inc. ZERPOL SYSTEM -
basic process flow diagram
; historif al °Peratin9 and maintenance data on the present Zerpol System and the system that
Results
Performance--
Evaluation of the Zerpol technology based on the 3 main objectives led to the following findings:
The
The average TDS of the clarified recirculated process water is about 10 000
concentraton , of the recirculated water is higher than what is allowed for dteb^
The measured values m PPM were: Ni - 10.1 (vs. limit of 3.98); Cyanide - 2.9 ppm (vs ^ 20 Thes
frSS K/nT I0"""6'" main^6 " by Pl'°neer because the^ are economically achevable and do not
00 oon ™m w th ?9 PrOC88S- he 3Verage TDS °f the steam condensate is about 500 ppm and
100,000 ppm for the steam condensate and boiler blowdown, respectively. The clarified water
120
-------
#28 A SUPERCRITICAL FLUID CLEANING STUDY: APPLICATION TO INSTRUMENT BEARINGS
Participants
The host for the test was the Honeywell Space Systems Group of Clearwater, Florida, with materials
and testing support from the Naval Air Station of Jacksonville, Florida. Battelle, Columbus, under contract
to EPA, provided project test personnel, gathered data and prepared the draft project report.
Technology/Testing
The goal of this study was to evaluate a supercritical fluid (SCF) cleaning system that could be used to
clean precision parts that traditionally have been cleaned by organic solvents.
Typical bearing cleaning methods include solvent cleaning, aqueous cleaning, and vapor degreasing.
Many commonly used chlorinated solvents such as 1,1,1-trichloroethane and CFC-113, will not be
manufactured after 1995 because of legislation stemming from evidence that these solvents deplete
stratospheric ozone. Still other chlorinated solvents such as trichloroethylene, perchloroethylene, and
methylene chloride present health hazards to persons using them, as well as constituting hazardous air
emissions targeted for reduction by EPA.
This study focused on SCF cleaning of instrument bearings for the following reasons: (1) users of
instrument bearings have relied heavily in the past on CFCs and other organic solvents and have been
aggressive in looking for alternatives to meet current and scheduled environmental regulations; (2)
assembly and testing of instrument bearings requires cleaning at various stages, for which the efficacy of
the cleaning must be very high and the potential for contamination must be very low; and (3) the high
value of instrument bearings merits an investment in finding and developing improved cleaning methods
In this study, a single-component fluid consisting of supercritical carbon dioxide (CO2) was used
as the cleaning medium. The system operating range is shown in Figure 1. By varying pressure of the
cleaning chamber and thereby density of fluid, the solvent power can be tailored to dissolving specific
types of contaminents.
The essential elements of the system consist of:
. CO2 source (compressed gas cylinder)
. Chiller to liquify CO2 gas
. Pressure pump to elevate line pressure
. Hot water bath to elevate line temperature to that of the extraction vessel
. Cylindrical extraction vessel (2.5-L capacity)
. UV detector to observe removed contaminants
. Pressure reduction valve
. Separator vessel
. CO2 flow indicator.
Samples to be cleaned are placed on a cylindrical parts rack that fits inside the extraction vessel.
The process is started by drawing CO2 from the gas cylinder, purging it of water and other higher melting
point compounds in a cold trap, then pressurizing and heating the CO2 to the same P-T conditions as in
the extraction vessel. Heat tape is wound around all critical fluid transfer lines and temperatures are
monitored at various points by thermocouples. SC CO2 flows through the cylindrical cleaning chamber
from the bottom, where it dissolves and carries away soluble substances out through the top. After
extraction, the CO2 and dissolved contaminants pass through a pressure reduction valve, where pressure
is dropped below PC, and then enter the separator vessel. As CO2 returns to the gaseous state, its
127
-------
TABLE 2. ECONOMIC EVALUATION
— —
Expenses Annual Cost Savings
chemical usage $10,000
labor/maintenance $33,000
waste disposal $15,000
energy usage (increase) . $13,000
net savings $45,000
taxes . $22,500
net savings after taxes $22,500
Conclusions--
t«at H ~^e Z?rP°i!!r0 discharge svstem can be used successfully at metal finishing shops where the
pToductqualTy ^^ C°ndensate can be ^^ulated and reused for rinsing without impairing
There are some potential drawbacks to the Zerpol system. The quality of the recirculated ororp^
water (10,000 ppm TDS) may not be acceptable for a,, typesof metal finishing" ^fonKK^
still be required in some critical rinsing operations. The increased workload on the boiler will increase
teh?SctCo0 a? ™y< f° Tde b°iler 'ife exPectancV due to Baling. Softeners are required ahead of
the system make-up if the calcium and magnesium in the system water supply exceed 5ppm.
The results of this investigation demonstrated that at this facility, the Zerpol zero discharge
system eliminates the discharge of an aqueous wastestream by allowing recycling of most of the process
water. The system facilitates the recovery of metals. Salts from the boiler blowdown are treated as
wastes in a standard water treatment facility.
The positive factors represent a reduction in the volume of waste generated at the facility throuqh
a series of process modifications that have stimulated additional source reduction initiatives at the facility
will be avTai?aSe
122
-------
Solvent PIOOM
SCF Ctoonlng
A**«mbly and Lubrication
Figure 2. Comparison of sovlent and SC CQ cleaning process steps used by Honeywell Space
Systems Groups for instrument bearing cleaning
TABLE 1. OPERATING CONDITIONS FOR SC CO, CLEANING
Parameter
Pressure (gauge)
Temperature
SC CO^ Density
SC CQ, Flow Rate:
(mass)
(linear)
(volume)
Scientific Units
240 bars
85* C
0.6 g/cm3
3.4 kg/hr
1 cm/min
80 cm3/min
Engineering Units
3500 psi
185^
40 Ib/ft3
7.4 Ibs/hr
0.4 in/min
5 in3/min
Time
1 -3 hours
1-3 hours
Product Quality Evaluation: The product quality evaluation consists of two parts: cleaning effectiveness
and material compatibility. Determination of product quality involved testing both steel test coupons and
actual bearings. The sample sets were artificially lubricated with several typical lubricants and then
cleaned by both the solvent system and the SC CO, system for comparison. Both sets of parts were
examined visually to determine gross cleaning. The test coupons were further analyzed by Fourier
transform infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy (XPS). Torque tests were
performed on the bearings to determine if they had been adequately cleaned.
129
-------
The cleaner used in this study, Petroferm BIOACT™ 285, was selected because it is
representative of this class of material. BIOACT is a mixture of high molecular-weight aliphatic esters and
can be categorized as a semiaqueous fluid. The cleaner is meant to be used without dilution and must be
rinsed with alcohol, such as IPA, rather than with water. The alcohol used is a technical grade that is at
least 98% IPA, with the remainder being water. The IPA evaporates rapidly due to its high vapor pressure
(3 mm Hg @ 68|DF).
Testing: The product quality evaluation involved analytical testing to ensure that the new technology
provided an acceptably cleaned product. Cleaning effectiveness was evaluated on the basis of visual
inspection by an experienced technician at Duffy. Additionally, for the purposes of this test, a more
quantitative approach v/as also tried to monitor cleaning effectiveness.
Two similar motors were selected from the shop's inventory, each was disassembled and a rotor
housing with stator and end cover were selected for testing. The two sets of motor parts were'
contaminated with an oil/carbon mixture, representative of the soils found on actual motor parts The
soiled parts were heated in a 105|DC oven for 16 to 24 hours (MIL-C-85570B) and then cleaned at regular
intervals. Cleaning performance was determined by measuring residual soluble surface material on the
parts. The parts were cleaned using the IBR cleaning system, then visually inspected by a Battelle
technician and by Duffy staff.
Thereafter, the parts were cleaned by agitation in a 1-L bath of hexane. The hexane was
evaporated onto platinum weigh dishes to determine nonvolatile matter according to ASTM D 1353 i e
the amount of residue remaining after cleaning. ' ' "'
Certain parameters for monitoring the condition of the cleaner and alcohol also were checked
including appearance, color, nonvolatile matter (alcohol only), specific gravity, and pH of a water extract of
the ester cleaner.
One of the product quality parameters was designed to show possible adverse effects of the
cleaner and IPA on wire insulation materials. Tests were conducted to evaluate whether the elastomers
are compatible with BIOACTTM 285. Swell ratio (by weight and thickness change, ASTM method D 2765)
was measured in small (approximately 2-inch-square) coupons of the elastomers Buna-N, Hypalon M
sihcone, and neoprene. These materials are used for electrical insulation on the wire leads of older electric
motors.
Waste volume was determined by measuring the volumes of spent cleaner and IPA after
completion of testing. The test was completed after cleaning about 108 parts over a twelve week period.
Results
Performance-
The data in Table 1 show that the residual soil levels on both sets of motor parts varied
consistently over the course of the study. Except for the first measurement (day 8), residue measurements
from day 37 to day 86 are within 1 to 2 g per set of motor parts. The higher values on day 8 are believed to
be due to removal of debris from the motors or to incomplete cleaning, because these results are not
consistent with the remainder of the cleaning runs. Higher residue measurements at 71 and 79 days are
believed to be caused by soil-loading of the ester cleaner.
124
-------
TABLE 2. POLYMERIC MATERIALS TESTED IN THIS STUDY
Common or
Trade Name
Teflon™
Delrin™
Meldin™
Phenolic
Chemical Name
Polytetrafluoroethylene (PTFE)
Polyoxymethylene, Acetal Resin (POM)
Porous Polyimide (PI)
Phenol-Formaldehyde Resin (PF)
Nominal Sample
Thickness (in)
0.064
0.12
0.21 (0.6 dia. rod)
0.265
SC CO2 cleaning did not result in a length change in the phenolic material but did in the Teflon™
and Delrin™ materials. In general, the polymer coupons increased in length. The magnitude of the
increase was greater clue to SC CO2 cleaning than from the alcohol and Freon™ cleanings. The length
changes were all less than 0.6%, therefore dependent upon the situation this change may not be
significant.
Pollution Prevention Potential-
At the site of this study, approximately 150 bearings are cleaned per year. Approximately 300 mL
of each of 5 solvents (Freon™, toluene, hexane, isopropyl alcohol and acetone) are used for cleaning
each bearing. This means that approximately 12 gallons of each solvent is generated as waste (some in
the form of air emissions) per year. Freon™-113) will no longer be manufactured after 1995. Toluene is
on EPA's list of 17 priority chemicals that was targeted for 33% voluntary reduction by 1992 and 50%
reduction by 1995. These emissions and wastes are substituted with CO2. CO2 is not generally
considered as a hazardous material or waste. It is, however, correlated with global warming. Additionally,
wastes can be produced as part of its manufacture and processing. Use of waste CO2 as distinct from
that manufactured from virgin material and for the express purpose for making CO2 also impacts the net
effect of P2. Sources such as lime kilns, for example, would not produce a net increase of CO2 while
utilizing a waste product from that operation.
Economics-
The economics of the SC CO2 cleaning process include capital and operating costs. Capital costs
include investment in equipment and installation. Operating costs include purchasing CO2, energy, labor,
and maintenance. These costs were determined by records of purchases and experiences in using the
system, provided by Honeywell. Table 3 compares the major operating costs of the two.
The SC CO2 system can process a maximum of 20 bearings per load, while the solvent system is
a small bench top unit that cleans one bearing at a time. The cost estimate for the SC CO2 system is
based on running a full load each time. This may not actually be feasible with a facility that cleans only
150 bearings a year. The annual operating cost per bearing would be higher for smaller production rates.
Honeywell recycles their Freon™ which would result in a lower purchase cost, but this would be offset by
the cost of recycling. In this scenario the purchase cost was used because the recycling costs were not
readily available. The disposal cost does not include the price of Freon™ disposal.
131
-------
about 180 gal of spent solvent can be collected for distillation and later reuse, and another 180 gal is
unrecoverable. In contrast, the new cleaning system generates 106.6 gal of spent solvent per year of
which 16.1 gal of IPA is unrecoverable due to evaporation, and 78.1 gal of liquid waste is produced.'
Air emissions need further study when evaluating the benefits of an ester- and alcohol based cleaning
system and making comparisons with other petroleum distillates or chlorinated organics. The petroleum
distillate solvent contains an unknown amount of toluene, ethylbenzene, and other hazardous air
pollutants. The chlorinated solvents are both volatile and toxic. The ester-based cleaner itself does not
result in any significant evaporation. Although the alcohol rinse has a high evaporation rate, its
constituents are not generally considered as hazardous as those found in petroleum solvents or
chlorinated solvents.
Economics-
Table 2 gives the annual operating costs of both the existing petroleum solvent cleaner and the
alternative ester-based cleaning system. The major operating cost of the new cleaning system is due to
the cost of the ester cleaner and IPA rinse. If a heating element is used, energy usage also needs to be
considered. Disposal costs may vary depending on the system currently in use. At this shop a contractor
retrieves the used petroleum solvent for recycling and supplies the shop with a clean recycled product for
use.
Under the ester/alcohol system, the same contractor supplies the cleaning and rinsing fluids
While these solvents could be recycled , the contractor does not have a recycling system in place.
Recycling of these fluids could favorably impact the price of the ester/alcohol system. Changes in
disposal costs and compliance requirements could also significantly change this equation.
TABLE 2. ANNUAL OPERATING COSTS
New Cleaning System
Ester: 51.4 gal @ $20.00 per gal $1,028
Isopropyl Alcohol: 55.2 gal @ $3.00 per gal $166
Disposal: 78.1 gal @ $2.50 per gal $195
Total3 $1,389
Existing Petroleum Solvent System
Solvent purchase and disposal $1,070
Totalb $1,070
b I°!a! 3oes not !'nclude labor. energy, and small maintenance costs
Total does not include cost of drying the parts, should faster drying be necessary.
The ester/alcohol system may be less expensive in comparison with chlorinated solvents. Actual
applications need to be considered for determining economics.
The full report, "Replacement Solvent Cleaner/Degreaser Study at Duffy Electric and Machine
Company, by Bruce M. Sass, et a!., will be available as an EPA series 600 report.
126
-------
#29 REPLACEMENT NON-METHYLENE CHLORIDE PAINT REMOVER
Participants
The host for the evaluation was the Tooele Army Depot, Consolidated Maintenance Facility, Tooele,
Utah, who also operated the Parts Chemical Cleaning System (PCCS). Battelle, Columbus Laboratories,
on contract to EPA, helped design the test program, supplied test personnel and equipment, and wrote the
draft report.
Technology/Testing
The focus of this study is on the Parts Chemical Cleaning System (PCCS), which is designed for
depainting, cleaning and applying conversion coatings to ferrous and nonferrous engine parts and
powertrain subassemblies. Application of conversion coatings is a surface preparation method to provide
corrosion protection and increase adhesion of the paint.
The PCCS is designed such that an automated overhead monorail transports baskets of parts through
tanks of paint remover and various rinses prior to application of conversion coatings. A pre-programmed
system controls the process by controlling the material handling equipment that immerses the baskets into
the tanks for predetermined dwell times, drains the baskets, and moves the baskets to succeeding tanks.
The rinsewater from the process contains paint remover, dissolved paint resins, pigments, and other
paint additives. This rinsewater is piped into a holding tank and then is later treated by a fixed-film
biological reactor prior to merging it with a stream that goes to the Industrial Waste Treatment Facility.
The system employs automatic controls to regulate tank solution levels, temperatures, agitation tank
ventilation, tank heating and solution filtration.
This study provides an investigation of an 86% N-methyl-2-pyrrolidome (NMP) and 14%
monoethanolamine (MEA) mixture as a substitute for methylene chloride solvent in an immersion paint
removal operation.
To evaluate product quality, test coupons were made and processed through the paint remover system
along with actual parts. An equal number of coupons were coated with heat resistant coatings and
chemical agent resistant coatings (CARC) respectively. The degree of paint removal from the coupons
was qualitatively evaluated. The baskets of actual parts were evaluated on a pass/fail basis. To evaluate
the pollution prevention potential of the new paint remover solvent system, three process streams were
evaluated. First, the paint removal solvent was evaluated for dissolved metals (Cd, Cr, Cu, Pb, Mn, Ni,
and Zn) and concentrations of NMP and MEA. The rinsewater entering the biological reactor was
analyzed for concentrations of metals, NMP, MEA, pH, total suspended solids, total organic carbon, and
chemical oxygen demand. After treatment by the biological reactor, the rinsewater was also analyzed for
the same analytes.
Results
The NMP/MEA solvent mixture removed the heat resistant coating from ten out of thirteen test
coupons. The mixture also performed well at removing the polyurethane topcoat of the CARC. The
topcoat was removed on all test coupons. The mixture demonstrated difficulty in removing the primer coat
of the CARC system. Percent removal ranged from 32% to 95%. Based on visual observations of the
test coupons, the NMP/MEA mixture removed the primer coat more readily on smooth surfaces than
rough surfaces.
133
-------
solvent power decreases substantially and contaminants drop out of solution and remain in the separator
vessel. The CO2 continues to flow out of the separator vessel through a flow indicator and to the
atmosphere. The operating conditions for the SC CO2 cleaning unit are listed in Table 1.
Temperature (*C)
I
I
«
I
Gas-Uquld
Equilibrium
.50 .75 1.0 1.25
Density (g/ml)
Source: Schn«id«r (1978).
Figure 1. Pressure-density diagram for pure CO2. Isotherms are solid lines; Cp is critical point;
shaded area is useful cleaning region
The Space Systems Group purchased supercritical fluid processing equipment from Liquid
Carbonic in 1990, and modified it extensively to perform cleaning operations. Supercritical fluid cleaning
is one technology that the Space Systems Group is evaluating as a substitute for organic solvents. A
comparison between solvent and SC CO2 cleaning processes is shown in Figure 7. The SC CO2 cleaning
unit (shown in Figure 1) is operated full time at the facility and now routinely handles a portion of the
cleaning workload. The cleaning system is still being evaluated by Honeywell, and improvements are
continually being considered and implemented.
128
-------
WASHINGTON STATE
Seven technologies were evaluated for the state of Washington. Robert Burmark of the Washington
Department of Environmental Quality (DEQ) was the Poject Officer for the state, participating in
identification of state pollution issues, locating host companies, preparation of Quality Assurance Plans
and, assisting with the demonstrations.
#30 RECYCLING ELECTRIC ARC FURNACE DUST: JORGENSEN STEEL FACILITY
Participants
The testing was performed at the Earle M. Jorgensen (EMJO Steel Company in Seattle. The developer
of the technology, Roger B. Ek and Associates, provided a pilot scale unit that used K061 electric arc
furnace dust generated by the plant. The developer provided test personnel to operate the unit. The
SAIC Inc., on contract to EPA, provided test personnel to design EPA's part of the test, obtain the data
and drafted a final report.
Technology/Testing
The steel-making industry produces a large amount of Electric Arc Furnace (EAF) dust as part of
normal production. A glass technology called Ek Classification™ (hereafter called "the Process") has
been developed by Roger B. Ek and Associates, Inc. (hereafter called "the Developer") to recycle this
listed waste (K061) and convert it, along with other byproducts of the steel-making industry (i.e., spent
steel slags, spent refractories, mill scale, and grinding swarf), into marketable commodities that'are
defined as nonleachable by TCLP protocols. These products may include colored glass and glass-
ceramics; ceramic glazes, colorants, and fillers; roofing granules and sand-blasting grit; and materials for
Portland cement production.
The goal of this project was to evaluate the effectiveness of the Process in generating a nonleachable
product from K061-listed waste. Three glass recipes were designed for use at EMJ, identified as Glass I,
II and III. The EPA test program focused on recipe II.
Due to the scope of the effort, the EPA work was restricted to the collection of two duplicate samples
from each of the solid products for the Glass II recipe (granular and castable).
The test furnace was located in the steel-melting area of the EMJ plant so that fugitive emissions could
be collected with the steel plant's dust collection system and routed back to the baghouse.
Gas burners were used to bring the furnace up to operating temperature (2,400 to 2,SOOT). This
operation required approximately 12 hr. Once the operating temperature was reached, the glass batch
was added. The electric heating system utilized two, commercial-sized, 1-1/4-in. diameter, molybdenum
electrodes.
The testing used natural gas as the primary melt energy. The purpose of the electric melting tests was
to establish melt conductivity, measure the amperage flow at constant voltage and select glass
temperature isothermal conditions. These data were used to determine the specifications for transformer
equipment (especially the operating voltage range) to be used in full-scale operations. The electrodes
also provided heat to maintain the furnace temperature between 2,400 and 2,500°F. Each batch
produced about 250 to 300 Ib. of molten product.
135
-------
The impact on the polymer materials used in bearings as ball retainers and oil seals was
c TM^, Po|ymer couPons were subjected to either SC CO2 cleaning or to one of two solvents (IPA or
Freon ) for 4.5 hours. Measurements of weight change (swell ratio) and length change were made for
instrument bearing cleaning.
Results
Results of the cleaning tests after SC CO2 cleaning, showed no contamination in any of the
coupons tested with the exception of the Mil-L-6085 contaminated coupons. An aliphatic ester was
detected at the O(D polarization angle. Unfortunately, the Mil-L-6085 contaminated coupon that was
cleaned by the solvent system was not analyzed at the O|D polarization angle. Therefore it is not certain
whether aliphatic ester would have been detected by FTIR on the solvent-cleaned coupon It can only be
S?£ , !^at SC C°2 Cleanin9 did not remove al1 of the ester lubricant. However, it is not know from
FTIR if SC C02 cleaning of Mil-L-6085 lubricant is comparable to solvent cleaning or inferior to solvent
cleaning. FTIR does show that SC CO2 cleaning removed PAO and PFPE lubricants from 440C bearina
surfaces as effectively as solvent cleaning, within the sensitivity of the method.
Bearing Tests -
A performance test of the cleaned product may be the most important test of the SC CO, cleaning
method. In this case, instrument miniature bearings were cleaned using the SC CO2 method to determine
cleaning effectiveness. Removal of lubricating fluids from bearings is expected to be less efficient
compared with cleaning flat coupons.
Spin-axis bearings with Teflon retainers were supplied by the Naval Air Station, Jacksonville
Florida. Torque tests had been performed on these bearings after they were received from the
manufacturer and again after they were cleaned and relubricated at Jacksonville Naval Air Station All of
the bearings were new, and had not been used other than for testing purposes. The startup and running
(or dynamic) torque tests are appropriate for spin-axis bearings, which are defined by MIL-STD-206B.
The procedure for testing torque involved running baseline torque tests on solvent cleaned
bearings and comparing these results with SC CO2 cleaned and relubricated bearings. These tests show
whether SC CO2 cleaning affects the measured torque, possibly due to incomplete cleaning swellinq of
ball retainers, or other effects. y
Polymer Compatibility Tests -
This portion of the product quality evaluation determined whether the SC CO2 cleaning method
has adverse effects on the polymer components of a product. These may include changes in
appearance, swelling of the material, or changes in length. Certain polymers that are commonly used in
bearings as ball retainers or as oil seals were tested to determine if they are adversely affected by SC CO
and CFC solvent cleaning. The polymeric materials tested are listed in Table 2. The coupons were *
subjected to SC CO2, Freon™-113, or isopropyl alcohol (IPA) for approximately 4.5 hours per each
procedure. Three types of measurements were conducted on the polymer coupons: appearance length
change and swell ratio. Sample coupons were cut from sheet stock in approximately 1 in x 1 in
dimensions for appearance and swell ratio tests. Sample coupons were cut from sheet stock in
approximately 1 in x 6 in pieces for the length change tests.
After removing the coupons from the CFC bath or SC CO2 chamber the coupons were inspected
visually under white light for color change, curling, thinning, and other obvious signs of damage No
changes were noted by visual inspection.
130
-------
Results
Samples analyzed by NET Pacific for EPA indicated low leachability characteristics for metals in the
final products as shown in Table 1. The leachable metal content in both the castable and the granular
samples was within the TCLP limits for all compounds for which they were analyzed. Barium, chromium,
lead, and zinc were the only compounds detected in either of the EPA samples. Comparison of these
data to those obtained by Sound Analytical Services, Inc. (see Table 1) produced similar results (for the
granular product only) even though the Developer's laboratory could not achieve the same detection limits
as the EPA laboratory.
TCLP analyses were performed on Glasses I and III by the Developer's laboratory. The results of
these analyses indicated that the products were within the TCLP leaching maximums.
Stack gas sampling data were previously gathered during earlier tests at the Oregon Steel Mill (OSM).
Although these data suggest acceptable air emissions, the data are of questionable quality because they
do not satisfy EPA stack testing protocols and standards.
TAE3LE 1. TCLP RESULTS AND COMPARISON TO
REGULATORY LIMITS FOR SAMPLES FROM EMJ
EPAHW
No.1
D004
D005
D006
D007
D008
D009
D010
D011
Contaminant
Arsenic
Barium
Cadmium
Chromium
Lead
Mercury
Selenium
Silver
Zinc
EPA
Castable
Sample2
(mg/L)
<0.0025
0.043
<0.0035
0.050
0.067
<0.000086
<0.001
<0.0092
0.95
EPA
Granular
Sample2
(mg/L)
<0.0025
0.025
<0.0035
0.13
0.120
<0.000086
<0.001
<0.0092
0.60
Developer
Granular
Sample
(mg/L)
<0.2
<0.1
<0.1
0.1
<0.1
<0.002
<0.3
<0.1
0.6
Regulatory
Level3
(mg/L)
5.0
100.0
1.0
5.0
5.0
0.2
1.0
5.0
NR
Cost estimates were performed for the OSM plant by the developer. A full-scale system producing 60
tons of glass/day, and operating 350 days/yr would require an initial cost of $10,500,000 for design,
construction and start up.
For a ten yr. period, the Process could produce a gross profit of $63,195, 000 while avoiding
$43,040,000 in disposal costs, for a total savings of $106 million, not including reduced liability benefits,
and avoidance of administrative costs for permits and managing of hazardous waste under the old
system.
The actual savings realized will depend on the types and amounts of the products sold. Estimates of
market conditions during the OSM test period indicated that the lowest value products were cement
137
-------
Pollution Prevention Imipact--
The NMP/MEA paint removal system eliminates the use of methylene chloride for this depainting
operation. As a result, this system prevents the disposal of 48,843 Ibs of used methylene chloride
annually. Since methylene chloride is no longer a part of the system, Toole can now regenerate the
activated carbon used in its building ventilation system. Prior to this innovation, the spent carbon material
was considered a hazardous waste. This change resulted in the saving of 240,000 Ibs of activated carbon
from disposal as hazardous. The fixed-film biological reactor removed 89% to 98% of the NMP.
TABLE 1. ECONOMIC EVALUATION DATA
Activities Annual Cost Savings
Process Operation $140,000
Activated Carbon Disposal $174,000
Payback Period Less than 1 year
NMP/MEA is a viable replacement for methylene chloride in the removal of heat resistant and chemical
agent resistant coatings. However, health effects considerations require that the choice between NMP
and methylene chloride be a more deliberate one. EPA's Office of Pollution Prevention and Toxics has
issued a report entitled Lifecycle Analysis and Pollution Prevention Assessment for NMP in Paint
Stripping. This report states that NMP may cause reproductive and developmental effects in humans.
The report further states that the primary risk of exposure to NMP is through dermal contact and that this
exposure can be effectively controlled through use of impervious protective gloves. U.S. EPA considers
methylene chloride to be a hazardous air pollutant because of its low exposure limit and high volatility.
Methylene chloride is also a suspected human carcinogen by the National Institute of Occupational Safety
and Health (NIOSH) and by the American Conference of Governmental Industrial Hygienist (ACGIH).
The full report, entitled "A Replacement Non-Methylene Chloride Paint Remover Study at Tooele Army
Depot, Tooele, Utah" by Bruce Sass, et al., will be available as an EPA series 600 project report.
134
-------
#31 LOW-VOLATILITY SOLVENT AND FILTRATION SYSTEM FOR MECHANICAL PARTS WASHING
Participants
The host for the mechanical parts washing project was the Titus-Will Ford dealership garage in
Tacoma WA. Mechanics at Titus-Will operated the equipment. Battelle, Columbus Laboratories, on
contract to EPA, provided test personnel, designed the test and drafted the final report.
Technology/Testing
The Breakthrough/Edge-Tek Filter System technology, which is produced by Inland Technology, was
evaluated in this study. It has two components: Breakthrough, a low-volatility solvent, composed mostly
of C11-C13 hydrocarbons with cleaning potential similar to that of mineral spirits, and a two-stage
filtration system built into the parts washing station (Figure 1).
Figure 1. Typical Breakthrough/Edge-Tek Filter System
The first stage is a coarse (80-^ m nominal pore size) reusable stainless steel filter, and the second is
a fine (0.1-^m nominal pore size) single-use filter. The filters are designed to remove suspended matter
in the solvent, thus lengthening the solvent's useful lifetime. There is some indication that the fine filter,
made of cellulosic material is also capable of removing grease and oil from the system. Heavier
hydrocarbons may separate as colloids and then be filtered out. Periodically- typically monthly- the
coarse filter can be removed and washed with the solvent and the paniculate can be collected for
disposal as solid waste; the fine filter can be discarded and a new filter installed. Approximately yearly
the material that accumulates in the bottom of the parts washing station must be removed and disposed
of properly.
139
-------
additives at $2 to $6/ton. The highest value products, such as glass ceramics and architectural tiles sold
from$175to$650/ton.
Conclusions were:
• The glass types tested, resulted in relatively non-leachable products
• For the metals of interest for K061 waste (cadmium, chromium, and lead), leachability values were
lower than those allowed under RCRA regulations for TCLP.
• The Process can utilizes other (non-listed) foundry wastes to replace constituents that would be
purchased as virgin additives for glass-making. Ideally this could result in both a conservation of
resources and recycling of both hazardous and non-hazardous wastes at the foundry.
• This project did not focus on investigating compliance issues in terms of air emissions and waste water
generated during batch charging, melting, quenching and drying of the three glass products. It is
believed that significant variation in emission species and concentrations are possible, due to the
specific application and associated operational procedures.
• Compliance issues should be evaluated on a case-by-case basis at least until full-scale data are
accumulated to better identify the variability associated with applying this technology.
The full report, titled "Recycling of Electric Arc Furnace Dust: Jorgensen Steel Facility," by Trevor W.
Jackson and Jamie Sue Chapman is available as report number EPA/600/R-95/007.
138
-------
Results
The initial batch of Breakthrough solvent required disposal 16 months after Titus-Will switched to the
Breakthrough/Edge-Tek Filter System. Due to faulty communications and the improper use of the system,
as described below, the contents — solvent and sludge — of a 30-gal parts washing station weighing
approximately 120 Ib., had to be disposed of prematurely. On an annual basis at this rate of use, this is
the equivalent to 90 Ib of waste per parts washing station.
A small number of used fine filter elements, each weighing approximately 1 Ib, also required disposal.
Therefore, the total amount of waste with Breakthrough as used at Titus-Will during the study was
approximately 100 Ib per parts washing station per year. Replacing the mineral spirits-based system with
Breakthrough thus reduced the volume of waste requiring disposal by 1500 Ib per year per station. The
total waste reduction cannot be estimated because no information is available for losses with the previous
system due to drag out and evaporation. However, the data on Breakthrough is favorable; through the
end of the study less than two gallons (approximately 16 Ibs.) of solvent had been added to the station due
to account for drag out and evaporation.
At the time of this test, a few of the mechanics wanted to replace the used Breakthrough solvent with
fresh because they were no longer convinced that the solvent was working as efficiently as it should.
Consequently, it was discovered that the mechanics were not adhering to vendor-recommended
procedures regarding the fine filter replacement. It was found later in the study that most of the
mechanics had removed the fine filters from the washing units, either replacing them with a coarser
substitute (e.g., rags or paper towels) or operating the unit with only the coarse filter installed. The latter
was the case for the unit used for sampling. While this prevented the determination of the maximum
Breakthrough solvent life, the 16 month period did provide a conservative data point for comparing waste
amounts and economics with the previous system using mineral spirits.
The results of flashpoint analyses on the sludge and on both fresh and used Breakthrough solvent are
presented in Table 1.
TABLE 1. SUMMARY OF RESULTS OF FLASHPOINT ANALYSES
Sample Location Flashpoint (°F)
Solvent in Reservoir 104
Sludge 122
Fresh Solvent 158
Regulatory Level 140
Tables 2 and 3 present the results of the total and the Toxicity Characteristic Leaching Procedure
(TCLP) extractable metals analyses. As seen in Table 2, the total metals concentrations were elevated
only in the sludge taken from the tank. This would be expected, because the paniculate matter should be
accumulating in the sludge. The TCLP-extractable metals analyses showed cadmium, chromium, and
lead concentrations below RCRA-regulated levels. The used solvent, however, does contain a
concentration of TCLP-extractable cadmium equal to the regulatory level and TCLP-extractable lead in
excess of the regulatory level; thus it would require treatment and disposal as a hazardous waste.
141
-------
Site Description--
Product Quality Evaluation-
140
-------
The equipment cost was supplied by the manufacturer and is shown for a 30-gal parts washing station
complete with the Edge-Tek Filter System, not including the solvent. Startup costs consist primarily of the
initial solvent supply, along with a short period of operator training. Working capital was estimated as a 1-
month supply of consumables, i.e., one fine filter. The contingency was estimated as a fixed percentage
of the equipment, materials, installation, and engineering costs combined. Note, however, that most of
these costs are zero, because the shop requires no modification for the new technology. The only utility
connection required is an electrical outlet supplying 110V and 15 amps. The Standard Depreciation,
Income Tax, Inflation, and Cost of Capital Rates were estimated.
The primary economic benefit of using the Breakthrough/Edge-Tek Filter System is the reduction in
waste disposal costs. There is a net reduction of 0.75 ton per year per station in wastes disposed of or
sent off site for recycling. This is a savings of $60 per month to dispose of this material.
A payback period of fewer than 4 years was found. At 4 years the return on investment (ROI) was 9.67%,
and after 10 years the ROI was greater than 29%.
Economically, the Breakthrough/Edge-Tek Filter System appears to be a viable waste reduction
technology. The system has a moderate payback period and moderate ROI over its life, given the inputs
used.
Discussion--
The evaluation of the Breakthrough/Edge-Tek Filter System marginally demonstrated that the combination
has the potential to reduce the amount of waste generated during parts cleaning. For this application the
process provided parts having an acceptable degree of cleanliness for the operations at the host site.
However, due to problems with system maintenance and sampling, the results did not support the claims
made by the vendor.
The full report, entitled "Low Volatility Solvent and Filtration System for Mechanical Part Washing" by
David P. Evers, et al. will be available as EPA 600 report series in the near future.
143
-------
TABLE 2. SUMMARY OF RESULTS OF TOTAL METALS ANALYSES
Sample Location
Solvent in Tank
Water in Tank
Sludge
Fresh Solvent
Cadmium
mg/L
1.0
0.75
23.7
<0.12
Chromium
mg/L
1.2
0.65
65.9
<0.27
Lead
mg/L
36
8.4
1,065
<1.5
Nickel
mg/L
1.5
<0.75
41.7
<0.58
Aluminum
mg/L
12.6
12.5
4050
<1.7
Iron
mg/L
166
281
25,050
1 8
TABLE 3. SUMMARY OF RESULTS OF TCLP METALS ANALYSES
Sample
Location
Solvent in Tank*
Water in Tank
Sludge
Fresh Solvent
Reg. Level
Cadmium
mg/L
1.0
0.318
0.183
<0.12
1.0
Chromium
mg/L
1.1
<0.007
<0.011
<0.27
5.0
Lead
mg/L
36
0.16
0.90
<1.5
5.0
Nickel
mg/L
1.7
0.157
0.14
<0.58
NA**
Aluminum
mg/L
<1.7
0.13
0.41
<1,7
NA**
Iron
mg/L
168
41.4
16.4
0.92
NA**
** NA = Not Applicable
Total Organic Carbon (TOC) analyses were conducted on the sample of water and sludge taken from
the tank. The TOC of the water sample was 26,000 mg/L, whereas the sludge TOC was 25,000 ppm.
Both values indicate that a considerable amount of organic matter accumulated in the respective samples.
The TOC of the water sample is high enough to require pretreatment before disposal to a POTW.
Economic Assessment--
Process economics of interest included capital costs, operating and maintenance costs, and waste
disposal costs. Capital costs included the cost of the equipment, the cost of installation, and any other
one-time costs associated with making the equipment operational. Operating and maintenance costs
included all those associated with day-to-day operation of equipment, such as detergent costs, water cost,
energy costs, maintenance materials (including spare parts) and labor to operate and maintain the
equipment. Waste disposal costs included on-site and off-site treatment costs or income.
An economic analysis was performed comparing the costs to install, operate, and maintain the
Breakthrough/Edge-Tek Filter System wit the previous system. Economic indicators such as payback
period and return on investment were calculated and used to estimate the economic benefits. The
evaluation included a combination of the available site-specific costs that were not business-sensitive to
the host company. Other costs were based on engineering judgments made by Battelle, based on typical
costs associated with the activity. A discussion of these follows.
142
-------
e-
8
|
I
en
145
-------
#32 POWER WASHER WITH WASTEWATER RECYCLING
Particiants
Si « ^e^^ Base
equipment during the test. Personnel f J , the W *™*'* their
« ,
^
144
-------
increasing detergent concentration or, lastly, replacing washwater and detergent. For the duration of the
use of the Power Washer, the wash water had not as yet been replaced.
Results
With the previous, manual spray-cleaning operation, approximately 170 gallons per month of water oil
and grease, and paniculate matter was disposed of as wastewater. Currently, the amount of material '
requiring disposal monthly is 25 gallons of paniculate waste. The oil and grease removed from the water
with the Clean Machine, on the order of 5 gallons per month, is recycled off-site. The volume of makeup
water has been reduced to approximately 75 gallons per month, the majority of which is lost as condensed
steam during Power Washer loading and unloading.
Since installing the Power Washer/Clean Machine equipment, Seattle Metro has also eliminated seven
solvent cleaning stations. Solvent waste production has been reduced by approximately 210 gallons per
month. The total reduction in waste volume (including water vapor air releases) approaches 275 aallons
per month, or about 2000 pounds. y
Waste Reduction Assessment--
There has been an overall reduction in hazardous waste volume with use of the Power Washer in
conjunction with the Clean Machine. This reduction is approximately 80%, when all wastestreams --
aqueous solvent, and sludge - are considered. In addition, the wastestreams generated are seqreqated
so that the waste oil and grease can be collected easily and recycled off site, the paniculate matter
disposed of properly, and the wastewater treated efficiently.
Economics of the system were based on data from Seattle Metro on the purchase price for the Power
Washer/Clean Machine equipment and estimates for installation and start-up costs as percentages of
capital cost. a
Use of the Power Washer has decreased labor hours for cleaning parts by approximately three-
quarters of a man-year. This was reflected as a productivity credit in the analyses because the operator is
now free to perform other duties. Detergent use has been reduced by over 80%.
Savings in purchases of solvent, water and natural gas are also realized. However there is a net
increase in electricity consumption, largely due to the fact that the unit is electrically heated.
The Mart Power Washer and Clean Machine combination can reduce the amount of waste generated
when compared to manual, high-pressure spraying or solvent cleaning. The system is capable of
providing parts with an acceptable degree of cleanliness for the application tested.
For this and similar applications, the process economics are reasonably attractive The return on
investment over a 10 year period was calculated at greater than 28%. Using natural gas for heating water
(vs. electricity) could further improve the economics.
If no oil film can be tolerated on the cleaned parts, such as painting operations, this system may not be
an acceptable cleaning method without additional cleaning.
The full report, entitled "Power Washer With Wastewater Recycling Unit," by David P Evers et al will
be available as an EPA report.
147
-------
o
I
I
I
I
s
3,
146
-------
The effectiveness of bicarbonate of soda blasting depends on optimizing a number of operating
parameters including nozzle pressure, standoff distance, angle of impingement, media flow rates water
pressure, and traverse speed.
The present study evaluated the bicarbonate of soda blasting technology, ARMEX®/ACCUSTRIP™
(see Figure 1), marketed by the CDS Group (Houston, Texas). The evaluation was conducted at
NASA/JCS's Ellington Field, which maintains and repairs a fleet of 37 aircraft.
Product Quality Evaluation--
Although the quality of the work for this bicarbonate blasting system was based on paint removal
without damage to the wheel surface that either modified metal performance or masked any cracks during
inspection, an additional concern was the anodized layer below the paint. This thin (around one ten-
thousand of an inch), electrochemical oxide layer is used to improve the corrosion resistance of the metal.
Because of the relative vulnerability of this layer between the paint and metal the condition of this layer
was used to determine the performance of the depainting process.
This study did not evaluate the effects of blasting on metal substrate damage and crack closure,
because of previous work published on that topic.
Results
About 30 gallons of wastewater were generated and collected in a vat during each of the two blasting
sessions. The mean values for the measured pollutants are presented in Table 1. The Cr concentration
did not meet the local discharge limits, so the wastewater could not be disposed of to the Publicly Owned
Treatment Works (POTW). Approximately 8 gallons of solid waste settled to the bottom of the vat after
bicarbonate blasting of 4 wheels. Metal concentrations measured are presented in Table 2. Only a very
small fraction of the metals was leachable under the Toxicity Characteristic Leaching Procedure (TCLP)
conditions (see Table 2). TCLP requires the waste to meet limits of 1.0 mg/L Cd, 5.0 mg/L Cr, and 5.0
mg/L Pb. No regulations had been set for Cu, Mn, Ni, and Zn. The wastewater in the rotoclone separator
contained less than detection limit of TSS and a very small amount of heavy metals, ranging from 0.005
mg/L of Cd to 0.39 mg/L of Zn. For the particular case tested, the wastewater could be sewered without
treatment.
Other considerations were hazards that the stripping technology might pose to workers. These
included toxic airborne particulate and unsafe noise exposures. Air quality in the vicinity of the blasting
operator was measured in terms of airborne metal concentrations. Noise levels were measured on a
sound-level meter and a dosimeter.
Air emissions were measured in the breathing zone of the operator and analyzed for Cd, Cr, Cu, Pb,
and Zn. The cloud of mist created around the blasting activity was maintained within the work area and
removed by a ventilation system consisting of an exhaust hood and a rotoclone dust separator.
The results of the airborne metal exposure study indicated that 8 hrs. time-weighted average (TWA)
exposure to the airborne metals were below specified OSHA and American Conference of Governmental
Industrial Hygienists (ACGIH) limits. Sound levels measured periodically in the operator's hearing zone
during the two blasting sessions, on the "A" - weighted scale, ranged from 76.8 decibels (dBA) to 120.0
DBA. Dosimetry samples integrated cumulative noise. If the actual work period were increased to a full 8
hours, the projected 8-hour TWAs would be 121.3 dba for the first test and 115.9 DBA for the second test.
A peak level of 146 DBA, the maximum level the dosimeter is capable of measuring, was recorded during
both periods sampled.
149
-------
#33 BICARBONATE OF SODA BLASTING TECHNOLOGY FOR AIRCRAFT WHEEL DEPAINTING
Participants
The Paint Stripping Shop at Ellington Field, National Aeronautics and Space
Administration/Lyndon B. Johnson Space (NASA/JCS) Center in Houston, Texas, hosted the test and
assisted in implementing the evaluations. Battelle, Columbus Laboratories, on contract to EPA, helped
design the test program, supplied test personnel and equipment, and wrote the draft report.
Technology/Testing
Bicarbonate of soda blasting is a relatively new process that is commercially available.
Compressed air delivers sodium bicarbonate media from a pressure pot to a nozzle where the media
mix with a stream of water. The media/water mixture impacts the coated surface and removes old
coatings from the substrate. The water dissipates the heat generated by the abrasive process, aids the
paint removal by hydraulic action, and reduces the amount of dust in the air. As another convenience,
the workers, do not need to prewash or mask the surface. The dust, unlike that of plastic media, is not
an explosive hazard, nor is sodium bicarbonate toxic in this form. The airborne participates generated
from the stripping operation, however, can contain toxic elements from the paint being removed.
Water line
Blast hose Blast nozzle
Figure 1. ACCUSTRIP SYSTEM™ with Wet Blast Head
148
-------
Economic Evaluation--
Cost comparisons were made for bicarbonate blasting vs. chemical stripping. Blasting times to strip
each wheel were measured during the test. NASA/JSC historical data were used to determine chemical
stripping times. The capital investment, operating costs, and payback period were calculated according to
the worksheets provided in the U.S. EPA Waste Minimization Opportunity Assessment Manual. The
results of the economic analysis indicated that a return on investment (ROI) greater than 15% (which is
the cost of capital) could be obtained in 4 years, or that the payback for NASA/JSC would be 4 years.
Conclusions/Recommendations-
The bicarbonate of soda blasting evaluation demonstrated that the blasting technology can effectively
strip paint from aircraft wheels. The blasting technology substantially reduced the number of man-hours
required for paint stripping in comparison to chemical stripping. The time saved was more than 95%.
The liquid waste accumulated in the vat exceeded the discharge limit for Cr and could not be sewered
into the POTW. The quantity to be shipped away as hazardous waste was about 7.5 gal/T-38 aircraft
wheel. The solid waste in the vat contained paint chips and debris, most of which was insoluble under the
TCLP conditions. The wastewater in the rotoclone separator could be sewered without treatment.
Although convenient for this application and within existing local limits, the source reduction of this
waste as well as reuse/recycling should be investigated in greater depth.
Although the exhaust ventilation system kept the heavy metal concentrations in the workspace below
OSHA and NASA limits, the opportunities for source reduction to minimize rotoclone wastewater should
be explored as well as possibilities for recycling and reuse of this water.
The operator of the blasting equipment was required to wear a full-face air-purifying respirator and
protective clothing. Although the present test results did not make this an OSHA requirement, previous
testing of this system produced chrome particulate concentrations that did. The added precautions are
recommended until a better understanding of the system is developed. Improved lighting for better
visibility at the work surface also is recommended.
The noise measurements indicated that, under the conditions encountered during this study, hazardous
noise exposures can result. Therefore, engineering control of noise exposure should be investigated.
The full report, entitled "Bicarbonate of Soda Blasting Technology for Aircraft Wheel Depainting" by
Abraham S.C. Chen et al., is available as EPA/600/R-94/127.
151
-------
TABLE 1. OIL AND GREASE, TSS, Ph, AND METAL
CONTAMINANTS IN WASTEWATER COLLECTED FROM THE VAT
Parameter
Oil and grease (mg/L)
TSS (mg/L)
PH
Cdb (mg/L)
Cr" (mg/L)
Cub (mg/L)
Pbb (mg/L)
Mnb (mg/L)
Nib (mg/L)
Znb (mg/L)
Mean
Concentration
49.1
253
8.367
0.033
8.0890
1.240
1.430
0.022
0.006
5.990
Local3
Discharge Limit
200
365
6-10
0.2
5.0
2.0
1.5
3.0
3.0
6.0
" Maximum allowable limits for grab samples, Industrial Waste Permit No. 1030, City of Houston,
Texas, March 10, 1989.
b Total metal.
TABLE 2. TOTAL AND LEACHABLE METALS IN SOLID WASTE
THAT SETTLED TO THE VAT BOTTOM
Metal
Cd
Cr
Cu
Pb
Mn
Ni
Zn
Field
Blank
0.50
0.69
1.30
1.70
0.19
0.50
1.90
Total Metal
Mean
Concentration
(mg/kg)
2.73
146.07
32.97
70.87
2.77
0.72
281.33
Field
Blank
0.0050
0.0127
0.0036
0.0190
0.0056
0.0050
0.0560
Leachable
Metal Mean
Concentration
(mg/L)
0.0303
2.2006
0.3927
0.5397
0.0023
0.0017
4.2840
Data have been corrected with field blank.
150
-------
Table 1 shows the characterization results. In appearance and color, the spent samples varied vastly
from the clear recycled and virgin samples. All the measured parameters showed a significant
improvement from spent to recycled samples but were not quite up to virgin grade. The water content
increase in the recycled samples was traced to a slight leakage from the water-cooled condenser. Site
personnel were able to correct the problem after the completion of testing.
Results
Of 55 gallons processed, 39 were recycled, 16 were residue, and 4 were lost to air emissions. This
was characterized as a typical run.
Table 2 shows the waste reduction achieved by distillation. Through recycling, large volumes of spent
solvent waste were reduced to small volumes of distillation residue, which was disposed of as RCRA
hazardous waste. MEK is a hazardous chemical listed on the Toxic Releases Inventory (TRI). The
solvent is on EPA's list of 17 chemicals targeted for 33% reduction by 1992 and 50% reduction by 1995.
TABLE 1 CHARACTERIZATION OF SOLVENT SAMPLES
Sample Appearance
Atmospheric Unit
(MEK)
Specific
Color3 Gravity
Nonvolatile
Matter
mg/100 ml
Conductivity
p mhos/cm
Water
Content
% by wt
Acid
Acceptance6
Purity
%c
Spent
Recycled
Recycled Dupd
Virgin
Dark Grey
w/secliment
Clear
Clear
Clear
•'
5
5
5
0.845
0.827
0.821
0.800
6,951
2.6
2.0
2.2
7.05
3.30
3.40
1.15
1.89
5.42
5.56
0.09
NA'
NA
NA
78.41
85.02
85.54
a On a scale of 5 to 500, with 500 being the darkest color. ASTM D1209 and D2108.
b Measured as equivalent NaOH wt% ASTM D2942.
0 Gas chromatography analysis based on ASTM D2804.
d Duplicate analysis of the same sample.
e Not comparable with standards. Sample was too dirty.
' NA = not analyzed
The economic evaluation compares the costs of recycling to conventional practice. Table 3 shows the
major operating costs associated with disposal and the atmospheric batch unit. For the unit, recycling
saved £ $10,000/yr. The purchase price of the atmospheric batch unit is $12,995. A detailed calculation
based on worksheets provided in the Facility Pollution Prevention Guide (EPA, 1992) indicated a payback
period of less than 2 years.
153
-------
#34 ONSITE SOLVENT RECOVERY WITH AN ATMOSPHERIC STILL
Participants
The host for the atmospheric still evaluation was Navistar International Transportation
Corporation, Plastics Division, Columbus, Ohio. Company personnel also operated the distillation
equipment. Battelle, Columbus Laboratories, on contract to EPA provided test personnel, designed the
test and drafted the final report.
Technology/Testing
Atmospheric distillation is the simplest technology available to recover liquid spent solvents.
Units that can distill as little as 5 gallons or as much as 55 gallons/batch are available. Some units can
be modified to operate under vacuum for higher-boiling solvents (>13Jf C). Contaminant components
with lower boiling points than the solvent or that form an azeotrope with the solvent cannot be separated
(without fractionation) and may end up in the distillate. The unit used in this study (Figure 1) was Model
LS-55D, manufactured by Finish Thompson, Inc. It was used by Navistar to recycle spent methyl ethyl
ketone (MEK) to clean spray painting lines between colors.
MJ/SV/V-
'Vapors y.-.;
• . ..'.«••»
Contaminated
Solvent
Condenser
J
Stillbag
Heated
Walls
Electric
Heat Source
Reclaimed
Solvent
Figure 1. Atmospheric distillation unit
The product quality objective for the distillation unit was to show that the recycled solvent was of
sufficient quality to reuse. One 55 gallon drum of spent solvent was processed in -12 hrs. The
distillation residue, often a relatively small fraction of the spent solvent, is disposed of as hazardous
waste.
The amount of residue left behind is a function of the application and not the distillation units. Samples
of the spent and recycled solvents were analyzed by standard ASTM methods to determine the
improvement in quality. Virgin solvent samples also were collected at each site and subjected to the
same tests for comparison.
152
-------
#35 ONSITE SOLVENT RECOVERY WITH VACUUM HEAT-PUMP DISTILLATION
Participants
The vacuum heat-pump still was tested at Cooper Industries, Belden Division, in Richmond,
Virginia. The Plant personnel operated the distillation equipment. Battelle, Columbus Laboratories, on
contract to EPA provided test personnel, designed the test and drafted the final report.
Technology and Testing
The vacuum heat pump unit was tested on spent methylene chloride (MC) at a site that
manufactures wires and cables. The MC is used for cold (immersion) cleaning of wires and cables to
remove markings (ink).
The vacuum unit tested, Model 040 is manufactured by Mentec AG in Switzerland and supplied
in the United States by Vaco-Solv Chicago, Inc. Its configuration is similar to a conventional vacuum
distillation system except that the pump, in addition to drawing a vacuum, functions as a heat pump
(Figure 1). No external heating or cooling is applied. The heat pump generates a vacuum for distillation
and compresses vapors for condensation. Because of this feature, the unit used 50% to 75% less
energy than conventional systems. Model 040 is suitable for solvents with boiling points up to 80~C.
Condensate Trap
Vapor Filter
•—Air
Overflow
Protection
*- Feed
Residue
Figure 1. Vacuum heat-pump distillation unit (Source: Vaco-Solv Chicago, Inc.)
155
-------
TARI F 9 WASTF RFHI IHTinM WITH ATMO.QPMFRir. QTII I
-Disposal Option-
Wastestream
Annual Volume
-Recycling Option-
Wastestream Annual Volume
Atmospheric
Spent MEK
Drums
Unit Test Site:
880 gallons
17 drums
Distillation residue 262 gallons
Still bags
Cooling Water 18
Drums
1 7 bags
,360 gallons
Item
Annual Usage
Report
The full report entitled "Onsite Solvent Recovery"
EPA/600/SR-94/026.
Unit Cost ($)
Annual
Disposal Option
Virgin solvent
Disposal
-labor
-drums
-disposal fee
Atmospheric Unit
Virgin Solvent
Operating labor
Routine Maintenance
-spare parts
-labor
Energy
Cooling water
Disposal
-labor
-drums
-residue disposal
-still bags
880 gal
8hr
17
900 gal
245 gallons
17 hours
r
12hrs
1,265kWh
18,360 gal
3
5
262 gal
17
10.50/gal
8/hr
40/drum
400/55 gal
Total
10.5/gallons
8/hours
86/ea
8/hr
0.4/Kwh
1/1 000 gallon
8/hr
40/drum
675/55 gallons
84/1 2 bags
Total
9,240
64
680
6,545
16,529
2,573
136
86
96
51
18
24
200
3,215
119
6,518
by Arun R. Gavaskar is Available as
154
-------
Table 1 shows the characterization results for samples from the vacuum unit. In appearance and
color, the spent samples varied greatly from the clear recycled and virgin samples All the measured
parameters showed a significant improvement from spent to recycled solvent but they did not auite
measure up to virgin grade.
Some performance characteristics of MC were also evaluated. The Ph of the water extract of the
recycled solvent was fairly close to the "virgin" value of 7. The spent sample Ph of 5 indicates the
presence of potentially corrosive components. The corrosion test on steel and aluminum (ASTM D2251)
yielded noticeable corrosion only in the case of the steel strip placed in the spent solvent sample No such
corrosion was evident due to the recycled solvent, indicating that recycling improved the quality. '
«i«ct ^ the ^aSt,6 reduction acnjeved. Through recycling, large volumes of spent solvent
waste were reduced to small volumes of distillation residue, which is disposed of as a RCRA hazardous
WclSIG.
TABLE 2. WASTE REDUCTION BY ATMOSPHERIC AND VACUUM UNITS
' ""^
• Disposal Option- -Recycling Option-
Wastestream Annual Volume Wastestream Annual Volume
Vacuum Unit Test Site:
Spent MC 3,000 gallons
Drums 55 drums
Distillation residue
Air emissions
Drums
Used oil
136 gallons
218 gallons
3 drums
1 gallon
unit .nctc «OQ;-J£CUTim unit
-------
was of
Results
at a faster rate than recommended by the manufacturer Because the un^sh,T "^ bein9 °perated
ors^^^^^
into the.., area ,he vapor ^asld thro^pTe
^^ wouid
TABLE 1. CHARACTERIZATION OF SOLVENT SAMPI F.g
Vacuum Unit (MC)
Spent Dirty grey-brown
Recycled Clear
Recycled Dupd Clear
Virfin Clear, tinge of yellow
— ——"•••« ihe darkest color
D Measured as equivalent NaOH wt% ASTM D2942
c Gas chromatography analysis based on ASTM D2804
d Duplicate analysis of the same sample.
e Not comparable with standards. Sample was too dirty
f NA-not analyzed.
156
-------
The testing consisted of loading machined steel parts into the unit and running the cleaning cycle,
which is largely automatic, as previously described. Cycle times, load weights and PCE concentrations
were measured around the outside of the LEVD, inside upon first opening of the lid and shortly thereafter,
and ambient concentrations in the work area. A pair of flame ionization detectors were used for these
measurements.
TABLE 1. LEVD CLEANING CYCLE
Stage
Solvent heat-up (once a day)
Solvent spray (optional)
Vapor fill
Degreasing
Condensation
Air recirculation
Carbon heat-up
Desorption
Adsorption
Vendor-Recommended
Time Settings
Variable3
10-180 sec
Variable"
20-180 sec
120 sec
1 20 sec
Variable0
60 sec
60-240 secd
Times Set for
This Testing
Variable3
not used
Variable13
60 sec
120 sec
120 sec
Variable0
60 sec
240 sec
a Requires p1 hr on days following overnight shutdown when sump solvent temperature drops to 70pC. After weekend
shutdowns, when sump solvent temperature drops to 20pC, it may take 1.5 hr for solvent to reach vapor temperature.
Timer on unit allows automatic heat-up.
b Depending on the workload mass and type of metal. Varied from 8.5 min for 165 Ib to 36.5 min for 915 Ib of steel parts.
0 Carbon heat-up took approximately 22.5 min during testing.
d At 60 sec, if monitor shows that chamber concentration is above 1 g/m3, then the adsorption stage proceeds to the full 240-
sec stage. This sequence repeats if necessary
Testing was conducted on the LEVD using perchloroethylene (PCE) solvent. Test runs were
conducted on machined steel parts with and without cutting oil on the parts. The pollution prevention
aspect of the LEVD was the main focus of this technology. The completely enclosed design of the
working chamber allows the potential for air emissions only when the cleaning cycle is complete and the
lid is opened. Any solvent vapor not evacuated from the chamber during condensation or adsorption
releases to the atmosphere.
Table 2 shows the total cycle times and emissions recorded from the LEVD by a flame ionization
detector (FID) probe inserted (for this test) into the working chamber below the designated vapor level.
FID measurements began during the adsorption stage and continued until after the lid was opened. A
second FID probe (ambient), positioned outside the unit near the lid seal, took continuous measurements
all around the unit during operation, with special emphasis around the lid to detect any leaks. Ambient
levels (3 to 4 ppm) in the indoor facility on the test days were observed.
159
-------
#36 ONSITE SOLVENT RECOVERY WITH LOW EMISSION VAPOR DECREASING
Participants
The Low Emission Vapor Degreaser (LEVD) was tested at Davidsburg, Michigan, which is a
manufacturing site for Durr Automation Inc. where the LEVD's are made. Durr personnel operated the
equipment. Battelle, Columbus Laboratories, on contract to EPA provided test personnel, designed the
test and drafted the final report.
Technology/Testing
The LEVD is used in Europe, where vapor degreasers are regulated as a point source. Previous
studies (Battelle, 1992) on conventional open-top vapor degreasers have shown that a large part of the
solvent (more than 90% in some cases) is lost through air emissions, which are considerable even though
vapor degreasers are required to have primary cooling coils (tapwater cooled) and a certain freeboard
height Air emissions are mainly workload-related, caused either by dragout of solvent on the workload
itself (and subsequent vaporization) or by disturbance in the air-vapor interface during entry and exit of the
workload. Other sources are convection and diffusion during startup, operation, idling, shutdown, and, to
a small extent, equipment leaks.
Air emissions are a concern for metal finishers because many solvents used in vapor degreasing
have been targeted by EPA in the 33/50 Program, in Clean Air Act Amendments and Environmental and
Occupational Safety and Health Administration (OSHA) regulations, which have become more stringent.
Pollution control devices available for conventional vapor degreasers include increased freeboard height,
refrigerated coils, and covers to eliminate drafts and reduce diffusion.
In contrast, LEVD units are completely enclosed, airtight units. This evaluation used Model 83S
(Sizel) manufactured in the United States by Durr Automation, Inc. Figure 3 shows its operation. Loads
can range from 330 to 1100 Ib (of steel parts) in this model. When the lid is shut and the unit is switched
on, compressed air hermetically seals the lid shut for the duration of the cycle.
Table 1 shows typical cleaning cycle stages. During "vapor fill," solvent vapors enter the chamber
from the outer jacket and degreasing begins. During "condensation," solvent vapors are condensed out
by a refrigerated cooling coil at the bottom of the chamber. During "air recirculation," the air-solvent
mixture is recirculated through a chiller to condense out more solvent. During "carbon heat-up," solvent
adsorbed in the previous cycle is released (desorbed) to the circulating air and condenses out in the
chiller During "adsorption," the chamber air is recirculated in the reverse direction - first through the
chiller and then through the carbon. Most residual solvent vapor in the cold air is adsorbed on the carbon.
A photoionization detector (PID) probe verifies that the chamber air has less than 1 g/m of solvent and
signals the air compressor to release the seal on the lid to end the cycle. If the chamber air has more
than 1 g/m3 of solvent, the cycle loops back to the desorption stage. The entire cycle is programmed and
requires no operator attention except to load and unload the workload. The LEVD also works as a
distillation unit to clean the liquid solvent in the sump. During distillation, the unit is switched on without
any workload in the chamber.
The shape of the parts may affect cycle time. Parts with recesses that can trap solvent should be
arranged in the basket so that the solvent liquid drains out. Other features offered by the vendor
(oscillating or rotating baskets) may need to be used. Otherwise, either the air recirculation stage time
must be increased, or the unit will loop into several adsorption cycles until the chamber concentration falls
below 1 g/m3.
158
-------
Results
Figure 2 shows how a typical LEVD cleaning cycle ends. The same pattern was evident in the
other runs. Time zero corresponds to the start of measurements when the FID probe in the working
chamber was activated. Just before the adsorption cycle ended, the chamber FID read 52 ppm, which
was below the targeted 1 g/m3 (150 ppm of PCE). When the lid was retracted, the chamber
concentration dropped sharply as the residual solvent vapor in the chamber dispersed. The ambient FID
probe showed a corresponding increase (to 6 ppm). Both FID readings soon stabilized to facility
ambient levels (3 to 4 ppm). The solvent concentration at the edge of the chamber opening dropped
from 6 ppm to ambient level in approximately 2.5 minutes (Figure 2), resulting in operator exposure well
below the OSHA limit.
In all the test runs, the solvent concentration was below the targeted 1 g/rrt3 (150 ppm PCE).
The volume of the working chamber is 0.6 m3. Assuming that all the residual solvent vapor (1 g/rrf
maximum) in the chamber is discharged to the ambient area, the typical air emission through the
opened top is 0.6 g (0.00132 Ib)/cycle or less. It takes 1 hour to clean 560 Ib of oiled steel parts.
Therefore, the air emission from this LEVD mode is 0.00132 Ib of solvent/hr.
In comparison, a typical conventional open-top vapor degreaser, cleaning at a similar rate (- 560
Ib of steel parts/hr) would emit approximately 0.147 Ib of solvent/ff /hr (EPA, 1989), or 0.662 Ib of sol-
vent/hr from its 4.5-fl2 opening during continuous operation.
10000
100
soo
600
Time (seconds)
Figure 2. Concentrations at the end of the cleaning cycle for Run 1
161
-------
Electric
Heat
Legend
Desorption Stage
Adsorption Stage
Liquid Solvent
Water
Figure 1. Low-emission vapor degreaser
(Source: Durr Automation, Inc)
TABLE 2. EMISSIONS FROM LEVD
Run
No.a
Mass of Steel
Parts (Ib)
Final Chamber
Concentration^
(ppm)
Total PCE
Emission0
(Ib/cycle)
Total
Cycle Time
(min)
Emission
Rate
(Ib/hr)
1
165
165
900
8
Target
165s
91 tf5
560
52
75
92
43
47
78
150
0.0005
0.0007
0.0008
0.0004
0.0004
0.0007
0.0013
67.5
50.5
4Cf
69
6GP
0.0011
0.0007
0.0005
0.0006
0.0006
0.0013
Runs 4, 7, and 9 were interrupted to allow other measurements.
At the moment when the seal on the lid is released.
Based on 150 ppm « 1 g/m3 of PCE and a chamber volume of 0.6 m3.
Normally the machine is programmed to release the lid when solvent concentration in the chamber falls below 1 g/m3 (150
ppm of PCE). This target was easily met in all the test runs.
Workload parts were dipped in cutting oil before the run.
The test parts were already hot from being used in previous runs when inserted into working chamber. Hence total cycle
times for these runs are lower than normally expected.
Expected cycle time for 560 Ib of steel parts (workload).
160
-------
Other cost/benefit factors should be taken into account when making economic decisions. The
LEVD does not require capital and operating expenditures for auxiliary equipment that may be required for
a standard conventional vapor degreaser (increased freeboard ratio, refrigerated coils, lip exhausts, room
ventilation) in order meet or anticipate increasingly stringent environmental and worker safety regulations.
The LEVD is a self-contained unit that requires no additional facility modifications to achieve significant
emission reductions.
Conclusions-
The LEVD evaluated in this study demonstrated good potential for pollution
prevention/waste reduction.
The LEVD reduced air emissions significantly compared to emissions from a conventional
vapor degreaser.
Longer payback period may not be attractive. However, this number may be significantly
impacted by such less quantifyable factors such as the Clean Air Act amendments, the
avoidance of auxiliary equipment that is needed with conventional degreasers as add-on
hardware, and by potential, downward trends in prices of the units with time.
The full report titled "On Site Solvent Recovery" by Arun R. Gavaskar, et al., is available as report number
EPA/600/R-94/026.
163
-------
Therefore, the LEVD reduces air emissions by more than 99% compared to air emissions from
the typical conventional open-top vapor degreaser (i.e., with a 0.75 freeboard ratio, primary cooling coil
electric hoist, and no lip exhausts) used in this calculation.
The pollution prevention potential of this unit is further enhanced by its ability to perform as a liquid
solvent distillation system for cleaning the sump solvent; this capability was not a part of this evaluation
When pollution prevention is an objective, the LEVD also affords greater production flexibility because it
has none of the idling losses between loads or downtime losses during shutdown of the conventional
degreaser.
Table 3 lists the LEVD's major operating costs and the operating costs for a conventional open-
top vapor degreaser with similar production capacity. With a vendor-quoted purchase price for the LEVD
of $210,000, and savings in annual total operating costs of £$25,000, mainly from reduced labor costs
(due to larger batch size) and less solvent required (due to solvent recovery). The LEVD pays for itself in
|D10 years.
TABLE 3. OPERATING COSTS FOR LOW-EMISSION VAPOR DECREASING
Item Annual
Volume
Conventional Deareaser
Operating Labor 4,000 hr
Electricity 25,500 kWh
Cooling water 480,000 gal
Maintenance:
-Labor 22 hr
-Materials
Net Solvent Loss 2,642 Ib
LEVD
Operating Labor 333 hr
Electricity 93,725 kWh
Maintenance
-Labor 262.5 hr
-Materials
Unit
Cost
$8/hr
$0.04/kWh
$1/1 000 gal
$8/hr
-
$0.71/lb
Total
$8/hr
$0.04/kWh
$8/hr
.
Total
Total
Cost
$32,000
$ 1 ,020
$ 480
$ 176
$ 88
$ 1 .876
$35,640
$ 2,664
$ 3,749
$2,100
$2,100
162
-------
The most significant retrofit was installation of an Enercon corona discharge treater.
Modifications to the Hudson/Sharp 48 inch, central impression, six-color, flexographic printing
press included upgrading drying capacities and using enlarged exhaust and supply fans.
Because of prohibitive costs, the Heinrich (W&H) press was not modified or retrofitted and was
not used in the water-based ink tests. Future plans would include replacing this press with one that could
accommodate the ancillary equipment required for water-based ink use.
Ink metering rolls were replaced to facilitate drying. Pumps were also replaced to accommodate
the new printing inks. Additional ductwork and noise abatement equipment were needed.
Table 1 shows the VOC emissions as a function of ink use, based on historical data.
TABLE 1. VOC EMISSIONS BASED ON INK USE, 1990
Ink used VOCs calculated
Month (Ib/wk) (Ib/wk)
April 3,038 2,111
May 1,681 1,700*
June 2,686 2,289
July 2,109 1,731
August 2,945 2,345
This va'ue is derived from the historical operational data and attributed to high makeup solvent use during the
event.
Table 2 provides information on the total pounds and percent of ink used, calculated VOC
emissions, and VOC emissions as a percentage of ink used for each of the four 1-week-long evaluation
periods.
A review of the quality assurance sheets indicates that the use of water-based inks typically did
not change product quality although some problems arose after customer use, depending upon the
ultimate use of the packaging, what the package contained, and the means by which the packages were
sealed. Heat and stress of the printed package material caused by the package folding and sealing
process at times resulted in a loss of ink adhesion.
Some combinations of water ink and solvent ink were incompatible. Water inks did not provide a
consistent opaque white for lamination to cover metallized films and resulted in "blocking" (or transfer of
print) when printing on Saran-coated materials, especially cellophane. In most cases, however,
depending on the surface printed, no difference was noted with the use of water-based inks.
Normal propyl alcohol added in small amounts (less than 1%) to prevent water ink foaming at the
ink pan and to assist in ink wetting was beneficial. Variations of the pressure sensitive "stickyback"
material used to attach the printing plates to the plate cylinder (solid versus cushioned stickyback) also
enhanced printed solid plate backgrounds without pin-holding. The plate material may also have an
effect. Photo polymer plates work well with water but are more expensive than rubber. Nylon plates are a
possible compromise with a longer life than rubber plates.
165
-------
ERIE COUNTY, NY
Five technologies were identified for Erie County. The Project Officer for Erie County was Paul
R^lncI P6 r6 ?mty D+f a,rtment of Environmental Planning, Division of Environmental Compliance
Because Ene County was the last to join the WRITE Program, three evaluations are completed while two
are still in the final report writing stage. For these, only the test program is described.
OF HAZARDOUS MATERIAL IN WIDE WEB FLEXOGRAPHIC PRINTING
PROCESS
Participants
The Lustrepnnt Company hosted the test and operated the equipment. Test design and
management was by E:rie County Environmental Compliance Services, (ECEECS) Buffalo New York
The test personnel were supplied by Recra Environmental, Incorporated on contract to ECECS.
Technology/Testing
A wide web flexographic printing firm substituted water-based inks for solvent-based inks when
manufacturing flexible packaging, using plastic sheet substrates (e.g., plastic bags for bread) The project
objectives were to evaluate the technical feasibility (particularly as related to process implementation and
performance), the economic effect, and the resulting change in VOC emissions achieved by fhe
^1103' evaluati°n was to quantify the reductjon in both volatile and liguid-phase solid
This is a study of the effectiveness and applicability of ink substitutions to reduce waste in a wide
web (greater than 16 in. wide) flexographic printing process. The Lustreprint Company prints flex" We
packaging whose products are used in the food and snack industry and in medical industrial and
consumer applications,, Printing is completed on a number of different web materials (commonlv
poypropyene (acrylic coated, Saran coated, and uncoated corona pretreated), cellophane (Saran coated)
po yes er (both metalhzed and unmetallized), polyethylene, and nylon (both Saran coated and uncoa°ed»
At the time of this study, Lustrepnnt used one Hudson/Sharp 48 in., central impression, six-color flexo-
press and one Hemrich (W&H) five-color, flexo, stack press.
New York's regulations reguire that a facility reduce overall plant emissions to within the
comphance level of 100 tons/yr. As an option, Lustreprint chose to reduce the use of solvent-based inks
and adhesives. The first step eliminated solvent-based adhesive used in laminating. This was followed
by a phase-in of water-based inks in the printing operation. The company goals are to reduce all volatile
organic air emissions to an extent that would eliminate the need for costly air abatement and permitting
and to eliminate all liquid-phase solid waste, at the facility.
To achieve these goals, ink use was monitored over four, one week long, study periods- 3 weeks
when both water-based and solvent-based inks were used and 1 week when only solvent-based inks were
used Historical data for emissions and waste generation were extrapolated for comparison with the
weekly experimental data. From the 4-week ink use and waste analysis data the VOCs released as
emissions from the printing process, could be calculated. A material accounting approach was used for
thee calculations. All liquid wastes generated during the test periods were segregated and analyzed for
percent volatile constituents. Substituting water-based inks required press modifications
164
-------
TABLES. VOC REDUCTION
Week
1
2
3
4
Total
ink (Ib)
2,363
2,254
2,252
1,237
Factored*
VOC (Ib)
1,772
1,634
1,633
897
Water ink
(%)
52.9
22.5
0.0
55.6
Reduced
VOC (Ib)
827.5
1,252.7
1,571.5
509.09
Reduction
(%)
53.3
23.4
0
43.3
* Calculated by taking 72.5% of the total ink quantity.
Waste Reduction-
Historically, 315 gal of solid waste was generated each month. This translates to approximately
one-and-one-half, 55 gal drum or 424 Ib/wk. The printing operations during Week 1 were an 87%
decrease from normal in solid waste generation (from 424 Ib to 55.5 Ibs); and 100% elimination of solid
waste generation in Weeks 3 and 4.
Note that much of this waste decrease can be attributed to factors other than the type of ink used.
The WRITE Program evaluation and the use of the waste generation from increased awareness of press
operators and deterred waste generation. This induced press operators to reuse solvent for additional
cleaning or reuse in the solvent inks.
Economic Analysis-
An economic analysis of the changeover from solvent to water-based ink is included as part of
this project.
Fixed, variable, and overhead costs are affected by this substitution and are considered. Fixed
costs include the purchase and installation of new equipment (primarily the Enercon corona discharge
treater) and costs for replacing equipment ancillary to the central impression cylinder press, such as
pumps, dryer upgrade, ink pans, etc.
Variable cost adjustments include the premium paid, or reduced cost, for water-based inks.
Disposal costs were calculated by using the amount of waste solvent ink generated in gals and the most
recent disposal cost figures provided by Lustreprint. Other variable costs included variations in labor
hours and utilities.
Overhead costs also play a role in determining the cost savings. Items such as the time
previously expended for regulatory compliance, insurance costs, employee equipment and safety training,
and OSHA compliance were expected to be reduced as a result of removing hazardous waste from the
shop floor. These potential cost savings were estimated from existing figures where available.
Based on these costs, payback period was calculated (Table 4).
The payback period could be further reduced by eliminating the solid waste disposal. With the
complete changeover to water inks and the planned purchase of an ink splitter, at approximately $8,000,
an additional savings for solid waste disposal would be possible. The payback period would then be
reduced 0.53 year.
167
-------
TABLE 2. INK AND VOC EMISSION DATA FOR 4-WEEK STUDY PERIOD
Measured Parameter
No. of inks
Solvent ink (Ib)
Water ink (Ib)
Total ink (Ib)
Solvent ink (%)
Water ink (%)
VOC emissions (Ib) (calculated)
VOC emissions (%) (% of ink total)
Waste (Ib)
Waste VOC content (Ib)
Weekl
" i . ...
23
1,112
1,251
2,363
47.1
52.9
827.5
35.0
55.6
54.3
-
Week 2
— — • ^— — __ — ___ __
32
1,746
508
2,254
77.5
22.5
1,251.7
55.5
20.0
4.7
=^ -=
"^^-^^^^— ^— ^i^-^^— ,
Weeks
33
2,252
0
2,252
100
0
1,571.5
69.8
0.0
0.0
=^*
Week 4
22
549
688
1,237
44.4
55.6
509.0
41.1
0.0
0.0
VOC Reduction-
P.rh n^f 'tUting ""^ Watter,basejd ink reduced the emissions generated from the printing process For
each percent increase ,n waterbased ink use, the calculated reduction in VOC emissions was 14 Ib (Table
thP nprrt f **?" iTl^'6 3' f°r ****** ^ and 2^ {he VOC 9enerati™ decreased in proportion to
the percentage of water-based ink used. A 52.9% water-based ink use resulted in a 53.3% reduction in
VOC em,ss,ons. S.m.larly, in Week 2, a water-ink use rate of 22.5% resulted in a VOC emissbn^duction
For Week 4, the corresponding reduction in VOC emissions was less significant' a 55 6% water
mk use rate reduced VOC emissions only 43.3%. Total ink use for Week 4 was 1 ,237 Ib of combtied
water ,nk and solvent mk This amount is approximately half that was used during the other 3 weeks of
W th th7 nUmbeL°f d?fTn! JnkS US6d in Week 4 is' however' comparable with that used !n Week
1 With the same number of mk changes at the printing stations and with each change requiring a
cleaning before adding new mk, the amount of cleaning make-up solvent relative to total ink use is
reduced
166
-------
#38 ULTRASONIC CLEANING AS A REPLACEMENT FOR A CHLOROFLUOROCARBON-BASED
SYSTEM
Participants
The host for the test, who also ran the vapor degreaser was Conax Buffalo Incorporated,
Cheektowaga, New York. Recra Environmental Incorporated, of Amherst, New York, on contract to Erie
County Environmental Services, provided test personnel, gathered the data and wrote the draft report.
Technology/Testing
Conax has been engaged in the design and manufacture of highly engineered, precision product
for industrial, aerospace, nuclear, fiber optic, and military applications. At Conax, stainless steel,
aluminum and copper parts coated with standard screw oils, water-based coolants, in-house shop dirt,
and metal shavings are in-house shop dirt, and metal shavings are cleaned in a series of cleaning and
rinsing tanks of modular design using a heated alkaline solution. Previously, cleaning activities involved
the use of two types of freon-based solvents that generated more than 10,000 Ib of fugitive emissions
annually from two vapor degreasers and two work bench stations.
Since 1990, chlorinated solvents and chlorofluorocarbons (CFC-113) including trichloroethylene,
1,1, 1-trichloroethane, trichlorotrifluoroethane (freon), and a freon/acetone mixture have been used at
Conax. The CFCs are used for both degreasing parts after machining, and cleaning parts prior to
assembly, shipment, or stock. Until recently, four operations within Conax utilized chlorinated solvents
and CFCs. These include machining centers parts cleaning, machine shop vapor degreasing, assembly
vapor degreasing, and final assembly cleaning.
The ultrasonic parts cleaning system was installed to avoid and eliminate the problems associated
with further CFC use.
The Miraclean* system used by Conax is designed and manufactured by Chautauqua Metal
Finishing Supply of Jamestown, New York. It is a modular design of cleaning and rinsing tanks,
employing an aqueous cleaning agent within the ultrasonic tank to accelerate and facilitate the cleaning
action (i.e., cavitation). Miraclean systems have a variety of available options such as additional rinse
tanks and dryer station to meet individual customer needs.
The ultrasonic cleaning system purchased by Conax entails six cleaning stations (see Figure 1).
The ultrasonic cleaning system was evaluated for 131 batches of parts ranging from large tubes
to pins and from 1 to several thousand parts/batch. Because this was considered typical production, the
results would be extrapolated to an annual basis. Average cleaning times and chemical addition
requirements were documented, and subjective quality control inspections were done on each batch.
Project forms developed for the project tracked the time in minutes for each batch at each station in the
cleaning process. Clean and rinse tank Ph was monitored along with the clean and heated rinse tank
temperatures. The number and description of parts in each batch were also listed on another project
form. Averages for processing times, Ph, and temperatures were calculated along with totals for a
breakout of batch part quantities from 1 to 15, 15 to 100, 100 to 1,000 and 1,000 + parts/batch.
169
-------
Variable
Initial
Investment
Projected
Savings
Payback
Period, year
Current process revisions
Adding an $8000 ink splatter
Full water-based ink conversion
$62,901
$70,901
$62,901
$ 24,587
$ 34,887
$117,078
2.56
2.03
0.54
Conclusions-
w'b
168
-------
TABLE 1. HISTORICAL AND PROJECTED
Year
1987
1988
1989
1990
1991*
1992t
Fugative Emission
(Ib)
25,215
32,990
12,819
10,876
6,900
3,450
Hazardous Waste
(Ib)
2,670
1,290
4,400
1,595
1,890
1,380
* Estimated for remainder of 1991.
t Projected for 1992.
Emissions generated at Conax originated from cleaning operation at the machining centers, the
assembly tables, and the Blakslee vapor degreaser. The elimination of CFC use at the machining centers
by substituting aqueous cleaners into the bench top ultrasonic cleaning units reduced emissions by 14,500
Ib/yr over a period from 1987 to 1990.
The elimination of the Blakslee vapor degreaser further reduced emissions to a projected total of
3,450 Ib/yr for 1992, a reduction of 68% from 1990. Total volatile emission reduction projections, from
1987 to 1992, are 86% from 25,215 to 3,450 Ib/yr resulting from the two operational changes.
An economic analysis of the changeover from CFC vapor degreasing using the Blakslee unit to
the Miraclean ultrasonic system utilizing an aqueous-based cleaning solution is included as part of the
project.
Fixed and variable costs have been considered as part of the evaluation. Fixed costs include the
cost for equipment arid installation of the Miraclean system. These costs include the ultrasonic
equipment, NEMA enclosure, three tank system, pumps, filter, sparger pump, tank covers, overhead
crane, supplies, and labor.
Variable costs included in the economic assessment were raw materials, power costs, sewer
fees, off-site disposal, water costs, and labor. Raw material cost was determined using 1990 cost data
and material use supplied by Conax. Labor cost was estimated using $15/hr as a basis. Sewer fees and
water cost information was supplied by Conax. Total operating costs were determined as a summation of
variable costs.
A total operating cost/batch of parts cleaned was also determined for comparison.
Based on these costs, annual savings and payback period for the new Miraclean system were
calculated for the project.
171
-------
Fresh '
Water Feed
\
Parts
\ Rinse Water
To Drain
Hot Rinse Tank
180°F
Second
Counterfbw
Rinse Tank
Ambient T°
First
Counterflow
Rinse Tank
Ambient 7*
Ultrasonic Clean
Tank
150-180°F
Figure 1. Conax's Miraclean System Schematic
Result?
estimated at 26 arum^/vr anrl 11*5/1 rvnn «->:n:«« i / , "" — •*• »"»'wvt.iwiiio
170
-------
#39 REMOVAL AND CONTAINMENT OF LEAD-BASED PAINT VIA NEEDLE SCALERS
Participants
The New York State Thruway Authority (NYSTA) provided two bridges for paint removal while
Pentek Incorporated, the developer of the technology provided the test hardware and operated it during
the test. Recra Environmental Incorporated, of Amherst, New York, on contract to Erie County
Environmental Services, provided test personnel, gathered the data and wrote the draft report.
Technology/Testing
The use of abrasive blasting with expendable grit media for removing lead-based paints from steel
structures has been the standard for many years, mainly because of its efficiency both in removing the
lead-based paint and in achieving the surface cleanliness and profiles required for subsequent coating
operations.
The process's generation of difficult-to-contain, airborne, lead-contaminated particulates presents
a high potential of lead exposure to workers and the local environment. Although sophisticated systems to
control or contain airborne particles would minimize the potential for environmental contamination, they
may result in more hazardous localized environments for workers and result in substantially higher overall
costs for lead-based paint removal operations. Additionally, using expendable abrasive grit to remove
lead paint generates excessive amounts of waste material that requires disposal as hazardous waste.
The industrial participants for this program were the New York State Thruway Authority (NYSTA)
and Pentek, Incorporated. The NYSTA is responsible for the operation and maintenance of the New York
State highway system. Pentek has been manufacturing dustless surface preparation equipment for use
by nuclear facilities and hazardous waste remediation/cleanup contractors since 1985.
The Pentek system is a form of power tool cleaning that combines material removal and
containment. The Pentek CORNER-CUTTER® (Figure 1), a hand-held needlegun for surface preparation
in tight spots and/or vertical and inverted horizontal steel or concrete surfaces, is one of three models of
surface preparation tools that Pentek manufactures.
Material is removed through the actions of pneumatically operated reciprocating steel cutting bits
made into needles that scarify and pulverize the paint or coating. The removed material is contained first
by using an adjustable shroud located at the tool's point of operation to localize containment, and second,
by transporting the contained materials via vacuum to an attached VAC-PAC containment vessel. The
vacuum head of the containment drum is equipped with high-efficiency particulate air (HEPA) filters that
prevent the escape of airborne dust.
Conventional abrasive blasting employs compressed air to propel expendable abrasive particles
against the surface to be cleaned, to produce a surface profile required by Standards and Specifications of
the Steel Structures Painting Council No. 6. The spent abrasive and paint debris are manually collected
for disposal, usually as hazardous waste.
173
-------
*AA A^ 1T w6 t0tM f'Xed °°StS f°r the Miraclean svstem according to information provided by Conax was
$44,411. Variable costs calculated for the two systems are listed in Table 2.
«, 0«v,,,yo projected using the aqueous ultrasonic system was calculated to be $27 178
, ,„=, ,«su,i«u in a per bath savings of $7.94 ($7.26/batch vs. $15.20/batch for vapor degreasinq)' A '
payback period for the system using the savings calculated and reported total costs was determined to be
Utility Costs
Labor Costs
Raw Material Costs
Water costs
Sewer Costs
Off-Site Disposal
Total Operating Costs
Freon Vapor
Degreaser ($)
1,559
8,205
33,939
1,780
6,200
370
52,053
Aqueous Ultrasonic
Cleaning System ($)
8,087
8,295
1,203
890
6,200
200
24.875
e lnS!f "ati0n °f an ultrasonic Parts Cleanin9 un't, utilizing water-based cleaners the
of vapor degreasing with solvent-based cleaners is possible without impacting cleaning quality.
cases,
9°°d'
Annual waste reduction realized was over 12,000 Ib when fugitive emissions are included.
An annual cost savings of $27,178 was calculated and results primarily from a reduction in raw
"UltraSO,nihc| Cleanin9 as a Replacement for a Chlorofluorocarbon-Based
.b. Kranz, et al., is available as report no. EPA/600/R-93/223.
172
-------
The abrasive blast media consisted of Ebony Grit 20, a non-silica, lead-free abrasive.
To minimize the potential for cross-contamination and to satisfy bridge painting schedules and
other logistical concerns, these comparative evaluations were conducted on two separate bridges. The
bridges had similar structures and closely matching coats of paint.
On Day 1 of the conventional abrasive blasting test on Bridge #10, background information
regarding the process was obtained and both cleanup activities from the prior day's work and setup
activities for work to be performed were observed and recorded. Background lead-in-air concentrations,
used as a baseline for both technology evaluations, were also monitored. On Day 2, work procedures
were observed. Monitoring of personal and area air was performed as well as taking data for
measurements to assess productivity and waste generation.
Information relative to time and labor requirements for daily cleanup and job site mobilization and
demobilization activities was also gathered. This information was integrated with job site observations to
estimate the man hours required and their associated costs.
The Pentek system was evaluated at NYSTA Bridge #1. Evaluation consisted of observing and
documenting mobilization, paint removal, and cleanup and demobilization activities.
Results
Equipment, equipment maintenance, vehicles, utilities and fuel, containment structures, and
personal protective equipment costs were not separately included in the evaluation. For simplicity and
uniformity, a standard labor rate of $15/hr was assumed for all labor classifications. Based on
demonstrated production rates, approximately eight Pentek systems, each using three CORNER-
CUTTERS®, would be needed to equal the production rate of the two-operator abrasive blasting system-
Labor requirements for support, mobilization, and demobilization were also higher for the Pentek
system, primarily because of the number of workers required. Cleanup labor costs were substantially
higher for the abrasive blasting process. These comparative labor costs are shown in Table 1.
TABLE 1. TOTAL ESTIMATED LABOR COSTS
Labor Category Abrasive Blasting ($) Pentek ($)
Paint Removal 1,500 18,450
Support 1,125 6,090
Mobilization 945 1,440
Demobilization 210 720
Cleanup 5460 240
Labor Total 9,240 26,940
175
-------
Removed paint
chips/dust/rust
pneumatc
operation housing
Adjustable containment
shroud
Substrate
Figure 1. Pentek CORNER-CUTTER" schematic
an,
174
-------
containment structures and engineered systems necessary when using abrasive blasting technology to
ensure worker health and safety and protection while under stricter air emission standards.
The full report, entitled "Removal and Containment of Lead-Based Paint via Needle Sealers" by
P.B. Kranz, is available as EPA 600/R-94/114.
177
-------
TABLE 2. HAZARDOUS WASTE GENERATION AND DiSPngAi COSTS
•Industry average for bulk waste including transportation
theoretical waste generated based upon .175 ton waste/ton of steel cleaned
tTheoret,cal waste generated based upon 11.5-mil. paint thickness and paint solids density of 66.3 lb/ft.=
TABLE 3. TOTAL COSTS
Catego
Labor
Materials
Hazardous waste
disDosal
!r?r ^e 8'hr time-weighted averages (TWA), the abrasive blasting data indicated O
^^^^£' « &%
'6ad °r res"irabte ^ and only
The Pentek system was significantly less efficient in removing paint (i e ft %r) esoeciallv the
The dustless needlegun system may be competitive when factoring in costs of sophisticated
176
-------
#41 FINISHING FABRICATED METAL PRODUCTS WITH POWDER COATINGS
Participants
The host for this evaluation was the Diversified Control, Inc. (DCI) of Orchard Park, NY. The
study, evaluation and preparation of the draft report was a joint effort by Erie County Department of
Environment and Planning, Division of Environmental Compliance Services (ECS) and Recra
Environmental, Inc. (RECRA).
Technology/Testing
The DCI facility manufactures junction boxes for the telecommunications and cable television
industries. The company has utilized wet, solvent-based epoxies, water-based coatings and most
recently, powder coating as a finishing process. Powder coatings used are primarily the polyester
thermosetting types.
The coatings are electrostatically applied to cold-formed, galvanized, steel coil stock that has
been cleaned, sealed, and then cured using an electric-resistance, infra-red process.
The evaluation compared product performance, environmental impact and economics of the three
technologies, wet organic solvent, water-based solvent and powder coating.
Measurements taken included discharge wastewater analysis for metals, oil sand grease, total
dissolved solids and Ph. The volume and characteristics of process solid waste were also be determined.
Solid waste characteristics includes ignitability, corrosivity, reactivity, and toxicity and TCLP determinations
for toxic metals and volatiles. Other data sources were company historical records and vendor data.
Results
The evaluation has been completed and the final draft is being prepared. The detailed results will
be published in the final report.
The final report, entitled "Finishing Fabricated Metal Products with Powder Coating" will be
published as an EPA/600 series report.
179
-------
#40 LOW-VOC WOOD FURNITURE COATINGS
Participants
Technology/Testing
Results
^
178
-------
SECTION 4
INDUSTRY INDEX
This section is a cross reference among industry/technology types, project titles, page numbers,
and project numbers used for the organizing the 41 projects. Table 1, presented below, is arranged on
the basis of an alphabetic listing by industry/technology types. The industry/technology types represented
are coating technology, depainting, electronics, metal plating and finishing, "other" (miscellaneous
industries), printing, surface cleaning and steel.
TABLE 1. INDUSTRY INDEX
Technology
C - Improved coating technology
(low- or no-solvent paints,
varnishes, improved
application)
D - Depainting (paint, varnish,
surface coat removal)
E - Electronics
Project Title
Evaluation of Five Waste Minimization
Technologies at the General Dynamics
Pomona Division; Robotics Painting
Low-VOC Wood Furniture Coatings
Finishing Fabricated Metal Products
with Powder Coatings
Evaluation of Five Waste Minimization
Technologies at the General Dynamics
Pomona Division; Plastic Bead Blast
Paint Stripping
Replacement Non-Methylene Chloride
Paint Remover
Removal and Containment of Lead-
Based Paint via Needle Sealers
Modifications to Reduce Drag Out at a
Printed Circuit Board Manufacturer
Sponge Rollers and Flow Controller for
Printed Circuit Board Manufacturing.
Carbon-Black Dispersion Preplating
Technology for Printed Wire Board
Manufacturing
Electronic Component Cooling
Alternatives: Compressed Air and
Liauid Nitroaen
Page
30
178
179
34
133
173
82
85
90
140
Project. #
05
40
41
06
29
39
18
19
20
25
181
-------
SECTION 3
REFERENCES
«nH P Re/ecrences used for this reP°rt are the project plans, Quality assurance plans final Project Reoort
and Project Summanes used to summarize the 41 technical evaluates under the WRITE pronram
Conclusions and recommendations presented in Section 1, Introduction are based on thP author's
•nterpretafons of the results as well as understandings reached during the S^ear pehod ol manaqha the
WRITE program and participating in conducting the projects for the state of Washington. managmg the
180
-------
Evaluation of ZERPOL (Zero Liquid
Discharge System) at Pioneer Metal
Finishing
119
26
O - Other
Q! Reduction of waste oil for
internal combustion
engines
Evaluation of Three Oil Filter Designs
for Pollution Prevention Effectiveness
17
02
O2 Metal working oils/fluids
recycling
Mobile Onsite Recycling of
Metalworking Fluids
106
23
O3 Recycling of spilled
oils/fluids, recycling of
sorbents
A Fluid Sorbent Recycling Device for
Industrial Fluid Users
110
24
O4 Spray painting line
cleaning solvent (MEK)
recycling
Onsite Solvent Recovery with an
Atmospheric Still
152
34
Pr - Printing
On-Site Newspaper Ink Recycling
44
Ink and Cleaner Waste Reduction
Evaluation for Flexographic Printers
62
WASTE EVALUATION OF SOY-
BASED INK AT A SHEET-FED
OFFSET PRINTER
78
Replacement of Hazardous Material in
Wide Web Flexographic Printing
Process
164
09
13
17
37
S- Surface cleaning
Evaluation of Five Waste Minimization
Technologies at the General Dynamics
Pomona Division; Freon Recovery
37
An Automated Aqueous Washer for
the Metal Finishing Industry
40
Ink and Cleaner Waste Reduction
Evaluation for Flexographic Printers
62
Evaluation of Ultrafiltration to Recover
Aqueous Iron Phosphating/Degreasing
Bath
74
Waste Evaluation of Soy-based Ink at
a Sheet-Fed Offset Printer
78
A Replacement Solvent
Cleaner/Degreaser Study at Duffy
Electric and Machine Company
123
07
08
13
16
17
27
183
-------
M- Metal plating or finishing
— ' • . _
Watts Nickel and Rinse Water
Recovery via an Advanced Reverse
Osmosis System
Evaluation of Five Waste Minimization
Technologies at the General Dynamics
Pomona Division; Electroplating Rinse
Water Reduction
Evaluation of Five Waste Minimization
Technologies at the General Dynamics
Pomona Division; Sulfuric Acid
Anodizing
Cadmium and Chromium Recovery
from Electroplatinq Rinsewaters
Nickel Recovery from Electroplating
Rinsewater by Electrodialysis
Chromate Recovery from Chromating
Rinsewater in the Metal Finishing
Industry
Alkaline Noncyanide Zinc Plating and
Reuse of Recovered Chemicals
Recycling Nickel Electroplating Rinse
Waters by Low Temperature
Evaporation and Reverse Osmosis
Evaluation of Ultrafiltration to Recover
Aqueous Iron Phosphating/
Degreasinq Bath
Modifications to Reduce Drag Out at a
Printed Circuit Board Manufacturer
Sponge Rollers and Flow Controller for
Printed Circuit Board Manufacturing
Carbon-Black Dispersion Preplating
Technology For Printed Wire Board
Manufacturinq
Evaluation of an Electrodialytic
Process for Purification of Hexavalent
Chromium
Substituting Cadmium Cyanide
Electroplating with Zinc Chloride.
14
20
25
48
54
58
65
70
74
82
85
90
94
101
01
03
04
10
11
12
14
15
16
18
19
20
21
22
182
-------
SECTION 5
WRITE PROGRAM REPORT LIST
The specific project reports and project report summaries may be obtained by ordering
through the Center for Environmental Research Information (CERI), or the National Technical Information
Service (NTIS) at the addresses indicated below.
CERI Publications Unit, US EPA
26 W. Martin Luther King Drive
Cincinnati, OH 45268
(513)569-7582
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
(703)487-4650
CALIFORNIA:
Watts Nickel and Rinse Water Recovery via an Advanced Reverse Osmosis System. C. Schmidt, et
al., EPA/600/R-93/150.
Evaluation of Three Oil Filter Designs for Pollution Prevention Effectiveness. Lisa M. Brown and
Robert Ludwig, to be available as an EPA/600 series report.
Evaluation of Five Waste Minimization Technologies at the General Dynamics Pomona Division:
Electroplating Rinse Water Reduction. Lisa M. Brown and Robert Ludwig, EPA/600/S2-91/067,
Evaluation of Five Waste Minimization Technologies at the General Dynamics Pomona
Division: Sulfuric Acid Anodizing. Lisa M. Brown and Robert Ludwig, EPA/600/S2-91/067.
Evaluation of Five Waste Minimization Technologies at the General Dynamics Pomona Division:
Robotics Painting.Lisa M. Brown and Robert Ludwig, EPA/600/S2-91/067
Evaluation of Five Waste Minimization Technologies at the General Dynamics Pomona Division:
Bead Blast Paint Stripping. Lisa M. Brown and Robert Ludwig, EPA/600/S2-91/067.
Evaluation of Five Waste Minimization Technologies at the General Dynamics Pomona Division:
Freon Recovery. Lisa M. Brown and Robert Ludwig, EPA/600/S2-91/067.
185
-------
St - Steel
— — .
A Supercritical Fluid Cleaning Study:
Application to Instrument Bearings
Low-Volatility Solvent and Filtration
System for Mechanical Parts Washinq
Power Washer with Wastewater
Recycling
Bicarbonate of Soda Blasting
Technology for Aircraft Wheel
Depainting
Onsite Solvent Recovery with Vacuum
Heat-Pump Distillation
Onsite Solvent Recovery with Low
Emission Vapor Degreasing
Ultrasonic Cleaning as a Replacement
for Chlorofluorocarbon-Based System
Mobile Onsite Recycling of
Metalworkinq Fluids
Recycling Electric Arc Furnace Dust:
Jorgensen Steel Facility
•" — — ^— .
127
139
144
148
155
158
169
— — — — ^—
106
135
1 1
28
31
32
33
35
36
38
23
30
184
-------
NEW JERSEY:
Mobile Onsite Recycling of Metalworking Fluids. Arun Gavaskar, et al., EPA/600/SR-93/114.
A Fluid Sorbent Recycling Device for Industrial Fluid Users. Abraham S.C. Chen, et al.,
EPA/600/SR-93/154.
Electronic Component Cooling Alternatives: Compressed Air and Liquid Nitrogen. Stephen C.
Schmidt, et al., EPA/600/SR-94/170.
Evaluation of ZERPOL (Zero Liquid Discharge System) at Pioneer Metal Finishing. Hanna J.
Saqa, will be available as an EPA/600 series report.
A Replacement Solvent Cleaner/Pegreaser Study at Duffy Electric and Machine Company. Bruce
M. Sass, et al., will be available as an EPA/600 series report.
A Supercritical Fluid Cleaning Study: Application to Instrument Bearings. Bruce M. Sass, et al., will
be available as an EPA/600 report.
Replacement Non-Methvlene Chloride Paint Remover. Bruce M. Sass, et al., will be available as
an EPA/600 report.
WASHINGTON:
Recycling Electric Arc Furnace Dust: Jorgensen Steel Facility. Trevor W. Jackson et al.,
EPA/600/R-95/007.
Low-Volatility Solvent and Filtration System for Mechanical Parts Washing. David P. Evers, et al.,
will be available as an EPA/600 series report.
Power Washer with Wastewater Recycling. David P. Evers, will be available as an EPA/600 series
report.
Bicarbonate of Soda Blasting Technology for Aircraft Wheel Depainting. Abraham S.C. Chen, et
al., EPA/600/R-94/127.
Onsite Solvent Recovery with: an Atmospheric Still Vacuum Heat Pump Distillation Low Emission
Vapor Degreaser. Arun R. Gavaskar, et al., EPA/600/R-94/026.
ERIE COUNTY, NY:
Replacement of Hazardous Material in Wide Web Flexographic Printing Process. P.B.Kranz, et
al., EPA/600/SR-93/149.
Ultrasonic Cleaning as a Replacement for Chlorofluorocarbon-Based System. P.B. Kranz, et al.,
EPA/600/R-93/223.
Removal and Containment of Lead-Based Paint via Needle Sealers. P.B. Kranz, et al.,
EPA/600/R-94/114
187
-------
CONNECTICUT:
GaVaskar' et al"
On-Site Newspaper Ink Recycling. Arun Gavaskar, et al., EPA/600/R-92/251
c Arun Gavaskar, et al.,
Nickel Recovery from Flantmnlating Rinsewater Dv Electrodialysis Arun Gavaskar et al to be
available as an EPA/600 series report. — '
Chromate Recovery from Chromatina Rinspyyater in the Metal Finishing industry Arun
et al., to be available as an EPA/600/ series report. -
ILLINOIS:
Fv*'Uati°n for Flgvnrjraphi. Prints Gary D Mj||erf
°f Re°nvemd r^mirals. Jacqueline M. Peden,
Recycling Nickel Electroplating Rinse Waters bv Low Temperature Evaporation and Reverse
Osmosis, Timothy C. Lindsey, EPA/600/R-93/160. -
Evaluation of Ultrafiltratinn to Recover Aqueous Iron Phnsphatina/Dgnrpaginn Rath Garv n
Miller, et al., EPA/600/SR-93/144. ' - - - M ' ^
Waste Evaluation of Soy-Rased Ink Rt a Sheet-Fed Offept Printn. Gary Miller, et al., EPAJ600/R-
MINNESOTA:
Man"fRrtlirPr' Teresa M- Harten et
f°r ™**d C^ "™* Man' 'fart' lrin^' wi" be available as
Carbon-Black Dispftrsinn PreD|atina Technology for Printed Wire Board Manufacturin
Folsom, et al., EPA/600/R -93/201. -- In
Evaluation of an Elflntmdialvtic Process for Purification of Hexavalent Chromium Dale Folsom
- " - - "' Udie roisom.
al., EPA/600/R-94/071 .
Substituting Cadmi
Substituting Cadmium Cyanide Electroplating with Zinc Chloride Electroplating B.C. Kim, et al.,
186
-------
SECTION 6
WRITE PROGRAM PERSONNEL
CALIFORNIA: U.S. EPA project officer - Lisa M. Brown (301) 975-5044
California Department of Toxic Substances, principal investigator -
Robert Ludwig
Hosts for technology evaluations - General Dynamics, Pomona Division,
Pomona, CA; Hewlett Packard, Sunnyvale, CA; Orange County Transit
Authority, Orange County, CA.
CONNECTICUT: U.S. EPA project officer - Lisa M. Brown (301) 975-5044
Connecticut Hazardous Waste Management Service, Connecticut
Technical Assistance Program, principal investigator - Rita Lomasney. ESSAR
Environmental Services The Connecticut Hazardous Waste Management
Service consultant - Sumner Kaufman.
Hosts for technology evaluations - Quality Rolling and Deburring,
Thomaston; The Hartford Courant; the Torrington Company, Torrington, CT;
Automatic Plating of Bridgeport, Bridgeport, CT.
ILLINOIS: U.S. EPA project officer - Paul M. Randall 513 569-7673
Illinois Hazardous Waste Research and Information Center, principal
investigator - Gary Miller. Hosts for technology evaluations - MPI Label
System Inc., University Park, IL; P&H Plating Company, Cook County,
IL; Graham Plating Company, Chicago,IL; R.B. White, Bloomington IL;
Office of Printing Services, University of Illinois, IL.
MINNESOTA: U.S. EPA project officer - Teresa Harten 513 569-7565
Minnesota Environmental Control Agency and Minnesota Technical
Assistance Program Agency, and Minnesota Technical Assistance
Program, University of Minnesota - principal investigators, Cindy
McComas and Paul Pagel.
Hosts for technology evaluations - Micom Inc, Brighton MN; Hutchinson
Technology Inc., Hutchinson, MN; McCurdy Circuits, Orange County, CA;
Paramax Inc. St. Paul, MN; SL Modern Hard Chrome, Camden, NJ;
Aeroquip Inc., Industrial Connectors Div., Van Wert, OH.
189
-------
NEW JERSEY:
U.S. EPA project officer - Johnny Springer 513 569-7542
New Jersey Department of Environmental Protection - principal
investigators, Mohamed Elsaady, Anthony Tomljanovic, New Jersey
Department of Environmental Protection and Energy - Norine Binder;
New Jersey Institute of Technology - Daniel Watts Hosts for technology
evaluations - Cook's Industrial Lubricants Inc., Linden, NJ; Newark Air
l-orce Base, OH; Pioneer Metal Finishing Company, Franklinville, NJ;
Duffy Electric and Machine Company, Chillicothe, OH; Safety-Kleen
Corporation, Elgin, IL; Honeywell Space Systems Group, Clearwater,
Tooele Army Depot, Consolidated Maintenance Facility, Tooele, UT.
WASHINGTON:
Robert Burmark.
U.S. EPA project officer - Ivars Licis 513 569-7718
State of Washington Department of Ecology - principal investigator
Hosts for technology evaluations - Earle M. Jorgensen Steel Company,
Seattle, A; Titus-Will Ford, Tacoma, WA; Municipality of Metropolitan '
Seattle, Atlantic Base Garage; Ellington Field, National Aeronautics and
Space Administration/Lyndon B. Johnson Space Center, Houston, TX;
Navistar International Transportation Corporation, Plastics Division,
Columbus, OH; Cooper Industries, Belden Division, Richmond, VA; Durr
Automation Inc., Davidsburgh, Ml.
ERIE COUNTY, NY:
U.S. EPA project officer - Paul Randall 513 569-7673
Erie County Department of Environmental Planning - principal
investigator Paul Kranz.
Hosts for technology Evaluations - Lustreprint Company, Erie County;
Conax Buffalo Inc., Cheektowaga, NY; New York State Thruway
Authority; Dinaire Corporation, Buffalo, NY; Diversified Control Inc
Orchard Park, NY.
EPA CONTRACTORS:
CA.
Battelle, Columbus OH; PEI, Cincinnati OH; SAIC, San Diego,
190
ol'.S. GOVERNMENT PRINTING OFFICE: 1995-650-006/22036
-------
-------
------- |