United States
          Environmental Pro;ect'-)n
          Agency
Office of Air Quality
Planning and Standards
Researcn Triangle Park NC 27711
EPA-450/4-84-007]
September 1985
         Locating And
         Estimating Air
         Emissions From
         Sources  Of
         Epichlorohydrin
KP 4SO/4
8/4-00 7 j

-------
                                EPA-450/4-84-007J
                                    September 1985
Locating And Estimating Air Emissions
   From Sources Of Epichlorohydrin
           US ENVIRONMENTAL PROTECTION AGENCY
                Office Of Air And Radiation
            Office Of Air Quality Planning And Standards
            Research Triangle Park, North Carolina 2771 1

                   September 1985

-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U.S. Environmental Protection
Agency, and has been approved for publication as received from the contractor. Approval does not signify that the
contents necessarily reflect the views and policies of the Agency, neither does mention of trade names or commercial
products constitute endorsement or recommendation for use
                                           EPA-450/4-84-007J

-------
                              TABLE OF CONTENTS

Section                                                               Page

   1      Purpose of Document 	   1

   2      Overview of Document Contents 	   3

   3      Background	   5

               Nature of Pollutant	   5

               Overview of Production and Use	   5

               References for Section 3	12

   4      Emissions from Epichlorohydrin Production 	  14

               Epichlorohydrin Production 	  14

               Inadvertent Production of Epichlorohydrin in
               Other Industrial Processes 	  24

               References for Section 4	25

   5      Emissions from Industries Which Use Epichlorohydrin
          as a Feedstock	27

               Production of Synthetic Glycerin 	  27

               Production of Epoxy Resins (Continuous
               Process)	31

               Production of Epoxy Resins and Other Products
               from Epichlorohydrin (Batch Process) 	  36

               References for Section 5	44

   6      Emissions from the Use of Epichlorohydrin-
          containing Products 	  46

               Use of Epoxy Resins	46

               Use of Synthetic Glycerin	48

               Use of Wet-Strength Resins	49

               Use of Elastomers	49

               References for Section 6	50
                                    111

-------
                        TABLE OF CONTENTS  (Continued)




Section                                                               Page




   7      Source Test Procedures	51




               Literature Review of Sampling Methods	51




               Literature Review of Analytical Methods	52




               References for Section 7	54
                                     IV

-------
                               LIST OF TABLES

Table                                                                 Page

  1       Synonyms and Trade Names for Epichlorohydrin	    6

  2       Summary of the Physical and Chemical Properties
          of Epichlorohydrin	    7

  3       Estimated Domestic Consumption of Epichlorohydrin
          in 1984	    9

  4       Description of Streams and Vents Illustrated in
          Figure 1 for the Production of Epichlorohydrin	16

  5       Sources of Fugitive Epichlorohydrin Emissions
          from Equipment Leaks in Epichlorohydrin
          Production/Finishing Facilities 	   21

  6       Emission Factors for the Release of Epichlorohydrin
          from Epichlorohydrin Production 	   23

  7       Emission Factors for the Release of Epichlorohydrin
          from Batch Processes Which Use Epichlorohydrin as
          a Feedstock	41

  8       Some Producers of Epichlorohydrin Products	43

  9       Epichlorohydrin Vapor Concentrations Above  Epoxy
          Resins at Various Temperatures Under Static
          Equilibrium Conditions	47

-------
                               LIST OF FIGURES

Figure

  1       Basic Operations That May Be Used in the
          Production of Epichlorohydrin from Allyl Chloride 	  15

  2       Basic Operations That May Be Used in the Production
          of Synthetic Glycerin from Epichlorohydrin	30

  3       Basic Operations That May Be Used in the Continuous
          Production of Epoxy Resins from Epichlorohydrin
          (Well-Controlled Facility)	33

  4       Flow Diagram for the Batch Production of Epoxy Resins ...  37

  5       General Flowsheet for the Production of
          Epichlorohydrin Elastomers	38
                                     VI

-------
                                  SECTION 1
                             PURPOSE OF DOCUMENT

     The U. S. Environmental Protection Agency (EPA), States, and local air
pollution control agencies are becoming increasingly aware of the presence
of substances in the ambient air that may be toxic at certain
concentrations.  This awareness, in turn, has led to attempts to identify
source/receptor relationships for these substances and to develop control
programs to regulate emissions.  Unfortunately, very little information is
available on the ambient air concentrations of these substances or on the
sources that may be discharging them to the atmosphere.

     To assist groups interested in inventorying air emissions of various
potentially toxic substances, EPA is preparing a series of documents such as
this that compiles available information on sources and emissions of these
substances.  This document specifically deals with epichlorohydrin.   Its
intended audience includes Federal, State, and local air pollution personnel
and others who are interested in locating potential emitters of
epichlorohydrin and making gross estimates of air emissions therefrom.

     Because of the limited amounts of data available on epichlorohydrin
emissions, and since the configuration of many sources will not be the  same
as those described herein, this document is best used as a primer to inform
air pollution personnel about (1) the types of sources that may emit
epichlorohydrin, (2) process variations and release points that may  be
expected within these sources,  and (3) available emissions information
indicating the potential for epichlorohydrin to be released into the air
from each operation.

-------
     The reader is strongly cautioned against using the emissions
information contained in this document to try to develop an exact assessment
of emissions from any particular facility.  Since insufficient data are
available to develop statistical estimates of the accuracy of these emission
factors, no estimate can be made of the error that could result if these
factors were used to calculate emissions from any given facility.  It is
possible, in some extreme cases, that orders-of-magnitude differences could
result between actual and calculated emissions,  depending on differences in
source configurations, control equipment, and operating practices.  Thus, in
situations where an accurate assessment of epichlorohydrin emissions is
necessary, source-specific information should be obtained to confirm the
existence of particular emitting operations, the types and effectiveness of
control measures, and the impact of operating practices.  A source test
and/or material balance should be considered as  the best means to determine
air emissions directly from an operation.

-------
                                  SECTION 2
                        OVERVIEW OF DOCUMENT CONTENTS

     As noted in Section 1, the purpose of this document is to assist
Federal, State, and local air pollution agencies and others who are
interested in locating potential air emitters of epichlorohydrin and making
gross estimates of air emissions therefrom.  Because of the limited
background data available, the information summarized in this document does
not and should not be assumed to represent the source configuration or
emissions associated with any particular facility.

     This section provides an overview of the contents of this document.  It
briefly outlines the nature, extent, and format of the material presented in
the remaining sections of this report.

     Section 3 of this document provides a brief summary of the physical and
chemical characteristics of epichlorohydrin, its commonly occurring forms,
and an overview of its production and uses.  A table summarizes the
quantities of epichlorohydrin consumed in various end uses in the United
States.  This background section may be useful to someone who needs to
develop a general perspective on the nature of the substance and where it is
manufactured and consumed.

     The fourth and fifth sections of this document focus on major
industrial source categories that may discharge epichlorohydrin air
emissions.  Section 4 discusses the production of epichlorohydrin and
Section 5 discusses the use of epichlorohydrin as an industrial feedstock in
the production of synthetic glycerin and epoxy resins.   For each major
industrial source category described in Sections 4 and 5, example process
descriptions and flow diagrams are given, potential emission points are

-------
identified, and available emission factor estimates are presented that show
the potential for epichlorohydrin emissions before and after controls
employed by industry.  Individual companies are named that are reported to
be involved with either the production and/or use of epichlorohydrin, based
on industry contacts and available trade publications.  Section 6 contains
information on possible releases of epichlorohydrin to air from the use of
materials containing trace epichlorohydrin levels.

     The final section of this document summarizes available procedures for
source sampling and analysis of epichlorohydrin.  Details are not prescribed
nor is any EPA endorsement given or implied to any of these sampling and
analysis procedures.  At this time, EPA has generally not evaluated these
methods.  Consequently, this document merely provides an overview of
applicable source sampling procedures, citing references for those
interested in conducting source tests.

     This document does not contain any discussion of health or other
environmental effects of epichlorohydrin, nor does it include any discussion
of ambient air levels or ambient air monitoring techniques.

     Comments on the contents and usefulness of this document are welcomed,
as is any information on process descriptions, operating practices, control
measures, and emissions information that would enable EPA to improve its
contents.  All comments should be sent to:

               Chief, Source Analysis Section (MD-14)
               Air Management Technology Branch
               U. S. Environmental Protection Agency
               Research Triangle Park, N. C.  27711

-------
                                  SECTION 3
                                 BACKGROUND

NATURE OF POLLUTANT

     Epichlorohydrin is a colorless, free-flowing, highly reactive liquid.
Its irritating odor has been likened to that of chloroform or garlic.  It is
both volatile and flammable.  It is soluble in most organic solvents and
forms azeotropes with many organic liquids.  It is slightly soluble in
petroleum hydrocarbons and in water.  Because of an asymmetric carbon atom
in the molecule, epichlorohydrin exists as an isomeric mixture with equal
amounts of the dextro- and levorotary forms.  Synonyms and trade names for
epichlorohydrin are given in Table 1; physical and chemical properties are
summarized in Table 2.  The Chemical Abstracts Service (CAS) registry number
for epichlorohydrin is 106-89-8.1>2

     Epichlorohydrin is not persistent in the environment, hydrolyzing in
several weeks.  Its atmospheric residence time, the estimated time in days
required for a given quantity to be reduced to 1/e (37 percent) of its
original value, is 5.8.3  At 20°C (68°F), its half-life in distilled water
is 8.0 days; in 3 percent sodium chloride, the half-life is 5.3 days.
                                                                          4
Epichlorohydrin also participates in free-radical photochemical reactions.

OVERVIEW OF PRODUCTION AND USE

     Epichlorohydrin is produced commercially in the United States by
chlorohydrating allyl chloride into isomeric glycerol dichlorohydrins,  which
are then dehydrochlorinated with alkali to form crude epichlorohydrin.
Crude epichlorohydrin can be used directly for the production of synthetic
                                   2
glycerin or refined for other uses.

-------
          TABLE 1.  SYNONYMS AND TRADE NAMES FOR EPICHLOROHYDRIN
l-chloro-2,3-epoxypropane




3-chloro-l,2-epoxypropane




(chloromethyl)ethylene oxide




2-(chloromethyl) oxirane




chloropropylene oxide




a-chloropropylene oxide




3-chloropropene 1,2-oxide




2-chloromethyl oxirane




ECH
glycidyl chloride



(chloromethy1)oxirane




3-chloro-l,2-propylene oxide




a-epichlorohydrin




D,L-a-epichlorohydrin




SKEKhG




1,2-epoxy-3-chloropropane




2,3-epoxypropyl chloride




glycerol epichlorohydrin

-------
TABLE 2.  SUMMARY OF THE PHYSICAL AND CHEMICAL PROPERTIES OF EPICHLOROHYDRIN2
Molecular Formula



Molecular Weight

Elemental Composition
Physical Properties

  Melting Point
  Freezing Point
  Boiling Point

  Density (g/ml, 20°C)

  Specific Gravity (20/20°C)
  Vapor Pressure (16.6°C)
                 (30°C)
  Concentration in Saturated Air
  (760 mmHG, 25°C)
  Coefficient of Thermal Expansion at 68°F
  Solubility
    Water (10*C)
    Water (20*0
  Pounds per Gallon (68°F)
  Flash Point (Tag open cup)
              (Tag closed cup)
  Autoignition Temperature
  Latent Heat of Vaporization (calc.)
  Odor Threshold in Air
  Surface Tension (20°C)
  Heat of Combustion
  Liquid Viscosity (25°C)
  Refractive Index (25°C)
  1 ppm at 25°C & 760 mmHg equivalent to
  1 mg/1 at 25°C & 760 mmHg equivalent to
  Heat Capacity (25°C)
                (100°C)
  Heat of Formation (25°C)
  Explosive Limits (volume I in air)
  Heat of Fusion (25°C)
CH2 	 CH-CH2C1
   V

92.53

C  - 38.94%
H  - 5.45%
Cl - 38.32%
0  «* 17.29%
-48.0°C
-57°C
116°C

d201.1812
 4
1.181
10 mmHg
22 mmHg

1.7%
0.000577 per °F

6.52%
6.58%
9.85
41CC
31°C
416°C
9060 cal/mole at the b.p.
10 ppm
37.00 dynes/cm
4524.4 cal/gm
0.0103 poise
nD1.4358
3.78 mg/m3
265 ppm
31.5 cal/mol°C
40.0 cal/mol°C
-35.6 Kcal/mol
3.8-21.0
2,500 cal/mol

-------
     Several alternative methods are being developed for producing
epichlorohydrin, but as of 1983 none are yet approaching commercial
application.   These methods include:
     •    Epoxidation of allyl chloride with:
          —peracids;
          —perborates;
          —tert-butyl hydroperoxide in the presence of vanadium,
            tungsten, or molybdenum catalysts;
          —a-phenylethyl hydroperoxide;
          —air or oxygen in systems which include aluminum-silver
            oxide (Al-Ag 0) or dimethyl phthalate-acetaldehyde.

     •    Chlorination of allyl alcohol to dichlorohydrins.

     •    Hydrochlorination of glycerol to chlorohydrins.

     •    Chlorination of acrolein to 2,3-dichloropropionaldehyde and
          reduction with sec-butyl alcohol to  2,3-dichlorohydrin.

     As of 1984, only two companies — Dow Chemical Company, Freeport, TX,
and Shell Chemical Company, Norco, LA — produced epichlorohydrin.  (Crude
epichlorohydrin from Shell Oil's Norco plant is finished at  Shell's Deer
Park, TX plant.)  Two epoxy resin manufacturers — Union Carbide and the
Plastics and Additives Division of Ciba-Geigy  — have had the capacity to
produce epichlorohydrin from purchased allyl chloride,  but as of 1975 had
                              6-9
not done so for several years.

     Domestic consumption of epichlorohydrin for 1984 is summarized in
Table 3.    Both quantity and percent of total  epichlorohydrin consumption
are given.  Crude epichlorohydrin may be used  directly for the production of
synthetic glycerin or it may be refined for other uses.  In  1982, more than
90 percent of the total U. S. production of unmodified epoxy resins was

-------


















3
-I

^.^
oo
^t
*™^

*^t
1— (


z
1-1
PS1
Q
^
P^
O
OS
o

33
U
P-,
w
o

a
o
1

*~I
z
o
u
u
H
§
O
Q

Q
W
H

^i
•— j
H


fl
bJ
L_^
53
H



























CD
0)
a
pt

^
o
'5



















U 3 B
•H U4 -H
to U -a
01 H-4 X
§0 .C
0
Q C M
O 0
a +J jz
4J O O
O 3 -t
H -0 Q.
O Id
M V4
a>



^
.A
x ^*
•"
u O
E — i
eg ^
3
^e?






u
3
•o
o
cu










•^^
a
oo
B
AJ
a
eg
u
•
a
01
.u
.*4
a
O
§
u

a
0)
4-i
§
•f-l


l_4
•
- 00
a B
OO-rf
.5 8
*J O
§•-1
IM
U
4) 00
U C
(Q *H
"4-4 "O
3 0
en S






*






s~*
r*
1

PUB
^

^

1
£






S
1
a



a
B -J
••< 0
a s
CU 0)
OS X
Cu
fr .2
§. "
Id


01
• - B .
ana
00 JS M
3 0. 01
b o B
•o ^ x
»-t i— •
. 01 0
moo.
o
•rf b 0)
•W O B
cj IM a
§j=
l-l 4J
O CU 01 .
O M hi B
•3so
» h -H
01 0) - <-i
00 JJ B 0
to m •-! 3
u a B -o
CU -4 01 o
> D. U U
cu a.
ji •- •«
^** o x a
•o o j: 01
o u — >
o a a --i
u^ ^ a
o a o
o ^ H
•H C (U
• W CO
CD c j-» 4J C
v aj c y -H
> 'H  JS fl 4J






CN







^.
C?
1

CO
^^
f^

1
s


X
a
u
ll
01
o
•u
B B
a -H
o x o'
a x u
— ox
0 a. -<
o t. > u e>
•H 01 O
" -= z x o
aw x -H
-c o) x o ~"
0, x ST B 0)
••* ^ O 01 J3
— 4 o a. ^ 4J
< cu u a. a
en
















00
B
M

a

14
01
a.
03
a.

^
0
u-t

•
x
a
3
•o
a
• r4
b
01
O.
eg
Cu






m







^^
m
I
o



p*
i
m







a
B
•H
a
01
at

00
c
01
u.
.u
en
cu
»
1
3
U
0
o
f— t
IM

4J
B
CU
u
eg
01
M
4J

u
01
cd
^

01
u
a


n
B
O


a
o •
•<-i a

v cu
a oo
— eg
u
00
U B
cu •-<
eg a
* "•






-









^^
CN
•^



vO







a
u
1
O
cu

B
CU
s
a
01
h
H
01
a
S

a
01
a
O
-C
n
01
r-4 •
.a a
0 C
cu
T3 00
B ha
a 01 B

o) gj a
1* "O Ol
•H M
> •«

W 3 X
o . er o
u a •-< ex
O -HO)
a -H
4J W b
cu .Q a o
^ eg o u-,
U **4 O
eg a. a
•TO B *•
cu SB*

m --i aw c
u w 3 «
ji £> n --i i.
a ^ u -o eg
a 3 S u
oo h a 01 cu
" > b
. . u •-<
a a eg 4^ 01
-H w 14-1 u a
« -H u a <3
0) 0) 3 01 -H
en J3 en at CK,






«N CN sO
rt O -i -H O 0








^^ ^N
c^ ^~\ m *™N
I «H 1 /*^ »H ^>
r^ v <•"> m v cs


-cr in CN in
1 • 1 — H • — H
ft O -H O
V V

a
cu

O
IH
3
cn
a
B U
a .c
Oi w
otf Id
a a
0> i— t b b
oo x 01 oi
6 C X u
eg oi u a
a j: o Id id cs
b u x 1
oi x J -i -i a!
§U U X X Cb
•a -o
U B -4 -I •-( —
a o x u 
-------
produced from refined epichlorohydrin.  Epoxy resins are cured either by
reaction with a cross-linking agent (hardener) or by self-polymerization
with the aid of a catalyst.  Epoxies are used in reinforced plastics,
casting, potting, encapsulation, molding compounds, protective coatings, and
  ...    11
adhesives.

     Other uses for epichlorohydrin include the production of
epichlorohydrin elastomers, glycidyl ethers, wet-strength resins, water
treatment resins, surfactants, solvents, adhesives, inks and dyes, asphalt
improvers, corrosion inhibitors, fumigants, flame retardents, sterilizing
agents, and pesticides.  About 5 percent of the epichlorohydrin produced in
the United States is exported to other countries.

     All emissions from the production and use of epichlorohydrin can be
broadly related to process vents, storage operations,  and fugitive losses
from pumps, valves, flanges, etc.

     According to a study based mainly on engineering  calculations and
assumptions about the composition of exhaust gas streams, fugitive emissions
are the largest source of epichlorohydrin, accounting  for 84 percent of the
total annual epichlorohydrin emissions.  Storage losses account for
12 percent and process vents only 4 percent.  (This distribution of
emissions will vary widely at individual facilities.)   Fugitive emissions
are the largest source of epichlorohydrin emissions because of the numerous
pumps, valves, etc., in most plants and because most other sources,
particularly process discharges, are generally well controlled.  It should
be noted, however, that fugitive emissions may be less important in batch
operations, such as those producing wet strength resins, elastomers, and
surfactants, because epichlorohydrin is not flowing through the process
                                                     12
components continuously, as was assumed in the study.

     There are no known uses of epichlorohydrin as a solvent; hence, no
solvent-related emissions would be expected from dispersed end use
applications as are common with many other organic chemicals.
                                     10

-------
     Some potential exists for volatile substances, including
epichlorohydrin, to be emitted from waste treatment, storage, and handling
facilities.  Reference 13 provides general theoretical models for estimating
volatile substance emissions from a number of generic kinds of waste
handling operations, including surface impoundments, landfills, land farming
(land treatment) operations, wastewater treatment systems, and drum
storage/handling processes.  Since no test data were available on
epichlorohydrin emissions from any of these operations at the time of
publication, no further discussion is presented in this document.  If such a
facility is known to handle epichlorohydrin, the potential for some air
emissions should be considered.
                                     11

-------
REFERENCES FOR SECTION 3

1.   Kirk-Othmer Encyclopedia of Chemical Technology.  Third Edition,
     Volume 5.  Chlorohydrins.   John Wiley and Sons.   New York, NY.   1980.
     pp. 858 - 864.

2.   Stanford Research Institute.  Chemical Economics Handbook.  SRI
     International,  Menlo Park, CA.  1978.

3.   Cupitt, L. T.  Fate of Toxic and Hazardous Materials in the Air
     Environment.  Environmental Sciences Research Laboratory, U. S.
     Environmental Protection Agency, EPA-600/3-80-084.  Research Triangle
     Park, NC.  August 1980.

4.   Syracuse Research Corporation.  Investigation of Selected Potential
     Environmental Contaminants:  Epichlorohydrin and Epibromohydrin.
     (Prepared for U.  S. Environmental Protection Agency, PB80-197585.)
     Syracuse, NY.  March 1980.

5.   McKetta, J. J., and W. A.  Cunningham, eds.  Encyclopedia of Chemical
     Processing and Design, Volume 8, Chlorohydrins.   Marcel Dekker, Inc.
     1979.

6.   Stanford Research Institute.  Chemical Economics Handbook:
     Epichlorohydrin.   SRI, Menlo Park, CA.  1975.

7.   Nonconfidential portions of a letter from R. R.  Erickson, Shell Oil
     Company, Deer Park, TX to David Beck, U.  S. EPA, Research Triangle
     Park, NC.  December 17, 1983.

8.   Nonconfidential portions of letter from W. L. Caughman, Jr., Shell  Oil
     Company, Norco, LA, to Jack R. Farmer, EPA, Research Triangle Park,  NC.
     October 13, 1983.

9.   Nonconfidential portions of letter from S. L. Arnold, Dow Chemical,
     U.S.A., Midland,  MI, to Jack R. Farmer, EPA, Research Triangle  Park,
     NC.  December 8,  1983.

10.  Assessment of Epichlorohydrin Uses, Occupational Exposure, and
     Releases.  Dynamac Corporation, Rockville, Maryland.  Prepared  for  the
     Office of Toxic Substances, U. S.  Environmental  Protection Agency,
     Washington, DC.  Prepared under EPA Contract No. 68-02-3952.
     July 1984.

11.  Chemical Products Synopsis.  Manville Chemical Products.   Cortland,  NY.
     December 1982.
                                     12

-------
12.   Memorandum entitled "Epichlorohydrin  Emissions  Summary:
     Epichlorohydrin Source  Assessment"  from Jeffrey A.  Shular, Midwest
     Research Institute, Raleigh,  NC,  to David  Beck, EPA,  Research  Triangle
     Park,  NC.   February 16, 1984.

13.   Farino,  W.,  et al.   Evaluation and  Selection  of Models  for Estimating
     Air Emissions from Hazardous  Waste  Treatment, Storage,  and Disposal
     Facilities.   EPA-450/3-84-020.  GCA Corporation,  Bedford, MA.
     December 1984.
                                     13

-------
                                  SECTION 4
                  EMISSIONS FROM EPICHLOROHYDRIN PRODUCTION

     Epichlorohydrin can be released to the atmosphere both during its
production and during its consumption as a raw material in other
manufacturing processes.  This section details the production of
epichlorohydrin and the emission factors associated with that production.
Manufacturing processes which use epichlorohydrin as a feedstock are
described in Section 5.

EPICHLOROHYDRIN PRODUCTION

Process Description

     Several processes have been developed for producing epichlorohydrin.  A
generalized process is described here, showing the basic operations
involved.  Figure 1 shows a process flow diagram of this generalized
process; Table 4 describes the streams and vents illustrated in Figure 1.
Shell Oil has indicated that their crude epichlorohydrin production process
differs from the generic epichlorohydrin process shown in Figure 1; however,
                                            2
details on how it differs were not provided.   The subheadings in the
following text correspond to the major component operations involved in
epichlorohydrin production.  Process variations discussed are those known to
be practiced by various manufacturers.

Allyl Chloride Production —

     Allyl chloride is derived from dry propylene by direct chlorination in
a high-temperature [^500°C (932°F)] gas phase reactor according to the
following reaction:
                                     14

-------
                                                        U-l
                                                         o

                                                         c
                                                         o
                                                         o

                                                        T3
                                                         O
                                                         I-l
                                                        Cu
                                                         c
                                                        i-l   0)
                                                            T3
                                                        •O  -H
                                                         01   )-i
                                                         en   o
                                                         co  E
                                                        ,e  o
                                                         4J  U
                                                            u-
                                                         CO
                                                         C  d
                                                         O -H
                                                        •H  h
                                                         J-l "O
                                                         CO  >-,
                                                         W -C
                                                         
-------
     TABLE 4.  DESCRIPTION OF STREAMS AND VENTS ILLUSTRATED IK
               FIGURE 1 FOR THE PRODUCTION OF EPICHLOROHYDRIN
Code Number
                 Description
  Stream
     1
     2
     3
     4
     5
     6
     7
     8

     9
    10

    11

    12
    13
    14
    15
    16
    17
    18
    19
    20
    21
    22
    23
    24
    25
    26
Chlorine feed
Water feed
Dilute hypochlorous acid
Allyl chloride feed
Dehydrochlorination reactor product
Separator underflow (3-5% dichlorohydrin)
Recycle to chlorine absorber (optional)
Alkali feed (sodium or calcium
  hydroxide or carbonate)
Dehydrochlorination reactor product
Azeotropic stream stripper overhead
  (26Z epichlorohydrin/vater azeotrope)
Liquid phase from azeotropic stripper com-
  bined with stripper bottoms (12)
Azeotropic steam stripper bottoms
Recycle to azeotropic steam stripper
Aqueous phase stripper bottoms
Organic phase from azeotrope
Overhead from organic phase stripper
Overhead recycle to organic phase stripper
Separator overhead condensate to vastevater
Organic phase stripper bottoms
Bottoms recycle to organic phase stripper
Purification column product
Overhead recycle to purification column
Final product epichlorohydrin
Purification column bottom
Bottoms recycle to purification column
Purification column bottoms to vastevater
                              Chlorine absorber vent
                              Chlorination reactor  vent
                              Dehydrochlorination reactor  vent
                              Azeotropic steam stripper  vent
                              Aqueous phase stripper vent
                              Organic phase stripper vent
                              Purification column vent
                              Storage tanks vents
                              Fugitive emissions, including valves,
                               flanges,  pump seals, etc.
                                  16

-------
     CH2=CH-CH3 + C12 —> CH2=CH-CH2C1 + HC1
     Propylene  Chlorine     Allyl    Hydrochloric
                            chloride     acid
The crude allyl chloride is fractionated and purified in several columns.
As Stream 4 it is fed to the chlorination reactor.

Hypochlorous Acid Production —

     Hypochlorous acid is produced from chlorine (Stream 1) and water
(Stream 2) in a packed tower chlorine absorber unit by the following
reaction:

     C12       +         H20       -->       HC10      +         HC1
   Chlorine             Water            Hypochlorous        Hydrochloric
                                             acid                acid
The absorber may be eliminated entirely by feeding gaseous chlorine and
water directly into the chlorination reactor (discussed next) along with the
allyl chloride.

Dichlorohydrin Production —

     Hypochlorous acid (Stream 3) and allyl chloride (Stream 4) are combined
in a chlorination reactor.  The chlorination occurs at atmospheric pressure
in the liquid phase:
2 CH=CH-CH2C1   +   2 HC10   -->   CH2C
Allyl chloride     Hypochlorous    1 ,2-dichlorohydrin     1 ,3-dichlorohydrin
                      acid               (70%)                   (30%)
The reactor product stream (Stream 5) is sent to a separator.  The separator
underflow  (Stream 6), which contains about 3 to 5 percent dichlorohydrin
                                     17

-------
isomers, is routed to the dehydrochlorination reactor.  Recycle of the
aqueous separator overflow (Stream 7) to the chlorination absorber is
optional.

Epichlorohydrin Production—

     An alkali (Stream 8) is added to Stream 6 in the dehydrochlorination
reactor.  The alkali can be sodium (or calcium) hydroxide or sodium (or
calcium) carbonate.  Use of a carbonate alkali greatly increases emissions
from the process because the large amount of CO- produced acts as a sweep
    1                                          L
gas.   A difference of opinion exists in the literature as to whether the
choice of alkali is a process option or a control option.  The
dichlorohydrins undergo dehydrochlorination and epoxidation according to the
following reaction:
CH C1-CHC1-CH2OH   +   NaOH   —>   E2C\~,^H~CH2C1   +   NaC1   +   H2°
                                      V
1,2-dichlorohydrin    Alkali        Epichlorohydrin      Salt      Water
                   e.g., Sodium                      e.g., Sodium
                     hydroxide                          chloride
An excess of alkali drives the reaction to completion.  The crude
epichlorohydrin product stream (Stream 9) from the reactor contains 3 to
                                                              1 3
5 percent epichlorohydrin, other reaction products, and water.  '    The crude
epichlorohydrin may be used directly in the production of glycerin or it may
be refined for use in other manufacturing processes.

Purification and Recovery of Epichlorohydrin —

     The crude epichlorohydrin stream (Stream 9) is purified first by
azeotropic steam stripping.  The overhead from the stripper (Stream 10), an
epichlorohydrin/water azeotrope with 26 percent water, is then  separated in
a liquid/liquid (1/1) separator into aqueous and organic phases.   The
aqueous phase from the 1/1 separator is combined with the bottoms from the
                                    18

-------
azeotropic steam stripper (Stream 12) and sent (Stream 11) to an aqueous
phase steam stripper.  The overhead from the aqueous phase stripper
(Stream 13) contains some epichlorohydrin and is therefore recycled to the
azeotropic steam stripper.  The bottoms from the aqueous phase stripper is a
wastewater stream (Stream 14).  One producer does not use an aqueous phase
stripper and considers Stream 11 to be a waste stream.

     A portion of the bottoms from the organic phase stripper (Stream 19) is
heated in a boiler and recycled to the stripper (Stream 20).  The remainder
of the bottoms is sent to the final purification column where purified
epichlorohydrin is fractionated, removed overhead,  and condensed
(Stream 21).  A portion of the product stream (22)  and a portion of the
bottoms stream (Stream 25) are recycled to the column.  The remainder of the
product stream (Stream 23) is the final product epichlorohydrin.  The
remainder of the bottoms stream (Stream 26)  is a wastewater stream.

Emissions

     The preliminary processes for production of allyl chloride,
hypochlorous acid, and dichlorohydrin involve no epichlorohydrin.  Hence, no
epichlorohydrin is emitted from either vents A or B or from any tanks,
valves, pumps, etc., used to store or transport materials in any of the
operations associated with Streams 1 through 8 in Figure 1.

     Most of the emissions from today's epichlorohydrin production
facilities are fugitive or storage losses.   Fugitive losses include those
from the numerous valves, flanges, pump seals, sampling ports, etc., found
in a production plant.  Fugitive emissions dominate because they are
numerous, and because other sources are fewer and/or well controlled.

     Many of the chemicals involved in the epichlorohydrin production
process are flammable and/or toxic; therefore, equipment is likely to be
well maintained for safety reasons.   One producer encloses all process and
                                     19

-------
tank sampling points in domes connected to a vacuum system.  Area monitors
detect leaks or spills of any chlorinated hydrocarbon.  The detection limit
for the monitors is less than 0.1 ppm.  In addition, personnel monitoring
and sampling of different areas of the plant are part of the industrial
u  •             6
hygiene program.

     Table 5 enumerates the types of fugitive epichlorohydrin emission
sources in plants which produce and/or finish epichlorohydrin.  These
sources and the various control methods used to minimize emissions from them
are described in Reference 7.

     In 1984, with the exception of one internal floating roof tank, all
epichlorohydrin at the production and/or finishing facilities was stored in
                 £ O Q
fixed roof tanks. ' '   One producer has the epichlorohydrin finishing
facility (refinery) separate from the production facility; therefore, crude
epichlorohydrin is shipped by boat from production to finishing.

     Process vent emissions constitute only a small fraction of the total
controlled epichlorohydrin emissions from a modern production facility.
Little information is available concerning the process conditions which
affect emissions of epichlorohydrin from process vents.   No information is
available on how (or whether) process upsets or startups affect these
emissions.

     The vent gas from the dehydrochlorination reactor (Vent C) is known to
contain some epichlorohydrin.  This vent can be the largest gaseous emission
source in the plant, but is not necessarily the largest  source of
epichlorohydrin emissions.  The composition of the vent  stream depends on
reactor design, operating conditions, and the type of alkali used.  If
carbonate is used as the alkali source, the large volume of carbon dioxide
formed entrains a large volume of epichlorohydrin vapors with it as it exits
the vent.  In this case, carbon adsorption or incineration is used to
control the vent VOC emissions.   However, hydroxide alkali may be used
instead to reduce greatly the vapors from this vent.
                                     20

-------




•— r
^*
H^
CO

*4*

w
"J


H
Z
W
»5£
&4
3 nJ
o* *
A
-S oo
O •
OS sO
fa co
w
CO M
Z H
O 1-4
H^ bj
CO M
CO O
I- <
.S fa
W

z z
3 5C
G CO
>-> 1— 1
X Z
O M
OS fa
o -^
— ; z
X O
c^!

e- u
W 3

a c
> 04

£—4
H- » 2

3 qi
j" r j i*"**)

r *\J
p—-t -*«
O O
C£
tft O
H-3 — :
O X
Qi U
|^3 t-H

C-O UJ



a

1-1

fv]
-^

CQ
^ti
H





|



1)

T*

U
D


e -^
•H C
-3 ki
>\ *n
J3 >\
O J3
u o
O ki
— O
~ .G
•*4 CJ
Q. -rt
Cjj o.
C
'•* 4_>
C
'ji n
4J CJ
c u
c a.
o
C. 4J
_O 00
u

u
.a
S
3
z






C
1 -J
« —

O
L4 W
4» 0)
fl

U 3*,





*f\ C
cj\ 11
1 U
\O Ut
r*» 01
O.

u
r- 3
I o
CM iy
^

4J
CNi CU
1 U
«M k4
— ' _«

C
O 11
— • ej
1 u
ui n


C 4-1
eg c
.= n
W CJ
en n
en a.
•^ in

















,,
jj.
,>\
H
ti
y
S
a.

3

'r]













•£ -T
1










O m
1 - —




CM CM
1




1 1





1 1





A
1 CM





_

y
C

«
p— ^ U
CO -U
o S
*& c \v
01 OJ i-4
* A J3
CJ U 3
eg at o
2* X &












en
rH
eg
n


a.
jp






i










i




i




I












I































en
^4
0
en
en
n
u
a.
o





Q.
00
u-\









S
"
— .


o




u-i
sr
fN




0






~
r^.
_

































en
n
00
c
eg

it.




y\ \c
CM e^
— •









f^- o




CM O
CO 0




o sr




c-l ffl







ex: >^














•o
-H
U 3
J! CJ-

C* i '
















en
n
4

>




m CM











1




I




I





1 1






in in














13
T-t
CJ 3
cn cer

^J ~-








en
n
n o
1-1 -*4
3 >
en 01
en £
01
LI w
3- n
r-l -^

< as




in
i










O
1 — .




in
I




1




CM
1






1 —














•o

3
en cr
•g -H
CJ -J




IT
i.
C

J~i
Zj
o
c
c
£
y

D

P.

u;




o o











O CTv




O O
rsj in




— O
_ -M





1






O r-.
1—













•73

3
a c-
S3 -H
w J




^
T3 C
:fl K
OJ >

•i-t *
_; ^
M

n 3
•o o.
c

,
e as
n •
O. 01

X
e c
— eg O

" '— i i-i efi *"
*g 4-» X 50

— i -J) S 11
n 4j ^ i> 3 —
j: C C eg
10 3 -H 00 01 0
O > C -C --J
-a cj u — i 4j g
c V >* n
eg n en ^ -4 j;
en o. O cj
— • n en -H u^
CJ H C3 — ' clO O
•H cu 3 c a
E en g -H
Oj • u
U C eg J3 3 M
3 M .H -a T3 5
o o. c « eg -4
O CO 4J 4J
50 .c eg T3 eg
" C CJ "4 C U
en -4 cu 3 eg n
>- j: S cj a.
n en p-4 u o
3 C CM cj 3 n
•a 1-1 o ej
O u-i cu eu cj c
i-i i-i J3 -3
o. c a n c
-4 c a n
c u n eg 'H 4-
•H -a u cj > c
t-i >> n n •* »
•O J= ^ en -a eg T3
>•, 3 4J c s n
js -4 on ao
O O 4-1 -rf j: oo 50
u rH eg yi ij c 3
O — j: en — i — i
J3 »4 S ^2 3 "~
cj o. • en w "o ^
-( n >> c -^ • o
Q. 4J CO r^ C T3
01 » -^ 0» 0 U "T3
X i-t ECO) —1 01 5)
• t- -H — 1 U 4J 3 -o
c/^ ej IM k4 c *0 e:;
«co nT3n u n —
n i- s .c n a, -;
e« C 3 O <*< O "3
oo,oc-4n — t 4j
3 -i o as -4 3 a.

ncj-^™S £ 3J£
n n 3 j= J o t.
J3O^ ij — 4 U O iJOl
4J O 0. 1*4 C CO Ll
en u. « n . j= -3
-4-a.n "So04j
^o — i — i 4j • eg c en

ncg co jj • b u oc

— t c/3 3 ui ^^ cn -e3 *j p-4
eg cj en •!) eg
3eg<-i^: eju o.nn

en eg j3>n Q.OC
ij 4j a. n oo n n
cc en . -4 -DC 2C_

see "S'-M ~ i =j
occgo oc u o.
O.O. 4Jin -HCOC
5 S VM 01 U ~4 C
OOO3CQ.-3 en
ncj> ~~^o •£ •*

cOOu Oen"ejO 31 30
j: eg to CO — -4 -3
cn-j4j noii»-c x —
ki c •— — 4 cjn-3 —
nnno. u— i 11-3
A "^ en s e^o O a. en en
S31I cgc^n C o Jl
3 — 1 U X ~4 O — n
ccja. oiceniJ -r^ ej-j
c n -H u c 4j rei
n -H u c -o o n -4 en u
en eg u u ej -o n -r.
j= c o en o co n _o -~_ 5

eg" " ja ej -o"~
21

-------
     The vent from the azeotropic steam stripper (Vent D) is not a large
source of epichlorohydrin emissions.  Either a thermal oxidizer or a wet
scrubber can be used for control of emissions from this point.  Organic
emissions from the aqueous phase stripper (Vent E)  can be controlled by
incineration, carbon adsorption, or wet scrubbing.

     The vent from the organic phase stripper (Vent F) can be a large source
of VOC emissions.  The percentage of epichlorohydrin in this stream is not
known.  The vent from the purification column (Vent G) releases only a small
quantity of VOC emissions, but the stream consists  primarily of
epichlorohydrin.  Emissions from these vents can be controlled by flares,
incineration, scrubbing, carbon adsorption,  or the  use of refrigerated vent
condensers.

     One producer routes all process emissions to a thermal oxidizer-NaOH
scrubber unit.  Control efficiency is reported by the company to be
99.99+ percent.  No epichlorohydrin has been detected from this source.
Assuming a 0.1 ppm detection limit for the method used, calculations by the
producer indicate an epichlorohydrin emission rate  less than 79 kg/yr if the
facility is operated at full capacity.

     Another producer apparently routes all process emissions from
production to a single stack controlled by a vent condenser.   The producer
                                                        9
reports zero epichlorohydrin emissions from this source.    Emissions from
the corresponding separate finishing operations (1  percent epichlorohydrin)
                                                            Q
are routed to incinerators rated as 99.99 percent efficient.    The stripper
bottoms stream (Stream 26) is known to contain some epichlorohydrin.
However, this stream is treated by hydrolysis and biotreatment before
disposal.  The amount of epichlorohydrin which escapes from this source  is
not known.

Emission Factors

     Table 6 presents available emission factor data for  epichlorohydrin
production.  This table represents industrywide totals.
                                     22

-------
        TABLE 6.   EMISSION FACTORS  FOR THE RELEASE OF EPICHLOROHYDRIN
                  FROM EPICHLOROHYDRIN PRODUCTION
                              Emission Factor                %
     Source                        (g/kg)               Total Emission
Process Vents
Q
Storage Facilities
Fugitive Sources
TOTAL6
0.00047b'f
0.15
0.62
0.78
0.06b
19.9
80.0
100.0

 Grams of epichlorohydrin emitted per kilogram of epichlorohydrin,.produced.
 Based on a nationwide annual production rate of 191 Gg (420 x 10  Ib)
 epichlorohydrin11 and nationwide emission totals from Reference 5.  As
 such, these factors do not necessarily represent emission rates from any
 particular facility.

 These factors represent epichlorohydrin emissions after controls (thermal
 oxidizers/NaOH scrubbers, vent condensers, and incinerators) reportedly
 effecting 99.99 percent removal.

 Includes storage tanks (mainly fixed roof),  transfer operations, etc.
 Storage emissions from glycerin manufacture  are also included.

 Includes valves, flanges, pump seals, sampling ports, etc.   Emission
 factors are approximated from average VOC emission factors  for  SOCMI
 process components and represent a relatively uncontrolled  facility
 where no significant leak detection and repair,programs are in  place
 to limit fugitive emissions.  One manufacturer  uses area monitors and
 employee exposure monitors to detect spills  or leaks and has equipped
 all process and tank sampling points with an enclosed dome  connected
 to a block vacuum system, which directs the  epichlorohydrin vapors to
 a thermal oxidizer.  No estimate is available of the the effectiveness
 of these measures.

p
 Independent numerical roundoff may affect totals.

 Reference 12 indicated that epichlorohydrin  emissions to air from reactor
 vent gas were 1.5 g/kg epichlorohydrin produced.  The reference does not
 indicate whether this situation is for controlled or uncontrolled
 emissions.  Besides the reactor vent gas, no other epichlorohydrin
 emission sources were identified.
                                     23

-------
     Data are not available to describe accurately the emissions from all
operations associated with epichlorohydrin production.  When inventorying
emissions of epichlorohydrin from such facilities, source-specific
information should be obtained to determine the existence of emitting
operations, control equipment, and emission levels.

Source Locations

     As of 1984, only two companies are producing epichlorohydrin:  Shell
Oil Company in Norco, LA, and Dow Chemical Company in Freeport, TX.  Crude
epichlorohydrin from the Shell facility is finished  (refined)  at Shell's
facility in Deer Park, TX.  Of the epichlorohydrin finished at the Deer Park
Facility, some is used directly there for the production of epoxy resins and
                                         269
some is sold as finished epichlorohydrin. ' '

INADVERTENT PRODUCTION OF EPICHLOROHYDRIN IN OTHER INDUSTRIAL  PROCESSES3

     Epichlorohydrin can be produced as a byproduct  during the manufacture
of a number of other chemicals.  Therefore, it can be emitted  not only from
processes which manufacture these other chemicals, but also (because it is a
contaminant in these chemicals) from processes which use these chemicals as
feedstock and from other uses (solvent, etc.) of these chemicals.  No
information is available on the amount of epichlorohydrin emissions from
this source.

     The most likely precursors of epichlorohydrin are:

          •    Allyl Chloride
          •    2,3 - Dichloropropene,
          •    1,3 - Dichloro -2-propanol,
          •    1,3 - Dichloropropene,
          •    Tris (Dichloropropyl)  Phosphate,  and
          •    Glycerin.
                                     24

-------
REFERENCES FOR SECTION 4

1.   Peterson, C. A., Jr.  Glycerin and Its Intermediates (Allyl Chloride,
     Epichlorohydrin, Acrolein, and Allyl Alcohol).   (Prepared for U.  S.
     Environmental Protection Agency, EPA-450/3-80-028e.)  IT Enviroscience.
     Knoxville, TN.  December 1980.

2.   Letter from R. R. Kienle, Shell Oil Company, Houston,  TX, to
     T. F. Lahre, U. S. EPA, Research Triangle Park,  NC.   February 7,  1985.

3.   Syracuse Research Corporation.  Investigation of Selected Potential
     Environmental Contaminants:  Epichlorohydrin and Epibromohydrin.
     (Prepared for U. S. Environmental Protection Agency, PB80-197585.)
     Syracuse, NY.  March 1980.

4.   Hydroscience, Inc.  Trip Report:  Dow Chemical Company,  Freeport, TX.
     1978.  As cited in Engineering Analysis of Epichlorohydrin Production
     Process, Preliminary Document.  (Prepared for U. S.  Environmental
     Protection Agency, Contract No. 68-02-3171.)  Radian Corporation,
     McLean, VA.  September 1981.

5.   Memorandum entitled "Epichlorohydrin Emissions  Summary:
     Epichlorohydrin Source Assessment" from Jeffrey  A.  Shular,  Midwest
     Research Institute, Raleigh, NC, to David Beck,  EPA, Research Triangle
     Park, NC.  February 16, 1984.

6.   Nonconfidential portions of letter from S.  L.  Arnold,  Dow Chemical,
     U.S.A., Midland, MI, to Jack R. Farmer, EPA, Research  Triangle Park,
     NC. December 8, 1983.

7.   VOC Fugitive Emissions in Synthetic Organic Chemicals  Manufacturing
     Industry—Background Information for Proposed Standards.
     EPA-450/3-80-033b.  U. S. Environmental Protection Agency.   Research
     Triangle Park, NC.  November 1980.

8.   Nonconfidential portions of letter from R.  R.  Erickson,  Shell Oil
     Company, Deer Park, TX, to David Beck,  EPA, Research Triangle Park, NC.
     December 27, 1983.

9.   Nonconfidential portions of letter from W.  L. Caughman,  Jr.,  Shell Oil
     Company, Norco, LA, to Jack R. Farmer,  EPA, Research Triangle Park, NC.
     October 13, 1983.

10.  Texas Air Control Board.  Permit Applications for Shell  Chemical
     Company, Deer Park, TX.  Austin, TX.

11.  Chemical Products Synopsis.  Manville Chemical Products.  Cortland, NY.
     December 1982.
                                    25

-------
12.   Assessment  of  Epichlorohydrin Uses, Occupational Exposure, and
     Releases.   Dynamac  Corporation,  Rockville, MD.  Prepared for  the Office
     of Toxic  Substances,  U.  S.  Environmental Protection Agency, Washington,
     DC.   Prepared  under EPA  Contract No. 68-02-3952.  July 1984.
                                     26

-------
                                   SECTION 5
      EMISSIONS FROM INDUSTRIES WHICH USE EPICHLOROHYDRIN AS A FEEDSTOCK

     About 85 percent of the epichlorohydrin produced in the United States
is used as a feedstock in the production of synthetic glycerin and epoxy
resins.  These production processes, along with the epichlorohydrin
emissions associated with them, are described in this section.  Other uses
of epichlorohydrin include the manufacture of elastomers, glycidyl ethers,
wet strength resins, surfactants, water treatment resins, and other
products.  (See Table 3).  Because of inadequate information, these other
products are not discussed in this section.  The locations of a few known
producers of these other products are listed at the end of the section.

PRODUCTION OF SYNTHETIC GLYCERIN

     Synthetic glycerin is a chemical intermediate used in the production of
materials such as alkyd resins (for paints), cellophane and meat casings,
tobacco (including triacetin), explosives and other military products,
drugs, toothpaste, cosmetics, monoglycerides and foods, and urethane foams.
The largest single use for synthetic glycerin is in the production of
tobacco materials, which consumes 18 percent of all synthetic glycerin
generated.  Synthetic glycerin is in direct competition for use with natural
glycerin.  In many cases synthetic glycerin is preferred because of its
lower moisture content.  Synthetic glycerin has been used exclusively for
polyols and urethane foams because of its lower moisture content.
                                     27

-------
Process Description





     Crude epichlorohydrin is piped directly from the dehydrochlorination


reactor of the epichlorohydrin production plant (Stream 9 of Figure 1).


Without further purification, the crude epichlorohydrin is blended with a


large volume of dilute aqueous sodium carbonate,  then heated in a hydrolyzer

                                                                          2
to convert epichlorohydrin to glycerin by the following reaction sequence.
     CH0 — CH-CH.C1 + H00 --> CH,-CH-CH0C1
       2   /    2      2      |  2 ,     2


          0                    OH  OH



     Epichlorohydrin   Water    a-monochlorohydrin



     2CH,-CH-CH_C1  +  Na.CO.  — >  2CH.-CH — CH0  +  2NaCl  +  C00 +  H-0
        2      2         23          2       ,  2                 22
                                                       Sodium   Carbon  Water

                                                      chloride  dioxide
OH OH OH 0
a-monochloro- Sodium
hydrin carbonate
CH--CH--CH,, + H00 — > CE,-
2 \/ 2 2 2
OH 0 OH
Glycidol
CH-CH2
OH OH
Glycidol Water Glycerin
     Another reaction sequence which can be used involves the addition of


aqueous sodium hydroxide.
CH0—CH—CH0C1

  V

Epichlorohydrin
     CH_— CH—CH0C1
                2
      OH   OH



   a-monochlorohydrin
                          H0  —>   CH.--CH—CHC1
Water



 NaOH
                     Sodium

                    hydroxide
  OH   OH



a-monochlorohydrin



->   CH2—CH—CH2



      OH    OH   OH



       Glycerin
                                                     NaCl
                                 Sodium

                                chloride
                                     28

-------
     No information is available concerning the predominance of either of
these sequences or concerning how significantly the choice of sequence affects
emissions from the process.  Further description here of the process assumes
the first sequence.

     A generalized flow diagram for the production of synthetic glycerin from
crude epichlorohydrin is given in Figure 2.  The production stream from the
hydrolyzer contains about 20 to 25 percent glycerin and 10 to 15 percent salt.
Excess sodium carbonate is neutralized with hydrochloric acid; the carbon
dioxide thus generated is captured in a carbon dioxide adsorber with dilute
sodium hydroxide as the absorbing liquid.  Aqueous sodium carbonate formed in
the absorber is routed to the sodium carbonate storage area for use in the
hydrolyzer.

     Multiple-effect evaporators remove some of the water from the product
stream, thus causing the salt to crystallize.  The resulting raw
glycerin/salt/water slurry is centrifuged to remove the salt crystals; a water
rinse in the centrifuge washes residual glycerin from the salt crystals.   The
centrifuged product stream at this point is about 45 percent glycerin.  A
repeat of the evaporation and centrifugation steps raises the glycerin
                                               2
percentage in the solution to about 85 percent.

     Finishing involves distillation, solvent extraction (with acetone
solvent) and, as the final step, activated carbon adsorption to remove trace
                            2
impurities and color bodies.

Emission Factors

     Most vents release inert gases and water vapor with no significant VOC of
any kind.  The vent stream from the CO- absorber is routed to the thermal
oxidizer/NaOH scrubber unit in the epichlorohydrin section.   This unit has  a
                          4
99.99+ control efficiency.    There is no evidence of emissions of
                                                                          3
epichlorohydrin from the process during upsets, i.e.,  system malfunctions.
                                    29

-------
PRODUC
GLYCERI







t—



c-1
o
CJ
CM
Z



r-

t w 4— tr co >
O co o M
~3 5 *~ 2
UJ «UJ


cc
UJ
< 	 ex.03
0 =
0°
CQ
t" „
9 8"


^Ul
kl '1 <
™ fsj^"
t
Q- n
ll

i

^



A j
0



j

j

CRUDE

k

k



k



L

L
Z
EPICHLOROHYDI
FINISHING

EVAPORATIOI>
AND
NTRIFUGATK
ALT REMOVA
o-
IT
UJ
N

Ul
Z


5*
i ^

HYDROLYZEP



                                                     4-1
                                                     o
                                                     o
                                                     3

                                                     O

                                                     Q.

                                                     OJ
                                                     OJ
                                                     CD
                                                     3    •
                                                          C
                                                     01  -r-l
                                                      E   o
                                                          >j
                                                     4J   O
                                                      «  r-4
                                                     X  Si
                                                     •U   O
                                                     4-1   O
                                                      a]   M
                                                      Jj  4-1
                                                      0)
                                                      o.  c
                                                      O  -H
                                                          (-
                                                      o   a>
                                                     •r-l   U
                                                      W   >,
                                                      03  r-t
                                                     M   eo
                                                      0)
                                                      S-i
                                                      3
                                                      60
                                                     •r-l
                                                     ft.
30

-------
     Glycerin production plants must be well maintained to protect personnel
                                                                      2
from the highly flammable and toxic chemicals involved in the process.   All
process and tank sampling ports are enclosed in domes connected to a vacuum
system.  Area monitors detect leaks or spills of any chlorinated hydrocarbons.
Personnel monitors and sampling of different areas of the plant are part of an
                           4
industrial hygiene program.
Source Locations

     Recently, Dow Chemical Company, Freeport, TX, has been the only U.S.
                                          4 5
producer of glycerin from epichlorohydrin. '

PRODUCTION OF EPOXY RESINS (CONTINUOUS PROCESS)

     The term "epoxy resin" applies to polymeric materials which contain
epoxide groups.  A curing or hardening agent converts the resin to a thermoset
material.  More than 90 percent of the total U.S. production of unmodified
epoxy resins is produced by reacting epichlorohydrin with bisphenol A
[2,2-di(4-hydroxyphenyl) propane].  These resins, known as diglycidyl ethers
of bisphenol A, may vary from low-viscosity liquids to high-melting solids,
depending on the ratio of epichlorohydrin to bisphenol A.  Production of the
liquids requires 0.68 Ib epichlorohydrin per pound of product;  production of
solids requires 0.47 Ib/lb.

     Other active hydrogen compounds are also reacted with epichlorohydrin to
form epoxy resins:  epoxy novolac resins require 0.73 Ib epichlorohydrin per
                                                      3
pound of product and phenoxy resins require 0.4 Ib/lb.   Because of the  small
volumes involved, epoxy and other resins are more likely to be  made in batches
rather than by continuous processes.   They are discussed in the next
subsection along with other products made in batches.
                                    31

-------
Process Description

     Figure 3 is a generalized flow diagram  for  the  continuous production of
epoxy resins from epichlorohydrin and bisphenol  A.   Raw materials are
contacted in a reactor to form an organic  resin  solution product and an
aqueous brine by-product.  Available reaction  data indicate  that the resin
forms by the following reactions.

2H.C — CHCH0C1  +  HO - (o) - C(CH_)0 - (o) -  OH   -->
  2 \   /    2            ^-s       3 2   \-V
     0
Epichlorohydrin     Bisphenol A
                          CH?C1CHCH70 - (o)  -  CCCH,.)..,  - (o)  -OCH?CHCH?C1   (A)
                            Z  I    Z     \—s       J  Z    \—J      Z |    2.
                               OH                                 OH
                                    Chlorohydrin Intermediate
CH0C1CHCH00 - Co) - C(CH.)0 - (o) - OCH0CHCH0C1   +   NaOH  —>
  2  I    2.    \—f       J 2.   ^—t      2. i    /
     OH                                 OH
          Chlorohydrin Intermediate                  Sodium
                                                   hydroxide
H0C—CHCH^O - (o^-C(CH.)9 - (o) - OCH^CH —  CH9  + 2NaCl  +   H00   (B)
 2 \ /    2.    \_y     j 2.   N—t      2.  \  /   2.                2.
   0                                     0

       Ether with Terminal Epoxy Groups            Sodium    Water
                                                  chloride
(n + 1)H2C—CHCH20-{o)-C(CH3)2-(o)-OCH2CH—CH2 + n HO-^)-C(CH3)2-{o)-OH —>
          0                              0

           Ether with Terminal Epoxy Groups                Bisphenol A

H0C—CHCH0	0-{o)-C(CH,)»-0-(o)	0-/o)-C(CH  ) -{oVoCH7CH~CH          (C)
 2. \ /    2.       \—f     J Z    *—'       >—'     -3 ^  >—'     i.  \ /  £
   0        •-                     -*n                          0

                   Epoxy  Resin
                                     32

-------
                                                o
                                                a
                                               1 O
                                                ac
                                                a.
 o    •
•H  ^-s
4J   >^
 U   4-J
 3  -H

 O  -H
 M   O

    U-l
 en
 3  T>
 o   o>
 3  rH
 C  r-t
•H   O
u   !-i
 C   -U
 O   C
 O   O
     O
 CU   I
                                                              fl   S
                                                              t._4  \.^

                                                              'O   C3
                                                              0)  -H
                                                              W   >-i
                                                              3  -a

                                                              a;  x
                                                              J3   O
                                                              6  x.
                                                                   CJ
                                                              W  -H
                                                              co   a.
                                                                   g
                                                              cn   o
                                                              C   ^
                                                              O  <4-t
                                                              •H
                                                              0)   CO
                                                              a,  aj
                                                              o   )->

                                                              o   >,
                                                              •H   X
                                                              t/3   O
                                                              co   a.
                                                              pa   ai
                                                              CO

                                                              01
                                                              3
                                                              &0
33

-------
     Whether the above reaction sequence applies to all manufacture of epoxy
resins from epichlorohydrin and bisphenol A is not known.  One company
mentions the addition of acetone and sodium carbonate to the reactor along
with the epichlorohydrin, bisphenol A and sodium hydroxide raw materials.
The acetone serves as a solvent.  The purpose of sodium carbonate is not
known.  Toluene, methyl ethyl ketone or methyl isobutyl ketone may also be
used as solvents.
     The following is a description of epoxy resin production by one
    facturer.   It is not known hi
industry; no other was available.
manufacturer.   It is not known how representative this description is of the
     The resin solution product is concentrated in an atmospheric flasher,
then extracted with water to remove residual by-product.   Finishing involves
vacuum flashing, high-vacuum evaporation, and polish filtration with
diatomaceous earth.

     Water from the extraction section, combined with the aqueous byproduct
from the reactor, is steam stripped to recover volatile organics.  These
organics are recycled to the feed preparation area along with the overheads
from the atmospheric flasher.  Overhead water from the steam stripper is
combined with makeup water from the vent recovery system to be used as solvent
by the extraction unit.

     The atmospheric flasher and steam stripper share an overhead system
vented to the vapor recovery system.  All other equipment in the extraction
section either is operated full of liquid (with no vent)  or is connected to
the overhead system.

Emission Factors

     For the well controlled facility shown in Figure 3,  the vapor recovery
vessel is the only source of process emissions.  One company reports
                                      34

-------
98 percent control efficiency for the vapor recovery system and epichloro-
hydrin emissions of 0.005 g/kg of product from this vent.    There is virtually
                                                                   Q
no chance for upset emissions of epichlorohydrin from this process.   Data on
epichlorohydrin process emissions from an uncontrolled facility were not
available.
     Fugitive emissions from pumps, valves, flanges, etc.  account for almost
88 percent of the total epichlorohydrin emissions from epoxy resin
           9
production.   Fugitive emissions of epichlorohydrin are minimized because of
its high flammability and toxicity.  At least one company  leak-proofs with
                                                                       Q
pressurized double seal systems all pumps which handle epichlorohydrin.    The
dominance of fugitive emissions is due in part to the controls in place  on
other emission sources.
     Several techniques are used in the industry for control of emissions from
storage and handling:

     •    Maintain storage tanks at constant level and high pressure relief
          setting;

     •    Vent storage tanks to the vapor recovery system;

     •    Vent storage tanks to a balloon header which breathes in and out;

     •    Use floating roof tanks;

     •    Control emissions from operational abnormalities  and refilling  after
          maintenance with a thermal oxidizer;  and

     •    Vent emissions to carbon adsorption unit if balloon system becomes
          overpressured.
                                     35

-------
One plant estimates epichlorohydrin storage emissions of 0.3 g/kg (control
methods used not given).  This same company predicts no epichlorohydrin
storage emissions when storage tanks are vented to a balloon header.  A carbon
                                                                    Q
adsorption unit is used if the balloon system becomes overpressured.   As an
industry average, storage emissions account for less than 4 percent of all
                                                       9
epichlorohydrin emissions from epoxy resin manufacture.
Source Locations

     As of 1983, only two companies were producing epoxy resins from
epichlorohydrin in a continuous process:  Shell Oil Company, Deer Park, TX,
and Dow Chemical Company, Freeport, TX.  Shell Oil also produces epoxy resins
by a batch process.  Shell's resin is known as EPON^  while Dow's is known as
D.E.R.®

PRODUCTION OF EPOXY RESINS AND OTHER PRODUCTS FROM EPICHLOROHYDRIN
(BATCH PROCESS)

     A number of products are made from epichlorohydrin by batch processes
because the amounts involved are small.  Among these are various epoxy resins,
phenoxy resins, polyamine and polyaminoamide (wet-strength) resins, glycidyl
ether (a surfactant), and elastomers.

     The epichlorohydrin is usually almost totally consumed in the reaction;
therefore, process vents are not a major source of emissions.   For this reason
and because process data are scarce, the following discussion focuses on the
epichlorohydrin feed facilities.

Process Description

     The industry practices described below are compiled from data collected
from several producers.    A general flow diagram for batch production of
epoxy resins is shown in Figure 4 while a probable flowsheet for
epichlorohydrin elastomer is illustrated in Figure 5.
                                     36

-------
  e
  e
• 9

I!

                         w to

                         "a
                         ui
                           11
                                                                                          C

                                                                                          W
                                                                                          0)
                                                                                          l-l
                                                                                          X
                                                                                          o
                                                                                          a,
                                                                                          C
                                                                                          o
                                                                                          o
                                                                                          3
                                                                                          T3
                                                                                          O
                                                                                          J-i
                                                                                          a.
                                                                                          
-------
                                                   C1J

                                                   o
                                                   _1
                                                   CO
                                                   to
                                                    C

                                                    s-
                                                    o
                                                    s-l
                                                    o
                                                   •-1
                                                    a,
                                                    0)

                                                   14-i
                                                    o


                                                    o
                                                    c
                                                    3
                                                   13
                                                    O
                                                    S-i
                                                    a.

                                                    0)
                                                   JS
                                                    o
                                                    OJ
                                                   "
                                                    3
                                                    O
                                                    0)

                                                   I
                                                    cu
                                                    S-,

                                                    00
38

-------
     Epichlorohydrin is commonly supplied to the facilities by railcar, then
transferred to a storage tank.  It may then be transferred to a weigh tank, to
a blend tank, or directly to the reactor.  Transfer is normally via closed
pipeline and sometimes under nitrogen blanket, although it may simply be drawn
from drums by vacuum.

     The number of process sources of epichlorohydrin emissions varies widely.
The reactor vent is the primary—and often the only—process source.  This
vent may be uncontrolled or it may be controlled by vent condenser, packed
tower scrubber, incinerator, pressure vacuum vent valve, or other device.
Several producers report epichlorohydrin emissions from this vent only while
the reactor is being charged; others report more continuous emissions.
Epichlorohydrin content in process streams can be reduced to less than
1 percent after the reactor.

     Only one epoxy resin manufacturer, as shown in Figure 4, provides for the
handling of epichlorohydrin stripped from the product:  this epichlorohydrin
is stripped from the product by vacuum distillation and stored in "wet" tanks
for reuse.  Epichlorohydrin in the "wet" tanks is covered by a layer of water,
which is eventually removed through an overflow weir, mixed with alkaline
wastewaters (which hydrolizes the epichlorohydrin) in a covered surge tank,
and released to the sewer.  The epichlorohydrin is recycled to the weigh
tanks.

     Other possible process sources for epichlorohydrin emissions include
transfer of the reaction mixture to another vessel and product refinement.
These sources are typically very minor.

     Epichlorohydrin emissions are also associated with feed storage and
handling.  Breathing losses are caused by expansion and contraction of the
vapors within the storage vessel.  Working losses are associated with the
filling and emptying of the vessel.  Fixed roof tanks are standard for storage
of the epichlorohydrin feed, though one plant used closed head drums.
                                     39

-------
Emissions from the tanks may be controlled by nitrogen blanket, water blanket,
or pressure-vacuum vent valve.  As mentioned above, epichlorohydrin is
normally transferred via pipeline or under nitrogen blanket.   Epichlorohydrin
is drawn from the drums by vacuum.

     Fugitive sources may account for much of the epichlorohydrin emissions
from these processes.  Some producers have no system for controlling leaks
from pumps, compressors, flanges, valves, and sample connections.  In some
plants, the vacuum pumps in epichlorohydrin service are equipped with liquid
seals which serve to reduce epichlorohydrin vapor discharges.   Two producers
report no fugitive emissions of epichlorohydrin.   One attributes this absence
of emissions to the control measures used, including sealed magnetic drives on
all pumps which transfer epichlorohydrin.  The other, a producer of polyamide
resins, attributes it to a negative pressure (3.92 in Hg.  abs.) within the
process.  At this negative pressure, any leakage  would be  into the process
from the atmosphere.

     Wastewater and solid waste streams containing epichlorohydrin may be
associated with these processes.  No analyses are available for
epichlorohydrin emissions from these sources.  However, because
epichlorohydrin is readily hydrolized, emissions  are presumed  to be minimal.

Emission Factors

     Table 7 presents epichlorohydrin emission factors for batch processes
which use epichlorohydrin as a feedstock.  These  emission  factors were
calculated from emission and throughput totals for the polyamide resin,
elastomer, and surfactant industries and are only broad averages, not
representative of any particular facilities.

     Fugitive emissions are by far the largest category of epichlorohydrin
emissions from these industries, constituting 87  percent of the total
emissions from this sector.  As indicated in Section 3, fugitive emissions may
                                     40

-------
       TABLE 7.  EMISSION FACTORS FOR THE RELEASE OF EPICHLOROHYDRIN
                 FROM BATCH PROCESSES WHICH USE EPICHLOROHYDRIN AS A
                 FEEDSTOCK
                                Emission Factor                 %
       Source                       (g/kg)               Total Emissions
Process Vents
Storage Facilities
Fugitive Sources
TOTAL
0.56
0.47
7.10
8.13
7
6
87
100

Grams of epichlorohydrin emitted per kilogram of epichlorohydrin used.
Calculated from aggregate emission and throughput totals for various resin,
elastomer, and surfactant producers which use epichlorohydrin as a raw
material in batch processes, as given in References 9 and 12.  These
factors do not represent the emission rates at any particular facilities.


Includes storage tanks and transfer operations.


Includes pump seals, compressors, flanges, valves, pressure relief devices,
sample connections and open-ended lines.  Calculations are based on the use
of average VOC fugitive emission factors for SOCMI process components
representing emissions from relatively uncontrolled facilities where no
significant leak detection and repair programs are in place for fugitive
emission control and were extrapolated to an average plant schedule of
300 days per year.  Because epichlorohydrin does not flow continuously
through process  components in batch processes, these factors may be
overstated.  Also, some manufacturers use various measures to detect and
reduce fugitive emissions.  Some use personal monitors on employees and
perform regular area monitoring.  One manufacturer uses sealed magnetic
drives on pumps in epichlorohydrin service while another maintains the
process system at negative pressure so all leaks from pumps, flanges, etc.
are vented back to the process. .No estimates are available of the
effectiveness of these measures.
                                    41

-------
dominate not necessarily because they are large,  but because other sources are
fewer and/or well-controlled.   Also, the fugitive emission factors given in
Table 7 may be overstated because they do not account for the fact that in
batch processes the epichlorohydrin is not flowing continuously through
                   9
process components.

Source Locations

     Table 8 gives the names and locations of some producers of epoxy resins
(batch), polyamide-epichlorohydrin resins, epichlorohydrin elastomers, and
surfactants made from epichlorohydrin.  No claim of completeness is made for
this table.
                                      42

-------
          TABLE 8.  SOME PRODUCERS OF EPICHLOROHYDRIN PRODUCTS
                                                              9,13
           Product-Company
  Location
     Epoxy Resins (Batch Process)
          Celanese Corporation

          Ciba-Geigy Corporation

          Union Carbide Corporation

          Shell Oil Company

     Polyamide-Epichlorohydrin Resins
          Borden


          Diamond Shamrock

          Hercules
          Georgia-Pacific


          Rohm and Haas

     Epichlorohydrin Elastomers
          Hercules

          B.F. Goodrich

     Surfactants
          Proctor & Gamble


     Flame Retardents
          Stauffer Chemical
Louisville, KY

Tom's River, NJa

Bound Brook, NJ

Deer Park, TX
Demopolis, AL
Sheboygan, WI

Charlotte, NC

Chicopee, MA
Hattiesburg, MS
Milwaukee, WI
Portland, OR
Savannah, GA

Peachtree City, GA
Eugene, OR

Philadelphia, PA
Hattiesburg, MS

Avon Lake, OH
Cincinnati, OH
Kansas City, KS
Gallipolis Ferry, WV
 Ciba-Geigy has announced plans to close this facility and add 45Gg
 (100 million Ibs) of capacity to its Mclntosh,  AL,  facility.  1

NOTE:  This listing is subject to change as market conditions  change,
facility ownership changes, plants are closed down,  etc.   The  reader should
verify the existence of particular facilities by consulting current listings
and/or the plants themselves.   The level of epichlorohydrin emissions  from
any given facility is a function of variables such as  capacity,  throughput,
and control measures, and should be determined through direct  contact  with
plant personnel.
                                     43

-------
REFERENCES FOR SECTION 5

1.   Assessment of Epichlorohydrin Uses, Occupational Exposure, and
     Releases.  Dynamac Corporation, Rockville, MD.  Prepared for the Office
     of Toxic Substances, U. S. Environmental Protection Agency, Washington,
     D.C.  Prepared under EPA Contract No. 68-02-3952.  July, 1984.

2.   Peterson, C. A., Jr.  Glycerin and its Intermediates (Allyl Chloride,
     Epichlorohydrin, Acrolein, and Allyl Alcohol).  (Prepared for U. S.
     Environmental Protection Agency, EPA-450/3-80-028e).  IT Enviroscience.
     Knoxville, TN.  December, 1980.

3.   Syracuse Research Corporation.  Investigation of Selected Potential
     Environmental Contaminants:   Epichlorohydrin and Epibromohydrin.
     (Prepared for U. S. Environmental Protection Agency, PB80-197585).
     Syracuse, NY.  March, 1980.

4.   Nonconfidential portions of  letter from S. L. Arnold, Dow Chemical,
     U.S.A., Midland, MI, to Dave Beck, U. S. EPA, Research Triangle
     Park, NC.  December 27, 1983.

5.   Chemical Products Synopsis.   Manville Chemical Products.  Cortland, NY.
     December, 1982.

6.   Bales, R. E.  Epichlorohydrin Manufacture and Use Industrial Hygiene
     Survey.  (Prepared for U.S.  Department of Health, Education, and
     Welfare, National Institute  for Occupational Safety and Health,
     Contract No. 210-75-00064.)   Tracer Jitco, Inc.   Rockville, MD.
     February, 1978.

7.   Texas Air Control Board.  Permit Applications for Shell Chemical
     Company, Deer Park, TX.  Austin, TX.

8.   Texas Air Control Board.  Permit Applications for Dow Chemical Company,
     Freeport, TX.  Austin, TX.

9.   Memorandum entitled "Epichlorohydrin Emissions Summary:
     Epichlorohydrin Source Assessment" from Jeffrey  A.  Shular,  Midwest
     Research Institute, Raleigh, NC, to David Beck,  EPA, Research Triangle
     Park, NC.  February 16, 1984.

10.  Nonconfidential portions of  letters submitted to J. R.  Farmer,  Emission
     Standards and Engineering Division, Office of Air Quality Planning  and
     Standards, U. S. EPA, Research Triangle Park, NC, by industrial users
     of epichlorohydrin.  October - December 1983.

11.  Letter from H. H. Flegenheimer, Celanese Corporation, Louisville, KY,
     to Dave Beck, U. S. EPA, Research Triangle Park, NC.   January 6,  1984.
                                     44

-------
12.   Memorandum entitled "Calculations  of  Gas  Velocities  and  Storage  and
     Fugitive Emissions:  Epichlorohydrin  Source Assessment"  from
     Jack R.  Butler and  Jeff  Shular,  Midwest Research  Institute,  Raleigh,
     NC,  to David Beck,  U.  S.  EPA,  Research Triangle Park,  NC.  February  17,
     1984.

13.   SRI  International.   1985  Directory of Chemical Producers - United
     States.  Menlo Park, CA.   1985.
                                     45

-------
                                   SECTION 6

         EMISSIONS FROM THE USE OF EPICHLOROHYDRIN-CONTAINING PRODUCTS

     Trace levels of epichlorohydrin residues may be contained in products
manufactured from epichlorohydrin feedstocks.  During the use of these
products, volatilization of the epichlorohydrin is possible under certain
temperature and pressure conditions, thereby resulting in potential
atmospheric emissions of epichlorohydrin.  The potential for such emissions is
discussed in this section in connection with the use of epoxy resins,
glycerin, elastomers, and wet-strength resins.

USE OF EPOXY RESINS

     Epichlorohydrin residue levels in epoxy resins have been determined from
resin manufacturers through the use of product technical bulletins and
material safety data sheets.  A Shell Oil Company technical bulletin on its
epoxy resins and reactive diluents (produced from epichlorohydrin) indicates
that trace levels of epichlorohydrin are contained in these products.   Many
EPON^ resins are sold under manufacturer's product specifications that the
                                                                 1 2
epichlorohydrin content cannot be greater than 5 ppm (by weight) . '    Shell
has indicated that epichlorohydrin levels in most EPOrh^ resins are generally
                           1 2
in the 1 ppm - 2 ppm range. '   However, specialty resins and reactive
diluents do exist that have epichlorohydrin levels ranging from 10 ppm -
1,400 ppm.

     One manufacturer has empirically determined the vapor concentrations of
epichlorohydrin above epoxy resins that would result from the exposure of
epichlorohydrin-containing resins to air under various temperature conditions.
These experimental results are shown in Table 9.  As expected, the higher the
                                    46

-------
    TABLE 9.   EPICHLOROHYDRIN VAPOR CONCENTRATIONS ABOVE EPOXY RESINS AT
              VARIOUS TEMPERATURES UNDER STATIC EQUILIBRIUM CONDITIONS1
                               Epichlorohydrin Levels  in the Resin,  ppm,  wt,

                                     10              5              1


   Temperature,  °C (°F)           Epichlorohydrin Level in Vapor,  ppm, v/v

     27 (80)                           0.6            0.3            0.06

     49 (120)                          2              1              0.2

                      NOTE:   60°C (140°F) Recommended  Handling Temperature
71 (160)
93 (200)a
116 (240)
138 (280)
149 (300)
5
12
26
50
64
2.6
6
13
23
32
0.5
1.2
2.6
5.0
6.4

93°C (200°F)  is the maximum recommended handling temperature.
                                    47

-------
temperature during use, the higher the potential for epichlorohydrin
emissions.  It should be noted that these experiments gave results indicating
that the epichlorohydrin concentrations potentially occurring are independent
of the method of application (i.e., pouring, spreading, or spraying) of the
resin.  Epichlorohydrin air concentrations were found, however, to be
dependent on the surface area of resin in contact with air.

     Several verifications have been made in industrial situations of
epichlorohydrin emissions from the use of epoxy resins.  In one case,
structural steel members were being coated with an epoxy paint by hand
spraying.  Measured epichlorohydrin concentrations in air in the building
                                                 3             3
where spraying was occurring ranged from 2.4 mg/m  - 138.9 mg/m .  In several
OSHA tests of industrial processes involving epoxy resins for plastics
production and processing, epichlorohydrin concentrations of 0.01 ppm -
3.8 ppm were measured in workplace air.   These examples indicate the
potential for epichlorohydrin emissions from epoxy resins that contain
epichlorohydrin residues.

USE OF SYNTHETIC GLYCERIN
     Epichlorohydrin residues in glycerin have been determined to be very low
and thus the potential for epichlorohydrin emissions from volatilization
during glycerin use is practically nonexistent.  Dow Chemical has tested its
synthetic glycerin products and found no epichlorohydrin residues (detection
limit of 1.5 ppm). '   Epichlorohydrin residues in synthetic glycerin are
minimal or nonexistent because any residual epichlorohydrin is hydrolyzed
during the production process.  In addition, several high temperature
distillations are performed to purify crude synthetic glycerin that are very
effective at eliminating any potential residues such as epichlorohydrin.
Theoretical yield calculations involving the chemical reactions taking place
during synthetic glycerin production have been performed to estimate potential
epichlorohydrin residue levels.  These calculations give an estimate of
epichlorohydrin levels in synthetic glycerin of less than 1 part per
trillion.1
                                     48

-------
USE OF WET-STRENGTH RESINS

     Although quantitative data on the levels of epichlorohydrin contained in
wet-strength resins could not be identified in the literature, some amount of
epichlorohydrin residue is projected to occur in aqueous wet-strength resin
solutions.   If epichlorohydrin is contained as a contaminant in the resins,
it would most likely be released as a vapor during application of the resin to
paper and during paper manufacture dewatering and drying steps.  No workplace
or other epichlorohydrin emissions data related to wet-strength resin use
could be found.

USE OF ELASTOMERS

     The potential for epichlorohydrin residues to be found in crude
epichlorohydrin elastomers is significant because they are prepared with an
excess of epichlorohydrin.  However, the reaction of the residual
epichlorohydrin with vulcanizing agents and stabilizer additives and the
adsorption of residual epichlorohydrin by carbon black filler help prevent
excessive releases of epichlorohydrin during elastomer storage and use.  No
quantitative data on epichlorohydrin residue levels in epichlorohydrin
elastomers could be found in the literature.

     Tests of workplace air at elastomer processing plants have not detected
any epichlorohydrin.  Tested process areas included elastomer weighing
                                                                 1 4
stations, elastomer extrusion, two-roll mill mixing, and molding. '
                                     49

-------
REFERENCES FOR SECTION 6

1.   Assessment of Epichlorohydrin Uses, Occupational Exposure, and Releases.
     Dynamac Corporation, Rockville, Maryland.   Prepared for the Office of
     Toxic Substances, U. S. Environmental Protection Agency,  Washington,  DC.
     Prepared under EPA Contract No. 68-02-3952.   July 1984.

2.   Shell Oil Company.  Shell Chemical Company Technical Bulletin SC:
     106-82:7-15.  1982.  Houston, Texas.

3.   Letter from Arnold, S., Dow Chemical to Parris,  G., Dynamac Corporation.
     January 1984.  Information on Dow's synthetic glycerin.

4.   Hercules Chemical.  Herclor^ Epichlorohydrin Elastomers,  Hazards
     Associated with Epichlorohydrin Monomer in the Handling and Processing of
     Herclor^ Elastomers.  Hercules Bulletin ORH-24D.  1983.
                                     50

-------
                                   SECTION 7

                            SOURCE TEST PROCEDURES

     EPA is investigating source sampling and analytical procedures for
epichlorohydrin air emissions, but none have yet been published or
recommended.  The sampling and analysis methods presented in this chapter for
epichlorohydrin emissions represent those that have been published in the
literature as viable methods.  The presentation of these methods in this
report does not constitute endorsement or recommendation, nor does it signify
that the contents necessarily reflect the views and policies of the U.S. EPA.

LITERATURE REVIEW OF SAMPLING METHODS

     Adsorption onto activated charcoal is a preferred sampling method for
epichlorohydrin.  Silica gel has also been used as the adsorbent.   The
National Institute for Occupational Safety and Health (NIOSH) method number
SI 18 for epichlorohydrin prescribes the use of standard commercial tubes
containing 150 g of 20/40 mesh activated carbon in two sections:  100 g in the
front section and 50 g in the rear.  The two sections are divided by 2 mm of
polyethylene foam.  Samples as large as 20 liters collected at 200 ml/min are
allowed.  The epichlorohydrin is desorbed from the charcoal with carbon
disulfide.  Water vapor interferes with sample collection by displacing the
organic vapors.

     Impingers or bubblers containing distilled water or dilute sulfuric acid
have also been used to collect epichlorohydrin vapors.   In one method, a
2-liter sample is drawn through two bubblers in series at a rate of
0.5 liter/minute.  Each bubbler contains 8 milliliters of water.  During one
                                    51

-------
test with about 5.2 ppm epichlorohydrin the efficiency of the first bubbler
was 80 percent.  The efficiency of the second bubbler was not given.

     Plastic bags, glass bottles, aluminum foil-polyester laminate bags and
teflon bags have been used with varying rates of success as collection
devices.  Sampling duration is usually from a few seconds to two minutes.  In
one test in which the epichlorohydrin concentration ranged from 5 to 27 ppm,
samples in teflon bags suffered a 20 to 26 percent loss of epichlorohydrin
after 24 hours.  Samples in aluminum foil-polyester bags suffered a 19 to 40
percent loss under the same conditions.

LITERATURE REVIEW OF ANALYTICAL METHODS

     Gas chromatography has become the method of choice for separation and
analysis of organic materials because it is sensitive, specific, and suitable
for analysis of samples collected on charcoal.   The National Institute for
Occupational Safety and Health (NIOSH) method for determination of
epichlorohydrin calls for a flame ionization detector and a column packed with
10 percent carbon disulfide.  The overall NIOSH method operates over the
concentration range 11.7 to 43.1 mg/m  with a relative error of 0.7 percent at
      2
5 ppm.   Any other compound with the same retention time as epichlorohydrin
will interfere, but the interference can be eliminated by changing the
                      3
separation conditions.

     Colorimetry is the most common wet method for determination of
epichlorohydrin concentration.  Several methods involve hydrolysis of the
epichlorohydrin to glycol, then oxidation to formaldehyde.  For aqueous
solutions of epichlorohydrin, both hydrolysis and oxidation are usually
accomplished with periodic acid.  If the sample was collected in dilute
sulfuric acid, hydrolysis is accomplished by the sulfuric acid and periodic
acid is added for oxidation.  The formaldehyde may be reacted with sodium
arsenite or ammonia and acetylacetone reagent, Schiff's reagent, or
phenylhydrazine and potassium ferricyanide to form colored complexes.  The
                                     52

-------
sodium arsenite/acetylacetone method can detect as little as 20 yg
epichlorohydrin.  The Schiff's reagent method can detect 0.01 to 0.1 mg
epichlorohydrin in a 6-ml solution.  The phenylhydrazine/potassium
ferricyanide method has been used for epichlorohydrin concentrations of from
0.45 to 14 mg/m  in air with maximum error of 0.3 percent.   The
ammonia/acetylacetone method can detect as little as 6 ppm epichlorohydrin
                                 2
with an error of about 2 percent.   Formaldehyde or any substance which could
yield formaldehyde, such as ethylene oxide or ethylene glycol, will interfere
with these methods.  Many aldehydes will interfere with the Schiff's reagent
method.
     Practical and detailed methods for quantitative determination of
epichlorohydrin concentrations with infrared spectroscopy have not been
developed.  One source indicates a minimum detection limit of 3000 ppm.
Another source indicates measurement of 10 ppm with ± 2 percent precision and
                                          2
accuracy.  Terminal olefins can interfere.
     Advanced techniques have been applied to qualitative, rather than
quantitative, determination of epichlorohydrin.   For example,  photoelectron
spectroscopy can distinguish among epichlorohydrin, epibromohydrin,
epifluorohydrin, and other halo-oxygen compounds.   It is not suitable for
                            2
quantitative determinations.
                                    53

-------
REFERENCES FOR SECTION 7

1.   National Institute for Occupational Safety and Health (NIOSH).   Criteria
     for a Recommended Standard.   Occupational Exposure to Epichlorohydrin.
     U. S. Department of Health,  Education and Welfare.  (No date.)

2.   Syracuse Research Corporation.   Investigation of Selected Potential
     Environmental Contaminants:   Epichlorohydrin and Epibromohydrin.
     (Prepared for U. S. Environmental Protection Agency,  PB80-197585).
     Syracuse, NY.  March 1980.

3.   National Institute for Occupational Safety and Health (NIOSH).   NIOSH
     Manual of Analytical Methods, Second Edition.  Part II:  Standards
     Completion Program Validated Methods, Volume 2.  U. S.  Department  of
     Health, Education, and Welfare, Cincinnati,  OH.  April 1977.
                                     54

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/4-85-007J
4. TITLE AND SUBTITLE
Locating And Estimating
Of Epichlorohydrin
2.
Air Emissions From Sources
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS


12. SPONSORING AGENCY NAME AND ADDRESS
Office Of Air Ouality Planning And Standards (MD 14)
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
September 1985
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT f\
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERS
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
    EPA Project Officer:  Thomas F.  Lahre
16. ABSTRACT
       To assist groups interested in inventorying  air  emissions  of  various
  potentially toxic substances, EPA is preparing a  series  of  documents  such as  this
  to compile available information on sources and emissions of  these substances.
  This document deals' specifically with epichlorohydrin.   Its intended  audience
  includes Federal, State and local air pollution personnel and others  interested
  in locating potential emitters of epichlorohydrin in  making gross  estimates  of  air
  emissions therefrom.

       This document presents information  on 1) the types  of  sources that may  emit
  epichlorohydrin, 2) process variations and release points that  may be expected
  within these sources, and 3) available emissions  information  indicating the
  potential for epichlorohydrin release into the air from  each  operation.
17.
a.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Epichlorohydrin
Emissions Sources
Locating Air Emission Sources
Toxic Substances
18.
DISTRIBUTION STATEMENT
b. IDENTIFIERS/OPEN ENDED TERMS

19. SECURITY CLASS /This Report)
20 SECURITY CLASS /This page/
c. COSATl Field/Group

21. NO. OF PAGES
60
22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION is OBSOLETE

-------
                                                         INSTRUCTIONS

   1.   REPORT NUMBER
       Insert the EPA report number as it appears on the cover of the publication.

   2.   LEAVE BLANK

   3.   RECIPIENTS ACCESSION NUMBER
       Reserved for use by each report recipient.

       TITLE AND SUBTITLE
       ''"itle should indicate clearly and briefly the subject coverage of the report, and be displayed prominently.  Set subtitle, if used, in smaller
        1 pe or otherwise subordinate it to mam title. When a report is prepared in more than one volume, repeat the primary title, add volume
         mber and include subtitle for the specific title.

   5.   REPORT DATE
       Earh report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
       jpi- oval, date of preparation,  etc.).

   6.   PERFORMING ORGANIZATION CODE
       Leave blank.

   7.   AUTHOR(S)
       Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.).  List author's affiliation if it differs from the performing organi
       zation.

   8.   PERFORMING ORGANIZATION REPORT NUMBER
       Insert if performing organization wishes to assign this number.

   9.   PERFORMING ORGANIZATION NAME AND ADDRESS
       Give name, street, city, state, and ZIP code. List no more than two levels of an organizational hueaichy.

   10.  PROGRAM ELEMENT NUMBER
       Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.

   11.  CONTRACT/GRANT NUMBER
       Insert contract or grant number under which report was prepared.

   12.  SPONSORING AGENCY NAME AND ADDRESS
       Include ZIP code.

   13.  TYPE OF REPORT AND PERIOD COVERED
       Indicate interim final, etc., and if applicable, dates covered.

   14.  SPONSORING AGENCY CODE
       Insert  appropriate code.

   15.  SUPPLEMENTARY  NOTES
       Enter information not included elsewhere but useful, such as:  Prepared in cooperation with, Translation of, Presented at conference of,
       To be published in, Supersedes, Supplements, etc.

   16.  ABSTRACT
       Include a brief ^200 words or less) factual summary of the most significant information contained in the report.  If the report Contains a
       significant bibliography or literature survey, mention it here.

   17.  KEY WORDS AND DOCUMENT ANALYSIS
       (a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
       concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.

       (b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
       ended terms written  in descriptor form for those subjects for which no descriptor exists.

       (c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COS ATI  Subject Category List. Since the ma-
       jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will  be specific discipline, area of human
       endeavor, or type of physical object. The apphcation(s) will be cross-referenced with secondary Field/Group assignments that will follow
       the primary posting(s).

   18.  DISTRIBUTION STATEMENT
       Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
       the public, with address and price.

   19. & 20. SECURITY CLASSIFICATION
       DO NOT submit classified reports to the National Technical Information service.

   21.  NUMBER  OF PAGES
       Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.

   22.  PRICE
       Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (Rev. 4-77) (Reverse)

-------