EPA-650/2-74-097
September 1974
Environmental Protection Technology Series
XwX-X'Xv
-------
EPA-650/2-74-097
VINYL CHLORIDE - AN ASSESSMENT
OF EMISSIONS CONTROL
TECHNIQUES AND COSTS
by
B.H. Carpenter
Research Triangle Institute
P.O. Box 12194
Research Triangle Park
North Carolina 27709
Contract No. 68-02-1325 (Task 17)
ROAP No. 21AUY-03
Program Element No. 1AB015
EPA Project Officer: Kenneth Baker
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
September 1974
-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
TABLE OF CONTENTS
1. INTRODUCTION 1
1.1 PURPOSE AND SCOPE 1
1.1.1 Process Types ...... 1
1.1.2 Process Emissions ..... 2
1.1.3 Emissions Controls 6
1.1.3.1 Recycling 6
1.1.3.2 Condensation 6
1.1.3.3 Compression 7
1.1.3.4 Adsorption 7
1.1.3.5 Incineration 8
1.1.3.6 Oxidation with Ozone 9
1.1.3.7 Scrubbing 9
1.1.3.8 Venting to Flares . 9
1.1.3.9 Longer Term Controls 9
1.1.4 Control Costs . 10
1.2 REPORT ORGANIZATION 10
2. VINYL CHLORIDE MONOMER 11
2.1 OVERVIEW OF VCM PROCESSES 11
2.2 HYDROCHLORINATION OF ACETYLENE 12
2.2.1 Emission Points 12
2.2.2 Controls and Costs 15
2.3 CHLORINATION-OXYCHLORINATION OF ETHYLENE (WITH AIR) AND
DEHYDROCHLORINATION 15
2.3.1 Emission Points 20
2.3.2 Controls and Costs 24
2.4 CHLORINATION-OXYCHLORINATION OF ETHYLENE (WITH OXYGEN) AND
DEHYDROCHLORINATION 27
2.4.1 Emission Points 27
2.4.2 Controls and Costs 31
2.5 DIRECT CHLORINATION OF ETHYLENE, AND DEHYDROCHLORINATION . . 31
2.5.1 Emission Points 31
2.5.2 Controls and Costs 40
3. POLYVINYLCHLORIDE 43
3.1 OVERVIEW OF PROCESSES ' . 43
3.2 SUSPENSION POLYMERIZATION 43
3.2.1 Emission Points 43
iii
-------
TABLE OF CONTENTS (Continued)
3.2.2 Controls and Costs 49
3.3 EMULSION POLYMERIZATION 49
3.3.1 Emission Points 49
3.3.2 Controls and Costs 55
3.4 BULK POLYMERIZATION 59
3.4.1 Emission Points 59
3.4.2 Controls and Costs 59
3.5 SOLUTION POLYMERIZATION 62
3.5.1 Emission Points 62
3.5.2 Controls and Costs 67
4. SUMMARY OF ASSESSMENT AND FUTURE NEEDS 71
4.1 RESULTS OF PRODUCTION-EMISSION ASSESSMENT 71
4.2 RESULTS OF SOURCE-CONTROL ASSESSMENT 72
4.2.1* VCM from Hydrochlorination of Acetylene (Process 1) .... 72
4.2.2 VCM from Chlorination-Oxychlorination of Ethylene (Using
Air) and Dehydrochlorination (Process 2) 73
4.2.3 VCM from Chlorination-Oxychlorination of Ethylene (Using
Oxygen) and Dehydrochlorination 73
4.2.4 VCM from Direct Chlorination of Ethylene and
Dehydrochlorination 73
4.2.5 VCM from Suspension Polymerization 74
4.2.6 VCM from Emulsion Polymerization 74
4.2.7 VCM from Bulk Polymerization 75
4.2.8 VCM from Solution Polymerization 75
4.3 RESEARCH AND- DEVELOPMENT NEEDS 75
iv
-------
LIST OF TABLES
1-1 VCM and PVC Production and VCM Emissions Estimates
1-2 Comparison of VCM Emissions by Clusters of VCM and
PVC Producers
2-1 VCM Plant Emission Data 11
2-2 Emission Rates for VCM Process 1 14
2-3 Controls and Costs for VCM Process 1 16
2-4 Emission Rates for VCM Process 2 22
2-5 Controls and Costs for VCM Process 2 25
2-6 Emission Rates for VCM Process 3 32
2-7 Controls and Costs for VCM Process 3 34
2-8 Emission Rates for VCM Process 4 38
2-9 Controls and Costs for VCM Process 4 41
3-1 VCM Emissions Data from PVC Plants 46
3-2 Emission Rates for PVC Suspension Process ......... 48
3-3 Controls and Costs for PVC Suspension Process 50
3-4 Emission Rates for PVC Emulsion Process ..... 56
3-5 Controls and Costs for PVC Emulsion Process 57
3-6 Production Capacities of Bulk Plants 59
3-7 Emission Rates for PVC Bulk Process 61
3-8 Controls and Costs for PVC Bulk Process 63
3-9 Production Capacities of Solution Plants 62
3-10 Emission Rates for the PVC Solution Process 66
3-11 Controls and Costs for PVC Solution Process 68
-------
LIST OF FIGURES
2-1 Simplified Flow Sheet Acetylene to Vinyl Chloride Process. . 13
2-2 Ethylene Dichloride from Ethylene and Chlorine 17
2-3 Ethylene Dichloride from Ethylene + HCL + Air 18
2-4 Dehydrochlorination of Ethylene Dichloride 19
2-5 Ethylene Dichloride from Ethylene and Chlorine 28
2-6 Ethylene Dichioride from Ethylene + HCL + Oxygen 29
2-7 Dehydrochlorination of Ethylene Dichloride 30
2-8 Ethylene Dichloride from Ethylene and Chlorine 36
2-9 Dehydrochlorination of Ethylene Dichloride . . 37
3-1 Simplified Flow Diagram PVC Suspension Process 44
3-2 Simplified Flew Diagram Batch Emulsion Polymerization of . .
Vinyl Chloride 53
3-3 Simplified Flow Diagram Bulk Polymerization of Vinyl
Chloride 60
3-4 Simplified Flow Diagram - Solution Polymerization
Process for PVC 65
vi
-------
ACKNOWLEDGMENTS
This document is a compilation and assessment of the vinyl chloride
emission sources and controls in the monomer and polymer segments of the
industry. The assessment was carried out under the direction of Dr. Dale
A. Denny, and the technical direction of Mr. Kenneth Baker. It is based
upon data presented by manufacturers in June 1974.
Major contributions were made by a team of scientists: Ben H. Carpenter
(Research Triangle Institute), Stanley E. Dale (A. D. Little, Inc.), J. J.
Kearney (RTI), Richard A. Markle (Battelle), and C. J. Santhanam (A. D. Little,
Inc.). At RTI, Kay Marr provided editing. Ronnie Davis and Walter Williams,
III, provided drafting and Anne Inscoe typed the manuscript.
Vll
-------
VINYL CHLORIDE
An Assessment of Emissions Control Techniques
and Costs
1. INTRODUCTION
1.1 PURPOSE AND SCOPE
The purpose of this study is to provide an assessment report on vinyl
chloride emissions from monomer and polymer processes currently existing in
the United States. The assessment is based on information supplied at this
time (June 18, 1974) by manufacturers in response to Mr. Don R. Goodwin's
letter of request, written under the authority of the Federal Clean
Air Act (42 U.S.C. section 1857 et seq.). Requested process flowsheets
were not yet available. Requested cost data generally were not clearly
defined. For these reasons, simplified flowsheets were prepared to describe
the sources of emissions for the different processes; these led to a
listing of sources, which served as a framework for first tabulating and
then assessing control methods. The study results are tentative. They
could be altered or contradicted by data that accrue as manufacturers
continue to respond. For the extent of data, however, the assessments
are believed to be valid.
1.1.1 Process Types
Vinyl chloride monomer (VCM) processes are of four types: (1) hydro-
chlorination of acetylene; (2) chlorination-oxychlorination of ethylene
(using air) and dehydrochlorination; (3) chlorination-oxychlorination of
ethylene (using oxygen) and dehydrochlorination; and (4) direct chlorination
-------
of ethylene. Two VCM production processes — hydrochlorination and dehydro-
chlorination are represented by the acetylene-hydrogen chloride reaction
HC=CH + HC1 - >- H
and the thermal dehydrochlorination of 1,2-dichloroe thane
C1CH2-CH2C1 - *• H2C=CHC1.
Two VCM plants use the hydrochlorination of acetylene process. The
plants using dehydrochlorination of ethylene dichloride (EDC) usually are
integrated with an EDC production unit, but the plant processes for producing
EDC differ. Nine plants use the balanced oxychlorination process: ethylene
is chlorinated by a mixture of pure chlorine and chlorine produced by the
oxidation of hydrogen chloride, which is recycled from the cracking of EDC.
One plant uses a modified balanced oxychlorination process: oxygen instead
of air is used to oxidize the hydrogen chloride to chlorine. Three plants
use the direct chlorination process: ethylene is chlorinated and dehydro-
chlorinated, but the hydrogen chloride from the dehydrochlorination step
is not recycled. The 15 plants have a production capacity of 6.8 billion
pounds of VCM.
Polyvinylchloride (PVC) production capacity at 37 plant sites totals
(by name) 4.75 billion pounds a year. There are four production processes —
(1) suspension polymerization (78% of total production); (2) emulsion
polymerization (12%); (3) bulk polymerization (6%); and (4) solution
polymerization (4%) .
1.1.2 Process Emissions
Emissions of gaseous VCM occur at both VCM and PVC resin plants. This
VCM is distributed into the atmosphere surrounding the plant sources. The
-------
emissions patterns depend on the amount of VCM released, the nature of the
plant area from which it is released, and the meteorological conditions.
The concentrations occurring in residential areas are not known at the
present time, but appear to be mostly below 1 part per million (ppm). If
control at lower air concentrations becomes necessary, this will have to be
accomplished by limiting the VCM emissions at the plant sources. The level
of VCM which may be considered safe has not been determined. Until April
of 1974, the acceptable level was approximately 500 ppm; since April, an
emergency standard of 50 ppm has been established.
Since the accumulated data for this study show that 1973 production
capacities were 6.8 billion pounds of VCM and 4.75 billion pounds of PVC
resin and that PVC manufacturers are operating at full capacity in 1974, it
must be assumed that present production is either equal to or greater than
the 1973 capacities. Analysis of the data so far supplied to EPA by VCM
and PVC producers show that total VCM emissions escaping to the atmosphere
is on the order of 156-210 million pounds per year. Of the total VCM,
over 90 percent is from PVC production facilities, as shown in Table 1-1.
VCM-producing companies are listed in Table 2-1. Included are their
geographical locations, population figures for adjacent communities, and
Table 1-1. VCM AND PVC PRODUCTION AND VCM EMISSIONS ESTIMATES, 1973
Plant
Type
VCM
PVC
Total
Production Capacity
(MM Ibs/yr)
6,800
4,750
VCM Emitted
(MM Ibs/yr)
14-20.4
142-190
156-210
VCM Emitted
(% of production)
0.2-0.3
3-4
-------
calculated VCM emissions levels. FVC producers are similarly listed in
Table 3-1.
The VCM emissions from PVC plants are at least 10 times as copious as
those from VCM plants. This comparison is based on data supplied by
relatively few (7) companies. The reporting companies state that the
emission figures were arrived at by measurements, estimates, and guesses.
In no case has it been possible to calculate overall material balances.
Thus the accuracy of the VCM emissions data cannot as yet be assessed.
The lower VCM emissions levels (pound per pound of product) at VCM
plants are partially offset by the fact that these plants usually have
appreciably larger production capacities than PVC plants. The net result
is that the absolute VCM emissions levels from PVC plants are usually 2
to 5 times higher than those of VCM plants.
All data received have shown a large fraction of total VCM emission
losses as fugitive losses (30-50 percent) with no breakdown of individual
emission points. Fugitive losses should be presented as an aggregate of
small losses from valves, packing glands, pump seals, agitator seals,
compressor seals, flange connections, safety valves, pining
leaks, and instrument connections. Total fugitive losses should be not
more than about 10 percent of total.
Two other features of the industry are significant in terms of po-
tential VCM concentration levels in the atmosphere near plants: (1) VCM
plants are to some extent clustered in areas along the Texas-Louisiana
Gulf Coast and (2) PVC plants tend to be located either close to/or
adjacent to VCM production sites. Clustering of plants is greatest near
Pasadena-Deer Park, Texas, and Baton Rouge, Louisiana. In Table 1-2,
-------
Table 1-2. COMPARISON OF VCM EMISSIONS BY CLUSTERS OF VCM AND PVC PRODUCERS
Company Name and Location
Name Plate Type
City Production VCM of
Population Capacity Emissions3 Proc-
(1970 census) (MM Ib/yr) (MM Ib/yr) essb
Allied Chemical
Dow Chemical
Ethyl Corporation
Ethyl Corporation (Indus
Chem Div)
Georgia Pacific
The Goodyear Tire &
Rubber Co (Chem Div)
Monochem, Inc
Total
Continental Oil
P.P.G. Industries
Shell Chemical
Certainteed Corp
Total
Diamond Shamrock Corp
(Chem Co Plas Div)
Ethyl Corporation
Shell Chemical
Tenneco Chemical
Union Carbide Corp
(Chem & Plas Div)
Total
Dow Chemical
Dow Chemical
Shintech
Total
Baton Rouge, La
Plaquemine , La
Baton Rouge, La
Baton Rouge, La
Plaquemine, La
Plaquemine, La
Geismar, La
Lake Charles, La
Lake Charles, La
Narco, La
Lake Charles, La
Deer Park, Tex
Pasadena, Tex
Deer Park, Tex
Pasadena, Tex
Texas City, Tex
Freeport, Tex
Oyster Creek, Tex
Oyster Creek, Tex
165,963°
7,739d
165,963C
165,963C
7,739d
7,739d
7,739d
77,998
77,998
77,998
12,773e
89,277e
12,773e
89,277e
38,908
ll,997f
ll,997f
ll,997f
300
340
270
180
200
100
300
1690
600
300
700
200
1800
250
150
875
250
200
1725
180
800
250
1230
0.9
1.0
0.8
7.2
8.0
4.0
0.9
22.8
1.8
0.9
2.1
8.0
12.8
10.0
0.5
2.6
0.8
8.0
21.9
0.5
2. A
10.0
12.9
B
DC
B
S
U
S
A
B
B
B
U
S&E
DC
B
A
S
DC
B
U
a Extrapolations based on estimated 0.3% VCM atmospheric emission loss from VCM
produced, and on 4% from PVC produced.
A, chlorination of acetylene; B, chlorination-oxychlorination, with HCL from cracking re-
cycled to oxychlorinacion; OL, direct chlorination; E, emulsion polymerization;
S, suspension polymerization; U, unknown (under construction).
c Baton Rouge Parrish (county) - 302,031.
Plaquemine Parrish (county) - 25,225.
8 Harris County - 1,741,912.
Brazoria County - 10&.312.
-------
geographically clustered VCM and PVC producers are listed together for
comparison.
1.1.3 Emissions Controls
Emissions control equipment used in the state of the art is often a
basic part of the processing system, thus it serves primarily to recover
a reactant or a product. State of the art controls are appraised herein
by using performance data from the VCM and the PVC manufacturers or by
comparing the emissions levels of one plant with and one plant without
controls. Controls so appraised include recycling, condensation with
refrigeration, compression, adsorption, incineration, oxidation with ozone,
scrubbing (absorption), and venting to flares.
1.1.3.1 Recycling
Recycling is used in the balanced EDC process to return EDC vapors
vented from oxychlorination units to the recovery units of the direct
chlorination systems and to recycle the vent stream from the VCM light-ends
column to the oxychlorination reactor. Recycling eliminates the need
for additional recovery systems and reduces the total number of emissions
sources, but it carries VCM back to early process stages where VCM emissions
would not be expected to occur otherwise.
1.1.3.2 Condensation
Since many process vent streams contain hydrocarbons, acids, and low-
molecular-weight chlorinated hydrocarbons as well as VMC, the condensation
via heat exchange is used for recovery of material and, often incidentally,
for emissions control. When vent streams contain water vapor, condensation
is carried out at the lowest practical temperature consistent with keeping
the water from freezing on the condenser. When vent streams do not contain
-------
water vapor, refrigeration can be used for more effective control. The
low boiling point of VCM, however, imposes a natural limit to the ultimate
effectiveness of condensation. Temperature and pressure limits must be
maintained to prevent hazardous reactions (explosions) in the recovery
system because VCM is a reactive gas. Due to its flammability, the gas
must be handled in nonoxygen-containing streams, and most of the recovery
efforts must be made without exposure to air. When vent gases from the
acetylene-hydrogen chloride reactor are cooled to below 40°F at 35 psig
to recover VCM and then scrubbed (water) to remove hydrogen chloride (HC1),
about 85 percent of the VCM is removed by condensation.
1.1.3.3 Compression
Compression is used in controlling emissions in transfer operations, in
reducing the costs of refrigerated condensation units, and in combination
with condensation to help control emissions from reactor units. Overhead
vapors from receiving tanks are compressed, condensed, and recycled to the
tank car (or barge) being unloaded. Vent gases from PVC reactors may
contain 80 percent VCM after passing a water-cooled condenser (85°F).
A refrigerated condenser (40°F) would reduce the VCM emissions by 35-40
percent; at lower temperatures, water vapor, if present in sufficient
concentration, tends to freeze on the condenser surfaces. Condensation
at 40°F with compression to 40 psig would reduce the VCM emissions by an
estimated 60-75 percent, and further compression to 80 psig would reduce
them an estimated 80-90 percent.
1.1.3.4 Adsorption
In general, adsorption is recognized as the most efficient control
method for achieving low concentrations of organic emissions in air. Use
-------
of such a "concentrating device" could drastically reduce energy costs
involved in the compression and refrigeration cycles normally required to
condense dilute streams of fugitive VCM efficiently. Adsorption on
activated carbon (or another sorbent) should effect an estimated 90 percent
or better reduction in the VCM content of vented process gases. Use of
adsorption requires a system with provisions for regeneration of the sorbent
bed (desorption) and provision for recovery or disposal of the VCM. Two
or more adsorbers may be installed in parallel for alternate use and
regeneration. Desorption of VCM with hot inert gases permits recovery
or incineration of the vapors. Desorption may be accomplished with steam,
followed by condensation and phase separation.
Adsorption is indicated for use as a final control for vent gases
already subjected to condensation and/or scrubbing. Little experience has
been reported in these uses, however, and concern has been expressed that
polymerization of VCM on the carbon surface might reduce its capacity
substantially.
1.1.3.5 Incineration
Incineration is used to oxidize hydrocarbon contaminants to carbon
dioxide and water. During combustion, chlorinated hydrocarbons with hydrogen-
to-chlorine ratios of at least 5:1 yield hydrogen chloride; those with ratios
less than this yield chlorinated hydrocarbons which are difficult to collect.
Downstream from the incinerator, either absorption equipment (e.g., caustic
gas scrubbers) or possibly a recovery process is required to extract the
hydrogen chloride produced during combustion of VCM, PVC, or other chlo-
rinated compounds. Both incinerators and afterburners can oxidize influent
VCM at nearly 100 percent efficiency. Cycling afterburners can be used
8
-------
with adsorbers to conserve fuel; during desorption, the afterburner
incinerates the stripped VCM. A regenerative heat exchanger can be used
to preheat the influent, conserve fuel, and defray costs.
1.1.3.6 Oxidation with Ozone
Ozone oxidation is considered appropriate only for streams with a
few ppm's concentration because this process is expensive. As a clean-up
method, it is appropriately used after application of other techniques to
achieve complete elimination of VCM.
1.1.3.7 Scrubbing
Scrubbing with high-impact water .streams is generally used to remove
hydrogen chloride and some EDC from vent streams. Absorbents other than
water (lean oil, red oil, kerosene, and chlorinated hydrocarbons) might
be practical for removing VCM from oxygen-free streams, but the scrubbing
system would be large and would have to be operated at several atmospheres
of pressure to permit practical solvent rates, so operating costs would be
high.
1.1.3.8 Venting to Flares
VCM vented to and burned in flares releases hydrogen chloride, which
is also a pollutant. Flares are, therefore, a questionable control
technique and should be considered primarily for emergency control. Since
flaring usually requires supplemental fuel, the cost varies with fuel prices,
1.1.3.9 Longer term Controls
Longer term controls are those the use of which requires development
effort and comparative evaluation. Included are process modifications and
washing with EDC. Some oxychlorination systems may profit from the
addition of a postchlorination unit which would presumably eliminate VCM
-------
by chlorinating it along with any unreached ethylene. In practice, vent
streams other than the oxychlorinator effluent would be fed to this unit.
From direct chlorination systems, eligible feeds include: the direct
chlorination reactor vent, the water-wash vents, and the light-ends column
vent. Crude EDC may be an effective absorbent for VCM in vent gases;
the possibility requires investigation.
1.1.4 Control Costs
Control costs presented herein were supplied by manufacturers.
Generally, only capital investment costs are given. In some cases, annual
operating and maintenance costs are included. No standardization of cost
elements was achieved because data were too sparse; the data received
appeared to have been based on estimates for the particular process using
cost elements (e.g., labor) appropriate for the location. When ranges of
cost are shown, they should be regarded as best approximations only.
1.2 REPORT ORGANIZATION
The assessment is divided into two main parts—VCM processes and PVC
processes. Each of eight major processes is discussed as a separate entity.
Emission points are identified using simplified flow diagrams. Their
controls are discussed in terms of the emission source characteristics,
and cost estimates are given where available. The findings are summarized,
and research and development needs are indicated.
10
-------
2. VINYL CHLORIDE MONOMER
2.1 OVERVIEW OF VCM PROCESSES
Table 2-1 lists VCM producers to show published production capacities
and VCM estimated emissions by process type and location.
Table 2-1. VCM PLANT EMISSION DATA
Name Plate Type
„ „ j T • Clty Production VCM of
Company Name and Location n . *.. _ ... - . . a n
f J Population Capacity Emissions Pro-
(1970 census) (MM Ib/yr) (MM Ib/yr) cessb
Allied Chemical, Baton Rouge, La
American Chemical, Long Beach, Calif
Continental Oil, Lake Charles, La
Dow Chemical, Freeport, Tex
Dow Chemical, Plaquemine, La
Dow Chemical, Oyster Creek, Tex
Ethyl Corporation, Baton Rouge, La
Ethyl Corporation, Pasadena, Tex
B. F. Goodrich, Calvert City, Ky
Monochem Inc, Geismar, La
P.P.G. Industries, Lake Charles, La
P.P.G. Industries, Guayanilla, PR
Shell Chemical, Deer Park, Tex
Shell Chemical, Narco, La
Tenneco Chemical, Pasadena, Tex
Total
165,963*:
358,633°
77,998
11,997*
7,739'
e
165,963°
89,277^
31,627n
7,739
77,998
12,773s
89,277g
300
170
600
180
340
800
270
150
1,000
300
300
575
875
700
250
6,810
0.9
0.5
1.8
0.5
1.0
2.4
0.8
0.5
3.0
0.9
0.9
1.7
2.6
2.1
0.8
20.4
B
B
B
DC
DC
B
B
DC
B
A
B
U
B
B
A
Extrapolations based on estimated 0.3% of VCM production lost to atmosphere.
A, chlorination of acetylene; B, chlorination-oxychlorination, with HCL from
cracking recycled to oxychlorination; DC, direct chlorination; U, unknown.
C Baton Rouge Parrish (county) - 302,031.
Los Angeles County - 7,032,075.
6 Brazoria County - 108,312.
Plaquemine Parrish (county) - 25,225.
8 Harris County - 1,741,912.
Population of Paducah, Ky.
11
-------
2.2 HYDROCHLORINATION OF ACETYLENE
Two plants are reportedly producing VCM by the hydrochlorination
of acetylene. This process accounts for 8 percent of the total VCM
production.
2.2.1 Emission Points
The VCM is produced by the hydrochlorination of acetylene (Figure 2-1)
using a mercuric chloride catalyst. The product gases are compressed,
cooled, and pumped to a purification system consisting of a light-ends
distillation column and a heavy-ends distillation column. The light-ends
are passed to a vent-gas reactor, from which the product is recycled to
the purification system; the heavy-ends are transferred to a temporary
storage tank and eventually incinerated.
The main sources of emissions are (1) the reactor condenser vent,
(2) the scrubber vent, (3) the VCM product-loading facility, and (4) the
heavy-ends storage vent. The vent-gas reactor vent and the scrubber are
continuous emission sources, the others are intermittent. The instantaneous
rate of loss in the loading system may be 1-2 times greater than that in
the purification system, but the VCM loss per million pounds of VCM produced
is greater in the purification system.
The emission sources listed in Table 2-2 correspond with operational
steps in a generalized hydrochlorination process. The data on plant
production and emission rates are calculated values; their reliability
cannot be estimated from information presently available. Fugitive sources
(pumps, pump maintenance, valves, pressure relief valves, samplers) were
not identified separately by the manufacturers, so the value quoted is an
estimated total. Estimates of emissions resulting from upset conditions
12
-------
VENT
VENT GAS REACTOR
VENTt-Q
i—Lizr
HCL
»
•-
kCETYLENE
REACTOR
1
P 1
* , .
1
^**
' ^
LIGHT ENDS
DISTILLATION
i
^-•^
, *
^JL
i
i
to
SCRUBBING LIQUID
SEWER
VENT CONDENSER
VENT
VENT
VCM STORAGE
HEAVY ENDS
DISTILLATION
©
HEAVY ENDS
STORAGE
VCM
TO DISPOSAL
FIGURE 2-1 SIMPLIFIED FLOW SHEET ACETYLENE TO VINYL CHUMIDE PROCESS
-------
Table 2-2. EMISSION RATES FOR VCM PROCESS 1: HYDROCHLORINATION OF ACETYLENE
1.
2.
3.
4.
5.
6.
PROCESS EMISSION SOURCE3
Vent-gas reactor vent , C
Vent-gas scrubber vent, C
Heavy-ends distillation column vent, I
VCM storage tank vent, I
Heavy-ends storage tank vent , I
VCM- loading facility, I
Total
PLANT
Ib VCM/MM Ib
produced
Production
3528
750
4278
A
Ib VCM/hr %VCM Comment on Plant Operation
emitted emissions
Sources Emitting VCM
117.5 61
13
117.5 74
Fugitive Sources Emitting VCM
Pumps, C
Pump maintenance, I
Valves, C
Relief valves, C
Flange connections
Samplers, C
Miscellaneous
Total
1500
50.0
26
C is continuous, I is intermittent.
-------
were generally not available, so no values appear in the table; furthermore,
the quantities of these emissions depend on the efficiency of operation and
quality of maintenance and thus would not be generally applicable from
plant to plant.
2.2.2 Controls and Costs
The level of VCM can be reduced with state of the art controls to
significantly below 50 ppm. Selection of a control method depends on the
size of the emission source and the reduction level desired. Fugitive
sources must be identified before making a choice of controls available
to reduce individual emissions to any specified level.
The state of the art controls should be applicable to existing
acetylene-hydrogen chloride plants. The phasing out of this acetylene-
dehydrochlorination process and the likelihood that state of the art
controls will suffice to bring this process under good control obviate the
need for consideration of long-term controls.
The costs of reducing VCM emissions to <50 ppm have been provided by
some producers and have been estimated for other emission sources for
which data were provided. The estimates consist mainly of equipment and
instrumentation costs. Installation and labor costs vary with plant makeup
and location and therefore have not been included. In general, the costs
for reducing a major VCM emission stream (100-500 Ibs/hr) by 99 percent
range from $50,000 to $100,000, as shown in Table 2-3.
2.3 CHLORINATION-OXYCHLOKINATION OF ETHYLENE (WITH ALR) AND DEHYDRO-
CHLORINATION
The balanced oxychlorination process (Figures 2-2, 2-3, 2-4) for the pro=-
duction of EDC is the principal one used in this country in the manufacture of
15
-------
Table 2-3. CONTROLS AND COSTS FOR VCM PROCESS 1: HYDROCHLORINATION OF ACETYLENE
Emission Source Control
Percent Control
Emission Cost
Reduction Estimate
Comment
1. Vent-Gas Reactor Vent: continuous stream; 40°F, 35 psi
Refrigerated condenser, HCL scrubber
Carbon bed adsorber
85
99
$50,000 Plant A.
$50,000 Cost + installation + adsorbent.
2. Vent-Gas Scrubber Vent: continuous stream; 50-100°F
3. Heavy-Ends Distillation Column Vent: intermittent stream
4. VCM Storage Tank Vent: intermittent stream
5. Heavy-Ends Storage Tank Vent: intermittent stream
6. VCM-Loading Facility: intermittent stream
Fugitive Sources: continuous stream
Pumps
Pressurized mechanical seals
Valves and relief valves
Diaphragm valves
Double lock valves
Samplers (collectors)
Miscellaneous
Preventive maintenance
95
95
50
90 $1,000
50 $40,000 Cost of 1/2 man-year labor.
-------
V) «$CHU$R1|NAT.ON ^ FROM DEHYDgOCHU)-
CHLORINE
ETHYLENE
EDC RECYCLE
SNIFF CHLORINE
-0
WASHED
CRUDE
STORAGE
FINISHING
COLUMN
EDC PRODUCT
STORAGE
PRIMARY WASH
SECONDARY
WASH
WATER FROM OXYCHUORINATION AND VCM
E
LIGHT ENDS
WASH COLUMN
WATER
STRIPPER
TO REACTOR OR SALES STORAGE
WASTE WATER
TO EDC RECOVERY SYSTEM
^*
13
HEAVY ENDS
STORAGE
HEAVY ENDS TAR
REMOVAL COLUVU
TAR STORAGE
*• DISPOSAL
FIGURE 2-2 ETHYLENE DICHLORIOE FROM ETHYLENE AND CHLORINE
-------
ETHYLENE
AIR.
HYDROGEN
CHLORIDE
oo
VENT
HOT WATER
WATER-
TO WASHWATER
STRIPPING COLUMN
»STEAM
f
TO WASH CRUDE STORAGE
t^
G
AIR-
WASH WATER
TO WASHWATER
STRIPPING COLUMN
REACTOR
FIGURE 2-3 ETHYLENE BICHLORIDE FROM ETHYLENE + HCL+ AIR
-------
RECYCLED TO
OXYCHLORNATION
CONDENSER
oo
COMPRESSOR
QUENCH
COLUMN
CRACKING FURNACE
EVAPORATOR
—N
PARTIAL CONDENSER
©
I _D
EPC QUENCH
STREAM
REB01LER
WASTE
WATER
I
VENT
PHASE
SEPARATOR
UQHTENDS
COLUMN
LxJ
REBCMLER
-------
VCM. Currently, nine plants using this procewa produce 5035 niLUiou
pounds a year, accounting for 74 percent oi" total VCM utagc and approxlmateJy
77.5 percent of the 20.4 million pounds a year of VCM emissions.
2.3.1 Emission Points
The VCM manufacturing operations discussed in this section use two
processes—direct chlorination for making EDC from ethylene and chlorine,
and oxychlorination for making EDC from ethylene, hydrogen chloride, and
oxygen from the air. The EDC intermediate product is cracked to VCM during
dehydrochlorination.
During direct chlorination, the inerts in th.e chlorine feed are
vented from the crude EDC (after refrigeration of the gas) and the catalyst
is washed away with water. During -the oxychlorination, about 98 percent of the
ethylene is converted to EDC, and the gases from the reactor are cooled
to condense the product, then washed with water (or an aromatic solvent)
to remove hydrogen chloride, and vented to the atmosphere.
Crude EDC produced by the two processes is combined with recycled
crude EDC from cracking recovery and subjected to final distillation to
remove light and heavy ends. Refined EDC may be sold. If processed to VCM,
pure EDC is transferred to the VCM section of the plant, where it undergoes
thermal dehydrochlorination (cracking) as it passes through empty or packed
tubes (tubes packed with charcoal, pumice, or similar materials) at 900-950CF
and at pressures of 50-300 psig. The reactor gases (VCM, EDC, and hydrogen
chloride) are quenched in a column and transferred to a light-end distillation
column, where the hydrogen chloride is stripped off and recycled to the oxychlo-
rinator. (In some plants, the hydrogen chloride is removed by scrubbing
with water; the noncondensibles are removed from the column top; and EDC,
20
-------
from the bottom.) Then, the product stream undergoes a number of
distillation steps to remove both heavy and light ends (a drying
column is sometimes included in this sequence). The final product,
normally VCM, is transferred to storage and finally to loading
facilities.
The VCM emissions occur not only during the dehydrochlorination but
also, to a lesser extent, during the manufacture of EDO; in the last case,
emissions occur because of side reactions and because of VCM recycling.
The main dehydrochlorination sources of VCM are the purification system
vents, the scrubber tower vent, and the loading facility vents. Losses
from the purification system are continuous; those from the loading system
are intermittent. The instantaneous rate of VCM from loading loss may be
1-2 times greater than that from the purification process, but the VCM
loss per pound of VCM produced is greater for purification. Emission
sources listed in Table 2-4 are sequenced to correspond with successive
steps in a generalized processing operation. Fugitive sources (pumps,
pump maintenance, valves, waste water, pressure relief valves, samplers,
etc.) are indicated separately, since their exact location throughout the
system will vary from plant to plant. The sources shown were chosen to
provide a checklist for tabulating and comparing data from several plants.
The practices for handling vent and waste streams—namely, the recycling,
the combining of vented gases into single streams, and the varying of
numbers of reactors and storage tanks—presented special problems for
tabulating available data.
The reliability of the data in the table depends on whether they have
been measured or calculated. The VCM concentrations, commonly measured
21
-------
Table 2-4. EMISSION RATES FOR VCM PROCESS 2: CHLORINATION-OXYCHLORINATION OF ETHYLENE
(WITH AIR) AND DEHYDROCHLORINATION
PROCESS EMISSION SOURCE3
PLANT A
Ib VCM/MM Ib % VCM
produced'' emissions
PLANT B
Ib VCK/MM Ib
produced''
Comment on
% VCM -Plant
emissions Operation
Production Sources Emitting VCM
EDC from Ethylene + Chlorine
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
EDC
14.
15.
VCM
16.
17.
18.
19.
Reactor condenser vent.C
Crude reactor vent.C
Crude prod prim wash vessel vent.C
Crude prod seedy wash vessel vent.C
Wash water, stripper, storage vent.C
Washed crude prod storage vent.C
Light-ends column condenser vent.C
Finishing column vent.C
Light-ends purification column vent, I
Refined EDC storage tank.C
Heavy-ends, tar-removal column vent.C
Heavy -ends storage tank, I
Tar storage tank, I
Total
from Ethylene + Hydrogen Chloride + Air
Reactor vent, I
Reactor efflu refrig cond vent.C
Total
from Dehydrochlorination of EDC
Light-ends distillation column vent, I
Heavy-ends distillation column vent, I
Heavy-ends, tar-removal column vent.C
Wastewater stripping column vent.C
Total
Combined total, vents 1,7,9,11,15,16
0(0)
0(0) c
9.6-19-2(1-2) 0.4-0.5
0(0)
9.6-58(1-6) 0.5-1.5
0(0)c
240-1500(25-120) 13-38
0(0)
0-12(0-1. 25)d 0-0.3
0(0)<=
480-960(50-100) 24.8-26.4
0(0)
0(0)
40-65.9
NA
154-230(16-24) 6-9
6-9
240-400(300-500) 10.3-13.2
14.1(2000) < 1
Part of vent 11
Part of vent 5
10.7-13.9
3768-7502(391-748)
0(0)
0(0)c
9.6-19.2(1-2)
0(0)
8.0-49(0.8-5)
0(0)c
240-1500(21-101)
0(0)
0-12(0-1. 25)d
0(0)c
480-960(42-84)
0(0)
0(0)
NA
140-205(14-22)
2880-4800(252-420)
13(1680)
Part of vent 11
Part of vent 5
3768-7502(328-626)
— —
—
< 1 Water wash used.
Normally not vented .
0.5-1.5
—
13-38
—
0-0.3
—
24.8-26.4
—
—
40-65.9
—
6-9
6-9
10.3-13.2 2 hr/day
< 1 3 min/mo
—
— —
10.7-13.9
—
-------
Table 2-4. EMISSION RATES FOR VCM PROCESS 2: CHLORINATION-OXYCHLORINATION OF ETHYLENE
(WITH AIR) AND DEHYDROCHLORINATION, Continued
tvj
10
PROCESS EMISSION SOURCE3
PLANT A
Ib VCM/MM Ib
produced**
PLANT B
7. VCM Ib VCM/MM Ib % VCM
emissions produced*3 emissions
Comment on
Plant
Operation
Production Sources Emitting VCM, Cont'd.
Scrubbing (vents 1,3,4.7,8,9,11,16,17)
20. Scrubber tank vent, C
21. Tank-car Loading
Purge, I
Loading arm, I
Tank-car slip gauge, I
Refrigerated vent, I
Total
Pumps, C
Pump maintenance, I
Valves, C
Relief valve, C
Cooling water, C
Samplers , C
Miscellaneous
Total
1540-2300(160-240) —
114(285)e
16(20)
48(120)
192(300)
370(440)
3.8(0.4)
240 (25)
10 (1.04)
7.7(0.8)
12 (1.25)
20.8(2.16)
10 (1.04)
304. 3(6. 69) r
9.6-20.
Fugitive
—
7-17
1540-2300(134-202)
96(239)
13(17)
40(100)
160(252)
3 309(367) 9.6-20.3
Sources Emitting VCM
3.8(0.4)
240 (25)
10 (1.04)
7.7(0.8)
10 (1.05)
17.5(1.82)
8.5(0.88)
297.5(5.99)r 7-17
2 hr/48 hr
2 hr/day
1 hr/day
16 hr/day
1 hr/mo
aC is continuous; I is intermittent. bValue in parentheses is Ib VCM/hr released. cVents 2,6, & 10 total 0-13 MSCFH.
^Every 2 days, the 8-hr regeneration totals 0-10 Ib. eTo remove excess oxygen before reloading. ^Total continuous;
excludes intermittent pump maintenance.
Note: Plant C, with production capacity of 2.14 MM Ib/day, has emission rate of 1368 Ib VCM/MM Ib produced
(or 122 Ib VCM/hr) for all sources except loading (830(1516); 4 hr/day, 100 day/yr) and the dock recovery
vent (1087(2738); 290 days/yr.).
-------
by gas chromatography, can be determined to 1 ppm_+ 50 percent on "grab-
bag" samples. The accuracy of the calculated data cannot be estimated
with the information now available. Estimates for process upsets were
not generally available. In one plant, pump maintenance was estimated as
one pump per month at a VCM loss of 25 pounds. A report of one annual
maintenance (during which all tanks, towers, and reactors were pumped out
and residues vented to the atmosphere) estimated that 10,000 pounds of
VCM had been lost into the environment in a 24-hour period; this procedure
is now being revised, and a solvent absorption of residual VCM might be
implemented.
Since VCM purification, storage, and transfer are carried out at high
pressures to eliminate the need for relatively high-cost refrigeration
systems, there exists a real potential for VCM losses from pressure leaks.
The quantity of material lost from smaller or larger leaks in valves,
pumps, and similar causes is variable from installation to installation
and is largely determined by the quality of plant construction and main-
tenance. At best, only an average figure with a wide range can be assigned
to this source of emission.
2.3.2 Controls and Costs
The control methods used depend on the level of emission and the pos-
sibility of recycling VCM to eliminate the emission source and to recover
lost VCM. The state of the art controls described in the Table 2-5 should
be applicable to any VCM production plants which have reported to this time,
since the oxychlorination process and the controls used in at least some
of these installations are basically similar. It is our belief that all
producers can reduce the emissions in their plants significantly below
24
-------
Table 2-5. CONTROLS AND COSTS FOR VCM PROCESS 2: CHLORINATION-OXYCHLORINATION
OF ETHYLENE (WITH AIR) AND DEHYDROCHLORINATION
Percent Control
Emission Source Control Emission Cost Comment on Control Type
Reduction Estimate
5. Wash Water, Stripper, Storage Vent: continuous stream; gas flow 300-2000 SCFH; 90°F
Carbon bed adsorber 100
Ozone oxidation 100
7. Light-Ends Column Condenser Vent: continuous stream; 8000-500,000 SCFH
Refrigerated condenser 50
Carbon bed adsorber 100
9. Light-Ends Purification Column Vent: intermittent stream; 21,000 Ib hot N-/8 hrs (10 Ib VCM)
Carbon bed adsorber 100 $50,000 Plus installation cost,
Ozone oxidation 100 purge, and recycle
15. Reactor Efflu Refrig Cond Vent: continuous stream; 13,000-15,000 SCFH; -13°F
Additional oxychlorinator 90-93 $800,000 Includes all but
Carbon bed, purge, recycle 100 $60,000 adsorbent cost
16. Light-Ends Distillation Column Vent: intermittent stream
Refrig cond and carb bed $30,000-
adsorber, or ozone oxid 100 $150,000
Incineration 95-98 $50,000 Cost depends on HC1 recovery.
17. Heavy-Ends Distillation Column Vent: intermittent stream
Incineration 93-95
20. Scrubber Tank Vent: continuous stream; 500,000-800,000 SCFH; 40-100°F
Refrigerated condenser 75-91 $30,000+
Incineration 93-95 Cost depends on HC1 recovery.
Refrigerated condenser and $30,000- Depends on installation type.
carb bed, purge, recycle 100 $150,000 Depends on installation type.
-------
Table 2-5. CONTROLS AND COSTS FOR VCM PROCESS 2: CHLORINATION-OXYCHLOR1NATION
OF ETHYLENE (WITH AIR) AND DEHYDROCHLORINATION (continued)
Emission Source Control
Percent
Emission
Reduction
Control
Cost
Estimate
Comment on Control Type
21. Tank-Car Loading: intermittent stream; variable flow; 50-90°F
Purge 50
Vapor collection adapter 90
Vapor collection, recycle
Loading arm
Incineration
Vapor collection adapter 90
Slip gauge
Thermal detector, recycle 95 $1,000
Magnetic gauge 100 $1,000
Refrigerated vent
Incineration, HC1 recovery 97
Generates HC1.
Cost per tank car loading line.
Cost per tank car.
Cost depends on HC1 quantity.
22. Dock-VCM Revocery Vent: intermittent stream; 2700 Ib/hr of loading
Refrigerated condenser
98
$50,000
25. Fugitive Sources: continuous and intermittent
Pumps
Pressurized mechanical seal 95
Pump maintenance
Valve and relief valve
Diaphragm valve 95
Double lock valve 50
Rupt disc under relief valve 50
Cooling water
Sampler
Vapor collectors 90
Miscellaneous 50
$1,000
$40,000
Cost for 1/2 man-year labor.
-------
50 ppm with state-of-the-art controls. However, thermal combustion of
VCM produces hydrogen chloride, which is usually absorbed in water with a
resulting production of hydrochloric acid; in operations without an outlet
for the disposal of hydrochloric acid, this control method may not be
applicable.
The need for development of longer term controls is not strongly
indicated, pending the evaluation of data for state-of-the-art techniques.
Scrubbing VCM process vents and washing equipment with EDC appears
appropriate. Postchlorination units have reached the commercial market,
and their value in reducing VCM emissions should be further investigated.
The costs of control methods depend on the magnitude of the emission
to be controlled and the level to which the emission is to be controlled.
These costs will vary from plant to plant, depending on the physical design
of the plant. Estimates supplied by the reporting organizations are given.
In only one instance have estimates been given by two manufacturers for
control of the same emission source by the same control device; in this
instance, the estimates were consistent within 7 percent.
2.4 CHLORINATION-OXYCHLORINATION OF ETHYLENE (WITH OXYGEN) AND DEHYDRO-
CHLORINATION
One plant is currently using oxygen instead of air in the balanced
hydrochlorination-oxychlorination process shown in Figures 2-5, 2-6, 2-7.
The annual capacity of this VCM source is 300 million pounds.
2.4.1 Emission Points
The emission sources for this process are identical to those described
in subsection 2.3.1. The volume of emissions are different because of the
substitutions of 100 percent oxygen for the air mixture (80% N2 + 20% 02>.
27
-------
CHLORINE
ETHYLEKE
EOC RECYCLE
SNIFF CHLORINE
WATER FROM OXYCHLORINATION AND VGM
E
WASTE WATER
TDREACTCM OR SALES STORAGE
®j®
LIGHT ENDS
WASH «>"*"
WATER
STRIPPER
«—STEAM
TO EOC RECOVERY SYSTEM
^>
13
HEAVY ENDS TAR
REMOVAL COLUM
^-1
TAR STORAGE
^DISPOSAL
HEAVY ENDS
FIGURE 2-5 ETHYLENE OICHLORIDE FROM ETHYLENE AND CHLORINE
-------
ETHYLENE
OXYGEN >
HYDROGEN
CHLORIDE
ro
10
WATER-
STEAM
OXYGEN **-
HOT WATER »>
WASH WATER
TO WASHWATE'R
STRIPPING COLUMN
TO
STRIPPING COLUMN
TO WASH CRUDE STORAGE
*
G
REACTOR
FIGURE 2-& ETHYLENE OICHU9RIDE FROM ETHYLENE -I- HCL+ OXYGEN
-------
RECYCLED TO
OXYCHLORWATION
CONDENSER
ocb
COMPRESSOR
QUENCH
COLUMN
CRACKING FURNACE
EVAPORATOR
...t
*— -.
PARTIAL CONDENSER
©
I
EOC QUENCH
STREAM
REBOILER
WA!
WATER
I
I
VENT
PHASE
SEPARATOR
LIGHT ENDS
COLUMN
[ T ^
<—I—»
REBOILER
COOLING
WATER
REFLUX
-*©
HEAVY ENDS
COLUMN
©-
LOADING
PURGE
LOADING ARM
STORAGE SLIP
REFRKS VENT
TO EOC RECYCLE:
COOLER^
STEAM
COOLING WATER
T
CFI
©
SCRUBBER
EDCFEED
FIGURE 2-7 DEHYDROCHLORINATION OF ETHYLENE OICHLORIDE
-------
The emission points designated in Table 2-6 are characterized by temper-
ature, flow rate (continuous or intermittent), and percentage of VCM
wherever data permit. The reliability of the data are provided at two
levels: emissions quantities measured by standard methods to 1 ppm +0.5 ppm,
and calculated and estimated values varying as much, as 100 percent.
2.4.2 Controls and Costs
Based on the similarity of this process to that using air, the level of
VCM emissions should be reduced with state of the art controls to 90-95 percent
of the present levels, to significantly below 50 ppm and within the 5-50 ppm
range. The state of the art controls (Table 2-7) are applicable to all plants
using the balanced oxychlorination process, with either oxygen or air as
the oxidant.
The use of oxygen instead of air is itself a long-term control, in
the sense that it eliminates the effect of nitrogen on emissions. Nitrogen
from air must be vented, under conditions that result in its being a carrier
for vinyl chloride vapors.
The costs of reducing an emission to a level significantly below
50 ppm will vary from plant to plant, depending on the makeup of the process,
cost of installation, and level of the emission. In general, considering
cost of equipment and instrumentation, the cost of a 99 percent reduction
of a principal emission source will range $50,000 to $100,000.
2.5 DIBECT CHLORINATION OF ETHXLENE, AND DEHXDROCHLORINATION
Direct chlorination and dehydrochlorination (Figures 2-8, 2-9) are
used to produce VCM when the manufacturing plant has other uses for the
hydrogen chloride byproduct. Three plants employ this approach to make
approximately 700 million pounds of VCM a year.
2.5.1 Emission Points
Table 2-8 identifies the emission sources for both the production of
31
-------
Table 2-6. EMISSION RATES FOR VCM PROCESS 3: HYDROCHLORINATION-OXYCHLORINATION OF
ETHYLENE (WITH OXYGEN) AND DEHYDROCHLORINATION
PROCESS EMISSION SOURCE'
,a
PLANT A
Ib VCM/MM Ib
produced
% VCM
emissions
Comment on
Plant
Operation
Production Sources Emitting VCM
N>
EDC from Ethylene + Chlorine
1. Reactor condenser vent, C
2. Crude reactor vent, C
3. Crude prod prim wash vessel vent, C
4. Crude prod seedy wash vessel vent, C
5. Wash water, stripper, storage vent, C
6. Washed crude prod storage vent, C
7. Light-ends column condenser vent, C
8. Finishing column vent, C
9. Light-ends purification column vent, I
10. Refined EDC storage tank, C
11. Heavy-ends, tar-removal column vent, C
12. Heavy-ends storage tank, I
13. Tar storage tank, I
Total
Oxygen
EDC from Ethylene + Hydrogen Chloride +
14. Reactor vent, I
15. Reactor efflu refrig cond vent, C
Total
VCM from Dehydrochlorination of EDC
16. Light-ends distillation column vent, I
17. Heavy-ends distillation column vent, I
18. Heavy-ends, tar-removal column vent, C
19. Wastewater stripping column vent, C
Total
Combined total, vents 1,7,9,11,15,16
-------
Table 2-6. EMISSION RATES FOR VCM PROCESS 3: HYDROCHLORINATION-OXYCHLORINATION OF
ETHYLENE (WITH OXYGEN) AND DEHYDROCHLORINATION (continued)
W
OJ
PROCESS EMISSION SOURCEa
PLANT A
Ib VCM/MM Ib
produced
% VCM
emissions
Production Sources Emitting VCM,
Scrubbing (vents 1,7,9,11,15,16)
20. Scrubber tank vent, C
21. Tank-car Loading
Purge, I
Loading arm, I
Tank-car slip gauge, I
Refrigerated vent, I
Total
Pumps , C
Pump maintenance, I
Valves, C
Relief valve, C
Cooling water, C
Samplers, C
Miscellaneous
Total
114(285)
16 ( 20)
48(120)
192(300)
370(440)
3.8( 0.4)
240(' 25)
10(1.04)
7.7( 0.8)
12(1.25)
20.8(2.16)
10(1.04)
304.3(6.69)
—
9.6-20.3
Fugitive Sources Emitting
—
7-17
Comment on
Plant
Operation
cont ' d .
2 hr/48 hr
2 hr/day
1 hr/ day
16 hr/day
VCM
C is continuous, I is intermittent.
Data from Table 2-4 are shown here; they are believed to be transferable to this plant.
-------
Table 2-7. CONTROLS AND COSTS FOR VCM PROCESS 3: CHLORINA1 J.ON-OXYCHLORINATION
OF ETHYLENE (WITH OXYGEN) AND DEHYDROCHLORINATION
Percent Control
Emission Source Control Emission Cost Comment on Control Type
Reduction Estimate
5. Wash Water, Stripper, Storage Vent: continuous stream
Carbon bed adsorber
Ozone oxidation
7. Light-Ends Column Condenser Vent: continuous stream
Refrigerated condenser
Carbon bed adsorber
9. Light-Ends Purification Column Vent: intermittent stream
Carbon bed adsorber
Ozone oxidation
u>
15. Reactor Efflu Refrig Cond Vent: continuous stream
Additional oxychlorinator
Carbon bed, purge, recycle
16. Light-Ends Distillation Column Vent: intermittent stream
Refrig cond and carb bed
adsorber, or ozone oxid
Incineration
17. Heavy-Ends Distillation Column Vent: intermittent stream
Incineration
20. Scrubber Tank Vent: continuous stream
Refrigerated condenser
Incineration
Refrigerated condenser and
carb bed, purge, recycle
-------
Table 2-7. CONTROLS AND COSTS FOR VCM PROCESS 3: CHLORINATION-OXYCHLORINATION
OF ETHYLENE (WITH OXYGEN AND DEHYDROCHLORINATION (continued)
Emission Source Control
Percent
Emission
Reduction
Control
Cost
Estimate
Comment on Control Type
21. Tank-Car Loading: intermittent stream; variable flow
Purge
Vapor collection adapter 90'
Vapor collection, recycle
Loading arm
Incineration 90
Vapor collection adapter
Slip gauge
Thermal detector, recycle 95
Magnetic gauge 100
Refrigerated vent
Incineration, HC1 recovery 97
$1000
$1000
Generates HC1
Cost per tank car loading line
Cost per tank car
Cost depends on HC1 quantity
22. Dock-VCM Recovery Vent: intermittent stream
Refrigerated condenser
98
$50,000
25. Fugitive Sources: continuous and intermittent
Pumps
Pressurized mechanical seal 95
Pump maintenance
Valve and relief valve
Diaphragm valve 95
Double lock valve 50
Rupt disc under safety valve 50
Cooling water
Sampler
Vapor collector 90
Miscellaneous 50
$1000
$40,000
Cost for 1/2 - man year labor
)ata from Process 1
-------
EuC
FROM DEHYDROCHLO-
RINATION /y-
'•©
CHLORINE
ETHYLENE
EOC RECYCLE
SNIFF CHLORINE
TO REACTOR OR SALES STORAGE
HEAVY ENDS TAR
REMOVAL COLUMI
HEAVY ENDS
STORAGE
,—» TO EOC RECOVERY SYSTEM
—>
13
TAR STORAGE
^DISPOSAL
FIGURE 2-8 ETHYLENE OICHLORIDE FROM ETHYLENE AND CHLORINE
-------
EVAPORATOR
t
C F|
©
EDC FEED
RECYCLED TO
HCL RECOVERY
CONDENSER
oo
COMPRESSOR - '
QUENCH
COLUMN
CRACKING FURNACE
.—'
•—•x
.—-'
".": \
•'—, \
CV
^™*
PARTIAL CONDENSER
©
EDC QUENCH
STREAM
REBOILCR
TO EDC RECYCLE
STEAM
^
COOLER
VENT
16
»STE
I
1
VENT
WA!
WATER
PHASE
SEPARATOR
LIGHT ENDS
COLUMN
REBOILER
COOLING
WATER
REFLUX
•©
HEAVY ENDS
COLUMN
STORAGE
LOADING
PURGE
LOADING ARM
SLIP GAGE
REFRIG VENT
COOLING WATER
FIGURE 2-9 DEHYOROCHLORINATION OF ETHYLENE DICHLORIDE
-------
Table 2-8. EMISSION RATES FOR VCM PROCESS 4: DIRECT CHLORINATION OF ETHYLENE, AND DEHYDROCHLORINATION
PLANT A
PLANT B
PROCESS EMISSION SOURCE
Ib VCM/MM Ib
produced**
% VCM
emissions
Ib VCM/MM Ib
produced"
% VCM
emissions
Comment on
Plant Operation
Production Sources Emitting VCM
OJ
oo
EDC from Ethylene+ Chlorine
1. Reactor condenser vent, C
2. Crude reactor vent, C
3. Crude prod prim wash vessel vent, C
4. Crude prod seedy wash vessel vent, C
5. Wash water, stripper, storage vent, C
6. Washed crude prod storage vent, C
7. Light-ends column condenser vent, C
8. Finishing column vent, C
9. Ligh-ends purification column vent, I
10. Refined EDC storage tank, C
11. Heavy-ends, tar-removal column vent, C
12. Heavy-ends storage tank, I
13. Tar storage tank, I
Total
VCM from Dehydrochlorination of EDC
16. Light-ends distillation column vent, I
17. Heavy-ends distillation column vent, I
18. Heavy-ends, tar-removal column vent, C
19. Wastewater stripping column vent, C
Total
Part of vent 11
Part of vent 3
Part of vent 11
Part of vent 3
-------
Table 2-8. EMISSION RATES FOR VCM PROCESS 4: DIRECT CHLORINATION OF ETHYLENE, AND DEHYDROCHLORINATION (continued)
VO
'PROCESS EMISSION SOURCE3
21. Tank-car loading
Purge, I
Loading arm, I
Tank-car slip gauge, I
Refrigerated vent, I
Total
PLANT A
Ib VCM/MM Ib
produced
114(285)
16(20)
48(120
192(300)
370(440)
PLANT B
% VCM Ib VCM/MM Ib
Demissions produced
Production Sources Emitting VCM
% VCM Comment on
emissions Plant Operation
,' cont'd
2 hr/48 hr
2 hr/day
1 hr/day
16 hr/day
4 hr/day, 100 day/yr
Pumps, C
Pump maintenance, I
Relief valve, C
Cooling water, C
Sampler, C
Miscellaneous
Total
Fugitive Sources Emitting VCM
(52)
(16)
(3)
(6)
C is continuous, I is intermittent; sources 14, 15, and 20 do not exist for this process.
JValue in parentheses is Ib VCM/hr released.
-------
EDC by direct chlorination and the dehydrochlorination of EDC to VCM. The
emission points are characterized by flow rate, temperature, total emissions,
and percentage of VCM to the extent data permit. Where data are available,
two levels of reliability can be indicated: "measured" data are considered
reliable and "estimated" data (based on material balances) are considered
to have an error of at least 100 percent. This information is to be given
in Table 2-8. (Note: no data are available at this time).
Those producers making EDC by direct chlorination only are reported
to have < 0.1 of the emissions associated with the use of oxychlorination.
They do however need to recycle EDC from their dehydrochlorination process
to the direct chlorination system for recovery. This causes VCM emissions
from the crude EDC-refining system.
2.5.2 Controls and Costs
Neither the reactor vent nor the crude reactor would be expected to
contain VCM, since the crude product is subjected either to wash purifi-
cation or to anhydrous purification; however, no data are available on
the control of these sources. It is at this point that EDC recycled from
the dehydrochlorination process would likely be introduced; again, data
on emission levels and controls are not available at this time.
While state of the art controls appear to be adaptable to most emission
sources, certain controls (especially adsorption) must be adapted with
careful consideration of their relation to the rest of the plant. Thus
one plant proposes to use adsorption instead of the present refrigerated
condenser on the vinyl chloride light-ends column. They find it appropriate
to desorb with ethane and to compress the stripped gases for recycle.
Controls and cost information are tabulated in Table 2-9, both to
the extent available.
40
-------
Table 2-9. CONTROLS AND COSTS FOR VCM PROCESS 4: DIRECT CHLORINATION OF ETHYLENE, AND DEHYDROCHLORINATION
Percent Control
Emission Source Control Emission Cost fl Comment on Control Type
Reduction Estimate
5. Wash Water, Stripper, Storage Vent: continuous stream
Carbon bed adsorber
Ozone oxidation
9. Light-Ends Purification Column Vent: intermittent stream
Carbon bed adsorber
Ozone oxidation
*- 16. Light-Ends Distillation Column Vent: intermittent stream
M _
Refrigerated condenser and
carbon bed adsorber
Incineration
17. Heavy-Ends Distillation Column Vent: intermittent stream
Vented to gas holder, recycle
21. Tank-Car Loading: intermittent stream
Purge
Vapor collection adapter 90a
Vapor collection, recycle
Loading arm
Incineration Generates HC1
Vapor collection adapter 90
Data are from Process 1.
-------
Table 2-9. CONTROLS AND COSTS FOR VCM PROCESS 4: DIRECT CHLORINATION OF ETHYLENE, AND DEHYDROCHLORINATION (continued)
Emission Source Control
Percent
Emission
Reduction
Control
Cost
Estimate
Comment on Control Type
21. Tank-Car Loading: intermittent stream
Slip gauge
Thermal detector, recycle 95
Magnetic gauge 100
Refrigerated vent
Incineration, HCL recovery 97
$1000
$1000
Cost per tank car loading line
Cost per tank car
Cost depends on HC1 quantity
22. Dock-VCM Recovery Vent: intermittent
10
Refrigerated condenser
98
$50,000
25. Fugitive Sources: continuous and intermittent
Pumps
Pressurized mechanical seal 95
Pump maintenance
Valve and relief valve
Diaphragm valve 95
Double lock valve 50
Rupt disc under safety valve 50
Cooling water
Sampler
Vapor collector 90
Miscellaneous 50
$1000
$40,000
Cost for 1/2 man-year labor
-------
3. POLYVINYLCHLORIDE
3.1 OVERVIEW OF PROCESSES
A typical PVC plant includes the following operations:
1. Receiving and storage of VCM and catalysts.
2. Polymerization of VCM: measuring and charging, and reaction.
3. Stripping and recovery: reactor blowdown and recovery, and slurry
handling and storage.
4. Centrifugation or filtration.
5. Drying.
6. Pneumatic conveying and storage.
7. Packaging and shipping.
8. Blending-
9. Waste treatment.
Compounding, which may be done in a PVC resin plant or in separate facilities,
is not included in this study.
Total PVC output for 1972-73 is attributed to four process types:
suspension polymerization, 78 percent; emulsion polymerization, 12 percent;
bulk polymerization, 6 percent; and solution polymerization, 4 percent.
All polymerizations in the United States are batch operations, which probably
contribute strongly to the severity of the VCM emissions problem.
3.2 SUSPENSION POLYMERIZATION
The 34 suspension plants now in operation are producing 75-85 percent
of the PVC produced.
3.2.1 Emission Points
Suspension polymerization (Figure 3-1) uses a water media, a suspension
agent such as polyvinyl alcohol, and an oil-soluble catalyst. A typical
recipe is 100 parts VCM, 200 of water, 0.1-0.2 of catalyst, and 0.005-0.0 of
suspension agent. The VCM-water ratio can be varied over a wide range; the
limiting factor is sufficient fluidity of the final PVC-water slurry, for
adequate dissipation of the heat evolved during polymerization. Oil-soluble
43
-------
WET AIR
»| MONOMER
STORAGE
TANK
WATER
SUSPENSION AGENT
AND OIL SOLUBLE
CATALYST
FIGURE-3.1
SIMPLIFIED FIDW DIAGRAM PVC SUSPENSION PROCESS
-------
catalysts such as dibenzoyl peroxide, acetyl benzoyl peroxide, or dibutyl
peroxide are preferred to the water soluble persulfate catalyst. Persulfate
is slower, and it produces particles too fine to be handled in conventional
filtering and drying equipment.
Commercial suspension resins show little or no retention on a 100-mesh
(U.S. standard) screen and from 30 to 80 percent retention on a 200-mesh
screen. Such a particle size is achieved by rapid agitation in glass-lined
autoclaves at 35 to 45°C. Vigorous agitation suspends VCM as fine droplets
in water, controls size, and gives a granular polymer for better filtering
and drying.
During polymerization, alkaline buffers such as sodium carbonate, bi-
carbonate, and phosphate have been used to maintain the pH at 5-8 to prevent
formation of hydrochloric acid. Use of an inert substance such as nitrogen
increases reaction rate, reduces the amount of hydrochloric acid evolved,
and thereby increases the stability of the polymer.
There is little published information on molecular weight distribution
and chain branching of vinylchloride polymers. However, chloride polymers
in general have narrow molecular distributions, especially compared to
those of polystyrene.
The major advantage of suspension polymerization is the excellent
heat transfer rate attainable in conventional equipment. The major dis-
advantage is that water must be separated from the polymer, usually by
centrifugation and drying.
Table 3-1 indicates 34 suspension plants producing 3.5 billion pounds
a year. In Table 3-2 the VCM emissions rates are given for the sources shown
in Figure 3-1; the table is restricted to those plants for which data
have been supplied.
45
-------
Table 3-1. VCM EMISSIONS DATA FROM PVC PLANTS
Ol
Company Name and Location
Air Products and Chemicals, Inc
Plastics Division
American Chemical Corp
Borden, Incorporated
Borden Chemical Division
Continental Oil Company
Conoco Plastics Division
Diamond Shamrock Corp
Diamond Shamrock Chem Co (Subsid Plas Div)
Ethyl Corp (Industrial Chemical Div)
The Firestone Tire and Rubber Company
Chemical Plastics Division
The General Tire & Rubber Co (Chem/Plas Div)
The B.F. Goodrich Company
B.F. Goodrich Chemical Company
B.F. Goodrich Chemical Company
B.F. Goodrich Chemical Company
B.F. Goodrich Chemical Company
The Goodyear Tire & Rubber Company
Chemical Division
Great American Chemical Corp
Keysor-Century Corporation
Monsanto Company (Poly and Petro-Chem Co)
National Starch and Chemical Corp
Occidental Petroleum Corp
Hooker Chemical Corp (Subsid Ruco Div)
City PVC
Population Capacity
(1970 census) (MM Ib/yr)
Calvert City, Ky
Pensacola, Fla
Long Beach, Calif
Illiopolis, 111
Leominster, Mass
Aberdeen, Miss
Oklahoma City, Okla
Delaware City, Del
Deer Park, Tex
Baton Rouge, La
Perryville, Md
Potts town, Pa
Ashtabula, Ohio
Long Beach, Calif
Henry, 111
Louisville, Ky
Avon Lake, Ohio
Pedricktown, NJ
Plaquemine, La
Niagara Falls, NY
Fitchburg, Mass
Saugus , Calif
Springfield, Mass
Meredosia, 111
Burlington, NJ
Hicksville, NY
31,627C
59,507
358,633e
1,122 \
32,939 /
6,157
366,481
2,024 I
12,7738J
165,963h
2,091
25,355
24,313
358,633e
2.6101
361,472
12,261
7,739J
85,615
43,343
e
163,905
1,178
ll,991k
48,075
120
50
125
285
220
80
250
180
130
140
100
125
125
275
125
130
100
100
40
35
150
10
180
10
Type
VCM of
Q
Emissions Proc-
(MM Ib/yr) essb
4.8
2.0
5.0
11.4
8.8
3.2
10.0
7.2
5.2
5.6
4.0
5.0
5.0
11.0
5.0
5.2
4.0
4.0
1.6
1.4
6.0
0.4
7.2
0.4
S
S
S
S&E
S&E
S&E
S&E
S&E
S&E&SL
(10)
S&E&
B(40)
SL(20)
S&E&
B(40)
S
S
S&E
E
S&B
(160)
-------
Table 3-1. VCM EMISSIONS DATA FROM PVC PLANTS, Continued
Company Name and Location „ , ^.
r J Population
(1970 census)
Olin Corp (Thompson Plastics Company Div)
Pantasoto Company
Pantasoto Company
Robintech, Inc
Stauffer Chemical Company (Plastics Div)
Tenneco Chemicals, Inc
Tenneco Plastics Division
Tenneco Plastics Division
Union Carbide Corp
Chemicals and Plastics Division
Uniroyal Chemicals, Inc.
Subtotal (plants in operation)
Certainteed Corporation
Georgia Pacific
Shintech
Subtotal (plants under construction)
Total
Assonet, Mass
Passaic, NJ
Point Pleasant, WVA
Painesville, OHio
Delaware City, Del
Burlington, NJ
Fleming ton, NJ
Pasadena, Tex
Texas City, Tex
So Charleston, WVA
Painesville, Ohio
Lake Charles, La
Plaquemine, La
Oyster Creek, Tex
55,124 \
6,122 /
16,536
2,024
ll,991k
3,917
89,2778
38,908
16,333
16,536
77,978
7,739,
11,997
Type
PVC VCM of
Capacity Emissions3 Proc-
(MM Ib/yr) (MM Ib/yr) essb
150
120
250
160
165
60
300
200
120
140
4,750
200
200
250
650
5,400
6.0
4.8
10.0
6.4
6.6
2.4
12.0
8.0
4.8
5.6
190.0
8.0
8.0
10.0
26.0
216.0
S
E
S
S
S&E
S&E
S&E
S&E
(120)
S&E
U
U
U
Extrapolations based on est'd 4.0% VCM emission from PVC production (industry total 224 Ib/yr) and recovery.
B bulk, E emulsion, SL solution, S suspension, U unknown; value in parentheses, est'd process production(MM Ib
Population of Paducah, Ky. Joint venture: Atlantic Richfield & Stauffer Chem. e Los Angeles County - 7,032,0
Oklahoma County - 526,805. 8 Harris County - 1,741,912. h Baton Rouge Parrish (county) - 302,031.
Henry County - 53,217. ^ Plaquemine Parrish (county) - 25,225. k Burlington County - 323,132.
-------
Table 3-2. EMISSION RATES FOR PVC SUSPENSION PROCESS
oo
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
ac
PROCESS EMISSION SOURCE3
VCM tank vent, C
Reactor emergency vent, I
VCM recovery still vent, C
Blend or surge tank vent, C
Wastewater, VCM recovery, C
Centrifuge vent, C
Dryer vent, C
Air conveyor vent, C
Storage bin vent, C
VCM loading, storage vents, I
VCM charging, I
Folymerizer cleaning, I
Packing/hopper car, I
System cleaning, I
Fugitive losses, C&I
Abnormal losses, I
Total
is continuous, I is intermittent.
PLANT A
PLANT B
Ib VCM/MM Ib % VCM Ib VCM/MM Ib % VCM
produced^ emissions produced'5 emissions
Part of vent 15
Part of vent 15
Part of vent 15
4280(48.2)
Part of vent 9
Part of vent 9
Part of vent 9
13,498(152.0)
Part of vent 15
Part of vent 15
3019(34.0)
1243(14.0)
Part of vent 15
17,467(196.7)
39,507(444.9)
or 3.5% VCM in PVC
bValue
Part of vent 15
Part of vent 15
Part of vent 15
11.0 3447(34.5)
Part of vent 9
Part of vent 9
Part of vent 9
35.0 7872(78.8)
Part of vent 15
Part of vent 15
7.8 3447(34.5)
3.2 300(3.0)
Part of vent 15
40-46 18,691(187.1)
33,757(337.9)
or 3.0% VCM in PVC
in parentheses is Ib VCM/hr.
11.5
23
10-12
1
50-58
Note: Plant C with a capacity of 175 MM Ib/yr emitted 7079 Ib VCM/MM lb(155 Ib VCM/hr) from vent 1
and 2283 Ib VCM/MM lb(50 Ib VCM/hr) from vent 4; emissions totaled 24.2% for vent 1 and 7.8% for
vent 4.
-------
3.2.2 Controls and Costs
Emissions may be reduced by process modifications, but these possi-
bilities are excluded from the summary table—not because modifications
are less important but because they can have impacts on product quality
and thus need more detailed evaluation to justify inclusion. Of all the
process modifications, vacuum stripping after polymerization is one of
the most important. At least two companies are practicing this method in
emulsion FVC polymerization, just as in suspension polymerization. This
should be considered a state of the art control. The cost for retro-
fitting vacuum stripping is guestimated at $250,000-$500,000 per unit.
Emissions may be combined; the combined airstrearns may he treated at
less cost than individual ones. Control methods may be combined and/or
used in sequence; for example, ozonation is expensive, but may be efficient
as a final step to destroy VCM. All of these should be considered in the
developmental stage and in need of evaluation.
Table 3-3 gives an assessment (qualitative) of state of the art control
techniques. Economics and control levels are included to the extent
they are available.
3.3 EMULSION POLYMERIZATION
The 11 emulsion plants operating at present contribute 12 percent of
the PVC being produced.
3.3.1 Emission Points
The emulsion process (Figure 3-2) is similar to the suspension process
(Figure 3-1). In emulsion, the particle size is smaller. Emulsifiers and
additives maintain a nonsettling emulsion during polymerization. . At least
four components are needed—water, VCM, water soluble initiators, and an
-------
Table 3-3. CONTROLS AND COSTS FOR PVC SUSPENSION PROCESS
Emission Source Control
Percent
Emission
Reduction
Control
Cost .
Estimate
Comment on Control Type
3. VCM Recycle Column Vent: continuous; 0.017-0.019 Ib VCM/lb prod°'d; 50-80°F; 80-150 psig; variable flow
Vacuum stripping
Carbon bed
Incineration
Absorption
Refrigeration
50-80
50-99
50-99
50-90
40-60
c d
4. Blend or Surge Tank Vent: continuous stream; 0.0043-0.0035 Ib VCM/lb prod ' ; 80-130°F, 0-10 psig; variable fl
Vacuum stripping
Recycle to compressor
Carbon bed
Ozone oxidation
Incineration
Absorption
Refrigeration
50-80
40-60
50-99
50-99
50-99
50-80
40-60
As a process change, may affect product quality.
5. Wastewater, VCM Recovery: continuous; NA ; variable flow
Stream stripping
Carbon bed
Ozone oxidation
50-90
50-99
50-99
6. Centrifuge Vent: continuous stream; NA ; 80-130°F; 0-5 inches of water, variable flow
Vacuum stripping
Recycle to compressor
Carbon bed
Ozone oxidation
Incineration
Absorption
Refrigeration
50-80
40-60
50-99
50-99
50-99
50-80
40-60
May affect product quality; surface filters may cause VCM emlss
-------
Table 3-3. CONTROLS AND COSTS FOR PVC SUSPENSION PROCESS (continued)
Percent Control
Emission Source Control Emission Cost . Comment on Control Type
Reduction Estimate
7. Dryer Vent: continuous stream; NA ; 100-150°F, 0-2 inches of water; flow—10,000-30,000 SCFM/dryer
Vacuum stripping 50-80
Higher air recycle 40-80
Carbon bed 50-99 Applied after water removal
Ozone oxidation 50-99
Absorption 50-90
8. Air Conveyor Vent: continuous; NA ; 70-120°F, 0-2 inches of water; variable flow
Vacuum stripping 50-80 —
Recycle 50-99
9. Storage Bin Vent: continuous; NA; 70-120°F; 0-2 inches of water; variable flow
Q
Vacuum stripping 50-80 As a process change, may affect product quality.
Recycle 50-95
Incineration 50-95
10. VCM Loading, Storage Vents: intermittent; NA ; 60-80°F, 20-50 psig; variable flow
Recycle 50-80 —
Absorption 50-90
11. VCM Charging: intermittent; NAd; 60-80°F, 25-50 psig; variable flow
Recycle, padded transfer ^50-90
Absorption >50-90
-------
Table 3-3. CONTROLS AND COSTS FOR PVC SUSPENSION PROCESS (continued)
Percent Control
Emission Source Control Emission Cost , Comment on Control Type
Reduction Estimate
12. Polymerizer Cleaning: intermittent; NA ; variable flow
High pressure cleaning systems coming into use.
13. Packing/Hopper Car: intermittent; NA; ambient temperature; 0 psig; variable flow
Vacuum stripping 50-80 As a process change, may affect quality.
Wood and recycle 50-60
14. System Cleaning: intermittent; NA; variable flow
Evacuate vessels 50-60 Evacuated before cleaning.
g
15. Fugitive Losses6: variable; highly variable; >^ 60% of total emission losses; low data reliability
16. Abnormal Losses : emergency or abnormal, NA; highly variable CO - 150 Ib/hr)
A i_
All reductions are with cost trade-offs. NA is not available.
Calculated VCM-production values (good data NA) include fugitive losses. Substantial upset losses possible.
e f
Range for VCM reduction after the reactor. Some upset losses possible.
Halves packing glands; pump, agitator, and compressor seals; safety and blowout valves;
screwed and welded piping leaks; flange and instrument connections.
Relief valve losses from VCM storage tanks, polymerizers, blowdown tanks, VCM-recovery compressor discharges crude-
VCM storage tanks, and VCM distillation system; operational error; tank or vessel rupture.
-------
Ul
U)
UNLOADING
LOSS
MONOMER TANK REACTOR
VENT EMERGENCY
VCM
WATER
INITIATOR •
EMULSIFIER
, OVERSIZE
/STORAGE VENT
OVERSIZE
MATERIAL RECEIVER
AIR
BAGGING
BGURE 3.2 SIMPLIFIED FLOW DIAGRAM BATCH EMULSION POLYMERIZATION OF VINYL CHLORIDE
-------
emulsifying agent. Water is the continuous phase; VCM is the discontinuous
phase. Low concentrations of emulsifiers (either anionic or cationic)
behave essentially as electrolytes because they are uniformly distributed
between the water-VCM phases. When the concentration is increased, the
surface tension between the phases 'decreases and the conductivity of the
mixture increases. Above a critical concentration, the surface tension
and conductivity change less rapidly and the emulsifier begins to agglomerate
into groups of 20-30 molecules called micelles. In the water phase, the
initiator forms a free radical which migrates to the micelles, combines
with the VCM molecules, and begins the polymer chain. That is, a polymer
particle starts to form, the emulsifier collects at the surface, and the
VCM molecules diffuse (from the dispersed VCM droplets) through the water
phase and through the emulsifier to the growing polymer chain.
As polymerization of VCM progresses and as the other polymer particles
grow, more emulsifier is needed at the particle surfaces. In some cases
when 1-2 percent conversion is reached, the emulsifier micelles disappear
and all of the emulsifier is concentrated at the particle surfaces; at
higher conversions, perhaps 60 percent, all of the VCM is in the PVC
phase. The number of FVC particles formed (i.e., their size) seems to be
controlled in the early stages of polymerization. Termination steps
involving coupling or disproportionation are rare because the number of
PVC chains for each particle is small. Hence, PVC produced by the emulsion
process tends to have high molecular weights. In principle, the process
can be continuous; however, only the batch operation is practiced in the
United States.
54
-------
Table 3-1 lists the 11 emulsion plants operating at present. Total
production figures are not yet available. In Table 3-4, the VMC emissions
are given as pounds of VCM per million pounds of PVC produced and as total
pounds of VCM released per hour. The VCM emissions sources and rates are
restricted to those plants for which data have been received. The data
show that VCM emissions are about 4 percent of the VCM feed; this emissions
level is similar to that observed for suspension polymerization. However,
data from additional plants are needed to draw a firm conclusion regarding
the range and reliability of the emission levels.
3.3.2 Controls and Costs
Control techniques available, for each emission source will-be categorized
and rated in summary Table 3-5 when data becomes available.
Emissions may be reduced by process modifications, but these possi-
bilities are excluded from the summary table—not because modifications
are less important but because they can have impacts on product quality
and thus need more detailed evaluation to Justify inclusion. Of all the
process modifications, vacuum stripping after polymerization is one of
the most important. At least two companies are practicing this method in
emulsion PVC polymerization, just as in suspension polymerization. This
should be considered a state of the art control. The cost for retrofitting
vacuum stripping is guestimated at $250,000-$500,000 per unit.
Emissions may be combined; the combined airstreams may be treated at
less cost than individual ones. Control methods may be combined and/or
used in sequence; for example, ozonation is expensive, but may be efficient
as a final step to destroy VCM. All of these should be considered in the
developmental stage and in need of evaluation.
55
-------
Table 3-r4. EMISSION RATES FOR PVC EMULSION PROCESS
in
1.
2.
3.
4.
5.
^6.
7.
8.
9.
10.
11.
12.
13.
14.
PROCESS EMISSION SOURCE3
Unloading losses, I
VCM feed tank vent, I
Reactor emergency vent, I
Slurry hold tank vent, C
Dryer vent, C
Air conveyor vent, C
Storage vent, C
Resin-screening vent, C
VCM recovery vent, C
Polymerizer cleaning, I
Packing/hopper car, I
System cleaning, I
Fugitive losses, C&I
Abnormal losses, I
Total
PLANT A
Ib VCM/MM Ib % VCM
produced^ emissions
Part of vent 9
Part of vent 9
219(1.0) 0.6
10,293(490) 25.6
26,958(1231) 76.0
2738(12.5) 6.8
40,208(183.6)
PLANT B
Ib VCM/MM Ib
produced**
Part of vent 9
Part of vent 9
3026(11.4)
35,570(134.0)
796(3.0)
2920(11.0)
42,312(159.4)
% VCM
emissions
7.2
84.0
1.9
6.9
C is continuous, I is intermittent.
'Value in parentheses is Ib VCM/hr.
-------
Table 3-5. CONTROLS AND COSTS FOR PVC EMULSION PROCESS
Percent Control
Emission Source Control Emission Cost
Reduction Estimate
Comment on Control Type
1. Unloading Losses: intermittent stream
2. VCM Feed Tank Vent: intermittent stream
3. Reactor Emergency Vent: intermittent stream
Ul
4. Slurry Hold Tank Vent: continuous stream
5. Dryer Vent: continuous stream
6. Air Conveyor Vent: continuous stream
7. Storage Vent: continuous stream
-------
Table 3-5. CONTROLS AND COSTS FOR PVC EMULSION PROCESS (continued)
Cn
00
Percent Control
Emission Source Control Emission Cost Comment on Control Type
Reduction Estimate
8. Resin-Screening Vent: continuous stream
9. VCM Recovery Vent: continuous stream
10. Polymerizer Cleaning: intermittent stream
11. Packing/Hopper Car: intermittent stream
12. System Cleaning: intermittent stream
13. Fugitive Losses: continuous and intermittent
14. Abnormal Losses: intermittent
-------
3.4 BULK POLYMERIZATION
The 3 bulk polymerization plants are producing 300 million pounds
of PVC a year.
3.4.1 Emission Points
In bulk polymerization (Figure 3-3), only VCM and small amounts of
additives are charged to the reactors; inert diluents such as water or
solvents are not used. The Pechiney-St. Gobain bulk polymerization process
consists of two stages. In the prepolymerizer, the additives are slurried
in monomer, which is then about 10 percent converted to insoluble polymer.
In the autoclave, VCM is polyerized in a "paste" phase to obtain 90 percent
conversion.
The advantages claimed are (1) no impurities in the final PVC and
(2) low capital and operating costs because separation of an inert diluent
is not required.
Table 3-6 shows that three plants are operating bulk polymerization
of VCM. Data from only one company is available. In Table 3-7, the VCM
emissions are given in pounds of VCM per million pounds of PVC produced and
as total pounds of VCM released per hour.
3.4.2 Controls and Costs
The economically attractive bulk polymerization may offer more
Table 3-6. PRODUCTION CAPACITIES OF BULK PLANTS
Company Location Amount
The B. F. Goodrich Company NA 40 MM Ib/yr
The Goodyear Tire and Rubber Company NA 80 MM Ib/yr
Occidental Petroleum Corporation Burlington, N.J. 180 MM Ib/yr
59
-------
VINYL CHLORIDE
RECYCLE
CONDENSER
COLLECTOR
VINYL CHLORIDE
FROMSTORAGE
OVERSIZE
MATERIAL
RECEIVER
FINISHED
MATERIAL
RECEIVER
DISPOSAL
RECOVERY
COMPRESSOR
n—gpi
BAGGING
MILL
BAGGING
FIGURE 3.3 SIMPLIFIED FLOW DIAGRAM BULK POLYMERIZATION OF VINYL CHLORIDE
-------
Table 3-7. EMISSION RATES FOR PVC BULK PROCESS
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
PROCESS EMISSION SOURCE3
VCM feed tank vent, I
Prereactor vent, I
Reactor, I
Reactor emergency vent, I
Resin receiver vent, C
Resin collector vent, C
Finished product storage vent, C
Oversized PVC storage vent, C
Second collector vent, C
Ground material vent, C
Fugitive losses, C&I
Total
Ib VCM/MM Ib
produced
7900
Part of vent 3
2800
0
2000
Part of vent 5
Part of vent 5
Part of vent 5
Part of vent 5
Part of vent 5
800
13,500
PLANT A
Ib VCM/hr
emitted
29.1
Part of vent 3
10.3
0
7.4
Part of vent 5
Part of vent 5
Part of vent 5
Part of vent 5
Part of vent 5
29.4
76.2B,c
3! VCM Comment
emissions
33
12
0
20
35
C is continuous, I is Intermittent.
bPVC emitted at 2800 Ib VCM/MM Ib produced (10.3 Ib VCM/hr emitted).
C0r 3.1% VCM in PVC.
-------
resistance to a reduction of emissions than suspension polymerization.
Dry steam may enhance the vacuum stripping action and reduce strongly
the VCM emission. For all other emissions, controls similar to those
described in sections 3.2 and 3.3 can be used (Table 3-8).
3.5 SOLUTION POLYMERIZATION
Three solution polymerization plants (Table 3-9) are producing 150
million pounds of PVC a year.
3.5.1 Emission Points
Solution polymerization (Figure 3-4) is often conducted in systems
where PVC is slightly soluble in the solvent and VCM is soluble. The
products tend to have low molecular weights because the solvents act to
some extent as chain transfer agents; therefore, straight comparisons with
the other polymerization methods are not possible.
Absence of water simplifies the recovery of the final product and
is the major advantage for this process. If a volatile solvent like
N-butane is used, unreacted VCM and solvent are easy to remove.
No data are available on solution polymerization. Table 3-10 provides
for future data on sources shown in Figure 3r4; it does not give any
actual data. Total losses are expected to be approximately 2-3 percent
Table 3-9. PRODUCTION CAPACITIES OF SOLUTION PLANTS
Company
The Firestone Tire and Rubber Company
The B.F. Goodrich Company
Union Carbide Corporation
Location
NA
NA
NA
Amount
10 MM Ib/yr
20 MM Ib/yr
120 MM Ib/yr
62
-------
Table 3-8. CONTROLS AND COSTS FOR PVC BULK PROCESS
Percent Control
Emission Source Control Emission Cost Comment on Control Type
Reduction Estimate
1. VCM Feed Tank Vent: Intermittent stream
2. Prereactor Vent: intermittent stream
3. Reactor: intermittent stream
4. Reactor Emergency Vent: intermittent stream
5. Resin Receiver Vent: continuous stream
6. Resin Collector Vent: continuous stream
-------
Table 3-8. CONTROLS AND COSTS FOR PVC BULK PROCESS (continued)
Percent Control
Emission Source Control Emission Cost
Reduction Estimate
Comment on Control Type
7. Finished Product Storage Vent: continuous stream
8. Oversized PVC Storage Vent: continuous stream
9. Second Collector Vent: continuous stream
10. Grounded Material Vent: continuous stream
11. Fugitive Losses: continuous and intermittent streams
-------
VCM
n BUTANE
MONOMER
TANK
VENT
REACTOR
EMERGENCY
VENT
MONOMER
STORAGE
TANK
n BUTANE
STORAGE
STEAM-
REACT
^— .
ON
'
/
RECOVERY-
VENT
&
r>
STEAM
STRIPPER
BLEND
TANK
FIGURE 3 4 SIMPLIFIED FLOW DIAGRAM
-SOLUTION POLYMERIZATION
PROCESS FOR PVC
AIR
AppmyE|
BANBURY
MIXER
F—H
DICER
RIBBON
BLENDER
LOADING
ROLL MILL
BREIFING
MACHINE
COMPOUNDING
-------
Table 3-10. EMISSION RATES FOR THE PVC SOLUTION PROCESS
PLANT A
PLANT B
PROCESS EMISSION SOURCE3
Ib VCM/MM Ib
produced
% VCM
emissions
Ib VCM/MM Ib
produced
% VCM
emissions
<*»
o\
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
VCM tank vent, C
Reactor emergency vent, I
VCM recovery still vent, C
Blend or surge tank vent, C
Solvent storage tank vent, C
Centrifuge vent, C
Dryer vent, C
Air conveyor vent, C
Storage bin vent, C
VCM loading, storage tank vents, I
VCM charging, I
Polymerizer cleaning, I
Packing/hopper car, I
System cleaning, I
Fugitive losses, C&I
Abnormal losses, I
Total
C is continuous, I is intermittent.
-------
VCM and 1 percent PVC fines on PVC produced. Reduction should be by
control steps similar to those indicated for the suspension process.
3.5.2 Controls and Costs
Presently, the solution process probably leads to a PVC recovery of
96-97 percent of VCM intake. Most likely, the actual VCM losses total
2-3 percent. These losses are not specified as to different vent streams,
Table 3-11 shows controls that are believed to be appropriate.
67
-------
Table 3-11. CONTROLS AND COSTS FOR PVC SOLUTION PROCESS
Emission Source Control
Percent Control
Emission Cost
Reduction Estimate
Comment on Control Type
1. VCM Tank Vent: continuous stream; inerts + VCM; -13°F, 60-150 psig
2. Reactor Emergency Vent: intermittent stream
Carbon bed adsorption
Incineration
Ozone oxidation
Absorption
50-90
50-99
50-90
With ort without water wash.
High cost anticipated.
3. VCM Recovery Still Vent: continuous stream; variable flow
ON
00
4. Blend or Surge Tank Vent: continuous stream
5. Solvent Storage Tank Vent: continuous stream
6. Centrifuge Vent: continuous stream
Vacuum stripping
Water sprayer
Total recycle with bleed stream
50-80
May affect product quality.
Effectiveness to be determined,
Long-range control type.
-------
Table 3-11. CONTROLS AND COSTS FOR PVC SOLUTION PROCESS (continued)
Emission Source Control
Percent Control
Emission Cost
Reduction Estimate
Comment on Control Type
7. Dryer Vent: continuous stream
Vacuum stripping 50-80
Carbon bed adsorption 50-99
Ozone oxidation 50-99
Absorption 50-80
Water sprayer
Total recycle with bleed stream
Effectiveness to be determined.
Long-range control type.
8. Air Conveyor Vent: continuous stream
10
Vacuum stripping 50-80
Carbon bed adsorption 50-99
Incineration 50-99
Water sprayer
Total recycle with bleed stream
Effectiveness to be determined,
Long-range control type.
9. Storage Bin Vent: continuous stream
Vacuum stripping 50-80
Carbon bed adsorption 50-99
Incineration 50-99
Water sprayer
Total recycle with bleed stream
Effectiveness to be determined.
Long-range control type.
10. VCM Loading, Storage Tank Vent: intermittent stream
Recycle
Absorption
50-80
50-80
-------
Table 3-11. CONTROLS AND COSTS FOR PVC SOLUTION PROCESS (continued)
Percent Control
Emission Source Control Emission Cost Comment on Control Type
Reduction Estimate
11. VCM Charging: intermittent stream
12. Polymerizer Cleaning: intermittent stream
13. Packing/Hopper Car: intermittent stream
14. System Cleaning: intermittent stream
15. Fugitive Losses: continuous and intermittent
Pumps
Pump maintenance
Valve and relief valve
diaphragm valves 50
packed seals 50
rupt disc under relief valve
Miscellaneous
preventive maintenance 50
16. Abnormal Losses: intermittent
-------
4. SUMMARY OF ASSESSMENT AND FUTURE NEEDS
4.1 RESULTS OF PRODUCTION-EMISSION ASSESSMENT
The VCM emissions total about 3000 pounds per million pounds of VCM
produced. To this amount should be added an on-the-average smaller, inter-
mittent VCM loss from the loading area; a 90 percent reduction of this
amount can be obtained relatively straightforwardly by refrigeration and/or
absorption of VCM in the vents that are heavy in appropriate solvents (EDC
for instance) and by combustion of the organics and HC1-removal in other
vent streams. A 99 percent reduction is about the limit of present-day
technology; to achieve this, the loading area would definitely have to be
policed for vent losses.
In PVC production, the pounds of VCM losses per pound of PVC are at
least an order of magnitude higher than those from VCM production. Most
producers report PVC production as 3-4 percent lower than VCM intake, and
some producers report 1-1.5 percent of PVC production lost as fines to
the atmosphere. Thus, the actual VCM losses (2-3%) result from the batch
operation of the polymerization and in the filtration and the drying of
the polymer. Direct reduction to 50 percent of the present level seems
possible, but a 90 percent reduction without process changes in some
existing PVC plants is likely to be beyond present control techniques and
acceptable costs. However, intensive stripping at the end of the suspension
reaction may achieve the 90 percent reduction at acceptable costs.
Controls appraised for VCM production sources include recycling of
vent streams, condensation with refrigeration, compression, adsorption
with carbon, incineration, oxidation with ozone, absorption (scrubbing),
and venting to flares. Monomer loading (and unloading) involves special
71
-------
controls: vapor collection adapters with recycling, thermal level detectors
with recycling, and magnetic gauges. The PVC production sources can
benefit from the same controls plus vacuum stripping of crude product,
steam stripping, and carrier airstream recycling. Fugitive sources require
use of better valves, packings, etc.
Cost data received from VCM and PVC production show wide variation
without direct comparison. Since the limited cost data (none for pollution
control) were usually based on modifications of the existing system, the
estimates received were simply recorded in the tables dealing with state
of the art controls in sections 2 and 3.
4.2 RESULTS OF SOURCE-CONTROL ASSESSMENT
A qualitative assessment of the potential applications of selected
controls was based on information presented to date (June 1974) by U.S.
industrial firms. The results of this assessment are summarized for each
of eight VCM and PVC production processes. All percentage reductions
of emissions are estimates.
4.2.1 VCM from Hydrochlorination of Acetylene (Process 1)
The vent-gas reactor vent, which is the main emissions source, accounts
for 60 percent of total emissions. Condensation at 40°F and 35 psi is now
being used. Addition of refrigeration would decrease emissions 50 percent;
addition of an HC1 scrubber should achieve 85 percent reduction; these,
combined with carbon adsorption, should reduce emissions 99 percent.
Recycling, incineration, ozone oxidation, and venting to flares do not
appear to be applicable.
Fugitive emissions equal about 25 percent of total emissions. Use of
diaphragm on double lock valves, replacement of packed pump seals with
72
-------
pressurized mechanical seals, use of vapor collectors on samplers, and
preventive maintenance should reduce these emissions 50-95 percent.
Tank-car loading accounts for about 13 percent of emissions. Inciner-
ation, with HCl recovery should reduce loading arm emissions 90 percent and
condenser vent losses 99 percent. Thermal level detectors with recycling
would reduce slip gages 95 percent and magnetic gages 100 percent. Vapor
collection adapters with recycling would reduce purge losses 50-90 percent.
4.2.2 VCM from Chlorination-Oxychlorination of Ethylene (Using Air) and
Dehydrochlorination (Process 2)
Major emission sources are the EDC light-ends condenser column vent
(13-38%), the heavy-ends, tar-removal column vent (25-26%), the VCM light-ends
distillation column vent (10-13%), and the tank-car loading (10-20%).
EDC light-ends column emissions could be reduced 50 percent by refrig-
erated condensers and 100 percent by either carbon adsorption or recycle to
a postchlorination unit. Heavy-ends column emissions are believed to be
controllable by incineration (90% reduction) and by adsorption (100%
reduction). For VCM light ends, either adsorption or ozone oxidation
would achieve 100 percent reduction. Tank-car loading controls for VCM
process 1 would apply here.
4.2.3 VCM from Chlorination-Oxychlorination of Ethylene (.Using Oxygen)
and Dehydrochlorination
Major emission sources and controls are believed to be essentially
the same as those for process 2. Use of oxygen would reduce the amounts
of vent streams from the oxychlorination process and thus would reduce
emissions somewhat.
4.2.4 VCM from Direct Chlorination of Ethylene and Dehydrochlorination
The controls and reductions appear to be the same as for process 2
73
-------
for VCM emissions. Incineration (light-ends column) with HCl-recovery by
scrubbing is about 90 percent efficient for avoiding HC1 emissions.
4.2.5 VCM from Suspension Polymerization
Fugitive emissions account for 12-46 percent of total emissions.
Vacuum stripping of the crude product would reduce these emissions 50-80
percent; carbon adsorption, 50-99 percent; condensation with refrigeration,
40-60 percent; incineration, 50-99 percent; and absorption, 50-90 percent.
Collectively, the dryer vents, the air conveyor vent, the storage
bin vent, and the centrifuge vent provide 35 percent of the total emissions.
Vacuum stripping could achieve 50-80 percent reduction, and either carbon
adsorption, ozone oxidation, incineration, steam stripping, or recycling to
compressors would reduce emissions 40-99 percent. Air recycle should
reduce these emissions 40-80 percent.
Blend surge tanks account for 11 percent of total emissions. Carbon
adsorption, ozone oxidation, and incineration should give 50-99 percent
reductions. Vacuum stripping and absorption are expected to give 50-80
percent; recycles to compressors, 40-60 percent.
4.2.6 VCM from Emulsion Polymerization
The dryer vent, the air conveyor vent, the storage vent, and the VCM
recovery vent appear to account collectively for up to 85 percent of total
emissions. Carbon adsorption, ozone oxidation, and steam stripping could
reduce these emissions 50-99 percent. Absorption might achieve 90 percent.
The recycle of airstreams could be 40-80 percent efficient.
Fugitive losses contribute 3 percent of the emissions and the resin
receiver vents contribute another 7 percent. For either source, vacuum
stripping, would effect 50-80 percent reduction; carbon adsorption, 50-99
74
-------
percent. Condensation with refrigeration would reduce either source
40-60 percent; absorption, 50-80 percent. Preventive maintenance, would
reduce fugitive losses 25-50 percent. The surge tank vents could be
recycled (40-60% reduction) or oxidized with ozone (50-99% reduction).
4.2.7 VCM from Bulk Polymerization
The VCM reactor vent (12 percent), fugitive emissions (35 percent),
and the combined resin receiver, collector, and storage (20 percent) are
the major emission sources. For the reactor vent, adsorption could achieve
50-90 percent reduction. Intensive maintenance is believed to be capable
of giving 50-75 percent reduction of the diverse, ill-defined fugitive
emissions (which need further study). The product-collection systems vents
could be water washed and then adsorbed (50-90% reduction), oxidized with
ozone (90% reduction), or incinerated (50-90% reduction).
4.2.8 VCM from Solution Polymerization
While no data are available at present, solution polymerization is
expected to have the emission characteristics of the suspension poly-
merization process and to respond similarly to the same controls.
4.3 RESEARCH AND DEVELOPMENT NEEDS
The information needs identified during this study include performance
data, basic data and cost data. Performance data are needed to quantify
PVC buildup during adsorption with activated carbon and to identify, if
possible, uses which would improve the effectiveness of this control.
If the proposed larger PVC reactors (120,000 gallons instead of the present
5-10,000) are put into use, emergency blowing may lead to high peak values
for VCM emissions. Not considered in this study is the effect of VCM
emissions (sometimes 1000 ppm) remaining in the PVC; release of most of
75
-------
this may cause emission problems during further processing. Also needed
are basic data on the effectiveness of sprayers in removing solids from
vent gases of the PVC processes and on the use of EDO as an absorbent
for vent gases of the VCM processes. The impact of process modifications
on the quality of PVC products must be determined more precisely before
the adsorption by such control techniques can be recommended.
The costs and benefits of the use of oxygen instead of air for the
oxychlorination of ethylene should be quantified. This technique should
be compared with the use of postchlorination units.
Cost data are needed that distinguish between capital equipment
investments made primarily for recovery of unreacted VCM and those
investments made solely for reduction of emissions. Possibly, the total
costs for reactor vent streams can be reduced by the value of the recovered
VCM. Investments already made should be distinguishable from new costs
for additional control equipment.
76
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing)
. REPORT NO.
EPA-650/2-74-097
3. RECIPIENT'S ACCESSION-MO.
4. TITLE AND SUBTITLE
Vinyl Chloride--An Assessment of Emissions
Control Techniques and Costs
5. REPORT DATE
September 1974
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
Ben H. Carpenter
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORG tNIZATION NAME AND ADDRESS
Research Triangle Institute
P.O. Box 12194
Research Triangle Park, NC 27709
10, PROGRAM ELEMENT NO.
1AB015; ROAP 21AUY-003
11. CONTRACT/GRANT NO.
68-02-1325 (Task 17)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 6/74-9/74
14. SPONSORING AGENCY CODE
IS. SUPPLEMENTARY NOTES
6. ABSTRACT
The report gives results of a survey of conceptual techniques applicable to
vinyl chloride monomer (VCM) emission reduction with respect to VC monomer and
polymer production. VCM emission points have been identified and quantified for
four types of monomer plants—hydrochlorination of acetylene, chlorination/oxychl-
orination of ethylene (with oxygen) and dehydrochlorination, and direct chlorination
of ethylene and dehydrochlorination--and four types of polymer manufacture--sus-
pension polymerization, emulsion polymerization, bulk polymerization, and solu-
tion polymerization. Levels of control achievable and estimated cost of listed con-
trol techniques are presented.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air Pollution
Vinyl Chloride
Polymers
Acetylene
Ethylene
Chlorination
Air Pollution Control
Stationary Sources
Monomers
13B
07C
07D
I. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21 NO. OF PAGES
84
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
77
------- |