EPA-600/2-76-032d
December  1976
Environmental Protection Technology Series
                                       SOURCE  ASSESSMENT:
                                            PHTHALIC ANHYDRIDE
                                                 (AIR  EMISSIONS)
                                        Industrial Environmenta! Research Laboratory
                                              Office of Research and Development
                                             U.S. Environmental Protection Agency
                                      Research Triangle Park, North Carolina  27711

-------
               RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development. U.S. Environmental
Protection  Agency, have been grouped into five series. These five  broad
categories  were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
     1.    Environmental Health Effects Research
     2.    Environmental Protection Technology
     3.    Ecological Research
     4.    Environmental Monitoring
     5.    Socioeconomic Environmental Studies

This report has been  assigned  to the ENVIRONMENTAL  PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new  or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
                    EPA REVIEW NOTICE

This report has been reviewed by  the U.S.  Environmental
Protection Agency, and approved for publication.   Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                       EPA-600/2-76-032d

                                       December 1976
        SOURCE ASSESSMENT:

        PHTHALIC ANHYDRIDE

           (AIR EMISSIONS)
                     by

         R. W. Serth and T. W. Hughes

        Monsanto Research Corporation
             1515 Nicholas Road
             Dayton, Ohio 45407
           Contract No. 68-02-1874
            ROAP No.  21AXM-071
         Program Element No. 1AB015
     EPA Project Officer: Dale A. Denny

 Industrial Environmental Research Laboratory
   Office of Energy, Minerals, and Industry
      Research Triangle Park, NC 27711
                Prepared for

U.S. ENVIRONMENTAL PROTECTION AGENCY
      Office of Research and Development
            Washington, DC 20460

-------
                          PREFACE

The Industrial Environmental Research Laboratory (IERL)  of
EPA has the responsibility for insuring that pollution con-
trol technology is available for stationary sources to meet
the requirements of the Clean Air Act, the Water Act and
the Solid Waste legislation.  If control technology is un-
available, inadequate, uneconomical or socially unacceptable,
then financial support is provided for the development of
the needed control techniques for industrial and extractive
process industries.  Approaches considered include:  process
modifications, feedstock modifications, add-on control de-
vices, and complete process substitution.  The scale of the
control technology programs ranges from bench to full scale
demonstration plants.

The Chemical Processes Branch of the Industrial Processes
Division of IERL has the responsibility for investing tax
dollars in programs to develop control technology for a
large number  (>500) of operations in the chemical industries.
As in any technical program, the first question to answer
is, "Where are the unsolved problems?"  This is a determina-
tion which should not be made on superficial information;
consequently, each of the industries is being evaluated in
detail to determine if there is, in EPA's judgment, suffi-
cient environmental risk associated with the process to
invest in the development of control technology.  This report
contains the data necessary to make that decision for the
air emissions from phthalic anhydride manufacture.

                              iii

-------
Monsanto Research Corporation has contracted with EPA to
investigate the environmental impact of various industries
which represent sources of pollution in accordance with EPA's
responsibility as outlined above.  Dr. Robert C. Binning
serves as Program Manager in this overall program entitled,
"Source Assessment," which includes the investigation of
sources in each of four categories:  combustion, organic
materials, inorganic materials and open sources.  In this
study of phthalic anhydride manufacture, Mr. Kenneth L. Baker
and Mr. Edward J. Wooldridge served as EPA Project Leaders.
                              IV

-------
                          CONTENTS

Section                                                Page
I         Introduction                                   1
II        Summary                                        3
III       Source Description                            14
          A.   Process Description                       I4
               1.   o-Xylene Based  Process              16
               2.   Naphthalene Based Process           33
          B.   Materials Flow                           53
               1.   o-Xylene Based  Process              53
               2.   Naphthalene Based Process           59
          C.   Geographical Distribution                59
IV        Emissions                                     66
          A.   Selected Pollutants                       66
               1.   o-Xylene Based  Process              66
               2.   Naphthalene Based Process           70
          B.   Location and Description                 71
               1.   Scrubber Vent                       71
               2.   Incinerator Stack                   72
               3.   Storage Tank Vents                  78
               4.   Flaker and Bagger Vent              82
               5.   Transport Loading Facility Vent     83
               6.   Catalyst Storage Hopper Vents       84
               7.   Fugitive Emissions                  85
               8.   Dual Thermal Incineration           87
          C.   Environmental Effects                    89
               1.   Definition of a Representative      89
                    Source
               2.   Emission Factors                    90
               3.   Source Severity                      92
               4.   Industry Contribution to Total      100
                    Atmospheric Emissions
               5.   Affected Population                 100
               6.   Growth Factor                       105

-------
                    CONTENTS  (continued)

Section                                                Page
V         Control Technology                           106
          A.   State of the Art                        106
          B.   Process Modifications                   109
               1.   The Rhone-Progil "Chauney '71"     110
                    Process
               2.   Maleic Anhydride Recovery          HO
               3.   Direct Production of Phthalates    H3
               4.   Alternate Feedstocks
VI        Growth and Nature of the Industry
          A.   Present Technology
          B.   Emerging Technology
          C.   Marketing Strengths and Weaknesses
VII       Appendixes                                   125
          A.   Storage Tank Calculations               126
          B.   Rationale for Not Considering All       133
               Species Listed in Table 21
VIII      Glossary of Terms                            I37
IX        Conversion Factors and Metric Prefixes
X         References
                              VI

-------
                       LIST OF FIGURES
Figure
  1       Breakdown of Phthalic Anhydride Pro-
          duction in the Continental United States
  2       BASF Process for Manufacture of Phthalic      17
          Anhydride from o-Xylene

  3       Organic Liquid Heat-Transfer System           27
  4       Reaction Mechanism for the Catalytic          30
          Oxidation of o-Xylene

  5       Badger-Sherwin-Williams Process for Manu-     35
          facture of Phthalic Anhydride from
          Naphthalene
  6       Schematic Diagram of Scrubber for Phthalic    44
          Anhydride Plant

  7       Schematic Diagram of Thermal Incinerator      47
          for Phthalic Anhydride Plant

  8       Phthalic Anhydride Plant Locations            65

  9       Maleic Anhydride Recovery Process            112

 10       Flow Diagram for Direct Production of        114
          Phthalates

 11       Uses of Phthalic Anhydride                   118
                               Vll

-------
                       LIST OF TABLES
Table
  1       Summary of Emission Data for a Represen-
          tative o-xylene Based Phthalic Anhydride
          Plant
  2       Summary of Emission Data for a Represen-       7
          tative Naphthalene Based Phthalic Anhydride
          Plant
  3       Source Severity Factors and Industry Contri-   8
          bution to Total Emissions for o-xylene Based
          Phthalic Anhydride Production (Controlled
          Emissions)
  4       Source Severity Factors and Industry Contri-  10
          bution to Total Emissions for Naphthalene
          Based Phthalic Anhydride Production (Controlled
          Emissions)
  5       Affected Population Summary                   13
  6       Stream Code for BASF Process Illustrated in   18
          Figure 2
  7       Summary of Tankage Requirements for a         25
          5.9 x 104 Metric Tons/Yr o-xylene Based
          Phthalic Anhydride Plant
  8       Heats of Reaction for the Oxidation of        32
          o-xylene
  9       Reactor System Heat Balance for Production    33
          of Phthalic Anhydride from o-xylene
 10       Stream Code for Badger-Sherwin-Williams       36
          Process Illustrated in Figure 5
 11       Typical Scrubber Material Balance for a       45
          Naphthalene Based Phthalic Anhydride Plant
 12       Heating Values of Organic Compounds in        46
          Scrubber Purge Stream
 13       Incinerator Material Balance for a Naph-      48
          thalene Based Phthalic Anhydride Plant
 14       Summary of Tankage Requirements for a         49
          5.9 x 10U Metric Tons/Yr Naphthalene Based
          Phthalic Anhydride Plant
 15       Typical Properties of Davison Grade 902       51
          Catalyst
 16       Heats of Reaction for the Oxidation of        52
          Naphthalene

                              viii

-------
                 LIST OF TABLES (continued)


Table                                                  Page
 17       Reactor System Heat Balance for Pro-          53
          duction of Phthalic Anhydride from
          Naphthalene

 18       Material Balance for a 5.9 x 104 Metric       54
          Tons/Yr o-xylene Based Phthalic Anhydride
          Plant

 19       Material Balance for a 5.9 x 104 Metric       60
          Tons/Yr Naphthalene Based Phthalic Anhy-
          dride Plant
 20       Phthalic Anhydride Plants                     64

 21       Possible Reaction Products from the           67
          Oxidation of Xylene

 22       Concentrations of Contaminatns in Switch-     73
          Condenser Off-Gas

 23       Typical Scrubber Vent Gas Composition for     74
          o-xylene Based Process
 24       Typical Scrubber Vent Gas Composition for     75
          Naphthalene Based Process

 25       Typical Incinerator Flue Gas Composition      76
          for o-xylene Based Process with Scrubber-
          Incinerator Combination
 26       Typical Incinerator Flue Gas Composition      77
          for Naphthalene Based Process with Scrubber-
          Incinerator Combination

 27       Typical Incinerator Flue Gas Composition      79
          for o-xylene Based Process Using Direct
          Thermal Incineration

 28       Typical Incinerator Flue Gas Composition      80
          for Naphthalene Based Process Using Direct
          Thermal Incineration

 29       Storage Tank Working and Breathing Losses     81
          for a 5.9 x 10^ Metric Tons/Yr Phthalic
          Anhydride Plant
 30       Frequently Used Heat-Transfer Fluids          86

 31       Fugitive Emissions from Phthalic Anhydride    87
          Plants

 32       Typical Flue Gas Composition for Dual         88
          Incinerator
                               IX

-------
                 LIST OF TABLES (continued)
Table
 33       Emission Factors for o-xylene Based
          Phthalic Anhydride Plants

 34       Emission Factors for Naphthalene Based        93
          Phthalic Anhydride Plants
 35       Source Severity Equations                     97

 36       Emission Heights for Representative Source    ^7

 37       Source Severity Factors for a Representative  98
          o-xylene Based Phthalic Anhydride Plant
 38       Source Severity Factors for a Representative  99
          Naphthalene Based Phthalic Anhydride Plant

 39       Nationwide Emissions of Criteria Pollutants  1°1
          from Phthalic Anhydride Industry (Controlled
          Emissions)
 40       Emissions of Criteria Pollutants from        102
          Phthalic Anhydride Industry by State
          (Controlled Emissions)

 41       Affected Population                          104
 A-l      Storage Tank Input Data for o-xylene         129
          Based Plant

 A-2      Storage Tank Input Data for Naphthalene      130
          Based Plant
 A-3      Storage Tank Calculation Summary for
          o-xylene Based Plant
 A-4      Storage Tank Calculation Summary for         132
          Naphthalene Based Plant
 B-l      Compounds Not Included in Study              134
                              x

-------
                       LIST OF SYMBOLS
Symbol                             Definition
  A                      Area
  C                      Diameter factor
 Cap                     Production capacity
  C.                     Production capacity of plant i
  D                      Tank diameter
  Dp                     Mean population density
 D                       County population density for plant i
  e                      2.72
  E                      Emission factor, g/kg
  E1                     Emission factor, Ib/ton
  F                      Hazard factor
  F                      Equivalent gasoline working loss
  F                      Paint factor
   P
  H                      Effective emission height
  H'                     Tank outage
  K                      Turnover factor
  L                      Total petrochemical loss, barrels
  L,                     Total petrochemical loss, pounds
  L                      Total equivalent gasoline loss
  L                      Equivalent gasoline breathing loss
  M                      Molecular weight
  N                      Number of turnovers per year
  P                      Vapor pressure
  Q                      Mass emission rate
  S                      Source severity
 t, t0                   Averaging times
  u                      Average wind speed
  V                      Tank capacity
  W                      Liquid density
                               XI

-------
                 LIST OF SYMBOLS  (continued)
Symbol
  x

  AT
 xmax


 xmax

 x"(x)
          Definition

Distance downwind from source
Average daily ambient temperature
change

Vertical dispersion coefficient

Maximum mean ground level concen-
tration (short term average)

Maximum mean ground level concen-
tration (long-term average)

Annual mean ground level concen-
tration
                               XII

-------
                         SECTION I

                       INTRODUCTION
Phthalic anhydride (PAN) is an important industrial chemical
which is principally used in the manufacture of plasticizers,
polyester resins, and alkyd resins.  Present production capa-
city in the United States is nearly 5 x 105 metric tons/yr.
Phthalic anhydride is manufactured by the catalytic vapor-
phase oxidation of either ortho-xylene.  (designated herein
as o-xylene) or naphthalene.  The o-xylene based plants
employ fixed-bed reactors, while the naphthalene based plants
use the fluid-bed process.  This document presents a detailed
study of the phthalic anhydride industry from the standpoint
of atmospheric emissions and their potential environmental
impact.

The major results of this study, summarized in Section II,
include emission factors for each species emitted to the
atmosphere from each emission point within a representative
phthalic anhydride plant.  Also tabulated are several factors
designed to measure the environmental hazard potential of
phthalic anhydride operations.  These include source severity
factors, the industry contribution to total atmospheric
emissions of criteria pollutants, and the population affected
by a representative plant.
 1 metric ton = 106 grams = 2205 pounds =1.1 short tons
 (short tons are designated "tons" in this document); other
 conversion factors and metric system prefixes are presented
 in Section IX.
                                1

-------
Detailed descriptions of phthalic anhydride manufacturing
processes are given in Section III.  Discussion is limited
to the BASF  (Badische Anilin und Soda-Fabrik) process for the
oxidation of o-xylene, and the Badger-Sherwin-Williams process
for the oxidation of naphthalene, since these two processes
dominate the industry.  Included are descriptions of each
major processing step, flow diagrams, process chemistry, and
material and energy balances.

Atmospheric emissions from phthalic anhydride plants are
discussed in Section IV.  The species known to be emitted
and/or produced by the processes are detailed, and each
emission point within the plant is described.  Compositions
and flow rates of streams emitted to the atmosphere are
provided.  A representative phthalic anhydride plant is
defined, and emission factors for such a plant employing
the best available control technology are given.  These
emission factors are then used to generate source severity
factors, the industry contribution to total emissions of
criteria pollutants, and the affected population.

Present and future aspects of pollution control technology
in the phthalic anhydride industry are considered in Section
V.  Several process modifications are described which are
likely to have an impact upon the industry in the near
future.

Economic and production trends in the phthalic anhydride
industry are addressed in Section VI.  The trends in each of
the industries that are major consumers of phthalic anhydride
are also analyzed.  Finally, estimates of phthalic anhydride
production through the remainder of the present decade are
discussed.

-------
                        SECTION II

                          SUMMARY
Phthalic anhydride is manufactured by the catalytic vapor-
phase oxidation of either o-xylene or naphthalene.   There
are currently 10 phthalic anhydride plants operating in the
continental United States, plus one in Puerto Rico, having
a total capacity of 4.95 x 105 metric tons/yr (5.44 x 105
tons/yr).  Of the plants in the continental U.S., seven uti-
lize o-xylene as a feedstock and the remaining three use
naphthalene.  All of the o-xylene based plants employ fixed-
bed reactors, while the three naphthalene based plants use
fluid-bed processes.  The population densities of the
counties in which the plants are located range from 30 to
4,905 persons per square kilometer.

Sources of atmospheric emissions within phthalic anhydride
plants include the scrubber vent(s) and/or incinerator
stack(s), storage tanks, the flaker and bagger vents, the
transport loading facility vent, and fugitive emissions.
The catalyst storage hopper vents are additional sources of
emissions in naphthalene based plants.

The main process waste gas from the phthalic anhydride switch
condensers is controlled either by a scrubber-incinerator
combination or by direct incineration.  The latter method
has the advantage of providing control of carbon monoxide
as well as the organic species in the waste gas.  Either

-------
 the  scrubber or the incinerator, or both, may also be used
 to control the waste streams from the product purification
 section of the plant.  Alternatively, a separate scrubber or
 incinerator may be used to control these streams.

 Storage tanks contain o-xylene or naphthalene, crude phthalic
 anhydride, and refined phthalic anhydride.  Available control
 devices include conservation vents on o-xylene and naphthalene
 tanks, and condensers or sublimation traps on phthalic
 anhydride tanks.  The latter tanks may also be vented to
 the  incinerator for control.

 In the flaking and bagging operations, liquid phthalic anhydride
 is solidified in the form of flakes and packaged in bags for
 shipment.  The vents from these operations are ducted to a
 cyclone and/or a baghouse for control.  Phthalic anhydride
 is also shipped in the liquid form.  At the transport loading
 facility, liquid phthalic anhydride is pumped into tank
 trucks or railway tank cars for shipment.

 Fugitive emissions include heat-transfer oil  (e.g., Dow-
 therm A®, a diphenyl-diphenyl oxide eutectic mixture) which
 escapes from the heat-transfer circuits during process
 upsets.  In the case of naphthalene based plants, vanadium
 oxide catalyst dust is also emitted during catalyst transfer
 operations in which the fluid-bed catalyst is pneumatically
 conveyed between the storage hoppers and the reactor.  Vents
 on the catalyst storage hoppers are additional sources of
 emissions during these operations.  The vents are equipped
with cyclone separators for emissions control.

Emissions data for o-xylene based plants are summarized in
Table 1.   The data correspond to a plant with a dual incin-
eration control system, which represents the best available
control technology.  The main process incinerator controls

-------
         Table 1.  SUMMARY OF EMISSION DATA FOR A REPRESENTATIVE
                 o-XYLENE BASED PHTHALIC ANHYDRIDE PLANT
Species emitted
Particulate
Main process incinerator
Secondary incinerator
Flaker and bagger
Sulfur oxides
Main process incinerator
Nitrogen oxides
Main process incinerator
Secondary incinerator
Carbon monoxide
Main process incinerator
Secondary incinerator
Maleic anhydride
Main process incinerator
Secondary incinerator
Phthalic anhydride
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Benzoic acid
Main process incinerator
Secondary incinerator
Diphenyl oxide6
Fugitive emissions
o-Xylene
Storage tanks
Formaldehyde
Main process incinerator
Secondary incinerator
Total hydrocarbonsS
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Emission factor, g/kg
Uncontrolled
b
~b
0.10

5.0 ± 25%
_c
~c

125.0 ± 20%
52.0 ± 20%
3.75 ± 83%
15.6 ± 20%
10.6 ± 53%
0.29 ± 10%
0.10
0.45 ± 10%
_b

3.12 ± 20%
1.25 ± 50%
0.016
0.20 ± 10%
2a\
72.8 ± 30%
18.8 ± 60%
0.49 ± 10%
0.1
0.45 ± 10%
0.116
Controlled
0.25 ± 50%
0.125 ± 50%
0.001

5.0 ± 25%
1.25 ± 50%
0.125 ± 25%

0.125 ± 50%
1.25 ± 50%
1.82 ± 25%
0.038 ± 90%
0.545 ± 25%
0.106 ± 60%
0.003 + 20%
0.001
0.005 ± 20%
_d

0.109 ± 25%
0.0125 ± 55%
_d
0.002 ± 20%
0.074f
2.6 ± 30%
0.16 ± 60%
0.005 ± 20%
0.001
0.005 + 20%
_d
Control
efficiency.
99

0


99.9
96.5
99
96.5
99
99
99
99

96.5
99

99
96.5
96.5
99
99
99
99
 Emission factor is defined as weight of emission per unit
 weight of phthalic anhydride product.
 Emission data not available.
 No emissions generated in uncontrolled process.
 Fugitive emissions are not controlled.
 Heat-transfer fluid, assumed to be Dowtherm A.
 Total aldehydes reported as formaldehyde.
^Includes all non-methane organic species.

-------
 the main  process waste  gas  and the secondary incinerator
 handles the waste  streams from the product purification sec-
 tion of the plant.  Error bounds on the emission factors
 were estimated wherever possible.

 Emissions data for a naphthalene based plant with a dual
 incineration  control system are presented in Table 2.  In
 arriving  at these values it was assumed that this control
 system, which is currently  in operation on an o-xylene based
 plant, could  be applied to  a naphthalene based plant without
 altering  the  control efficiency of the system.  The accuracy
 of  the emission factors in  Table 2 is believed to be gener-
 ally poorer than that of the values listed in Table 1, as
 reflected in  the estimated  error bounds.
                        i
 In  order  to quantify the hazard potential of phthalic an-
 hydride operations, a severity factor was defined which
 represents the ratio of the mean maximum ground level ex-
 posure to the hazard level  of exposure for a given pollutant
 from a given  source.  Using Gaussian plume dispersion theory
 together  with the controlled emission factors in Tables 1
 and  2, severity factors were calculated for o-xylene and
 naphthalene based plants having production capacities of 5.9
 x 101* metric  tons/ yr (6.5  x 10H tons/yr) .  The results are
 summarized in Tables 3  and  4.  Also listed are the annual
mass  emissidns from a single plant and from the phthalic
 anhydride industry, and the percentage contribution of the
 industry  to the total mass  emissions of criteria pollutants
 (particulate, NO , SO ,  CO, and hydrocarbons).  These values
                X    X
are all based on the controlled emission factors listed in
Tables 1  and  2, and, hence, represent hypothetical emissions
which would occur if all plants employed the best available
control technology.

-------
Table  2.    SUMMARY  OF  EMISSION  DATA  FOR  A  REPRESENTATIVE
          NAPHTHALENE  BASED  PHTHALIC  ANHYDRIDE  PLANT
Species emitted
Particulate
Main process incinerator
Secondary incinerator
Flaker and bagger
Nitrogen oxides
Main process incinerator
Secondary incinerator
Carbon monoxide
Main process incinerator
Secondary incinerator
Phthalic anhydride
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport and loading
Fugitive emissions
Maleic anhydride
Main process incinerator
Secondary incinerator
Benzoic acid
Main process incinerator
Secondary incinerator
Naphthoquinone
Main process incinerator
Secondary incinerator
Naphthalene
Storage tanks
Vanadium oxide catalyst
Catalyst storage
Fugitive emissions
Diphenyl oxide
Fugitive emissions
Formaldehyde
Main process incinerator
Secondary incinerator
Total hydrocarbons^
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Emission factor,3 g/kg
Uncontrolled
b
~b
0.10
_ c
_ c
50.5_c
20.4
9.8
0.37
0.10
0.45b
7.0
4.9
1.56_b
0.69
6.6
0.60
0.41L
_D
0.016
*A
31.8
26.5
0.97
0.1
0.45
0.116


+ 50%
± 40%
± 70%
± 10%
+ 200%
± 80%
± 40%
± 40%
± 70%
± 10%



+ 79%
- 35%
± 70%
± 10%
± 10%
Controlled
0.25 ± 70%
0.125 ± 70%
0.001
1.25 ± 70%
0.125 t 40%
0.05 ± 70%
1.25 ± 70%
0.71 i 50%
0.10 ± 80%
0.004 ± 20%
0.001
0.005 .
d
0.24 + 200%
0.05 ± 90%
0.05 + 50%
_D
0.02 ± 50%
0.07 ± 80%
0.006 + 20%
0.01 _,
d
_d
0.074^
D
1.09 + 85%
40%
0.27 ± 80%
0.01 t 20%
0.001
0.005 + 20%
Control
efficiency,
99

99.9
96.5
99
99
99
99
96.5
99
96.5
96.5
99
99
97.5

96.5
96.5
99
99
99
99
             Emission factor  is defined as weight of emission per
             unit weight of phthalic anhydride product.
             Emission data not available.
             No emissions generated in uncontrolled process.
             Fugitive emissions are not controlled.
             Heat-transfer fluid, assumed to be Dowtherm  A.
             Total aldehydes  reported as formaldehyde.
            "includes all non-methane organic species.

-------
                           Table  3.   SOURCE SEVERITY FACTORS AND INDUSTRY CONTRIBUTION TO TOTAL EMISSIONS FOR O-XYLENE BASED
                                                     PHTHALIC ANHYDRIDE PRODUCTION (CONTROLLED EMISSIONS)
00
Species emitted
Particulate (total)
Main process incinerator
Secondary incinerator
Flaker and bagger
Sulfur oxides (total)
Main process incinerator
Nitrogen oxides (total)
Main process incinerator
Secondary incinerator
Total hydrocarbons (total)
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Carbon monoxide (total)
Main process incinerator
Secondary incinerator
Phthalic anhydride (total)
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Maleic anhydride (total)
Main process incinerator
Secondary incinerator
Emissions
- from all -
plants,3
metric
tons/yr
113



1,500

413


865






413


198






557


Emissions
from represent-
ative plant, "
metric tons/yr
22.2
14.8
7.4
0.06
295.0
295.0
81.2
73.8
7.4
170.3
153.4
9.4
0.30
0.06
0.30
6.8
81.2
7.4
73.8
39.0
32.2
6.3
0.18
0.06
0.30

109.6
107.4
2.2
Severity
factor,

0.0095
0.019
0.0038

0.14

0.14
0.061

0.23
0.056
0.0071
0.0088
0.044
4.0

0.000053
0.0021

0.28
0.20
0.024
0.050
0.24


5.4
0.45
Industry contribution to total emissions, c %
Nationwide
0.00063



0.0050

0.0018


0.0034






0.00042












California
0.0014



0.0492

0.0032


0.0051






0.0006












Illinois
0.0033



0.0249

0.0144


0.0161






0.0022












Louisiana
0.0039



0.1200

0.0133


0.0061






0.0010












New
Jersey
0.0148



0.0641

0.0166


0.0208






0.0028












Texas
0.0040



0.0393

0.0062


0.0077






0.0012












                 Emissions not calculated  for  individual  emission points.
                bA representative plant  is defined  to  be  one  having  a production capacity of 5.9  x lO1*  metric  tons/yr  (6.5  x  101*  tons/yr)
                CIndustry contribution calculated only for  criteria  pollutants.
                 Includes all non-methane  organic species.

-------
          Table 3 (continued).
                                SOURCE SEVERITY FACTORS AND  INDUSTRY  CONTRIBUTION TO TOTAL EMISSIONS FOR O-XYLENE BASED
                                             PHTHALIC ANHYDRIDE  PRODUCTION (CONTROLLED EMISSIONS)

Species emitted
Benzoic acid (total)
Main process incinerator
Secondary incinerator
Diphenyl oxide6 (total)
Fugitive emissions
o-Xylene (total)
Storage tanks
Formaldehyde (total)
Main process incinerator
Secondary incinerator

Emissions
from all
, . a
plants,"
metric
tons/yr
36


4.8

0.6

22


Emissions
from represent-
ative plant,
metric tons/yr
7.1
6.4
0.74
0.94
0.94
0.12
0.12
4.4ff
4.4f
-9


Severity
factor,

0.0065
0.0030

18.5

0.00022
f
0.074
-9

Industry contribution to total emissions, %
Nationwide











California











Illinois











Louisiana











New
Jersey











Texas











 Emissions not calculated for individual emission points.
 A representative plant is defined to be one having a production  capacity
 of 5.9 x 10" metric tons/yr  (6.5 x lO1* tons/yr)
 Industry contribution calculated only for criteria pollutants.
 Heat-transfer fluid, assumed to be Dowtherm A.
 Total aldehydes reported as formaldehyde.
"Data not available.

-------
             Table 4.  SOURCE SEVERITY FACTORS AND INDUSTRY CONTRIBUTION TO TOTAL EMISSIONS FOR
                   NAPHTHALENE BASED PHTHALIC ANHYDRIDE PRODUCTION  (CONTROLLED EMISSIONS)
Species emitted
Particulate (total)
Main process incinerator
Secondary incinerator
Flaker and bagger
Nitrogen oxides (total)
Main process incinerator
Secondary incinerator
Total hydrocarbons (total)
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Carbon monoxide (total)
Main process incinerator
Secondary incinerator
Phthalic anhydride (total)
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Maleic anhydride (total)
Main process incinerator
Secondary incinerator
Benzoic acid (total)
Main process incinerator
Secondary incinerator
Naphthoquinone (total)
Main process incinerator
Secondary incinerator
Emissions
from
all plants, a
metric tons/yr
56



206


224






195


123





44


7.6


13


Emissions
from represent-
ative plant, "
metric tons/yr
22.2
14.8
7.4
0.06
81.2
73.8
7.4
88.1
64.4
15.9
0.59
0.06
0.30
6.8
76.8
3.0
73.8
48.4
41.9
5.9
0.24
0.06
0.30
17.2
14.2
3.0
3.0
3.0
_e
5.3
1.2
4.1
Severity
factor

0.0095
0.019
0.0038

0.14
0.061

0.10
0.095
0.014
0.0088
0.044
4.0

0.000021
0.0021

0.36
0.20
0.032
0.050
0.24

0.72
0.60

0.0030
_ G

0.15
2.1
Industry contribution to total
emissions, c %
Nationwide
0.00031



0.0009


0.0009






n. 00020

















New
Jersey
0.0102



0.0114


0.0074






0.0018

















Pennsylvania
0.0023



0.0049


0.0182






0.0038

















aEmissions not calculated for individual emission points.
''Representative plant is defined to be one having a production capacity
 of 5.9 x 10" metric tons/yr  (6.5 x 101* tons/yr)
clndustry contribution calculated only for criteria pollutants.
dIncludes all non-methane organic species.

eData not available.

-------
         Table 4 (continued).   SOURCE SEVERITY FACTORS AND INDUSTRY CONTRIBUTION TO TOTAL EMISSIONS FOR
                   NAPHTHALENE BASED PHTHALIC ANHYDRIDE PRODUCTION (CONTROLLED EMISSIONS)


Species emitted
Naphthalene (total)
Storage tanks
Vanadium oxide catalyst (tota
Catalyst storage
Fugitive emissions
Diphenyl oxide (total)
Fugitive emissions
Formaldehyde (total)
Incinerator I
Incinerator II
Emissions
from
all plants,
metric tons/yr
0.9

.) 1.5


2.4

II9


Emissions
from represent-
ative plant, b
metric tons/yr
0.35
0.35
0.59
0.59

0.94
0.94
4.4JJ
4.49



Severity
factor

0.0058

1.4


18.5

v.u,-,

Industry contribution to total
emissions , %
Nationwide










New
Jersey










Pennsylvania










 Emissions not calculated for individual emission points.
 Representative plant is defined to be one having a production capacity
 of 5.9 x 10" metric tons/yr  (6.5 x 101* tons/yr)
 Industry contribution calculated only for criteria pollutants.
 Heat-transfer fluid, assumed to be Dowtherm A.
9Total aldehydes reported as formaldehyde.

-------
Phthalic anhydride production in 1973 totaled 4.658 x 105
metric tons  (5.12 x 105 tons) in the United States.  Pro-
duction in 1978 is expected to total 4.99 x 105 metric tons
(5.49 x 105 tons).  Thus, assuming that the same level of
control exists in 1978 as existed in 1973, emissions from
the phthalic anhydride industry will increase by 7% over
that period; i.e.:
            Emissions in 1978 _ 4.99 x 105
            Emissions in 1973   4.658 x 1(P ~

The average number of persons exposed to high contaminant
levels from phthalic anhydride operations was estimated and
designated as the "affected population."  The calculation
was made for each species emitted and for each emission point
within a representative plant for which the severity factor
exceeds 0.1 or 1.0.   The results are presented in Table 5.
The largest values of affected population are 525,000 persons
for o-xylene based production and 9,200 persons for naphtha-
lene based production.
                              12

-------
                                 Table  5.   AFFECTED POPULATION
                                      (number  of persons)
Compound
Sulfur oxides
Main process incinerator
Nitrogen oxides
Main process incinerator
Total hydrocarbons
Main process incinerator
Fugitive emissions
Phthalic anhydride
Main process incinerator
Secondary incinerator
Transport loading
Maleic anhydride
Main process incinerator
Secondary incinerator
Diphenyl oxide
Fugitive emissions
Formaldehyde
Main process incinerator
Naphthoquinone
Main process incinerator
Secondary incinerator
Vanadium oxide catalyst
Catalyst storage
g-Xylene based plant
No. of persons
where
X/F > 1.0
0
0
0
40
0
0
0
44,000
0
300
0
a
a
_a
No. of persons
where
X/F > 0.1
6,700
5,600
9,600
500
19,000
3,100
140
525,000
8,800
3,400
15,000
a
_a
_a
Naphthalene based plant
No. of persons
where
X/F > 1.0
_a
0
0
6
0
0
0
0
0
50
0
0
500
40
No. of persons
where
X/F > 0.1
_a
870
0
77
4,100
460
22
9,200
1,800
540
2,300
1,300
7,000
650
U)
       No emissions.

-------
                        SECTION III
                    SOURCE DESCRIPTION

A.   PROCESS DESCRIPTION

The first commercial phthalic anhydride process was patented
by Badische Anilin und Soda-Fabrik  (BASF) in Germany in 1896.
The process utilized the liquid-phase oxidation of naphtha-
lene in concentrated sulfuric acid in the presence of mercury
sulfate.1  In 1917, a catalytic vapor-phase process for the
oxidation of naphthalene was developed in both the United
States and Germany.  Phthalic anhydride was first produced
by the vapor-phase oxidation of xylene in 1946 in the United
States.2  o-Xylene has now become the preferred raw material
due to its lower cost and somewhat better yield ratio compared
with naphthalene.  At present, all units in the world, with
the exception of Progil in France, use vapor-phase processes.
In 1975, 67% of the phthalic anhydride produced in the
continental United States was obtained from o-xylene and
33% from naphthalene as shown in Figure 1 (and in Table 20)-3'1
1Ockerbloom, N. E.  Xylenes and Higher Aromatics, Part 3:
 Phthalic Anhydride.  Hydrocarbon Processing.  22:162-166,
 September 1971.
2Landau, R., and H. Harper.  Phthalic Anhydride.  Chemistry
 and Industry  (London).  July 29, 1961, p. 1143-1152.
3Phthalic Anhydride.  Chemical Marketing Reporter.   205:9,
 March 4, 1974.
4Anderson, E. V.  Phthalic Anhydride Makers Foresee Shortage
 Chemical and Engineering News.  53:10-11, June 30, 1975.
                               14

-------
                                        PHTHALIC ANHYDRIDE PLANTS
                                                10 PLANTS
                       o-XYLENE BASED
                         7 PLANTS
                      67%  OF CAPACITY
                                  NAPTHALENE BASED
                                      3 PLANTS
                                   33% OF CAPACITY
tn
            BASF PROCESS
               4 PLANTS
            53% OF CAPACITY
VON HEYDEN PROCESS
     3 PLANTS
       OF CAPACITY
BADGER-SHERWIN-WILLIAMS
 PROCESS     2 PLANTS
     24%  OF CAPACITY
OTHER PROCESS
   1 PLANT
 9%  OF CAPACITY
      Figure  1.   Breakdown  of phthalic  anhydride production in the  continental United States

-------
1.   o-Xylene Based Process

Phthalic anhydride is presently produced in the United States
from o-xylene by the von Heyden process, which is licensed
in somewhat different forms by Chemiebau and Lurgi, and by
the BASF process, developed in the late 1960's.  In 1975,
79% of the o-xylene based production in the continental
United States involved use of the BASF process and  21%
utilized the von Heyden process. 3'H  Figure 2  is a  schematic
flow diagram for the BASF process, which is described in
the following subsections.  The numbered streams in Figure
2 are identified in Table 6.  Since the gross features of
the von Heyden process differ only in minor ways from the
BASF process, the former will not be described per se.

a.   Chemistry - The vapor-phase oxidation of o-xylene to
phthalic anhydride on a vanadium oxide catalyst can be
represented by the following equation:
                         catalyst        r  „   Q + ^^
                                            0
         o-xylene   oxygen             phthalic    water
                                       anhydride

The principal side reactions are the result of the further
oxidation of phthalic anhydride to maleic anhydride, carbon
oxides, and water.  Based on the work of Bernardini and
Ramacci,5 the main sequence of reactions in the oxidation
of o-xylene is as follows:
5Bernardini, F., and M. Ramacci.  Oxidation Mechanism of
 o-Xylene to Phthalic Anhydride.  Chimica e 1'Industria
 (Milan).  481:9-17, January 1966.
                              16

-------
                                                                      ATMOSPHERIC EMISSIONS
STEAM
i-i O^. <" *
JS _*•© ' 	 1 " (§)^j®
"l 1
A B 	 G I j»

1 STEAM \_J
r&£ ' ©
PROCESS <£fcs^ ^rjLJ—
— 	 	 	 ^->- O-XYLENE STORAGE TANK VENT GAS


tt K • )>
-t^STEAM *
-*=rWATER
H ^ , 	
® ( T^o"
V /^ •
1
±=
®

J\ — ^
L
®

1 0|L. TO SECONDARY
t 4 SCRUBBER OR-
'~*-f~_~M~~W — | INCINERATOR
TO s^uTaTRT ^ ®fVACUUM
TO SECONDARY 5 _ /-^rt?}
trannnro -<---) W STFAM -^ — ^f
OR INCINERATOR --^-"-^ (|) WATER — »•
©„ P
N !3L
'
r 1
fi r~^i
Q OIL R
LJ U« , 	 „
1 	 ^TITV- *
IT
'OIL'
BASF PROCESS FOR THE MANUFACTURE OF
PHTHALIC ANHYDRIDE FROM o-XYLENE
A. O-XYLENE STORAGE L. INCINERATOR
B. 502 STORAGE M. HEATER
C. O-XYLENE PREHEATER N. CRUDE PRETREATMENT TANK
D. COMPRESSOR 0. PRECOOLER
E. AIR PREHEATER P. STRIPPER
F. REACTOR IFIXED BEOI Q, EVAPORATOR
G. '.IOLHVS ALT HEAT EXCHANGER R. EVAPORATOR
H. WASTE HEAT BOILER S. RECTIFIER
1 SWITCH ,0'!CF\s[R5 T. PHTHALIC Ar.h '• JRIG'C STORAGE
J. CRUDE PRODUCT STORAGE I. FLAKER
K, SCRUBBER V. BAGGING MACHINE

STEAM
WATER
®.

®
•*- WATER
K *^}
^^r « ,, ®^ CRUDE PHTHALIC ANHYDRIDE
r STORAGE TANK VENT GAS
f 	 ® n
'• -®,-J S ®
IVACUUM ^ AIR
'ff\| 	 ^. L _
^1L ^ -*^ Fua

TO SECONDARY
INCINERATOR
, . W. REFINED PHTHALIC ANHYDRIDE
STORAGE TANK VENT GAS
A® (3)
T^ ^^ ^ o A*CO inn D»rriro i/tur p«c
-*1 	 1


i .A
© B
j
PHTHALIC ANHYORIOt
*" IKANSFOKI
LOADING FACILITY
— ^- TO WAREHOUSE
^" TO TANK CARS

Figure  2.   BASF process for manufacture of phthalic anhydride from o-xylene

-------
                Table 6.  STREAM CODE FOR BASF PROCESS ILLUSTRATED IN FIGURE 2
    Stream
                  Identification
00
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
o-Xylene feed
Sulfur dioxide
Filtered air
Reactor feed
Reactor product
Boiler effluent
Crude product
Condenser off-gas
Scrubber vent
Scrubber liquid purge
Crude PAN product
Pretreatment exhaust
Pretreated crude
Stripping column exhaust
Stripping column overhead
Rectifying column feed
Rectifying column vacuum exhaust
Rectifying column bottom product
Distillation light ends
                                     Stream
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
               Identification
Refined PAN
Water to steam generator
Generated steam
Water to waste-heat boiler
Steam from waste-heat boiler
Scrubber makeup
Incinerator stack gas
Water to cooling coil
Steam from cooling coil
Water to cooling coil
Steam from cooling coil
Incinerator fuel
Combustion air
Flaker and bagger vent
o-Xylene storage vent
Crude product storage vent
Refined PAN storage vent
Loading facility vent
Fugitive emissions

-------
o-xylene
o-methyl-
benzyl
alcohol
o-tolual-
dehyde
o-toluic
acid
                                 0
                                   O
                                C02, CO, H2O
                            HC-
phthalide
 phthalic
 anhydride
     -Cx
      II
     0
  maleic
  anhydride
Direct oxidation of o-xylene or any of the intermediates
to carbon oxides and water can also occur.  A certain
amount of bypassing of intermediates in the sequence is
also possible.  In addition, many other reactions are
possible.  For example, Bernardini and Ramacci presented a
scheme containing 21 compounds and 31 reaction arrows.5

b.   Feed Materials - Feed materials used  in the BASF
process consist of o-xylene, air, and sulfur dioxide.  The
xylene feed  (stream 1 in  Figure  2) contains  95% by  weight
o-xylene, the remainder consisting of meta-  and para-
xylenes  (designated as m- and p-xylenes).  Only the o-
xylene can form phthalic  anhydride.  The m-  and p_-xylenes
are reportedly converted  to  carbon oxides  and water under
commercial reactor conditions.6   However,  Bhattacharyya
 6Friedrichsen,  W.,  et al.   Production of Phthalic
  Anhydride.   British Patent No.  1,082,326 (to BASF),
  September 6,  1967.
                               19

-------
and Gulati7 state that the products of p_-xylene oxidation
include maleic anhydride, p_-tolualdehyde, p_-toluic acid,
terephthalic acid, and p-benzoquinone, while m-xylene oxi-
dation yields maleic anhydride, m-tolualdehyde, isophthalic
acid, and p_-benzoquinone.

Filtered air  (stream 3) is compressed to a gage pressure
of 48.2 kPa to 55.2 kPa and preheated to 149°C using steam.
The liquid xylene feed is also preheated and vaporized by
injection into the hot air stream.  A small amount  (0.5-2.5%
by weight6) of sulfur dioxide  (stream 2) is added to the
feed stream in order to maintain catalyst activity.

c-   Reactor - In the BASF process, the weight ratio of air
to xylene in the feed stream  (stream 4) to the reactor  (unit
F) is 25, which is equivalent to 1 mole percent xylene  (weight
ratios as low as 20 and as high as 34 have been reported in
the literature)-8  Excess air is employed to ensure that the
mixture is below the lower explosive limit of 1.5 mole
percent o-xylene.9  After removal of unvaporized feed, the
reactants are fed to the fixed-bed reactor, which operates
at about 380°C within a range of 300-390°C.6'9
 The active  form of the catalyst contains  a mixture  of V+5
 and V+k.  The role of the S02  is to  adjust the  redox con-
 ditions so  as to maintain the  v+Vv+5 ratio within  an
 optimum range of from one to nine.6
7Bhattacharyya/ S. K., and I. B. Gulati.  Catalytic Vapor-
 Phase Oxidation of Xylenes.  Industrial and Engineering
 Chemistry.  5£: 1719-1726, December  1958.
8Spitz, P. H.  Phthalic Anhydride Revisited.  Hydrocarbon
 Processing.  £7:162-168, November 1968.
9Schwab, R. F., and W. H. Doyle.  Hazards in Phthalic
 Anhydride Plants.  Chemical Engineering Progress.  66:49-
 53, September 1970.                                —
                              20

-------
The BASF reactor has a capacity of approximately 1.44 x 101*
metric tons/yr and a diameter of 4.2 m.8  The reactor con-
tains 9,948 catalyst-filled tubes which are 25.4 mm in dia-
meter and 3 m long.8   (A larger version of this reactor con-
taining approximately 13,000 tubes is also used in some plants.)
A typical BASF plant consists of four reactors which can be
split to operate on two production trains.10  A molten salt
(sodium-potassium nitrate-nitrite eutectic)9 is circulated on
the shell side of the reactor to remove the heat produced in
the exothermic reaction.  The molten salt passes through an
external heat exchanger (unit G) where high pressure steam is
produced.  The gases leave the reactor at 375°C (stream 5)
and pass through a waste heat boiler (unit H) for additional
steam generation.

The conversion of the xylene in the reactor is 100%, with
a maximum yield of 1.03 kg phthalic anhydride per kilogram
of 95% o-xylene feed.11  (The theoretical yield is 1.39 kg
per kilogram of 100% o-xylene.1)

d.   Switch Condensers - Due to the excess (567%)  of air
employed in the reactor, the partial pressure of the
phthalic anhydride in the effluent gas stream is such that
the dew point is below the melting point  (130.8°C) of
phthalic anhydride.  Hence, the product condenses as a solid.
This operation is carried out in a parallel bank of tubular
condensers (unit I) which are alternately heated and cooled
by separate heat-transfer oil streams on an automatically con-
trolled cycle.  During the cooling portion of the cycle,
phthalic anhydride crystallizes on the outer surfaces of
the finned tubes.  During the heating portion of the cycle,
the solid phthalic anhydride is melted and then transferred
to the crude product storage tank  (unit J).  A typical
10Phthalic Anhydride:  Return to Overcapacity.  Chemical and
  Engineering News.  4j3:18-19, March 15, 1971.
 Phthalic Anhydride by Vapor-Phase Oxidation.  The Oil and
  Gas Journal.  7JL:92, March 12, 1973.
                                21

-------
operation1.2 employs a sequence of nine individual condensers,
with six condensing and two melting at any given time.  The
ninth unit is on standby to allow for periodic cleaning and
maintenance without disruption of service.  Residual gases
from the condensers  (stream 8) at a temperature of 66°C and
gage pressure of 26  kPa are sent to a water scrubber  (unit  K)
or, in  some plants,  directly  to an incinerator  (unit L).

e.   Product Purification  - The crude product, which contains
99% to  99.5% phthalic anhydride as phthalic acid,11 is  stored
at 149°C under  atmospheric pressure.  The refining operation
consists of two steps:  a  heat treatment step followed  by a
vacuum  distillation.  The  crude product  (stream 11) passes
through a preheater  (unit  M)  and then to the pretreatment
tank  (unit N),  where it is held at an elevated temperature
under vacuum  for 8 to 12 hours.  The purpose of the heat
treatment is  to decompose  color-forming compounds and to
convert them  to higher molecular weight materials that  can
be separated  from the main product by distillation.  At the
same  time, dissolved phthalic acid is dehydrated to the
anhydride, and  the associated water and other low-boiling
materials, such as maleic  anhydride and benzoic acid, are
partially evaporated and removed through the vacuum jet
ejector exhaust stream  (stream 12).  This exhaust stream  is
sent  to either  the scrubber  (unit K) or a separate control
device  (which may be another  scrubber or another incinerator).

The pretreatment tank is discharged to a continuous distil-
lation  system consisting of a stripping column  (unit  P) and
a rectifying  column  (unit  S), both of which operate in  the
range of 2.67 kPa to 26.7  kPa absolute pressure.  The discharge
 i2Riley,  H.  L.   How to  Design and Operate Fluidized-Catalyst
   PA Plants.   Hydrocarbon Processing and Petroleum Refiner.
   42:167-172,  June 1963.
                               22

-------
stream passes through a precooler (unit O) and enters the
stripping column at about 186°C.  Maleic anhydride and benzoic
acid are separated in the overhead stream (stream 15) which
is sent to the main incinerator (unit L) or to a secondary
control device (via stream 19)  for disposal.  The bottom
stream from the stripping column passes to the evaporators
(reboilers, units Q and R).  Part of the stream is recycled
to the stripping column, and the remainder  (stream 16) enters
the rectifying column  (unit S).

The bottom stream  (stream 18) from the rectifying column
(the residue) is sent to the main incinerator or a secondary
incinerator for disposal.  Phthalic anhydride  (99.99%) is
taken overhead, condensed, and sent to the product storage
tank (unit T) where it is stored at atmospheric pressure
and 149°C.

The phthalic anhydride in the storage tanks can be pumped
to tank cars for shipment as a  liquid,  or it  can be  sent to
a flaking machine  (unit U) for solidification.  The  flaked
product is weighed and packaged in 36-kg bags by a bagging
machine (unit V).

f.   Incinerators - The switch-condenser off-gas may be
treated either by a scrubber-incinerator combination or by
direct incineration.  These operations  are  common to both
the o-xylene and naphthalene based plants.  The scrubber-
incinerator combination is discussed in Section III.A.2.g,
"Naphthalene Based Process," since the  available data are
for a naphthalene based plant.  A dual  incineration  system,
which is currently in operation on a 5.9  x  10H metric
tons/yr o-xylene based plant, is described  below.  Either
system should be equally applicable to  o-xylene and  naphtha-
lene based operations.
                              23

-------
The dual incineration system employs two thermal incinerators,
one of which treats the switch-condenser off-gas while the
other treats all of the waste streams from the product puri-
fication section of the plant.  However, in the plant
currently using the system, the switch-condenser off-gas is
handled by two parallel incinerators, one on each of the two
production trains.  The total phthalic anhydride design
capacity of this system is 8.84 x 101* metric tons/yr.
The two parallel  incinerators incorporate waste-heat boilers
which produce steam at 400°C and a gage pressure of  4,480
kPa.13  The incinerators operate at  700°C to 760°C and the
combusted gases exit  from  the stacks at 250°C.  The  reported
control efficiency of this unit is 99.9% for carbon monoxide
and 96.5% for combined organics when operating at 760°C.13

The secondary incinerator  used to treat the waste streams
                                                            •
from the product  purification section of the plant is designed
to handle overheads and residues, and burns natural  gas to
maintain an operating temperature of 650°C.  The reported
control efficiency achieved by the unit is 99% for combined
organics.13

g.   Storage Tanks -  The feedstock,  crude product, and re-
fined product storage tank requirements for a typical 5.9 x 104
metric tons/yr o-xylene based phthalic anhydride plant are
summarized in Table 7.13

Conservation vents can be  used on o-xylene storage tanks to
minimize breathing losses.  The breather valves are  set to
open between 1.0  kPa  and 1.47 kPa pressure.1H  o-Xylene is
 13Personal  communication,  H.  M.  Lacy,  Monsanto Company.
   April  7,  1975.
 1£tFawcett,  R.  L.   Air Pollution  Potential of Phthalic Anhy-
   dride  Manufacture.   Journal of the Air Pollution Control
   Association.   2£: 461-465,  July 1970.
                              24

-------
       Table 7.  SUMMARY OF TANKAGE REQUIREMENTS FOR A
           5.9 x 104 METRIC TONS/YR o-XYLENE BASED
                  PHTHALIC ANHYDRIDE PLANT
Tank
no.
1
2
3
4
5
6
7
8
Material stored
o-Xylene
o-Xylene
Crude product
Crude product
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Capacity,
m3
329
3,331
360
360
341
341
379
757
Turnovers
per year
230
23
65
65
72
72
44
44
        aDetailed storage tank calculations  are shown in
         Appendix A.
stored at ambient temperature and has a vapor pressure of
from 0.13 kPa to 1.47 kPa under these conditions.  The tanks
may also be blanketed with nitrogen to prevent the formation
of an explosive mixture in the vapor space above the liquid
xylene.

Crude and refined phthalic anhydride storage tanks are
maintained at 149°C to 160°C, at which temperatures the
vapor pressure is 2.27 kPa to 3.33 kPa.  The tanks are
blanketed with dry nitrogen to prevent the entry of oxygen
and water vapor.  The former creates a fire hazard while
the latter results in the hydrolysis of phthalic anhydride
to phthalic acid.  Conservation vents are not used on these
                                25

-------
tanks because the phthalic anhydride vapors tend to solidify
and plug the vents.  Control can be maintained by venting
the tanks to the incinerator or by the use of condensers
or sublimation boxes in which the phthalic anhydride is
removed from the vent stream by condensation and solidifi-
cation.

h.   Heat-Transfer Circuits - The BASF process employs three
heat-transfer fluid systems:   (1) a molten salt system which
removes heat from the reactor  (units F and G in Figure 2);
(2) an organic fluid  (e.g., Dowtherm A,  a diphenyl-diphenyl
oxide eutectic) system that heats and cools the switch con-
densers  (units I); and  (3) an organic fluid system which
services the product purification section (units M, N, O, Q,
R).  Molten salt heat exchange systems and organic systems
are described in the literature.15'16"22
15Uhl, V. W., and H. P. Voznick.  Molten Salt as a Heat
  Transfer Medium.  Chemical Engineering Progress.   59:33-
  35, May 1963.
16Fried, J.  R.  Heat-Transfer Agents for High-Temperature
  Systems.   Chemical Engineering.  8jO: 89-98, May 28,  1973.
17Seifert, W. F., L. L. Jackson, and C. E. Sech.  Organic
  Fluids for High-Temperature Heat-Transfer Systems.
  Chemical Engineering.  79^:96-104, October 30, 1972.
18Purdy, R.  B., et al.  Indirect Heating with Aromatic Oils.
  Chemical Engineering Progress.  _5_9_: 43-46, May 1963.
19Conant, A. R., and W. F. Seifert.  Dowtherm Heat Transfer
  Medium.  Chemical Engineering Progress.  5J^:46-49,  May 1963
20Petersen,  D. E., and R. K. Bedell.   UCON Heat Transfer
  Fluid.  Chemical Engineering Progress.  59^36-39,  May 1963.
21Davis, W.  J., and P. G. Benignus.  Therminol FR-2  Heat
  Transfer Systems.  Chemical Engineering Progrsss.
  59^:39-42, May 1963.
22Frikken, D. R., K. S. Rosenberg, and D. E. Steinmeyer.
  Understanding Vapor-Phase Heat-Transfer Media.  Chemical
  Engineering.  82^86-90, June 9, 1975.
                                26

-------
Details of the heat-transfer  circuits employed in phthalic
anhydride plants  are  not  available.   However,  the essential
features of the organic fluid systems are illustrated in
the basic liquid-phase heat-transfer system shown in Figure 3
Heat-transfer fluid is pumped from the direct-fired heater
to the user system.   A surge  tank is located on the suction
side of the pump  in order to  accommodate thermal expansion
of the liquid during  startup  and surging due to the venting
of line-trapped steam.
                                      VENT TO
                                     ATMOSPHERE
                                  SURGE
                                  TANK
NATURAL GAS
   OR
  FUEL OIL
VALVE
          HEATER

                          RELIEF
                          VALVE
                                                    USER SYSTEM
                          VALVE
       Figure 3.  Organic  liquid heat-transfer system

Surge tanks are normally designed for maximum temperatures
of 50°C to 60°C in order to  provide a cold-fluid seal to
the atmosphere, thereby minimizing fluid oxidation.  In
order to achieve these temperatures,  tanks are not insula-
ted, and may be water jacketed.   The surge tank is vented
to the atmosphere when the vapor pressure of the fluid is
sufficiently low to permit an  unpressurized system and the
fluid is sufficiently stable to  air oxidation.  Higher
                                27

-------
vapor pressure fluids require system pressurization, which
can be accomplished by blanketing the surge tank with an
inert gas.

i.   Waste Water Streams - One source of waste water in
phthalic anhydride plants is the steam boiler blowdown, which
amounts to 0.52 kg per kilogram of phthalic anhydride.13
Another source of waste water in some plants is the secondary
scrubber used to control the waste streams from the product
purification section of the plant.  The liquid purge from
this scrubber must either be incinerated or sent to a waste
water treatment facility.

j.   Plant Shutdown, Turnaround, and Start-up - Data on
o-xylene based phthalic anhydride plant shutdowns, turn-
arounds, and start-ups were obtained from industry.13
Phthalic anhydride plants are shut down on an average of
once per year.  During start-up, the reactor is brought to
operating temperature by heating the process air in a natural
gas heater.  The mixture of hot air and combustion gases is
passed through the reactor and then vented directly to the
atmosphere.

Start-up or other operating upsets can result in emissions
of heat-transfer fluid (designated "oil" in Figure 2) from
surge tank vents in the heat-transfer circuits associated
with the switch condensers and with the product purification
section of the plant.  The emissions are in the form of an
aerosol mist, which reportedly settles in the immediate
vicinity of the emission point due to the low vapor pressure
of the heat-transfer fluid.  These emissions are estimated
to be 0.016 g/kg phthalic anhydride, of which 95% is estimated
to originate from the switch-condenser circuit.13
                               28

-------
k.   Catalyst - Many variations of the vanadium oxide
catalyst are possible.6  In one example,6 the catalyst
consists of 3.75 wt % vanadium oxide (as V2O5) and 21.25
wt % potassium pyrosulfate on a titanium dioxide carrier
(75 wt %).   To activate the catalyst, about 65% of the
vanadium is converted to the tetravalent state by heating
in air containing SO2.  Under the reaction conditions, the
vanadium oxide-potassium pyrosulfate mixture is present as
a melt in the pores of the Ti02 carrier which has a grain
size of 0.2 mm to 0.5 mm.

In a study of vanadium oxide catalyzed oxidation reactions,
Simard, et al.23 concluded that the organic oxidations
take place at the surface of the catalyst through ionic
reactions.   Based on this observation, one possible mechanism
for the reaction is indicated in Figure 4.  Simard, et al.
postulated that the surface layer of the catalyst is
composed of oxygen ions and pentavalent vanadium ions in
the geometrical arrangement shown.  In Step 1, an o-xylene
molecule is adsorbed on the catalyst surface.  In Step 2,
a hydrogen ion is removed from the hydrocarbon by a surface
oxygen ion, and an electron from the hydrocarbon is trans-
ferred to a vanadium ion.  In Step 3, a covalent bond is
formed between the available carbon atom and the hydroxide
group and an electron is transferred from the organic
molecule to a vanadium ion.  In the fourth step, the
organic molecule (o-methylbenzyl alcohol) is desorbed from
the surface leaving a vacant site on the surface.  Following
this, an oxygen atom from an oxygen molecule  in the gas
phase would fill the vacant site and be reduced to a minus
23Simard, G. L., et al.  Vanadium Oxides as Oxidation
  Catalysts.  Industrial and Engineering Chemistry.
  47_:1424-1430, July 1955.
                                29

-------
        0.367 nm
                                            0.367 nm
                               OO
                                                   O    O     O
                                                                               0.367 nm
                     STEP 1
                                                        STEP 2
                                                                                           STEP 3
OJ
O
                             0.367 nm
   O    O     O
O©  O    O  o O
   O o O o  O
O ©  O    O  o O
   O    O     O
                                                                           (IN THE GAS PHASE)
                                          STEP 4
               Figure  4.   Reaction mechanism for  the catalytic oxidation of o-xylene

-------
2 valence while the two adjacent V+Lt ions would be oxidized
to V+5.  Further oxygen atoms would then be added to the
organic molecule in a similar fashion.

Alternatively, the reaction could proceed via oxygen ions
adsorbed on the catalyst surface rather than incorporated
in the lattice.  In either case, electrons must be trans-
ferred from donor sites in the catalyst to oxygen atoms in
order to form oxygen ions.  The catalytic activity of the
vanadium oxide is due in part to the fact that it is a
semiconductor  (n-type) which facilitates these electron
transfer processes.24

The following three-stage mechanism for the oxidation of
sulfur dioxide on a vanadium oxide catalyst has been pro-
posed by Glueck and Kenney:25
             V205-S03 + S02 -

             (vosOiJ2 - ^V2o4*so3 + so3

             V2O4-SO3 + 1/2 02 - ^V2O5

In the above equations, V2O5-SO3 represents vanadium pentoxide
in the melt associated with sulfur trioxide formed from the
decomposition of the pyrosulfate anion.  The result of the
first two steps is the reduction of vanadium from the
pentavalent to the tetravalent state.  This provides an
explanation of the role of the sulfur  dioxide in maintaining
the v+Vv+5 balance in the catalyst.
24Cullis, C. F.  Heterogeneous Catalytic Oxidation of
  Hydrocarbons.  Industrial and Engineering Chemistry.
  5_9:18-27, December 1967.
25Glueck, A. R., and C. N. Kenney.  The Kinetics of  the
  Oxidation of  Sulphur Dioxide Over Molten Salts.
  Chemical Engineering Science.   2^:1257-1265,  1968.
                                31

-------
1.   Reactor System Heat Balance - The oxidation of o-xylene
is an exothermic chemical reaction.  The standard heats of
reaction for the main reaction and principal side reactions
were calculated on the basis that all of the compounds
involved are in the vapor state, and the results are listed
in Table 8.  The heat released in a commercial reactor has
been estimated to be 17.1 MJ/kga phthalic anhydride formed,26
and this value agrees well with an estimate based on the
data in Table 8.
 Table  8.  HEATS OF  REACTION FOR THE OXIDATION OF o-XYLENE
               Reaction
                                         Heat of reaction
MJ/kg
MJ/kg mol
o-Xylene +  302 •*• phthalic anhydride
                             +  3H20
o-Xylene +  7. 5O2 •*• maleic anhydride
                      +  4C02 +  4H20
o-Xylene +  10.502 ->  8C02 +  5H2O
12.5
28.2
41.2
  1,286
  2,983
  4,387
The heat of reaction is used to generate steam in the molten-
salt heat exchanger and in the waste-heat boiler.  Part of
this steam is used to satisfy process requirements and the
remainder is available for export.  A representative energy
balance for a phthalic anhydride plant is given in Table  9.
 ;1 MJ = 1 x 106 joules = 948 Btu  (see Section  IX).
26Schwartz, W. A., et al.  Engineering and Cost Study of Air
  Pollution Control for the Petrochemical Industry, Volume  7:
  Phthalic Anhydride Manufacture from Ortho-xylene.  Air
  Products and Chemicals, Inc., Houdry Division.  U.S.
  Environmental Protection Agency.  Research Triangle Park.
  Publication EPA-450/3-73-006-g, July 1975.  108 p.
                               32

-------
   Table 9.   REACTOR SYSTEM HEAT BALANCE FOR PRODUCTION  OF
              PHTHALIC ANHYDRIDE FROM o-XYLENEa
                  Heat  in
MJ/kg PAN
     Exothermic heat of reaction
     Feed preheat
     TOTAL
   17.1
    3.8
   20.9
                  Heat out
MJ/kg PAN
     Steam generation
       Reactor internal cooling
       Waste heat boiler
     Reactor heat losses
     Switch condensers
     Incremental effluent heat content
     TOTAL
   10.6
    6.0
    0.1
    3.3
    0.9
   20.9
 Basis:
     Table 18 Material Balance.
     Xylene feed at 26.7°C,  air  at 26.7°C.
     Feed preheated to 149°C.
     Reactor outlet temperature, 375°C.
     Condenser effluent gas  at 65.6°C.
     Effluent from waste heat boiler at  163°C.
 Difference in heat content  between condenser effluent
 streams and reactor feed streams.
2.   Naphthalene Based Process

All of the phthalic anhydride currently produced from
naphthalene in the United States is made via the fluid-bed
process.  Two of the three plants now in operation, which
account for 73% of the naphthalene based production, use
the Badger-Sherwin-Williams process, first commercialized
in the early 1960's.  The Koppers Company plant in Bridge-
ville, Pennsylvania, uses a fluid-bed process developed by
                              33

-------
American Cyanamid, which built and operated the plant in the
1950's.  This process was never licensed, and the plant was
eventually taken over by Koppers when they acquired all of
Cyanamid's facilities at the Bridgeville location.8  Figure
5 is a schematic flow diagram for the Badger-Sherwin-Williams
process, which is described below.  Process streams numbered
in Figure 5 are identified in Table 10.
a.   Chemistry - The vapor-phase oxidation of naphthalene to
phthalic anhydride on a vanadium oxide catalyst can be
represented by the following equation:
                                       O + 2C02 + 2H2O
1+4 1/2 02
naphthalene    oxygen          phthalic   carbon   water
                               anhydride  dioxide
The  side reactions are the consequence of the further
oxidation of phthalic anhydride to maleic anhydride, carbon
oxides, and water.  Based on the work of loffe, et al.,27
the  main sequence of reactions in the oxidation of naphthalene
is as follows:
 27loffe, I. I., et al.  Kinetics and Vapor Phase Oxidation
  Mechanism of Aromatic Hydrocarbons.  VI.   Zhurnal
  Fizicheskoi Khimii  (Journal of Physical Chemistry).
  29^692-698, 1955.
                              34

-------
                                                                                                                                  ATMOSPHERIC EMISSIONS
U>
                                                                                                                                • NAPTHALENE STORAGE TANK VENT GAS
       PROCESS
       AIR
                      PROCES .
                                 STEAM -
                                 WATER -
                                        1
                    STEAM
                                         ?
   1®

TmFRESH^G
^"*	r*UJ4TCD(^	
                                       WATER
                     VACUUM TO
                     SCRUBBER OR 1,
                     INCINERATOR.-1-1
                         OIL
                       VACUUM TO
                      SCRUBBER OR
                      INCINERATOR
                            (1?
              A. NAPTHALENE STORAGE
              B. COMPRESSOR
              C. AIR PREHEATER
              D. REACTOR (FLUID BED)
              E. CATALYST FILTER
              F. STEAM GENERATOR
              G. PARTIAL CONDENSER
              H. SWITCH CONDENSERS
              I  CRUDE STORAGE
                                                    *- SCRUBBER VENT GAS



                                                       INCINERATOR STACK GAS


                                                    >-FUGITIVE EMISSIONS



                                                       CATALYST STORAGE HOPPER VENTS



                                                       CRUDE PHTHALIC ANHYDRIDE
                                                       STORAGE TANK VENT GAS


                                                       REFINED PHTHALIC ANHYDRIDE
                                                       STORAGE TANK VENT GAS

                                                    *-TRANSPORT LOADING FACILITY VENT
                                                                   PHTHALIC ANHYDRIDE
                                                                   TRANSPORT LOADING
                                                                       FACILITY
                                                                                                                -»- TO TANK CARS
                                                                                                                              -+- aAKER AND BAGGER VENT
        J. SCRUBBER
        K. INCINERATOR
        L, HEATER
        M. CRUDE PRETREATMENT TANK
        N. PRECOOLER
        0. DISTILLATION COLUMN
        P. PHTHALIC ANHYDRIDE STORAGE
        Q. FLAKER
        R. BAGGING MACHINE
                                                                                                        TO WAREHOUSE
BADGER-SHERWIN-WILLIAMS PROCESS FOR MANUFACTURE OF
PHTHALIC  ANHYDRIDE FROM NAPTHALENE
                            Figure  5.    Badger-Sherwin-Williams  process   for  manufacture  of
                                                  phthalic  anhydride  from  naphthalene

-------
    Table 10.  STREAM CODE FOR BADGER-SHERWIN-WILLIAMS
              PROCESS ILLUSTRATED IN FIGURE 5
Stream
   Identification
Stream
   Identification
   2
   3
   4

   5

   6

   7
   8

   9
  10

  11

  12

  13

  14

  15

  16
Naphthalene

Air
Reactor product
Crude from partial
condenser
Partial condenser
off-gas
Crude from switch
condenser
Crude product
Switch-condenser
off-gas
Scrubber vent
Scrubber makeup

Scrubber liquid
purge
Crude PAN
Pretreatment vacuum
exhaust
Distillation column
feed
Distillation column
vacuum exhaust
Distillation column
light ends
  17

  18
  19
  20

  21

  22

  23
  24

  25
  26

  27

  28

  29

  30

  31

  32
Distillation column
bottom product
Refined PAN
Incinerator stack gas
Incinerator fuel

Combustion air

Water to steam
generator
Generated steam
Cooling water

Steam
Flaker and bagger
vent
Naphthalene storage
vent
Crude product
storage vent
PAN storage vent

Catalyst storage
hopper vents
Loading facility
vent
Fugitive emissions
                              36

-------
                                             o
                                              II
                H
                                0
                                II
                                c>
                                             C\
                                               O
                                             •cx
naphthalene  1,4-dihydroxy-   1,4-naphtho-     phthalic
           naphthalene      quinone         anhydride
                                    	~CO2,  CO,  H2O
                         HC	Cx
                                 II
                                O
                       maleic  anhydride
Direct oxidation of naphthalene  or  any of the intermediates
to carbon oxides and water  can also occur.   A certain amount
of bypassing of intermediates in the sequence is also
possible.  In addition,  a reaction  path involving
1,2-naphthoquinone as  an intermediate has been reported in
the literature.28

b.   Feed Materials -  Feed  materials used in the Badger-
Sherwin-Williams process consist of naphthalene and air.

Two of the naphthalene based plants use petroleum naphtha-
lene while the  other two use desulfurized coal-tar
 28Shelstad,  K.  A.,  J.  Downie, and W. F. Graydon.  Kinetics
  of  the  Vapor-Phase Oxidation of Naphthalene Over  a
  Vanadium Catalyst.  Canadian Journal of Chemical  Engin-
  eering.  3£:102-107, August 1960.
                               37

-------
naphthalene.  Typical compositions of petroleum and coal-
tar naphthalene are:29'30
                                        Wt %
                          Naphthalene
Coal-tar naphthalene      92   - 96
Petroleum naphthalene     99.0 - 99.7
  Sulfur
0.5 - 1.0
  0.001
Tetralin
    2
    0
Present-day catalysts are capable of handling naphthalene
feedstocks with sulfur content as high as 1%.12'31  However,
the sulfur in coal-tar naphthalene is principally in the
form of thionaphthene, which is converted to maleic anhydride
and sulfur oxides rather than phthalic anhydride.12'32'33
The resultant loss in yield of phthalic anhydride (1% sulfur
is equivalent to 4.2% thionaphthene) is apparently the pri-
mary reason for the use of desulfurized naphthalene.

The reaction products of the tetralin  (1,2,3, 4-tetrahydro-
naphthalene) present in coal-tar naphthalene are not known.
The most likely products, however, are maleic anhydride,
carbon oxides, and water.  Some tetralin may also be
29Graham, J. J., and P. F. Way.  Phthalic Anhydride by
  Fluid Bed Process.  Chemical Engineering Progress.
  5£: 96-100, January 1962.
30Personal communication, L. B. Evans, U.S. Environmental
  Protection Agency  (Data originally supplied by U.S.
  Steel Corp.).  June 12, 1975.
31Chopey, N. P.  Fluid-Bed Phthalic Anhydride.  Chemical
  Engineering.  6J9:104-106, January 22, 1962.
32Riley, H. L.  Design of Fluidized Reactors for Naphtha-
  lene Oxidation:  A Review of Patent Literature.  Trans-
  actions of the Institution of Chemical Engineers  (London)
  3J7:305-313, 1959.
33Improvements Relating to the Oxidation of Aromatic
  Hydrocarbons.  American Cyanamid Company.  British
  Patent No. 850,817, January 2, 1957.
                              38

-------
converted to phthalic anhydride via a mechanism such as
the following:
                H 0-OH
tetralin    1-hydroperoxy-    <*-tetralone    1,4-naphtho-
            tetralin                         quinone
                            O
                            II
                         x?-VC\
                         £)C /°
                         •
-------
particles in the fluidized bed reactor.29' 34  Within the
fluid bed, a uniform temperature is maintained  (to ± 5°C)
in the range of 340°C to 380°C.3k  Reactor pressure is set
by the backpressure of equipment downstream.

The reaction vessel itself is simply a large container of
sufficient size to hold the required amount of  catalyst.
Single reactors having capacities of more  than  4.5 x 104
metric tons/yr are feasible with the fluid-bed  process.8

The catalyst bed is supported on a gird plate having holes
through which the fluidizing air is blown.  Naphthalene  feed
is introduced into the fluid bed through nozzles.  Adequate
space above the normal bed level is provided to allow for
catalyst  settling before the gases leave the reactor.  The
ratio of  bed depth to diameter  is about 3:1 and the contact
time during reaction is 10 to 20 seconds.29

Removal of the heat produced by the exothermic  reaction  is
accomplished by means of cooling tubes located  directly  in
the catalyst bed.  Water is circulated in  these tubes and
steam is  thereby generated directly without the need of  a
secondary heat-transfer fluid.

The conversion of naphthalene in the reactor is 100% with
a yield of 0.97 kg phthalic anhydride per  kilogram of
naphthalene feed.35  (The theoretical yield is  1.16 kg per
kilogram  of 100% naphthalene.1)
 34Graham, J. J.  The Fluidized Bed Phthalic Anhydride
  Process.  Chemical Engineering Progress.  66:54-58,
  September 1970.
 35Phthalic Anhydride  (Sherwin-Williams/Badger).  The Badger
  Company, Inc.  Hydrocarbon Processing.  4_6:215, November
  1967.
                              40

-------
d.   Catalyst Filter - Effluent gases leaving the reactor
are cooled to about 260°C before entering the catalyst filter
unit (unit E) in order to prevent secondary reactions taking
place on the filters.3  The entrained catalyst particles are
separated from the gas stream by specially designed porous
ceramic filter elements.  Process air is used to periodically
blow back the filters, and the catalyst particles are returned
directly to the reactor bed.  Fiber glass filters, used in
some processes, are described in detail in Reference 12.

e.   Condensers - Due to the low air to naphthalene ratio
employed in the fluid-bed process, the crude phthalic
anhydride is recovered both as a liquid and a solid.  Between
40% and 60%29 of the product is obtained directly as a
liquid in the partial condenser  (unit G) and the remainder
is condensed as a solid in the switch condensers  (units H).
The latter operate in the same manner as those described
for the BASF process.  However, the partial condensation of
product plus the reduced air rate greatly reduces the load
on the switch condensers with a concomitant reduction in
size and cost.

The tail gases (stream 8) leave the switch condensers at
66°C and are sent either to a water scrubber  (unit J) or
directly to the incinerator (unit K).  The low air rate
 The cooling may be accomplished in a quenching section
 within the reactor itself, rather than in an external heat
 exchanger as shown in Figure 5.  The quenching section is
 simply a second fluidized bed located above the main bed
 and separated from it by a grid plate.  Further details of
 this reactor scheme may be found in Reference 36.
  Rousseau, W. P. C.  Production of Phthalic Anhydride.
  U.S. Patent No. 3,080,382 (to the Badger Co.), March
  5, 1963.
                              41

-------
makes direct incineration more economical than in the fixed-
bed process.

The crude product is stored as a liquid at 149°C and atmos-
pheric pressure  (unit I).

f.   Product Purification - Crude phthalic anhydride  (stream
12) from the crude product storage tank passes through a
preheater  (unit  L) and then to the pretreatment tank  (unit
M) in much  the same manner as in the BASF process.  Treating
chemicals consisting of  maleic anhydride  (1 g per kilogram
PAN) and sodium  hydroxide  (0.05 g per kilogram PAN) are
added to promote the pretreatment process.37' In the
Badger-Sherwin-Williams  process the final product is obtained
by means of a batch distillation column which operates at
an absolute pressure of  2.66 kPa.  The light ends  (stream 16)
are taken off at a reflux ratio of 40 and sent to the
incinerator for  disposal.  The main cut is then taken at a
reflux ratio of  1, condensed, and sent to the phthalic
anhydride storage tank  (unit P), where it is held at  149°C
and atmospheric  pressure.  The residue, or bottoms  product,
is sent to  the incinerator.

The refined phthalic anhydride  (99.7% minimum) can  be
pumped to tank cars for  shipment in liquid form, or it can
be sent to  flaking and bagging machines  (units Q and  R) for
shipment as a solid.
 37Personal communication, L. B.  Evans,  U.S.  Environmental
   Protection Agency  (Data originally  supplied by Union
   Carbide Corporation).  June  12,  1975.
                               42

-------
g.   Scrubber and Incinerator -

(1)   Scrubber - The characteristics of the effluent gas from
the switch condensers impose a number of constraints on the
design of a scrubber for phthalic anhydride plants:

     •  The efficient removal of aldehydes and maleic
       anhydride requires a multistage rather than a
       single-stage unit.14'38

     •  Since the liquid effluent from the scrubber must
       be incinerated to avoid a water pollution problem,
       a recycle system is necessary to minimize the
       volume of the effluent stream.

     •  The organic compounds present in the condenser
       tail gas are highly corrosive, thus requiring
       the use of corrosion resistant materials of
       construction.

A scrubber designed to meet the above requirements is
illustrated in Figure 6.38   (The process streams in Figure 6
are numbered for later reference to the material balance
given in Table 11.)  The unit consists of two stages, each
of which contains a conventional fluid-bed packing on a
supporting grid.  The two stages are separated by a conical
shaped deflector plate and collection tray.  The unit is
designed to operate between 35°C and 40°C with an organic
removal efficiency of 98% to 99%.  This high efficiency  is
possible because most of the organics are solids at the
scrubber operating conditions.  The material of construc-
tion is 316 stainless steel.
38Ferrari, D. C., and C. G. Bertram.  Method and Apparatus
  for the Removal of Organics from Chemical Waste Gases.
  U.S. Patent No. 3,624,984  (to the Badger Co.), December
  7, 1971.
                                43

-------
                       VENT GAS
        MIST ELIMINATOR
                            \
  CONDENSER
  TAIL GAS
RECYCLE _


2"d STAGE
PURGE ,

^ t
k-


Xto*.

M

•*•-*• TT^f
2nd
STAGE
/ \
*?F "^^
1st
STAGE
v/
.i^^—' rnrni ifl
^ FREsH V\


ist STAGE
^ RECYCLE



1st STAGE PURGE
TO INCINERATOR
       Figure 6.  Schematic diagram of scrubber  for
                phthalic anhydride plant38
The condenser tail gas enters the  first  stage  at approxi-
mately 66°C.  The scrubbing liquor for this  stage consists
of the purge from the second stage and the first stage
recycle stream.  The liquor is  slurry containing 10% to 12%
by weight of organic solids.  The  concentration of organic
compounds, both dissolved and in the form of slurry solids,
ranges from 10% to 50% by weight depending on  the composi-
tion of the gas being treated.  A  portion of the recycle
stream is continuously purged and  sent to the  incinerator
for disposal.

In the second stage, the gas is scrubbed with  a dilute
solution  (0.5% to 3% by weight) of organic pollutants
(chiefly maleic acid).  Fresh water is added to the second
                               44

-------
stage at a rate sufficient to replace the water removed
from the scrubber in the two exit streams.  The makeup
water combines with the second  stage recycle  stream to form
the second stage scrubbing liquor.  The  scrubbed gas  stream
passes through a mist eliminator and is  then  vented to the
atmosphere at 38°C.

Table 11 gives a typical  scrubber material  balance, obtained
from the Badger patent,38 for a naphthalene based  phthalic
anhydride operation.  In this example, only the condenser
effluent stream is fed to the scrubber.   In practice, the
    Table  11.   TYPICAL  SCRUBBER MATERIAL  BALANCE  FOR A
         NAPHTHALENE  BASED  PHTHALIC  ANHYDRIDE  PLANTa
Stream No. :
Description:
Temperature, °C:
Component
Air
H20
Maleic anhydride
Phthalic anhydride
Benzoic acid
Naphthoquinone
Maleic acid
Phthalic acid
TOTALS
1
Condenser
tail gas
66
2
Makeup
water
16
3
Vent gas
38
4
1st stage
purge
38
Flow rates, kg/hr
222,273
4,545
291
141
11
11


227,272

6,272






6,272
222,273
9,545
4.6
2.3
0.5
0.9


231,826

1,202


11
11
339
155
1,718
 Blanks indicate no mass flow of component
exhaust gases from the steam ejectors in the purification
section of the phthalic anhydride plant may also be fed to
the scrubber.  In addition, it will be noted that sulfur
                               45

-------
and carbon oxides are included with the air, since they
behave as inert gases in the scrubbing operation.  The flow
rates in this example correspond to a production rate of
approximately 5.68 x 104 metric tons/yr.

 (2)  Incinerator - Incinerators used in phthalic anhydride
plants are of two types:   direct flame  (thermal) units and
catalytic units.  Catalytic  incinerators  are auto-thermal
and operate at temperatures  of 427°C to 482°C.  The  catalyst
is platinum, platinum family or platinum  activated alumina
on a metal ribbon mesh  or  ceramic  base.14  Thermal incinerators
operate  at temperatures of 700°C to  982°C and  consume  fuel
 equivalent to 279 kJ per kilogram  of phthalic  anhydride
 produced.38
 A thermal  incinerator38 is shown  schematically in  Figure 7-
 (The  process  streams  in Figure  7  are numbered  for  later
 reference  to  the material  balance  given  in Table  13.)  The
 primary  feed  stream  to  the incinerator is the  liquid purge
 from  the first  stage of the scrubber.   Other feed  streams
 may  include  the residue (liquid),  light  ends (gas),  and
 ejector  exhaust (gas)  streams  from the product purification
 section  of the  plant.   Heating values  of the main organic
 components in the  scrubber purge  stream are listed in
 Table 12.   The  heating  value of the liquid residue stream
 is estimated to be  23.3 MJ/kg.38
       Table  12.   HEATING VALUES  OF  ORGANIC  COMPOUNDS  IN
                     SCRUBBER PURGE  STREAM
              Compound
           Maleic  acict
           Phthalic  acid
           Benzoic acid
           Naphthoquinone
  Gross heat of
combustion, MJ/kg
       13.8
       21.6
       26.5
       29.1
                                46

-------
                             TO
                          ATMOSPHERE
                  RESIDUE
                   PITCH
            LIGHT ENDS
           AND EJECTOR
             EXHAUST
       LIQUID PURGE
       FROM SCRUBBER
                                       V  |  ®   COMBUSTION
                                                  AIR
                                                 FUEL GAS
            ATOMIZING STEAM
               OR AIR
    Figure 7.  Schematic  diagram of thermal incinerator
              for  phthalic  anhydride plant38
The liquid purge  stream is atomized immediately before
entering- the combustion area of the incinerator by either
a steam or an air stream.   Fuel gas (e.g., methane) and
combustion air  (25%  in excess of the stoichiometric amount)
are introduced  into  the incinerator in a conventional manner
to provide an operating temperature of 760°C to 870°C.   Heat
recovery is achieved by means of a coil which can be used
either to preheat the combustion air (as shown in Figure 6)
or to generate  steam.

The incinerator reportedly attains an efficiency of 99.9%
conversion of organic materials.  A material balance on  the
incinerator38 is  shown in  Table 13.  The feed stream in  this
example is the  scrubber purge stream of Table 11.
                               47

-------
           Table  13.
INCINERATOR MATERIAL BALANCE FOR A NAPHTHALENE BASED PHTHALIC
                  ANHYDRIDE PLANT3
Stream No. :
Description:
Temperature, °C:
Component
Maleic acid
Phthalic acid
Benzoic acid
Naphthoqu inone
H2O
Oxygen
Nitrogen
Methane
C02
Organic s
TOTAL
1
Scrubber
purge
38
2
Atomizing
steam
186
3
Fuel
gas
16
4
Combustion
air
16
Flow rates, kg/hr
339
155
11
11
1,202





1,718




572





572







30


30





823
2,717



3,540
5
Stack gas
760





2,006
155
2,717

982
0.0059
5,860
00
        Blanks  indicate no mass flow of component.

-------
h.   Storage Tanks - Table 14 summarizes the feedstock,  crude
product, and refined product storage tank requirements for a
5.9 x lO4 metric tons/yr naphthalene based phthalic anhydride
plant.30  The crude and refined phthalic anhydride is stored
under the same conditions as described previously for o-xylene
based plants.  Naphthalene storage tanks are maintained at
85°C to 100°C, at which temperatures the vapor pressure is
1.33 kPa to 2.67 kPa.  Emission of naphthalene vapor can be
controlled by means of conservation vents.
        Table 14.  SUMMARY OF TANKAGE REQUIREMENTS FOR A
           5.9 x 104 METRIC TONS/YR NAPHTHALENE BASED
                    PHTHALIC ANHYDRIDE PLANT
Tank
no.
1
2
3
4
5
6
7
8
9
Material stored
Naphthalene
Naphthalene
Naphthalene
Crude product
Crude product
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Capacity,
m3
3,785
3,785
1,893
333
1,514
182
182
568
568
Turnovers
per year
1
13
26
148
2
130
130
2
40
 i.   Catalyst Storage Facilities  - The  fluid-bed process
 involves two catalyst storage  facilities which  are not
 shown in Figure 5.  During the blow-back cycle  of the
 catalyst filter  (unit E), the  fines  fraction of the  fluid-
 bed catalyst is separated and  removed from the  system as
 "spent" catalyst.  The fines fraction consists  primarily  of
 catalyst particles less  than 40 ym in diameter.  The spent
 catalyst stream is sent  to a tangential cyclonic separator
 where the catalyst particles are  removed from the entraining
 air stream with a removal efficiency of 98.5%.37  The
                                49

-------
recovered catalyst flows by gravity to a drum, approximately
0.2 m3 in volume, which is designated the spent catalyst
storage facility.  The exhaust gas stream from the cyclone
is vented to the atmosphere.  This operation is intermittent
and totals approximately 50 hr/yr.37

During periods of reactor shutdown for routine or emergency
maintenance, the catalyst is removed from the reactor and
transferred pneumatically to a catalyst storage hopper.
This operation is performed an average of four to eight
times per year, and has a duration of about 12 hours.37

The hopper vent is equipped with a cyclone separator that
recovers entrained catalyst with an efficiency of 90%.37
The recovered catalyst is returned to the storage hopper,
while the exhaust gas stream from the cyclone vents to the
atmosphere.

j.   Heat Transfer Circuits - The Badger-Sherwin-Williams
process employs two organic fluid heat-transfer systems:
(1) the circuit that heats and cools the switch condensers
(units H); and (2) the circuit which services the product
purification train (units L, M, N, O).  The fluid-bed
reactor is water cooled (units D and F).  The organic fluid
systems are similar to those discussed previously for the
BASF process (Section III.A.l.h.).

k.   Plant Shutdown,  Turnaround, and Start-up - Naphthalene
based phthalic anhydride plants are shut down for emergency
or routine maintenance an average of four to eight times
per year.37  During these periods, the fluid-bed catalyst
is transferred pneumatically from the reactor to the catalyst
                                50

-------
storage hopper as described above.  These transfer operations
result in emissions of catalyst particles from the storage
hopper vent and from fugitive sources.  No data are available
on emissions of heat-transfer fluid (designated "oil" in
Figure 5) from the heating and cooling circuits during start-up
and other process upset conditions.  However, these emissions
are presumably similar to those experienced by o-xylene based
plants, which have been described previously.

1.   Catalyst - Davison Grade 902 Catalyst3 is a finely
divided vanadium oxide catalyst on a silica gel base used
for the direct air oxidation of naphthalene to phthalic
anhydride in fluid-bed reactors.  Typical chemical and
physical specifications are given in Table 15.

 Table 15.  TYPICAL PROPERTIES OF DAVISON GRADE 902 CATALYST
Chemical analysis
Volatile at 374 °C
V205
K2SOk
S03
SiO2
Physical analysis
Bulk density
Surface area
Average pore diameter
Particle size distribution
£20 ym
>20 ym but <40 ym
>40 ym but ^80 ym
>80 ym
Wt %
1
9
29
12
50
Value
195 kg/m3
40 m2/g
3 mm
Wt %
12
11
28
49
 W. R. Grace & Company, Davison Chemical Division.  A  newer
 version of this catalyst, Davison Grade 906,  is  reported  in
 Reference 37.  Grade 906 contains 3.5% V205 by weight.
                                51

-------
The primary purpose of the K2SO4 modifier is to slow down
the reaction rate and prevent over-oxidation of the naph-
thalene.29  The Badger-Sherwin-Williams reactor uses
0.6 gram of catalyst per kilogram of phthalic anhydride
produced.37

m.   Reactor System Heat Balance - The oxidation of naph-
thalene is an exothermic chemical reaction.  The standard
heats of reaction for the main reaction and principal side
reactions were calculated on the basis that all of the
compounds involved are in the vapor state, and the results
are listed in Table 16.  Published figures for the heat
released in a commercial reactor are 16.3 to 20.9 MJ/kg
phthalic anhydride.9  A value of 17 MJ/kg phthalic anhydride
was estimated in Reference 26, and is in good agreement with
an estimate based on the data in Table 16.
      Table 16.
HEATS OF REACTION FOR THE OXIDATION
     OF NAPHTHALENE
                 Reaction
Naphthalene + 4.5O2 -»• phthalic anhydride
                           + 2C02 + 2H2O
Naphthalene + 9O2 -»• maleic anhydride
                       + 6CO2 + 3H2O
Naphthalene + 1.5O2 -> naphthoquinone + H2O
Naphthalene + 1202 •*• 10CO2 + 4H2O
                                            Heat of reaction
                           MJ/kg
MJ/kg mol
                            14.0
                            28.4
                             3.7
                            39.6
  2,732
    395
     51
    547
Steam is generated by circulating water through the reactor
cooling coils, and also in the partial condenser.  Part of
this steam is used to satisfy process requirements and the
remainder is available for export.  A representative energy
balance for a naphthalene based phthalic anhydride plant  is
presented in Table 17.
                                52

-------
 Table 17.   REACTOR SYSTEM HEAT BALANCE  FOR PRODUCTION  OF
           PHTHALIC ANHYDRIDE FROM NAPHTHALENES
                    Heat in
      Exothermic heat of reaction
      Air preheat

      Total
                   Heat out
      Steam generation
        Reactor internal cooling
        Partial condenser
      Reactor heat loss
      Reactor effluent cooler
      Switch condensers
      Incremental effluent heat content

      Total
                                          MJ/kg PAN
                                            17.0
                                             1.5

                                            18.5
                                          MJ/kg PAN
                                            13.2
                                             2.0
                                             0.1
                                             1.6
                                             1.2
                                             0.4

                                            18.5
       Basis:
         Material balance given in Table 19.
         Naphthalene feed at 93°C/ air at 26.7°C.
         Air preheated to 149°C.
         Reactor outlet temperature, 371°C.
         Condenser effluent gas at 66°C.
      Difference in heat content between condenser
       effluent streams and reactor feed streams.
B.
MATERIALS FLOW
1.   o-Xylene Based Process

The flow diagram for the BASF process was given in Figure 2.
Table 18 is a material balance for a typical 5.9 x 104 metric
tons/yr plant, based on data obtained from Reference 26.
                              53

-------
                       Table 18.   MATERIAL  BALANCE FOR A  5.9 x  101* METRIC  TONS/YR
                                   O-XYLENE BASED  PHTHALIC  ANHYDRIDE PLANT
Stream No. :
Description:
Temperature, °C:
Gage pressure, kPa:
Component
Sulfur dioxide
Carbon monoxide
Carbon dioxide
Nitrogen
Oxygen
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Xylene
m- and p_-Xylene
Misc. organics
Particulate
Water
Total
1
Xylene
27
0
2
Sulfur
dioxide
27
3
Air
27
0
4
Reactor
feed
149
50
5
Reactor
product
375
6
Boiler
effluent
163
7
Crude
product
149
0
8
Condenser
off-gas
66
26
Plow rates, kg/hr








7,098.6
334.6


7,433.2
34.1











34.1



140,897.3
42,804.1






1,996.8
185,698.2
34.1


140,897.3
42,804.1



7,098.6
334.6

1,996.8
193,165.5
34.1
1,095.9
3,784.6
140,897.3
32,819.1
7,517.7
591.8
51.8

34.1

6,339.1
193,165.5
34.1
1,095.9
3,784.6
140,897.3
32,819.1
7,517.7
591.8
51.8

34.1

6,339.1
193,165.5



34.5

7,350.4
276.4
31.4

34.1


7,726.8
34.1
1,095.9
3,784.6
140,862.8
32,819.1
167.3
315.4
20.4



6,339.1
185,438.7
Ul
        a
        Blanks indicate data not available.
        Blanks indicate no mass flow of exponent.
        ^Listed in Table 21.

-------
                  Table 18  (continued).   MATERIAL BALANCE FOR A  5.9 x  10U  METRIC TONS/YR
                                     O-XYLENE  BASED PHTHALIC ANHYDRIDE PLANT
Stream No. :
Description:
Temperature, °C:a
Gage pressure, kPa:'
Component
Sulfur dioxide
Carbon monoxide
Carbon dioxide
Nitrogen
Oxygen
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Xylene
m- and p_-Xylene
Misc. organics
Particulate
Water
Total
9
Scrubber
vent
38
0
10
Scrubber
liq. purge
38

11
Crude
PAN
149
0
12
Pretreatment
exhaust

-98.6
13
Pretreated
crude


14
Stripping
col. exhaust

-98.6
15
Stripping col.
overhead

-98.6
16
Rectifying
col. feed
186

Flow rates, kg/hr
34.1
1,095.9
3,784.6
140,897.3
32,819.1
3.2
8.2
6.8


4.6C
7,680.9
186,334.7





193.6
321.8
13.6



3,866.8
4,395.8



34.5

7,350.4
276.4
31.4


34.1

7,726.8



34.5

31.8
14.6




2,181.8e
2,262.7





7,318.6
261.8
31.4


34.1

7,645.9































7,280.0
14.5



34.1

7,328.6
U1
01
         Blanks indicate data not available.
         Blanks indicate no mass flow of component.
         Primarily solids from make-up water.
         B2181.8 kg/hr H20 condensate from steam ejector.

-------
               Table 18  (continued).   MATERIAL  BALANCE FOR  A 5.9  x 10U  METRIC  TONS/YR
                                  O-XYLENE BASED PHTHALIC ANHYDRIDE  PLANT
Stream No. :
Description:
Temperature, °C:
Gage pressure, kPa:
Component
Sulfur dioxide
Carbon monoxide
Carbon dioxide
Nitrogen
Oxygen
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Xylene
m- and p_-Xylene
Misc. organics0
Particulate
Water
Total
17
Rectifying
column vac.
exhaust

-98.6
18
Rectifying
col . bottom
product


19
Distillation
light Ends


20
Refined
PAN
149
0
21
Water to
steam gen.
16

22
Generated
steam
213
1,910
Flow rates, kg/hr


















7.3



34.1


41.4





38.6
247.3
31.4

•


317.3





7,272.7
14.6





7,287.3











28,527.3
28,527.3











28,527.3
28,527.3
23
Water
to waste-
ht. boiler
16













16,263.6
16,263.6
24
Steam
from waste-
ht. boiler
157
476












16,263.6
16,263.6
(Jl
         Blanks indicate data not available.
         Blanks indicate no mass flow of component.
        CListed in Table 21.

-------
         Table  18
(continued).   MATERIAL  BALANCE FOR A  5.9 x  104 METRIC TONS/YR
        O-XYLENE BASED PHTHALIC ANHYDRIDE PLANT
Stream No. :
Description:
Temperature, °C:
Gage pressure, kPa:
Component
Sulfur dioxide
Carbon monoxide
Carbon dioxide
Nitrogen
Oxygen
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Xylene
m- and p_-Xylene
Misc. organics
Particulate
Water
Ethane
Methane
Nitrogen oxides
Total
25
Scrubber
makeup
16

26
Incinerator
stack gas
927
0
27
Water to
cooling coil
16

28
Steam from
cooling coil
129
160
29
Water to
cooling coil
16

30
Steam from
cooling coil
129
160
31
Incinerator
fuel
16

32
Combustion
Air
27
0
Flow rates, kg/hr











3,026.8



3,026.8

18.2
2,759.1
9,660.9
395.4


3.6


3.2f
4,992.3


1.4
17,834.1












































"*•






















30.5








65.9
299.1

395.5



9,630.9
2,916.8






136.4



12,684.1
 Blanks indicate data not available.
^Blanks indicate no mass flow of component.
 Primarily solids from make-up water.

-------
             Table 18  (continued).   MATERIAL BALANCE FOR A 5.9 x 104  METRIC TONS/YR

                               O-XYLENE BASED PHTHALIC ANHYDRIDE PLANT
Stream No . :
Description:
Temperature, °C:
Gage pressure, kpa
Component
Sulfur dioxide
Carbon monoxide
Carbon dioxide
Nitrogen
Oxygen
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Xylene
m- and p_-Xylene
Misc. organics
Particulate
Water
Total
33
Flaker
and bagger vent
27
0
34
o-Xylene
stg. vent
27
0
35
Crude product
stg. vent
149
0
36
PAN
stg. vent
149
0
37
Loading
facility vent
27
0
38
Fugitive
emissions
27
0
Flow rates, b kg/hr





0.007







0.007








0.014




0.014





0.008







0.008





0.02







0.02





0.03







0.03














1/1
00
       Blanks indicate no mass flow of component.

-------
Sources of atmospheric emissions from the BASF.process are:
the scrubber vent (stream 9) and/or the incinerator flue
gas (stream 26),  xylene storage (stream 34),  crude product
storage (stream 35),  refined product storage (stream 36),
flaker and bagger vent (stream 33), transport loading
facility vent  (stream 37),  and fugitive emissions  (stream 38)
In addition, plants which employ a secondary scrubber or
incinerator to control waste streams from the product puri-
fication section of the plant will have emissions from
these devices.

2.   Naphthalene Based Process
The flow diagram for the Badger-Sherwin-Williams process was
given in Figure 5.  Table 19 is a material balance for a
5.9 x 104 metric tons/yr plant, based on data obtained from
Reference 26.

Sources of atmospheric emissions from the Badger-Sherwin-
Williams process are:  the scrubber vent  (stream 9), and/or
the incinerator flue gas (stream 19), naphthalene  storage
(stream 27), crude product storage  (stream 28), refined
product storage (stream 29), flaker and bagger vent  (stream
26), transport loading facility vent  (stream 31),  catalyst
storage hopper vents  (stream 30), and fugitive emissions
(stream 32).

C.   GEOGRAPHICAL DISTRIBUTION

There are currently eight companies manufacturing  phthalic
anhydride at ten locations  in the continental United States,
plus one in Puerto Rico.  Table 20 lists the manufacturers
and plant capacities, and the plant locations are  shown in
Figure 8.  The population densities of the counties  in which
the plants are located range from 30 to 4,905 persons/km2-
                                59

-------
             Table 19.   MATERIAL BALANCE  FOR A  5.9 x  104 METRIC TONS/YR
                       NAPHTHALENE BASED PHTHALIC ANHYDRIDE PLANT
Stream No. :
Description:
Temperature, °C: a
Gage pressure, kPa:
Component
Naphthalene
Phthalic anhydride
Maleic anhydride
Naphthoquinone
Misc. organics
Oxygen
Nitrogen
Carbon dioxide
Carbon monoxide
Water
Total
1
Naphthalene
93
0
2
Air
27
0
3
Reactor
product
260

7,497.3



37.7





7,535.0





22,130.0
68,197.3



90,327.3

7,492.3
101.4
53.2
37.7
12,225.4
68,197.3
7,030.4
366.4
2,356.4
97,860.5
4
Crude from
partial
condenser
149
Flow rates










5
Part. cond.
off-gas

, b kg/hr










6
Crude from
switch cond.
149











7
Crude
product
149
0


7,344.1
50.4
48.2
37.7
17.3
20.0


65.5
7,583.2
8
Switch-cond.
off-gas
66


148.2
50.9
5.0

12,208.2
68,177.3
7,030.4
366.4
2,290.9
90,277.3
 Blanks indicate data not available.
3Blanks indicate no mass flow of component.

-------
     Table 19  (continued).   MATERIAL BALANCE FOR A 5.9  x 10" METRIC TONS/YR
                     NAPHTHALENE BASED PHTHALIC ANHYDRIDE PLANT
Stream No. :
Description:
Temperature, °C:
Gage pressure, kPa:
Component
Naphthalene
Phthalic anhydride
Maleic anhydride
Naphthoguinone
Misc. organics
Oxygen
Nitrogen
Carbon dioxide
Carbon monoxide
Water
Total
9
Scrubber
vent
38
0
10
Scrubber
makeup
16
11
Scrubber
liq. purge
38
12
Crude
PAN
149
0
13
Pretreatment
vac. exhaust
-98.6
14
Dist. col.
feed
186
15
Dist. col.
vac. exhaust
-98.6
16
Dist. col.
light ends
-98.6
Flow rates, kg/hr





12,208.2
68,177.3
7,030.4
366.4






















7,344.1
50.4
48.2
37.7
17.3
20.0


65.5
7,583.2

18.2



17.3
20.0


65.5
121.0





















5.0
35.9







40.9
Blanks indicate data not available.
Blanks indicate no mass flow of component.

-------
               Table  19 (continued).   MATERIAL BALANCE FOR A 5.9  x 1014  METRIC  TONS/YR
                               NAPHTHALENE BASED PHTHALIC  ANHYDRIDE PLANT
Stream No. :
Description:
Temperature , °C :
Gage pressure, kPa:
Component
Naphthalene
Phthalic anhydride
Maleic anhydride
Naphthoquinone
Misc. organics
Oxygen
Nitrogen
Carbon dioxide
Carbon monoxide
Water
Methane
Ethane
Particulate
Nitrogen oxides
Total
17
Dist. col.
bottom prod.

18
Refined
PAN
149
0
19
Incinerator
stack gas
927
0
20
Incinerator
fuel
16
21
Combustion
air
27
0
22
Water to
steam gen.
16
23
Steam from
steam gen.
213
1,910
24
Cooling
water
16
Flow rates, kg/hr

48.2

48.2
37.7









134.1

7,272.7
14.6











7,287.3



















































35,154.6




35,154.6









35,154.6




35,154.6









5,340.9




5,340.9
to
        Blanks indicate data not available.
        Blanks indicate no mass flow of component.

-------
            Table  19 (continued).  MATERIAL BALANCE  FOR A 5.9  x 104 METRIC  TONS/YR
                           NAPHTHALENE BASED PHTHALIC ANHYDRIDE PLANT
Stream No. :
Description:
Temperature, °C:
Gage pressure, kPa:
Component
Naphthalene
Phthalic anhydride
Maleic anhydride
Naphthoquinone
Misc. organics
Oxygen
Nitrogen
Carbon dioxide
Carbon monoxide
Water
Total
25
Steam
135
207
26
Flaker
and bagger
vent
27
0
27
Naphthalene
stg. vent
90
0
28
Crude prod.
stg. vent
149
0
29
PAN
stg. vent
149
0
30
Catalyst storage
hopper vents
27
0
31
Loading facility
vent
27
0
32
Fugitive
emissions
27
0
Flow rates, kg/hr









5,340.9
5,340.9

0.007








0.007
0.044









0.044

0.012








0.012

0.014








0.014











0.03








0.03










u»
        Blanks indicate no mass flow of component.

-------
                                          Table 20.  PHTHALIC ANHYDRIDE PLANTS3'
Number
1
2
3
4
5
6
7
8
9
10
—
Company
Allied Chemical
BASF Wyandotte
Exxon Corp .
Koppers Co.
Koppers Co.
Monsanto Co.
Monsanto Co.
Std. Oil Calif.
Stepon Chem. Co.
U.S. Steel
Occidental Petroleum
Nominal
capacity,
10* metric
tons/yr
15.9
59.0
40.8
40.8
79.4
40.8
59.0
22.7
22.7
68.0
45.4
Location
El Segundo, Calif.
S. Kearny, N.J.
Baton Rouge, La.
Bridgeville, Pa.
Cicero, 111.
Bridgeport, N.J.
Texas City, Texas
Richmond, Calif.
Millsdale, 111.
Neville Island, Pa.
Arecibo, P.R.
County
population
density,
persons/km2
662
4,905
233
842
2,197
64
160
152
112
30
39
Raw
material
o-Xylene
o-Xylene
o-Xylene
Desulf. naphthalene
o-Xylene
Petro naphthalene
o-Xylene
o-Xylene
o-Xylene
Desulf. naphthalene
o-Xylene
Process
Chemiebau (von Heyden)
BASF
BASF
Own (fluid bed)
BASF
Badger-Sherwin-Williams
BASF
Lurgi (von Heyden)
Chemiebau (von Heyden)
Badger-Sherwin-Williams
Chemiebau (von Heyden)
                                        The following plants are not in operation:
—
—
—
—
—
—
—
Union Carbide Co.
Allied Chemical
Allied Chemical
W. R. Grace
Reichhold Chemicals
Reichhold Chemicals
Sherwin-Williams
45.4
45.4
15.0
34.0
13.6
45.4
9.1
Institute and South
Charleston, W. Va.
Frankford, Pa.
Ironton, Ohio
Fords, N.J.
Elizabeth, N.J.
Morris, 111.
Chicago, 111.
95
5,861
47
714
2,020
22
2,197
Petro naphthalene
Naphthalene
Naphthalene
Naphthalene
Naphthalene
o-Xylene
Naphthalene
Badger-Sherwin-Williams
Badger
Own (fixed bed)
Badger-Sherwin-Williams
Badger-Sherwin-Williams
Unknown
Badger-Sherwin-Williams
Numbers refer to code shown in Figure 8.

-------
        Figure 8.  Phthalic anhydride plant locations

The total U.S. capacity in 1975 was 4.95 x 10s metric tons/yr
In addition to this, a 2.27 x 10*4 metric tons/yr expansion of
the U.S. Steel Plant at Neville Island, Pennsylvania is
scheduled for completion in 1975.u
                                65

-------
                        SECTION IV
                         EMISSIONS

A.   SELECTED POLLUTANTS

1.   o-Xylene Based Process

A complete list of the possible reaction products from the
oxidation of xylene, according to the work of Bernardini and
Ramacci,5 and Bhattacharyya and Gulati,7 is given in Table 21,
In these experimental laboratory studies, gas chromatography
and various other quantitative analytical techniques were
employed to identify the reaction products.  In addition to
the compounds shown in Table 21, emissions from o-xylene
based phthalic anhydride plants contain sulfur oxides,
nitrogen oxides, particulates, and diphenyl oxide (from
Dowtherm® heat-transfer fluid).  Of the above mentioned
compounds, the following are either known or suspected of
being emitted on the basis of previously published data:llf»26
o-xylene, phthalic anhydride, maleic anhydride, benzoic
acid, o-tolualdehyde, formaldehyde, acetaldehyde, phthalalde-
hyde, acrolein, carbon monoxide, sulfur oxides, nitrogen
oxides, particulates, and diphenyl oxide.  Hence, these
compounds were selected for study in this program.  The
rationale for not considering the remaining materials listed
in Table 21 is presented in Appendix B.
                               66

-------
      Table 21.  POSSIBLE REACTION PRODUCTS FROM THE

                    OXIDATION OF XYLENE
         Compound
Formula
o-Methylbenzyl alcohol
o-Tolualdehyde
o-Toluic acid
Toluene
o-Hydroxymethylbenzoic acid
Citraconic anhydride
Phthalaldehyde
      COOH



      CH2OH
   O
   II

 •c"c\
 II   0
                                           O
      CHO



      CHO
                               67

-------
  Table 21 (continued).  POSSIBLE REACTION PRODUCTS FROM
                  THE OXIDATION OF XYLENE
         Compound
   Formula
Maleic anhydride
                                      O
                                    II   0
Phthalide
                                          CH2
Phthalaldehydic acid
                                          COOH
                                          CHO
Benzoic acid
     COOH
Phthalic anhydride
Phthalic acid
     COOH
                                           OOH
Carbon oxides and water
CO, CO2, H2O
o'-Carboxylphenyl-o-
  methylphenyl acetate
     CHo	COO
                                          CH3   HOOC
                              68

-------
  Table 21  (continued).  POSSIBLE REACTION PRODUCTS FROM
                  THE  OXIDATION OF XYLENE
         Compound
 Formula
o'-Methylphenyl-o-
~ methylphenyl acetate
bis(o-Methylphenyl) methyl
  ether
           )-CH2


            H3C
 ,2-bis(o-Methylphenyl)  ethane
    CH2	CH2


    'CH3   H3C
Formaldehyde
HCHO
Acetaldehyde
CH3CHO
Acrolein
                                        H
p_-Tolualdehyde
p_-Toluic acid
CHO
CH3


COOH
                                        CH-
                               69

-------
  Table 21 (continued).  POSSIBLE REACTION PRODUCTS FROM
                  THE OXIDATION OF XYLENE
         Compound
Formula
Terephthalic acid
   COOH
                                           COOH
p_-Benzoguinone
m-Tolualdehyde
   CHO
Isophthalic acid
   COOH
                                              'COOH
2.   Naphthalene Based Process

A reaction scheme for the oxidation of naphthalene, similar
to that for the oxidation of o-xylene, has not been published.
In addition to the reaction sequence given in Section  III.A.2.b.
involving 1,4-naphthoquinone, Shelstad, et al.,28 have reported
                              70

-------
a path involving 1,2-napththoquinone as an intermediate.
Thus, 1,2-dihydroxynaphthalene and 1,2-naphthoquinone are
expected reaction products.  In addition, the compounds
listed in Table 21 may also be formed by the oxidation of
naphthalene.  Furthermore, transfer operations involving
the fluid-bed catalyst result in emissions of vanadium
oxide catalyst dust.  Hence, the species selected for
detailed study in this program include those selected for
the o-xylene based process, with the addition of naphthalene,
the naphthoquinones, the dihydroxynaphthalenes, and vanadium
oxide catalyst dust.


B.   LOCATION AND DESCRIPTION

The sources of atmospheric emissions within phthalic anhydride
plants are:  the main process scrubber vent and/or the
incinerator stack, storage tank vents  (feedstock, crude
product, and refined product), the flaker and bagger vent,
the liquid product loading facility vent, and fugitive
emissions.  The catalyst storage hopper vents are an addi-
tional source of emissions in naphthalene based plants.
Those plants which employ a secondary scrubber or incinerator
to treat the waste streams from the product purification
section of the plant also have emissions from these units.
Each of the above sources is discussed below.


1.   Scrubber Vent

The scrubber vent gas (stream 9 in Figures 2 and 5) is a
continuous source of emissions consisting of scrubbed
process off-gas from the switch condensers.  Typical
concentration ranges for the major contaminants in the
condenser off-gas, as given by Fawcett,14 are shown in

                               71

-------
Table 22.  The corresponding flow rates for a 5.9 x 101*
metric tons/yr plant are in agreement with the material
balances given in Tables 18 and 19  (which are based on the
Houdry data26) with one exception:  that is, the flow rate
of maleic anhydride for a naphthalene based plant, which is
51 kg/hr in Table 19.


The organic compounds are removed from the condenser off-gas
stream with a maximum efficiency of 98% to 99% in the
scrubber,38 while the carbon and sulfur oxides are vented to
the atmosphere.  Typical compositions for the scrubber vent
gas stream are presented in Tables 23 and 24 for o-xylene
and naphthalene based plants, respectively.  These values
are based on the material balance data of Tables 18 and 19
and, in the case of naphthalene based plants, the assumption
of a scrubber efficiency of 98% for organics.


2.   Incinerator Stack

The incinerator flue gas (stream 26 in Figure 2 and stream 19
in Figure 5) is another continuous source of emissions.  The
incinerator is used to burn the liquid purge from the scrubber
or, in the case of plants employing direct incineration, the
switch-condenser off-gas.  The waste streams from the product
purification section of the plant and the phthalic anhydride
storage tank vents may also be fed to the incinerator.
Typical compositions of the stack gas are presented in Tables
25 and 26 for o-xylene and naphthalene based plants employing
a scrubber-incinerator combination.  These values are based
on the material balances given in Tables 18 and 19 and, in
the case of naphthalene based plants, on the assumption of
organic removal efficiencies of 98% and 99% in the scrubber
and incinerator, respectively.  The results for plants

                              72

-------
             Table 22.   CONCENTRATIONS OF CONTAMINANTS IN SWITCH-CONDENSER OFF-GAS14
Material
Phthalic anhydride
Maleic anhydride
Benzoic acid
Aldehydes (as CH20)
Carbon monoxide
Carbon dioxide
Sulfur dioxide
Naphthoquinone
Concentration ,
ppm by vol.
40-200
100-600
5-40
10-100
1,000-10,000
6,000-50,000
50-200
10-30
Flow rate,3
kg/hr
39-193
64-383
4-32
2-20
180-1,800
1,740-14,500
21-84
5.1-15
Uncontrolled
emission factor,
g/kg
5.4-26.5
8.8-52.6
0.55-4.4
0.28-2.8
24.8-248
239-2,000
2.9-11.5
0.7-2.1
U)
      For a production rate of 5.9 x 104 metric tons/yr.
      Only present when naphthalene is the feedstock.

-------
    Table 23.
TYPICAL SCRUBBER VENT GAS COMPOSITION
FOR O-XYLENE BASED PROCESS
               Production rate = 5.9 x 104 metric tons/yr
               Temperature     = 38°C
               Gage pressure   = 0 kPa
Component
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Tolualdehyde
Phthalaldehyde
Formaldehyde
Acetaldehyde
Acrolein
Carbon monoxide
Carbon dioxide
Sulfur oxides
Nitrogen
Oxygen
Water
Particulate
Others
Total
Concentration ,
wt %
0.0017
0.0044
0.0036
a
_a
a
a
_a
0.59
2.03
0.018
75.62
17.61
4.12
0.0025
a
100.0
Average
flow
rate,
kg/hr
3.2
8.2
6.8
a
a
a
a
a
1,095.9
3,784.6
34.1
140,897.3
32,819.1
7,680.9
4.6
a
186,334.7
Emission
factor,
g/kg
0.44
1.13
0.94
a
_a
a
a
_a
150.7
520.4
4.69
19,373.
4,513.
1,056.
0.63
_a
25,631
Not reported in data source.
                             74

-------
    Table 24.  TYPICAL SCRUBBER VENT GAS COMPOSITION
              FOR NAPHTHALENE BASED PROCESS

               Production rate = 5.9 x 10H metric tons/yr
               Temperature     = 38°C
               Gage pressure   = 0 kPa
Component
Phthalic anhydride
Maleic anhydride
Benzoic acid
Naphthoquinone
o-Tolualdehyde
Phthalaldehyde
Formaldehyde
Ace t a Idehyde
Acrolein
Carbon monoxide
Carbon dioxide
Sulfur oxides
Nitrogen
Oxygen
Water
Particulate
Others
Concentration ,
wt %
_a
_a
a
_a
a
a
_a
a
a
a
_a
a
a
_a
a
_a
a
Average
flow
rate,
kg/hr
3.0
1.0
0.2
0.15
b
b
b
_b
_b
366.
7,030.
0.0
68,177.
12,208.
_b
4.6
_b
Emission
factor,
g/kg
0.41
0.14
0.03
0.02
b
b
b
_b
_b
50.4
967-
0.0
9,378.
1,679.
_b
0.63
_b
Not calculated because data lacking for one or more
major components.
Not reported in data source.
                              75

-------
 Table 25.
TYPICAL INCINERATOR FLUE GAS COMPOSITION FOR
   O-XYLENE BASED PROCESS WITH
SCRUBBER-INCINERATOR COMBINATION
               Production  rate  =  5.9 x 104 metric tons/yr
               Temperature     =  927°C
               Gage pressure    =  0  kPa
Component
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Tolualdehyde
Phthalaldehyde
Formaldehyde
Acetaldehyde
Acrolein
Carbon monoxide
Carbon dioxide
Sulfur oxides
Nitrogen oxides
Nitrogen
Oxygen
Water
Methane
Ethane
Particulate
Others
Total
Concentration ,
wt %
_a
a
0.020
_a
_a
_a
_a
_a
0.10
15.47
_a
0.0079
54.17
2.22
27.99
_a
_a
0.018
_a
100.0
Average
flow
rate,
kg/hr
_a
_a
3.6
_a
_a
_a
_a
_a
18.2
2,759.
_a
1.4
9,661.
395.4
4,992.
a
_a
3.2
_a
17,834
Emission
factor,
g/kg
_a
_a
0.50
_a
_a
_a
_a
_a
2.50
379.
_a
0.19
1,328.
54.4
686.
_a
_a
0.44
_a
2,451
Not reported in data source,
                              76

-------
 Table 26.
TYPICAL INCINERATOR FLUE GAS COMPOSITION FOR
 NAPHTHALENE BASED PROCESS WITH
SCRUBBER-INCINERATOR COMBINATION
               Production rate = 5.9 x  ID4 metric tons/yr
               Temperature     = 927°C
               Gage pressure   = 0 kPa
Component
Phthalic anhydride
Maleic anhydride
Benzoic acid
Naphthoquinone
o-Tolualdehyde
Phthalaldehyde
Formaldehyde
Acetaldehyde
Acrolein
Carbon monoxide
Carbon dioxide
Sulfur oxides
Nitrogen oxides
Nitrogen
Oxygen
Water
Methane
Ethane
Particulate
Others
Concentration ,
wt %
_a
_a
a
a
a
_a
a
a
_a
_a
_a
_a
a
_a
a
_a
_a
_a
_a
a
Average
flow
rate,
kg/hr
1.75
0.87
0.15
0.07
_b
_b
_b
_b
_b
18.2
_b
_b
1.4
_b
_b
_b
_b
_b
3.2
_b
Emission
factor,
g/kg
0.24
0.12
0.02
0.01
_b
_b
_b
_b
_b
2.50
_b
_b
0.19
_b
_b
_b
_b
_b
0.44
_b
Not calculated because data  lacking  for  one  or more
major components.
Not reported in data source.
                              77

-------
employing direct incineration are presented in Tables 2713
and 28.26

The composition of the incinerator flue gas depends upon
incineration temperature, combustion efficiency, type of
incinerator (thermal or catalytic), and feed material
composition.  Incinerator temperature determines the nitrogen
oxides content of the stream due to oxygen fixation of
nitrogen, while combustion efficiency governs the quantities
of feed materials burned.  Feed material composition deter-
mines the overall composition of the stack gas.

Removal efficiencies of up to 99% of organic materials are
reported for a thermal incinerator in series with a water
scrubber.38  In the case of direct thermal incineration of
the switch-condenser off-gas, efficiencies of 96.5% for
organic material and 99.9% for carbon monoxide have been
attained.13  One difficulty with the latter method of control
is that the entire contaminant load exhausts to the atmos-
phere during periods of incinerator flame-out, which occur
an average of six times per year.13  Nevertheless, the
available data indicate that direct thermal incineration
is the best overall method of control at the present time
(see Section V.A).

3.   Storage Tank Vents

The emissions from feedstock, crude product, and refined pro-
duct storage tank vents (stream 34, 35, and 35 in Figure 2;
streams 27, 28, and 29 in Figure 5) for a 5.9 x 10** metric
tons/yr production rate have been estimated and are presented
in Table 29.  These estimates were obtained using the data
in Tables 7 and 14 together with the empirical correlations
                               78

-------
 Table 27.
TYPICAL INCINERATOR FLUE GAS COMPOSITION FOR
  o-XYLENE BASED PROCESS USING
   DIRECT THERMAL INCINERATIONl3
               Production rate =  5.9  x  104 metric  tons/yr
               Temperature     =  250 °C
               Gage pressure   =  0 kPa
Component
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Tolualdehyde
Phthalaldehyde
Forma Idehyde
Acetaldehyde
Acrolein
Carbon monoxide
Carbon dioxide
Sulfur oxides
Nitrogen oxides
Nitrogen
Oxygen
Water
Methane
Ethane
Particulate
Others
Concentration ,
wt %
_a
_a
_a
a
_a
a
a
_a
a
a
a
_a
_a
a
a
_a
_a
a
a
Average
flow
rate,
kg/hr
4.0
13.2
0.8
_b
_b
_b
_b
_b
0.9
_b
36.4
9.1
_b
_b
_b
_b
_b
1.8
_b
Emission
factor,
g/kg
0.55
1.82
0.11
_b
_b
_b
_b
_b
0.125
_b
5.0
1.25
_b
_b
_b
_b
_b
0.25
_b
Not calculated because data  lacking  for  one  or more
major components.
Not reported in data source.
                              79

-------
  Table  28.
TYPICAL INCINERATOR FLUE GAS COMPOSITION FOR
 NAPHTHALENE BASED PROCESS USING
   DIRECT THERMAL INCINERATION26
                Production rate = 5.9 x 104 metric tons/yr
                Temperature     = 250°C
                Gage pressure   = 0 kPa
Component
Phthalic anhydride
Maleic anhydride
Benzoic acid
Naphthoquinone
o-Tolualdehyde
Phthalaldehyde
Formaldehyde
Acetaldehyde
Acrolein
Carbon monoxide
Carbon dioxide
Sulfur oxides
Nitrogen oxides
Nitrogen
Oxygen
Water
Methane
Ethane
Particulate
Others
Concentration /
wt %
_a
_a
_a
_a
_a
_a
_a
_a
_a
a
_a
_a
a
_a
_a
a
_a
a
_a
_a
Average
flow
rate,
kg/hr
5.2
1.8
0.4
0.15
_b
_b
_b
_b
_b
0.4
_b
_b
9.1
_b
_b
_b
_b
_b
1.8
_b
Emission
factor/
g/kg
0.71
0.24
0.05
0.02
_b
_b
b
b
_b
0.05
_b
_b
1.25
_b
_b
_b
_b
_b
0.25
_b
 Not  calculated  because  data lacking for one or more major
 components.
3
 Not  reported  in data  source.
                              80

-------
                 Table 29.   STORAGE TANK WORKING AND BREATHING LOSSES FOR A 5.9 x 104 METRIC TONS/YR
                                               PHTHALIC ANHYDRIDE PLANT
Process
raw material
o-Xylene


Naphthalene


Species
o-Xylene
Phthalic anhydride
(crude)
Phthalic anhydride
(refined)
Naphthalene
Phthalic anhydride
(crude)
Phthalic anhydride
(refined)
Uncontrolled
emission
rate,
kg/hr
1.38
0.80
2.04
4.36
1.16
1.40
Uncontrolled
emission
factor,
gAg
0.19
0.11
0.28
0.60
0.16
0.19
Controlled
emission
rate,
kg/hr
0.014
0.008
0.020
0.044
0.012
0.014
Controlled
emission
factor ,
g/kg
0.002
0.001
0.003
0.006
0.002
0.002
Control
efficiency,
%
99
99
99
99
99
99
00

-------
for petrochemical losses from storage tanks which have been
formulated by the American Petroleum Institute.39"43  The
correlations give uncontrolled emission rates; the controlled
emission rates listed in Table 29 are based on a control
efficiency of 99%.44  The estimated accuracy of the empirical
correlations is ±10%.39~43  Calculations are presented in
the Appendix.

Emission control devices presently in use comprise conser-
vation vents on o-xylene and naphthalene tanks, and condensers
or sublimation boxes on crude and refined phthalic anhydride
tanks.  Phthalic anhydride tanks may, alternatively, be vented
to the incinerator for control.  However, it is common practice
to vent some or all of the storage tanks directly to the
atmosphere.

4.   Flaker and Bagger Vent

The vent from the flaking and bagging operations (stream 33
in Figure 2; stream 26 in Figure 5) is a continuous source
"Evaporation Loss from Fixed Roof Tanks.  American Petro-
  leum Institute.  New York.  API Bulletin No. 2518.  1962.
  38 p.
40Use of Variable Vapor Space Systems to Reduce Evaporation
  Loss.  American Petroleum Institute.  New York.  API
  Bulletin No. 2520.  1964.  14 p.
H1Petrochemical Evaporation Loss from Storage  Tanks.
  American Petroleum Insittute.  New York.  API Bulletin
  No. 2523.  1969.  14 p.
42Evaporation Loss from Floating Roof Tanks.  American
  Petroleum Institute.  New York.  API Bulletin No. 2517.
  1962.  13 p.
4 Evaporation Loss in the Petroleum Industry - Causes and
  Control.  American Petroleum Institute.  New York.  API
  Bulletin No. 2513.  1959.  57 p.
44Personal communication, R. G. Lunche and A. B. Netzley,
  Los Angeles County Air Pollution Control District.
  March 28, 1975.
                               82

-------
of particulate emissions.  This stream, which contains
phthalic anhydride as the only contaminant, is ducted to a
cyclone separator or baghouse for recovery of the solid
product.  For a 5.9 x 104 metric tons/yr plant, the uncon-
trolled emission rate for this stream is estimated to be
0.7 kg/hr of phthalic anhydride, with a corresponding emis-
sion factor of 0.1 g/kg.26  The controlled emission rate is
0.007 kg/hr, with a corresponding emission factor of 0.001
g/kg, assuming that a control efficiency of 99% is achieved
with a bag filter.44

5.   Transport Loading Facility Vent

The liquid product transport loading facility vent  (stream
37 in Figure 2; stream 31 in Figure 5) is a continuous source
of atmospheric emissions.  The only contaminant in this
stream is phthalic anhydride.  Using the emission factor
data listed in Reference 45, the uncontrolled emission rate
for this stream is estimated to be 3.3 kg/hr of phthalic
anhydride, based on an emission factor of 0.45 g/kg loaded,
a production rate of 5.9 x 10H metric tons/yr, and 100% of
product shipped in liquid form.  A control efficiency of 99%
was assumed, based on the data in Reference 45.  This yields
a controlled emission rate of 0.03 kg/hr and a controlled
emission factor of 0.005 g/kg phthalic anhydride produced.
The accuracy of these values is considered to be equivalent
to that for storage tanks, i.e., ±10% for uncontrolled
emissions.46
45Air Pollution Engineering Manual, 2nd Edition.  Danielson,
  J. A.  (ed.).  Environmental Protection Agency.  Research
  Triangle Park.  Publication No. AP-40.  May 1973.  987 p.
^Personal communication.  W. Fitzgibbons, Standard Oil of
  Ohio  (SOHIO).  June 23, 1975.

                               83

-------
6.   Catalyst Storage Hopper Vents

The vents on the storage facilities for active and spent
catalyst are intermittent sources of emissions from
naphthalene based plants.  The emissions consist of the fines
fraction (primarily particles <40 ym in diameter) of the
vanadium oxide catalyst, which is described in Table 15.
Both vanadium oxide and silica dusts are respiratory irri-
tants, and vanadium oxide can be toxic if taken internally.34
Emissions occur during catalyst transfer operations in which
the catalyst is conveyed pneumatically from the reactor
system to the storage hoppers.  Cyclone separators are
employed on the hopper vents to recover the entrained catalyst
particles from the air stream.

The uncontrolled emission rate from the spent catalyst
storage vent in a 4.08 x lO4 metric tons/yr plant is 295 kg/
hr.37  The operation is intermittent with a total duration of
approximately 50 hr/yr.  The uncontrolled emission factor is
therefore 0.36 g/kg.  The controlled emission rate is 4.6
kg/hr, and the corresponding emission factor is approximately
0.005 g/kg.  Thus, the control efficiency is 98.5%.

The uncontrolled emission rate from the active catalyst
storage vent in a 4.08 x 104 metric tons/yr plant is 22.7
kg/hr.37  The emissions occur an average of four to eight
times per year and have a duration of approximately 12 hours
per occurrence.  Thus, the uncontrolled emission factor is
approximately 0.05 g/kg.  The controlled emission rate is
2.3 kg/hr,  and the corresponding emission factor is 0.005
g/kg.  The control efficiency is therefore 90%.

The uncontrolled and controlled emission factors for the
combined catalyst storage vents are 0.41 g/kg and 0.01 g/kg,
respectively, resulting in a combined control efficiency of
97.5%.
                              84

-------
7.   Fugitive Emissions

Fugitive emissions from phthalic anhydride manufacture
originate from a number of sources:  pump seals, flanges,
valves, and compressor seals.  In addition to these losses
that occur under normal operating conditions, operating
upsets can result in emissions of fluid from the heat-
transfer circuits associated with the switch condensers and
the product purification section of the plant.  A typical
heat-transfer fluid used in phthalic anhydride plants is
Dowtherm A, a eutectic mixture of diphenyl oxide  (73.5%) and
diphenyl (26.5%).  This material has a very noticeable odor
which has been described as that of rose-geranium.14  Diphenyl
oxide has an odor threshold of between 0.01 ppm and 0.001 ppm
by volume.l4

The most frequently used organic heat-transfer fluids are
listed in Table 30 together with their compositions and
usable temperature ranges.16'17  Also listed are the thres-
hold limit values (TLV's).47  Of the fluids listed, Mobil-
therm Light® and Therminol 66® are known to be used in
phthalic anhydride plants.13  Emissions of these fluids
are in the form of aerosol mists, which reportedly settle
in the immediate vicinity of the source.13  It is estimated
that 95% of these emissions originate from the switch-
condenser heat-transfer circuit.13
47Threshold Limit Values for Chemical Substances and
  Physical Agents in the Workroom Environment with Intended
  Changes for 1975.  American Conference of Governmental
  Industrial Hygienists.  Cincinnati.  1975.  97 p.

                               85

-------
           Table 30.  FREQUENTLY USED HEAT-TRANSFER FLUIDS
      Name
Dowtherm A

Dowtherm E
Dowtherm G
Dowtherm H
Dowtherm J
Humbletherm 500
Mobiltherm Light
Mobiltherm 600
Mobiltherm 603
Therminol 44
Therminol 55
Therminol 60
Therminol 66
Therminol 77
Therminol 88
Therminol FR-1
Ucon 50-HB-280X
       Composition
Diphenyl-diphenyl oxide
  eutectic
o-Dichlorobenzene
Di- and tri-aryl ethers
Aromatic oil
Alkylated aromatic
Aliphatic oil
Aromatic oil
Alkylated aromatic
Paraffinic oil
Modified ester
Alkylated aromatic
Aromatic hydrocarbon
Modified terphenyl
Polyphenyl ether
Mixed terphenyl
Polychlorinated biphenyl
Ether of polyalkylene
  oxide
                                              Usable
                                           temperature
                                              range,
 16 to  400

-18 to  260
-10 to  345
-10 to  290
-75 to  300
-20 to  315
-30 to  205
-20 to  315
-18 to  315
-50 to  220
-18 to  315
-50 to  315
 -7 to  345
 16 to  370
145 to  425
 -4 to  315
-18 to  260
             TLV,
             g/m3
0.001

0.300
0.009
0.007
0.009
0.005
 Composition and usable  temperature data  for most of the heat-transfer
 fluids  are reprinted with permission from Chemical Engineering, May 28,
 1973, Copyright (c),  McGraw-Hill, Inc.,  New York, N.Y.  10020
 Other trade names include Thermex, Therm-S, Diphyl, and Therminol VP-1.
In the  Badger-Sherwin-Williams process,  product  purification
is a batch operation.   Hence,  fugitive  emissions  can  arise
from the batch filling  and dumping of process  vessels.   In
addition, emissions  of  vanadium oxide catalyst dust may occur
during  catalyst loading and  unloading operations.  Estimates
of fugitive  emissions are summarized  in Table  31.13'26   It
has been assumed that the heat-transfer fluid  is  Dowtherm A,
which is listed as diphenyl  oxide, its  major component.
                                  86

-------
Table 31.  FUGITIVE EMISSIONS FROM PHTHALIC ANHYDRIDE PLANTS
             Compound
     Phthalic anhydride
     Maleic anhydride
     Benzoic acid
     Naphthoquinone
     Diphenyl oxide
     Vanadium oxide catalyst
             r
     o-Xylene
     Naphthalene
     Formaldehyde
     Acetaldehyde
     Phthalaldehyde
     o-Tolualdehyde
     Acrolein
     Dihydroxynaphthalene
     Combined hydrocarbons
 Average
emission
  rate,
  kg/hr*
   0.12
   0.73
Emission
 factor,
  g/kg
   0.016
   0.10
      For a 5.9 x 104 metric tons/yr plant.
      Naphthalene based plants only.
     Co-Xylene based plants only.
      Does not include diphenyl oxide.
     Note:  Blanks indicate data not available,
8.   Dual Thermal Incineration

Available data13'26 indicate that the best presently feasible
method of control utilizes direct thermal incineration of the
switch-condenser off-gas together with a secondary thermal
incinerator to treat the waste streams from the product puri-
fication section of the plant  (see Section V.A).  Typical compo-
sitions and emission rates for the latter unit are presented
in Table 32 for an o-xylene based plant.13  This unit treats
                             87

-------
          Table  32.  TYPICAL FLUE  GAS COMPOSITION
                   FOR DUAL INCINERATOR

                 Production rate  =  5.9  x  104 metric tons/yr
                 Temperature      =  760°C
                 Gage pressure    =  0  kPa
Component
Phthalic anhydride
Maleic anhydride
Benzoic acid
o-Tolualdehyde
Phthalaldehyde
Formaldehyde
Acetaldehyde
Acrolein
Carbon monoxide
Carbon dioxide
Sulfur oxides
Nitrogen oxides
Nitrogen
Oxygen
Water
Methane
Ethane
Particulate
Others
Total
Concentration ,
wt %
0.010
0.0037
0.0012





0.11
5.55

0.011
55.47
8.32
30.51


0.011

100.0
Average
flow
rate,
kg/hr
0.8
0.3
0.1





9.1
455.

0.9
4,545.
682.
2,500.


0.9

8,194.
Emission
factor,
g/kg
0.11
0.04
0.0125





1.25
62.6

0.125
625.
93.8
344.


0.125

1,127.
Note:  Blanks indicate data not reported.
                              88

-------
all of the waste streams from the product purification section
of the plant, and reportedly achieves a control efficiency
of 99% for combined organics.  The applicability of such a
unit to naphthalene based plants is uncertain since product
purification is a batch operation in these plants.  However,
no fundamental barrier to its use in naphthalene based plants
is apparent.

C.   ENVIRONMENTAL EFFECTS

1.   Definition of a Representative Source

For the purpose of assessing the source severity, a repre-
sentative o-xylene based phthalic anhydride plant is defined
to be one using the BASF process and having a production
capacity of 5.9 x 10k metric tons/yr.  The BASF process is
specified because it is the newest and most widely used
technology, accounting for 79% of the o-xylene based pro-
duction in 1974.3'4  The standard size BASF plant has two
production trains of two reactors each, and has a capacity
of 5.9 x 104 metric tons/yr.  The seven o-xylene based
plants operating in the United States range in capacity
from 1.59 x lO4 to 7.94 x 104 metric tons/yr with a mean
value of 4.28 x 104 metric tons/yr.

A representative naphthalane based phthalic anhydride plant
is defined to be one using the Badger-Sherwin-Williams pro-
cess and having a production capacity of 5.9 x 104 metric
tons/yr.  The Badger-Sherwin-Williams process is the dominant
fluid-bed technology and accounted for 73% of the naphthalene
based production in 1974.3/4  There is no standard size for
the Badger-Sherwin-Williams fluid-bed reactors.  Hence, the
representative source capacity was chosen so as to place
the source severity calculations on the same basis as that
used for o-xylene based production.  The three naphthalene
                             89

-------
based plants range in size from 4.08 x 101* to 6.8 x 101* metric
tons/yr with a mean production capacity of 4.99 x 104 metric
tons/yr.

The source severity is assessed on the basis that the best
available control technology is applied to the representative
source.  This implies that the switch-condenser off-gas is
controlled by direct thermal incineration and that a secondary
thermal incinerator is employed to treat all of the waste
streams from the product purification section of the plant.
(See Section V.A.)  The dual thermal incineration system is
currently used in two of the four plants employing the BASF
process, and will probably be installed in other plants in
the future.  It is also assumed that all storage tank vents,
as well as the transport loading facility vent, are fully con-
trolled.  In addition, the flaker and bagger vent is assumed
to be equpped with a suitable bag filter.

2.   Emission Factors

Emission factors for an o-xylene based plant equipped with a
dual incineration control systemare given in Table 33.  The
data were obtained from References 13 and 26, except as noted.
The control efficiencies of the two incinerators were measured
only for total organic material.  In order to obtain controlled
emission factors of individual organic species it was assumed
that they were controlled with this same efficiency.  Error
bounds on the emission factors were estimated wherever possi-
ble.  The bounds on uncontrolled emission factors for the two
incinerators were taken directly from Reference 13.  Those
for storage tanks and transport loading were obtained from
References 39-43, and Reference 46, respectively.  The bounds
on controlled emission factors were taken to be somewhat
higher in order to reflect the uncertainty in the control
efficiencies of individual species.  The threshold limit
values (TLV's of the individual species,  as given in Refer-
ence 47, are also listed in Table 33.
                              90

-------
        Table  33.  EMISSION FACTORS FOK o-XYLENE BASED PHTHALIC ANHYDRIDE PLANTS
Species
Partioulate
Main process incinerator
Secondary incinerator
Flaker and bagger
Sultur oxides
Main process incinerator
Nitrogen oxides
Main process incinerator
Secondary incinerator
Carbon monoxide
Main process incinerator
Secondary incinerator
Maleic anhydride
Main process incinerator
Secondary incinerator
Phthalic anhydride
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Benzoic acid
Main process incinerator
Secondary incinerator
Diphenyl oxide
Fugitive emissions
o-Xylene
Storage tanks
Formaldehyde
Main process incinerator
Secondary incinerator
Acetaldehyde
Main process incinerator
Secondary incinerator
Phthalaldehyde
Main process incinerator
Secondary incinerator
o-Tolualdehyde
Main process incinerator
Secondary incinerator
Acrolein
Main process incinerator
Secondary incinerator
Total hydrocarbons
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
TLV,
g/n>3
0.010



0.813

0.009


0.055


0.001


0.006





0.050


0.001

0.435

0.003


0.180


J


J


0.00025


0.067






T
Emission factor, g/kg
Uncontrolled

_ a
^
0.10

5.0 ± 25%

_
— C

125.0 ± 20%
_ c

52.0 ± 20%
3.75 ± 83%

15.6 ± 20%
10.6 ± 53%
0.28 ± 10%e>1
0.10
0.45 ± 10%g
3

3.12 ± 20%
1.25 ± 50%

0.016
p
0.19 ± 10%
h i
n ^ U J 1
" a





a
"* a

a
a

a

-

72.8 ± 30% ,
18.8 ± 60% '
0.47 ± 10%
0.1 g
C.459
0.116
Controlled

0.25 ± 50%
0.125 ± 50%
0.001

5.0 + 25%

1.25 ± 50%
0.125 + 25%

0.125 ± 50%
1.25 ± 50%

1.82 ± 25%
0.038 ± 90%

0.545 ± 25%
0.106 ± 60%
0.003 ± 20%
0.001
0.005 ±h20%9
n

0.109 ± 25%
0.0125 t 55%
.
-

0.002 ± 20%
i
Confrol
efficiency. %


- .
99D

0




99.9
-
A
96.5°
99d
,
96. ,5
99b
99b
99b
9q9

o
96ti5
99


b
99
A
0.0741 a 96.^"
-
a
a


"a

a
a

a
"a


2.6 ± 30%
0.16 ± 60%
0.005 ± 20%
0.001 q
0.005y h















96.5
99 b
99 D
992
99

 Data not available.
 Estimate based on data from
 Reference 44.
CNo emissions generated in uncontrolled
 process.
 Assumed equal to va!ue measured for
 total organics in Reference 13.
Calculated using empirical correlations
 given in References 39-43.
 Value is for refined product storage
 cnly; crude product storage tanks are
 vented to the main process incinerator.
"value for loading racks given in
 Reference 45.
 Fugitive emissions are lot controlled.
1Total aldehydes reported as formaldehyde
 in Reference 44.
JTLV not defined.
klncludes all non-methane organic species.

 Includes 3.2 g/kg unspecified organic-
 residues.
                                               91

-------
Emission factors for a naphthalene based plant equipped with
a dual incineration control system are presented in Table 34.
The data were obtained from Reference 26, except as noted.
It was assumed that this control system could be applied to
a naphthalene based plant without having its control efficiency
altered.  It was also assumed that the particulate and nitro-
gen oxides emissions from the two incinerators would be the
same as the corresponding values for an o-xylene based plant.
Error bounds for emission factors from the two incinerators
were assigned on the basis of (1) the corresponding values
for an o-xylene based plant;  (2) the fact that the accuracv
of the data for naphthalene based production is believed to
be generally poorer than that for o-xylene based production;
and (3) the ranges for uncontrolled emission factors given
in Table 22.  Error bounds on emission factors for storage
tanks and transport loading are the same as those for o-xylene
based production.

3.   Source Severity

In order to obtain a quantitative measure of the hazard po-
tential of phthalic anhydride production, the source severity,
S, is defined as:
                          S =                           (1)

where x    is the maximum time-averaged ground level concen-
       max
tration of each pollutant emitted from a representative plant,
and F is defined as a primary ambient air quality standard
for criteria pollutants  (particulate, SO , NO , CO and hydro-
                                        X    X
carbons) , while for non-criteria pollutants,

                   F = TLV- 8/24 -0.01, g/m3

The factor 8/24 adjusts the TLV for continuous rather than
workday exposure, and the factor of 0.01 accounts for the
fact that the general population is a higher  risk  group than
healthy workers.
                              92

-------
      Table  34.
EMISSION  FACTORS FOR  NAPHTHALENE  BASED
 PHTHALIC ANHYDRIDE PLANTS26
Species
Particulate
Main process incinerator
Secondary incinerator
Flaker and bagger
Nitrogen oxides
Main process incinerator
Secondary incinerator
Carbon monoxide
Main process incinerator
Secondary incinerator
Phthalic anhydride
Main process incinerator
Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
Maleic acid
Main process incinerator
Secondary incinerator
Benzoic acid
Main process incinerator
Secondary incinerator
Naphthoquinone
Main process incinerator
Secondary incinerator
Naphthalene
Storage tanks
TLV,
g/m3
0.010



0.009


0.055


0.006






0.001


0.050 J


0.0004k


0.050

Emission factor, g/kg
Uncontrolled

a
-
0.10


-

50.5 ±C50%
-

20.4 ± 40%
9.8 ± 70% ,
0.35 ± 10%e'T
0.10
0.45 ±a!0%9
-

7.0 + 200% 1
4.9 ± 80%

1.56 ±a40%
-

0.69 ± 40%
6.6 ± 70%
p
0.60 ± 10%
Controlled

0.25 ± 70%
0.125 ± 70%
0.001

1.25 ± 70%
0.125 ± 40%

0.05 ± 70%
1.25 ± 70%

0.71 ± 50%
0.10 ± 80%
0.0004 ± 20%
0.001
0.005 ±h20%9
-

0.24 + 200% 1
0.05 ± 90%

0.05 ±,50%
d

0.02 ± 50%
0.07 ± 80%

0.006 ± 20%
Control
efficiency, %


- .
99




99.9
-
A
96 5 d
"h
"b
99b
99 9
•

96y5d
99°
A
96.5°


96^,5
99
h
99°
 Data  not available.
 Estimate based on data from Reference 44.
 No emissions generated in uncontrolled process.
 Assumed equal to value measured for  total organics  in Reference 13.
 Calculated using empirical correlations given in References 39-43.
 Includes both refined and crude product storage  tanks.
 Value for loading racks given in Reference 45.
 Fugitive emissions are not controlled.
 Uncontrolled emission factor of 7.0  g/kg, obtained  from Reference  26, is believed  to
 be low; compare Table 22.
 Estimate based on the value of 0.080 for benzene.
k
 TLV for naphthoquinone assumed equal to the value for quinone.
                                          93

-------
    Table  34  (continued).   EMISSION  FACTORS  FOR NAPHTHALENE
                  BASED PHTHALIC  ANHYDRIDE  PLANT26
Species
Vanadium oxide catalyst
Catalyst storage
Fugitive emissions
Diphenyl oxide
Fugitive emissions
Forma Idehyde
Main process incinerator
Secondary incinerator
Ace ta Idehyde
Main process incinerator
Secondary incinerator
Phthalaldehyde
Main process incinerator
Secondary incinerator
o-Tolualdehyde
Main process incinerator
Secondary incinerator
Acrolein
Main process incinerator
Secondary incinerator

Dihydroxynaphthalene
Main process incinerator
Secondary incinerator
Total hydrocarbons
Main process incinerator

Secondary incinerator
Storage tanks
Flaker and bagger
Transport loading
Fugitive emissions
TLV,
0.00056


0.001

0.003


0.180


-h


_n


0.00025


n



0.067





Emission factor, g/kg
Uncontrolled

0.41
a

0.016

2 -i b , in
a

a
a

-
a
a
"a
"
a
~a


a
"a

3.18 + 79%
- 35%
26.5 ± 70%P
0.95 ± 10%
0'.459
0.116
Controlled

0.01
_a

_

0.0741"
a

a
"a

-
a
a
~a

a
"a
~

a
"a

1.09 + 85%
- 40%
0.27 ± 80%
0.01 ± 20%
0.001
0.0059 h
_»
Control
efficiency, %

97.5



j
96. 5d

















96.5

99b
nn b
99 g
99y

 Data not available.
 Estimate based on data from Reference 44.
 Assumed equal to value measured for total organics in Reference 13.
 Fugitive emissions are not controlled.
 Calculated using the rule for mixtures given in Reference  47, the  catalyst composition
 given in Table 15, and the following individual TLV's:  V205, 0.00089 g/m3; Sib2,  0.0003
 g/m3 for total dust, respirable and nonrespirable; S03,  0.013 g/m3.  The TLV for V205
 was obtained by multiplying the listed value of 0.0005 g/m3 V205 as V by the factor
 (182 g V205)/(102 g V)  = 1.78  In lieu of other information, the TLV of K2S01( was
 assumed to be large compared with those of the other components.  Thus,
              TLV
                 catalyst     0.09     0.5     0.12
                           0.00089   0.0003   0.013
                                                  0.00056 g/m3
n
Total aldehydes reported as formaldehyde in Reference 44.
TLV not defined.
Includes all organic species.
Includes 5.2 g/kg unspecified organic residues.
Thus,  the source severity  represents the  ratio  of  the  maxi-
mum mean  ground  level  exposure  to the  hazard level of  ex-
posure for a given  pollutant.
                                       94

-------
The maximum ground level concentration, x   / is calculated
                                         in 3.x
according to Gaussian plume dispersion theory:
                         v
                         x
                          max     2 -
                                TTH eu
where  Q = mass emission rate, g/sec
       u = average wind speed, m/sec
       H = effective emission height, m
       e = 2.72

Equation (2) yields a value for a short-term averaging time
during which the Gaussian plume dispersion equation is valid.
The short-term averaging time was found to be 3 minutes in a
study of published data on lateral and vertical diffusion.^8
For a continuously emitting source, the maximum mean ground
level concentration for time intervals between 3 minutes and
24 hours can be estimated from the relation:49
                                      'to
                                          0. 17
                         xmax   xmax  t
where  t  = the averaging time
       tQ = the short-term averaging time  (3 min.)

For non-criteria pollutants, the averaging time, t, is 24
hours.  For criteria pollutants, the averaging times are
those used in the definition of the primary ambient air
quality standards.  The only exception is NO , for which the
                                            X
48Nonhebel, G.  Recommendations on Heights for New Industrial
  Chimneys.  Journal of the Institute of Fuel.  33:479-511,
  July 1960.
49 Turner, D. B.  Workbook of Atmospheric Dispersion Estimates,
  U.S. Department of Health, Education, and Welfare, Public
  Health Service.  Cincinnati.  Publication No. 999-AP-26.
  1969.  64 p.

                              95

-------
primary standard averaging time is one year.  Since Equation
(3) is not valid for averaging times of this magnitude, the
calculation of x    for NO  is based on Equation (5.13) of
                max       x
Reference 49, which estimates the annual average ground level
concentration .
Insertion of the national average wind speed of 4.5 m/sec into
the above equations leads to the severity factor equations
listed in Table 35.  The emission heights which were used in
the calculations are given in Table 36.  The heights for the
two incinerators were obtained from Reference 13, that for the
f laker and bagger from Reference 50, and the others were
estimated.  The emission rates, Q, were obtained from the
controlled emission factors in Tables 33 and 34.  The resulting
severity factors are tabulated in Tables 37 and 38 for o-xylene
and naphthalene based plants, respectively.  In these calcula-
tions, it was assumed that Gaussian plume dispersion theory is
equally valid for all emissions, irrespective of their chemical,
physical, or topological characteristics.

The largest value of the severity factor in Tables 37 and 38
is 18.5 for diphenyl oxide.  This value is based on the
assumption that Dowtherm A heat-transfer fluid is used in the
plant.  However, it is clear from the TLV's listed in
Table 30 that the severity factor for this emission is highly
dependent on the particular fluid used.  Thus, assuming the
same emission rate, the severity factor would be 37.0 for
Therminol FR-1® and 0.06 for Dowtherm E.

The above severity factors for diphenyl oxide emissions and
vanadium oxide catalyst emissions from catalyst storage
50Personal communication, L. B. Evans, U.S. Environmental
  Protection Agency  (Data originally supplied by Stepan
  Chemical Company).  June 12, 1975.
                              96

-------
            Table 35.  SOURCE SEVERITY EQUATIONS
            Pollutant
          Particulate
          SO
            x
          NO
            x
          Hydrocarbons
          CO
          All others
S (dimensionless)
70 QH~2
50 QH-2
315 OH'2-1
162.5 QH~2
0.78 QH~2
5.5 QH~2  (TLV)
                 Q = emission rate, g/sec
                 H = emission height, m
               TLV = threshold limit value, g/m3
    Table 36.  EMISSION HEIGHTS FOR REPRESENTATIVE SOURCE
             Source of emissions
           Main process incinerator
           Secondary incinerator
           Storage tanks
           Flaker and bagger
           Transport loading
           Fugitive emissions
           Catalyst storage
                                            Emission
                                             height,
               m
             61.0
             30.5
             15.2
              6.1
              6.1
              3.1
             12.2
facilities were calculated by treating these intermittent
sources as continuous sources with their respective annual
emission rates.  Severities calculated with emission rates
based on the average duration of these emissions would be
considerably higher.  For instance, S = 216 for vanadium
oxide catalyst emissions.  Although the duration of the
diphenyl oxide emissions is not known, the severity calcula-
ted on this basis would be correspondingly higher.
                              97

-------
          Table 37.  SOURCE SEVERITY FACTORS FOR A  REPRESENTATIVE O-XYLENE BASED PHTHALIC ANHYDRIDE PLANT
Compound
Particulate

Sulfur oxides

Nitrogen oxides
Carbon monoxide
b
Total hydrocarbons
Phthalic anhydride

Maleic anhydride

Benzoic acid

Diphenyl oxide

o-Xylene

Formaldehyde

Acetaldehyde

Phthaldehyde

o-Tolualdehyde

Acrolein
Source severity
Main process
incinerator
0.0095

0.14

0.14
0.000053

0.23
0.28

5.4

0.0065
a

a

d
0.074
c

c

c

c

Secondary
incinerator
0.019
a
\A

0.061
0.0021

0.056
0.20

0.45

0.0030
a

a

c

c

c

c

c

Storage
tanks
a
a
u
a

a

0.0071
0.024
a

a

a


0.00022
a

a

a

a

a

Flaker and
bagger
0.0038
a
vl
a

a

0.0088
0.050
a

a

a

a

a

a

a

a

a

Transport
loading
a
a
u
a

a

0.044
0.24
a

a

a

a

a

a

a

a

a

Fugitive
emissions
a
a
u
a

a

4.0
c
c

c


18.5
C

C

C

C

C

C

10
00
        No emissions.

        Includes  all  non-methane organic species.

        "Data not  available.
        Total aldehydes  reported as  formaldehyde.

-------
                                Table 38.  SOURCE SEVERITY FACTORS FOR A REPRESENTATIVE NAPHTHALENE BASED
                                                        PHTHALIC ANHYDRIDE PLANT
VD
Compound
Particulate

Nitrogen oxides

Carbon monoxide
b
Total hydrocarbons
Phthalic anhydride

Maleic anhydride

Benzoic acid

Naphthoquinone

Diphenyl oxide

Vanadium oxide
catalyst

Naphthalene

Formaldehyde

Acetaldehyde

Phthalaldehyde

o-Tolualdehyde

Acrolein

Dihydroxynaphthalene
Source Severity
Main
process
incinerator
0.0095

0.14

0.000021

0.10
0.36

0.72

0.0030

0.15
a-

a


a

d
0.074
c

c
\*
c

r
\*
c
\*
Secondary
incinerator
0.019

0.061

0.0021

0.095
0.20

0.60
a


2.1


a


a

c

c

r
Vrf
c

r
\*
r'
\*
Storage
tanks
a
a

a


0.014
0.032


a

a
u


a



0.0058
a

a



a

a

a

Flaker
and
bagger
0.0038


a


0.0088
0.050


a





a


a

a

a



a

a

a

Transport
loading
a


a


0.044
0.24
a

a

a

a
u
a


a

a

a

a

a

a

a

Catalyst
storage
a


a

a

a
a

a

a

a


1.4

a

a

a

a

a

a

a

Fugitive
emissions
a
a
w
a


4.0
c
c

c

c


18.5
c


c

c

c

c

c

c

c

                   No emissions.
                  tancludes all non-methane organic species.
"Data not available
                                                                     Total aldehydes reported as formaldehyde.

-------
4 .    Industry Contribution to Total Atmospheric Emissions

The mass emissions of criteria pollutants (particulate, SO ,
                                                          X
NO , CO, and hydrocarbons) resulting from phthalic anhydride
  X
production were calculated using the controlled emission
factors from Tables 33 and 34 together with the production
capacity data from Table  20.  The appropriate emission fac-
tor was multiplied by the production capacity nationwide and
for each state in which phthalic anhydride plants are  located
The total mass emissions  from all sources nationwide and for
each state were obtained  from Reference 51.  The percent
contributions to the total emissions resulting from phthalic
anhydride production were computed using these values.  The
results are presented in Table 39 for nationwide emissions,
and Table 40 for individual state emissions.

5.    Affected Population

A measure of the population which is exposed to a high
contaminant concentration due to a representative phthalic
anhydride plant can be obtained as follows .  The values of
x for which
                              = 0.1 or 1.0              (4)
                          r

are determined by iteration.  The value of x(x)f tne annual
mean ground level concentration, is computed from the
equation : 4 9
               X(x) =
                       a ux
                        z
A /_5\2
2 I a.
(5)
 511972 National Emissions Report.  U.S. Environmental
  Protection Agency.  Research Triangle Park.  Publication
  No. EPA-450/2-74-012.  June 1974.
                              100

-------
              Table 39.
NATIONWIDE EMISSIONS OF CRITERIA POLUTANTS FROM PHTHALIC
  ANHYDRIDE INDUSTRY (CONTROLLED EMISSIONS)
Material
emitted
Carbon monoxide



Sulfur oxides



Particulates



Nitrogen oxides



Hydrocarbons



Source
All sources
Phthalic anhydride production
o-Xylene based
Naphthalene based
All sources
Phthalic anhydride production
o-Xylene based
Naphthalene based
All sources
Phthalic anhydride production
o-Xylene based
Naphthalene based
All sources
Phthalic anhydride production
o-Xylene based
Naphthalene based
All sources
Phthalic anhydride production
o-Xylene based
Naphthalene based
Emissions,
metric tons/yr
97.6 x 106a
608
413
195
30.2 x 106a
1,500
1,500
0
18.0 x 106a
169
113
56
22.4 x lO63
619
413
206
25.3 x 106a
1,089
865
224
% Contribution
100.
0.00062
0.00042
0.00020
100.
0.0050
0.0050
0
100.
0.00094
0.00063
0.00031
100.
0.0028
0.0018
0.0009
100.
0.0043
0.0034
0.0009
o
    aData obtained from Reference 51.
    ^Includes all non-methane organic species.

-------
Table  40.
EMISSIONS OF CRITERIA POLLUTANTS FROM  PHTHALIC ANHYDRIDE  INDUSTRY
             BY  STATE (CONTROLLED EMISSIONS)
Material
emitted
Carbon monoxide





Sulfur oxides





Particulate





Nitrogen oxides





Hydrocarbons





State
California
Illinois
Louisiana
New Jersey
Pennsylvania
Texas
California
Illinois
Louisiana
New Jersey
Pennsylvania
Texas
California
Illinois
Louisiana
New Jersey
Pennsylvania
Texas
California
Illinois
Louisiana
New Jersey
Pennsylvania
Texas
California
Illinois
Louisiana
New Jersey
Pennsylvania
Texas
Total g
emissions,
106 metric
tons/yr
8.25
6.43
5.65
2.88
3.74
6.91
0.39
2.05
0.17
0.46
2.94
0.75
1.01
1.15
0.38
0.15
1.81
0.55
1.67
0.98
0.42
0.49
3.02
1.31
2.17
1.83
1.92
0.82
0.89
2.22
Emissions from
phthalic anhydride
plants,
metric tons/yr
53
141
56
134
142
81
192
510
204
295
0
295
14
38
15
38
41
22
53
141
56
137
149
81
111
295
118
231
162
170
Phthalic
anhydride
contribution, %
0.0006
0.0022
0.0010
0.0047
0.0038
0.0012
0.0492
0.0249
0.1200
0.0641
0.0
0.0393
0.0014
0.0033
0.0039
0.0250
0.0023
0.0040
0.0032
0.0144
0.0133
0.0280
0.0049
0.0062
0.0051
0.0161
0.0061
0.0282
0.0182
0.0077
       Data obtained from Reference 51.
       ^Includes all non-methane  organic species.

-------
where   Q = emission rate, g/sec
        H = effective emission height, m
        x = downwind distance from source, m
        u = average wind speed  (4.5 m/sec)
       a  = vertical dispersion coefficient,  m
        z

For atmospheric stability class C  (neutral conditions) ,
a  is given by:52
                     a  = 0.113 x°-911                  (6)
                      z
The affected area is then computed as

                    A = Tr(x22 - xL2) , km2               (7)

where xj and x2 are the two roots of Equation  (4).
The (capacity weighted) mean population density, D  , is
calculated for each plant type  (o-xylene and naphthalene)
as follows:
                     E CIDP.
               Dt> = —^=Tn	 ' persons/km2             (8)
where   C.  = production capacity of plant i
       Dp  = county population density for plant i

The product A-D  is designated the "affected population."
52Eimutis, E. C., and M. G. Konicek.  Derivations of
  Continuous Functions for the Lateral and Vertical Atmo-
  pheric Dispersion Coefficients.  Atmospheric  Environment,
  £:859-863, 1972.

                              103

-------
                        Table 41.  AFFECTED POPULATION
                              (number of persons)
Compound
Sulfur oxides
Main process incinerator
Nitrogen oxides
Main process incinerator
Total hydrocarbons
Main process incinerator
Fugitive emissions
Phthalic anhydride
Main process incinerator
Secondary incinerator
Transport loading
Maleic anhydride
Main process incinerator
Secondary incinerator
Diphenyl oxide
Fugitive emissions
Formaldehyde
Main process incinerator
Naphthoquinone
Main process incinerator
Secondary incinerator
Vanadium oxide catalyst
Catalyst storage
o-Xylene based plant
No. of persons
where
X/F > 1.0
0
0
0
40
0
0
0
44,000
0
300
0
a
a
_a
No. of persons
where
X/F > 0.1
6,700
5,600
9,600
500
19,000
3,100
140
525,000
8,800
3,400
15,000
_a
_a
_a
Naphthalene based plant
No. of persons
where
X/F > 1.0
_a
0
0
6
0
0
0
0
0
50
0
0
500
40
No. of persons
where
X/F > 0.1
_a
870
0
77
4,100
460
22
9,200
1,800
540
2,300
1,300
7,000
650
No emissions.

-------
The affected population was computed for each compound and
each source for which the severity factor, S, exceeds 0.1.
The results are presented in Table 41.  The mean population
density for O-xylene based plants is 1,667 persons per
square kilometer, and for naphthalene based plants it is
261 persons per square kilometer.  For x/F - 0.1, the total
number of persons affected by a representative o-xylene
based plant is 525,000 persons, while the total number
affected by a representative naphthalene based plant is
9,200 persons.  For 7/F - 1-0, the values are 44,000 persons
for an o-xylene based plant and 500 persons for a naphthalene
based plant.

6.   Growth Factor

In 1973, 4.658 x 105 metric tons of phthalic anhydride was
produced in the United States.53  As discussed in Section VI,
production is expected to total 4.99 x 105 metric tons in
1978.   Thus, assuming that the same level of control technology
exists in 1978 as existed in 1973, the emissions from the
phthalic anhydride industry will increase by 7% over that
period; i.e.:

              Emissions in 1978 _ 4.99 x 105
              Emissions in 1973 ~ 4.658 x 105  ~
 3Naphthalene Feedstock Outlook Mixed.   Chemical and
  Engineering News.   _52:14-17,  July 8,  1974.
                               105

-------
                         SECTION V

                    CONTROL TECHNOLOGY
A.   STATE OF THE ART

At the present time, all phthalic anhydride plants employ
pollution control devices to treat the switch-condenser
off-gas/ which constitutes the greatest potential source of
air pollution from phthalic anhydride manufacture.  The
control devices in use are water scrubbers and/or inciner-
ators, either thermal or catalytic.  The large air to feed
ratio, particularly in o-xylene based plants, makes the
scrubber-incinerator combination more economical than
incineration alone due to the large volume of gas that must
be handled.

Organic removal efficiencies of up to 98% to 99% in the
scrubber and 99.9% in the incinerator are achieved.38
However, carbon monoxide and sulfur oxides are not controlled
and are emitted to the atmosphere via the scrubber vent.
Further treatment of the scrubber vent gas (e.g., by a CO
boiler or gas adsorption) would be expensive due to the
low concentration (approximately 0.6% by weight) of pollu-
tants in this stream.  The principal advantage of direct
incineration of the switch-condenser off-gas is that it
allows the carbon monoxide to be controlled.  Control
efficiencies of 99.9% can be achieved for this contaminant
                               106

-------
with an organic conversion efficiency of 96.5%.13  The
sulfur oxides are again vented to the atmosphere, however.

The scrubber and/or incinerator can also be used to control
the waste streams from the product purification section of
the phthalic anhydride plant.  Alternatively, a separate
control device may be employed for this purpose.  The best
control system presently in use employs direct thermal
incineration of the switch-condenser off-gas together with
a second thermal incinerator that handles the waste streams
from the product purification section.  Water scrubbers are
also used in some plants to treat some of the latter streams.

The dual incineration system provides better overall emissions
control than the scrubber-incinerator system or other systems
being used for the following reasons:

      (1)  The removal efficiency for total organics is 96.5%
          for the dual incineration system being used at two
          phthalic anhydride plants.  The scrubber-incinerator
          system is capable of achieving 97% removal of total
          organics in existing commercial plants, but the dual
          incineration system has a better potential for
          increased organics removal through increased incin-
          eration temperatures in the existing plants.
          Theoretical calculations indicate that the organics
          removal efficiency will be 99% when operating at a
          temperature of approximately 860°C instead of
          760°C.5lf  The scrubber-incinerator system does not
          demonstrate this flexibility in existing plants.
 54Chi, C. T., and T. W. Hughes.  Technical and Economic
   Evaluation of Phthalic Anhydride Plant Air Pollution Con-
   trol.  Monsanto Research Corporation.  Dayton.  Prelimi-
   nary Draft Report.  U.S. Environmental Protection Agency,
   Contract 68-02-1320, Task 25.  August 1976.
                              107

-------
     However,  increasing  the  flame  temperature  in the
     dual incineration system by 100°C  will  increase
     the NO  emissions by about  15%.  The NO source
           x                                x
     severity at 760°C is 0.14;  at  860°C it  will be
     0.16 (0.14  x 1.15).

(2)   Direct thermal  incineration of the switch-condenser
     off-gas removes 99%  of  the  carbon  monoxide.  The
     scrubber incinerator system cannot control carbon
     monoxide emissions since the incineration  is per-
     formed on the scrubber purge liquid  (blowdown)
     and not on  the  scrubber  vent gas.  The  carbon
     monoxide source severities  for the dual incinera-
     tion system and the  scrubber-incineration  system
     on  an o-xylene  based plant  are 0.000053 and 0.21,
     respectively.   For a naphthalene based  plant, the
     severities  are  0.000021  for dual incineration and
     0.084 for the scrubber-incineration combination.

     The percent contribution to nationwide  carbon
     monoxide emissions is 0.002% for dual  incineration
     versus 0.047% for the scrubber-incinerator system.
     The contributions to total  carbon  monoxide emissions
     in  New Jersey (the largest  of  the  statewide contri-
     butions)  are 0.0047% for the dual  incineration
     system and  0.43% for the scrubber-incinerator
     combination.

(3)   The use of  a secondary  (dual)  incinerator  for
     control of  waste streams from  the  product  puri-
     fication section of  the  plant  is preferable to use
     of  a secondary  scrubber  to  control these  streams.
     The dual incinerator has a  high (99%)  control
     efficiency  for  organics  and does not generate an
     additional  liquid waste  stream requiring  disposal
     as  does scrubbing.
                       108

-------
     (4)   The use of a secondary incinerator for the product
          purification waste streams is also superior,  from
          an operating standpoint,  to using the main process
          incinerator for control of these streams.   The
          product purification waste streams vary in flow
          rate,  physical properties, and composition from
          the switch-condenser off-gas.  The secondary  in-
          cinerator is designed for optimal control  of
          product purification wastes while the main process
          incinerator is designed for optimal control of the
          switch-condenser off-gas.

Emissions from p_-xylene storage tanks are controlled by the
use of conservation vents.  Phthalic anhydride storage tanks
are controlled by means of condensers or sublimation traps,
or by venting to the incinerator.  Naphthalene storage tanks
can also be controlled by means of conservation vents.   In
most plants, however, either some or all of the storage tanks
are vented directly to the atmosphere.

Cyclones and baghouses are used to control the emission of
phthalic anhydride dust from the flaking and bagging oper-
ations.  In the case of naphthalene based plants, cyclones
are also used on the new and spent catalyst storage hopper
vents to control the emission of vanadium oxide catalyst
dust during catalyst transfer operations.

B.   PROCESS MODIFICATIONS

A number of modifications to the basic phthalic anhydride
process have been proposed in the recent literature.  Since
these modifications will potentially have an effect on
pollution control methodology with respect to phthalic
anhydride production, they are discussed in this  section.
                               109

-------
1.   The Rhone-Progil "Chauney '71" Process

A new process for the production of phthalic anhydride from
o-xylene has been developed by Rhone-Progil in France. 55  A
1.5 x 104 metric tons/yr plant has been in operation since
1971 in Aisne, France, and the process has been licensed to
Resins/ Inc.  (Philippines) and to Reposa  (Spain).  The flow
diagram for the new process is very similar to that for the
BASF process  (Figure 2), with the following differences:

     • No sulfur or sulfur dioxide is required in the
       reactor feed to maintain catalyst activity.  Hence,
       this source of pollution is eliminated.

     • The switch-condenser off-gas is controlled by
       means of a catalytic incinerator without an inter-
       mediate scrubber.  This control method is econo-
       mically more attractive here than with the BASF
       process due to a lower air to o_-xylene feed
       ratio.  (Specific figures have not been published.)

     • The liquid waste stream from the distillation
       section is further processed to produce solid
       pellets,  which are disposed of by burning.

It is claimed55  that the exhaust gases from this process
"meet all existing and forecasted regulations."

2.   Maleic Anhydride Recovery

The UCB (Brussels, Belgium) process for the recovery of
maleic anhydride from the scrubber liquid waste stream is
55Zimmer, J. C.  New Phthalic Anhydride Process.  Hydro-
  carbon Processing.  5_3_: 132-134, November 1974.
                              110

-------
described in Reference 56.  A similar process has been
developed by BASF.57  Instead of being sent to the inciner-
ator for disposal, the scrubber liquid purge stream is sent
to the continuous maleic anhydride recovery system shown
schematically in Figure 9.  The recovery process can be
divided into three steps:  concentration, dehydration, and
distillation.  The first step removes the organic compounds
from the scrubber purge stream in two stages.  In the first
stage, part of the water is evaporated at a closely controlled
temperature to avoid any crystallization of organics.  In
the second stage, the remaining water is flashed off in an
evaporator, leaving a bottom stream of liquid organics.
The water removed in both stages, which contains some residual
organic material, is condensed and recycled to the scrubber.

In the second step, the liquid organic stream is sent to a
dehydration unit where the organic acids are thermally
dehydrated to the anhydrides.  The maleic anhydride and
dehydration water are distilled off and separated by a
selective condensation process.  The water is recycled to
the scrubber, while the crude maleic anhydride is sent to
the vacuum distillation column for purification  (Step 3).
The refined maleic anhydride (97.7%) is taken overhead and
the bottom stream from the distillation column is recycled
to the dehydrator.  If maleic anhydride of greater purity
is required, a further purification step can be added.
 The recovery of the water is also important because sodium
 ion-free water must be used in the process.  Sodium ions
 catalyze the polymerization of maleic anhydride, and this
 reaction can result in an explosion.
56Weyens, E.  Recover Maleic Anhydride.  Hydrocarbon
  Processing.  S3:132-134, November 1974.
57Wirth, F.  Recover MA from PA Scrubber Water.  Hydro-
  carbon Processing.  5_4_: 107-108, August 1975.
                               Ill

-------
                         -CONCENTRATION	*f«	DEHYDRATION	*|«	Dl STILLATION
FROM PAN SCRUBBER

If
3
E
5
P
I
J

^

               RESIDUES TO BURNER
                                               UJ
                                               o
                                                 k	1
                                                     L-,-
                                                                  = 3
                                                                  o _
                                                                  < t-
                                                                                 TO PAN SCRUBBER
                                                                                   MALEIC ANHYRIDE TO STORAGE
                      Figure  9.   Maleic  anhydride recovery  process56

-------
The major components in the liquid waste stream from the
dehydrator are phthalic anhydride, benzoic acid, citraconic
anhydride, and maleic anhydride.  This stream can be disposed
of in a conventional burner.  Hence, a phthalic anhydride
plant incorporating this modification does not require an
incinerator for waste disposal.

3.   Direct Production of Phthalates

The single largest use of phthalic anhydride is in the
production of ortho-phthalates, which are used as plasti-
cizers.  Phthalates are manufactured by reacting phthalic
anhydride with the desired alcohol in the presence of
sulfuric acid, which catalyzes the reaction.  Wu and Maa58
have suggested a modification of the phthalic anhydride
process, by which the phthalate, rather than phthalic
anhydride, is produced.  A schematic flow diagram of the
proposed process is shown in Figure 10.  The reactor gases
(stream 5), instead of being sent to switch-condensers, are
fed to an absorption tower where they are contacted with
an alcohol-sulfuric acid mixture  (stream 6).  The phthalic
anhydride is absorbed in the alcohol and reacts to form the
monoester.  Thus, separation of phthalic anhydride from the
gas mixture and esterification to monoester are carried out
in a single step.  The monoester-alcohol-sulfuric acid
mixture (stream 8)  is sent to a fractionation tower where
the reaction proceeds further to form the diester.  The
phthalate is taken off as the bottom stream (stream 9) from
the fractionation tower, while the overhead stream (stream
10) is condensed and sent to a separator where the water
formed in the esterification reaction is removed.  The
58Wu, W. H., and J. R. Maa.  Make Phthalates Direct,
 ^Hydrocarbon Processing.  5^:117-118, April 1974.

                             113

-------
PROCESS (1
          A  o XYLENE STORAGE
          B  S02 STORAGE
          C. O-XYLENE PREHEATER
          D. COMPRESSOR
          E. AIR PREHEATER
          F. REACTOR
          G. ABSORPTION TOWER
          H. FRACTIONATION TOWER
          I  SEPARATOR
                                    ALCOHOL
                                    H2S04 "

                                     WASTE
                                      GAS
                                                         H
PHTHALATES
WATER RICH
  PHASE
         Figure 10.   Flow diagram for  direct production of  phthalates
                                                                           58

-------
organic-rich phase  (stream  11)  from the separator is recycled
to the absorption and  fractionation towers.

The waste gas stream from the absorption tower  (stream 7) is
expected to have a composition  similar to the switch-condenser
off-gas in the standard phthalic anhydride process.  Hence,
scrubbing and/or incineration would be required as control
measures for this stream.   The  principal advantage of the
modified process, from the  standpoint of air pollution control,
is that the emissions  from  the  product purification section
of the standard phthalic anhydride plant would be eliminated.
Emissions of heat-transfer  fluid from the switch-condenser
heat-transfer circuits would also be eliminated, as well as
phthalic anhydride emissions from the flaking and bagging
operations.  Phthalic  anhydride emissions from storage tanks
and the transport loading facility vent would be replaced by
phthalate emissions.   The other waste stream from the modified
process is the water-rich phase  (stream 12) from the separa-
tor.  This stream is expected to contain, as contaminants,
alcohol, sulfuric acid, phthalic acid, maleic acid, benzoic
acid, phthalate and other organics.  Hence, this stream pre-
sents a serious water pollution control problem.

4.   Alternate Feedstocks

Phthalic anhydride processes based on feed materials other
than o-xylene and naphthalene are possible and may become
important in the future should  supplies of the latter
materials become scarce.   Two alternate raw materials that
have been used are methyl naphthalene, which must be de-
methylated prior to use,  and acenaphthlene.59  Both of these
compounds are obtained from coal tar.  In fact, coal tar con-
tains more methyl naphthalenes than naphthalene.
59Austin, G. T.  Industrially Significant Organic Chemicals,
  Part 8.  Chemical Engineering, 24^107, July 22, 1974.
  \
115

-------
                         SECTION VI
              GROWTH AND NATURE OF THE INDUSTRY

A.   PRESENT TECHNOLOGY

Phthalic anhydride is currently produced in the United States
from o-xylene via the fixed-bed process and from naphthalene
via the fluid-bed process.  The BASF process is the most
recent and most widely used fixed-bed process, while the
Badger-Sherwin-Williams process is the dominant fluid-bed
process.  In recent years, o-xylene has become the preferred
feedstock due to its lower unit product cost, and all new
plants built since 1968 use o-xylene.  This trend is further
reflected in the fact that since 1968, 10 naphthalene based
plants in the U.S. have been shut down and naphthalene based
capacity has declined from 2.39 x 105 metric tons/yr in 1968
to 1.5 x 105 metric tons/yr in 1975.1'3'4   (However, a
2.27 x 104 metric tons/yr expansion of the U.S. Steel naph-
thalene based plant is scheduled for start up in 1975.4)
This trend could be reversed in the future if rising petro-
leum prices should result in a more favorable price for
coal-tar naphthalene than for o-xylene.  Since the fixed-bed
plants are capable of operating with either o-xylene or
naphthalene, a change of feedstock could be made both quickly
and easily.
                              116

-------
B.   EMERGING TECHNOLOGY

Two significant improvements  in the fixed-bed o-xylene
based process have been made  by Rhone-Progil.55  First,  in
the Rhone-Progil process sulfur dioxide  is not required  in
the reactor feed stream to maintain catalyst activity.
Hence, this source of air pollution is eliminated.  Second,
a lower air to xylene feed ratio has been achieved  (although
a specific figure has not been published), thus reducing the
volume of waste gas that must be treated.  This reduction in
the volume of waste gas improves the economics of treatment
by direct incineration, as opposed to treatment by  a  scrubber-
incinerator combination.  Direct incineration has the advan-
tage of controlling the emission of the  carbon monoxide
that is present in the waste  gas stream.

Other recent innovations in phthalic anhydride production
involve the recovery of maleic anhydride55 and benzoic
acid55 as by-products from the waste gas  stream.  The pro-
duction cost of maleic anhydride by the  UCB recovery
process56 is less than the raw material  cost alone  for the
conventional process using benzene.  In  addition, the cost
of process equipment is less  than for a  benzene based plant
of the same capacity.  With increasing costs for petro-
chemicals, it is expected that by-product recovery  from
phthalic anhydride production will become even more
economically attractive in the future.

C.   MARKETING STRENGTHS AND  WEAKNESSES

Figure 11 shows schematically the usage  of phthalic anhydride
in the United States. 60  Each of the major product  categories
is discussed below.
60 chemical Origins and Markets, 4th Edition.  Menlo Park,
  Stanford Research Institute, Chemical Information Services,
  1967.   99 p.
 \
                              117

-------
oo










PHTHALIC a.
ANHYDRIDE









nniiALAtt PU-TICIZCR- » [PLASTICIZERS FOR POLYVINU CHLORIDE
'Tli^ElHYLHEXYLrPHTHALATE 1 AND COPOLYMER RESINS
OmSOOECYL) PHTHAL4TE
SlJJm^HTHmu"" » pLASTICIZER FOR POLYVINYL ACETAH
BiETHYL PHTHiLATEn 	 . " 1 *«» CELLULOSICS
|_ DIMETHYL PHTHALATEj V. *—
^\^ ^PPLASTICIZERS FOR POLYVIHYL ICETATE
[AND CELLULOSICS: INSECT REPELLANTS

^ IIHS1TIJR4TFD ^
""I POLYESTER RESINS \

rTcTRArHinon AHH TFTRiRB(wn~! Rl<*C RET*ROAHTS IN POLYeSTER RESIHS,
1 TETRACHLORO- AND TETRABRCMO- POLYURETt1AMrFOAH3 5URFACE ^A*TIHQ^
^i PHTHALIC ANHYDRIDES AND ACIDSl 1 r-Tm" i.
1 —

// 	 '1


MOLDED PLASTICS
POLYESTER RESINS
_. „,„,„, rilTI,.LrlN pHEOICIHALS
«. PIIEHOLPHTIIALriN » j^H m[um

^ nillHI?ArHHf! '1-niHYnBnvYAHTHRAOIIIHnHFl 	 »— OYFS
                                                                                                                             ODKSTWKTIOK APPLICATIONS
                                                                                                                             DAT HULLS
                                                                                                                             UH3PORTATIOJ4 APPLICATIONS
                                                                                                                          •-CASTIHG RESIHS, AUTOMOTIVE PUTTIES
                                                                                                                            DYES
                                                                                                                            PLASTICIZERS
                                                                                                                          •-pHTHALOCYANIHE DYES
                                                                                                                           JAZO AHD INDIGO ID DYES
                                                                                                                            «IHYL ANTHRANIUTE —
                                                                                                                                                     •*- PERFUME
                                            Figure  11.    Uses  of  phthalic  anhydride
                                                                                                            60

-------
The largest single use of phthalic anhydride is in the
manufacture of plasticizers, which accounts for 50% of the
phthalic anhydride produced in the U.S.3   Plasticizers are
diphthalates produced by the esterification reaction of two
moles of alcohol with one mole of phthalic anhydride in the
presence of sulfuric acid, which catalyzes the reaction.
                                        O
                                        II
                           H2SO4    /^C
            0 + ROH + R'OH	  [   |T       + H2O
The major phthalic anhydride plasticizers, with their per-
centages of total phthalic plasticizer production, are:1

              di(2-ethylhexyl)phthalate  (40%)
              di(isodecyl)phthalate  (15%)
              di(iso-octyl)phthalate  (9%)
              n-octyl-n-decyl phthalate  (6%)

About 87% of these products are used to  impart low-temperature
flexibility, resilience, high impact strength, and good
electrical properties to polyvinyl chloride polymers and
copolymers  (with vinyl acetate).  Hence, their future growth
is tied directly to that of non-rigid PVC.

Raw material shortages, the slump in the automobile industry,
and concern over the health hazards of vinyl chloride monomer
could seriously affect the growth of non-rigid PVC.  For
example, new legislation restricting worker exposure to
vinyl chloride monomer may curtail output and force much
                             119

-------
higher resin prices.61  In fact, production for the first
quarter of 1975 was down nearly 40% compared to the same
period in 1974. 4  In spite of these concerns, however, the
industry is planning at least a one-third increase in PVC
resin capacity by the end of 1976.62

The above caveats notwithstanding, overall plasticizer pro=
duction is expected to increase at an average annual rate of
5% in the period 1975 to 1980.  Phthalate ester plasticizers
will ikely increase at a slightly lower level due to an
increasing demand for more flame-retardant vinyls, which will
result in a greater use of phosphate plasticizers.  In
addition, some competition is expected from specialty plasti-
cizers such as the trimellitates, the adipates, and the
sebacates, in applications where improved properties are im-
portant.  Historically, phthalate ester plasticizers ex-
perienced an average growth rate of 9% in the period 1963 to
1973.63

The second largest use of phthalic anhydride is in the manu-
facture of unsaturated polyester resins, which accounts for
24% of domestic phthalic anhydride production.3  Unsaturated
polyesters are made from unsaturated organic acids (as their
anhydrides)  and glycols, and are crosslinked with mono-
mers such as styrene, vinyltoluene, and diallyl phthalate.
A typical reaction involving maleic and phthalic anhydrides
is:
6 Edwards, P.  Chemicals '75/Aliphatics.  Chemical Marketing
  Reporter.  207:40-42, January 6, 1975.
62Burke, D. P.  Forecast '75 Riding it Out.  Chemical Week.
  116_: 17-24, January 8, 1975.
53Synthetic Organic Chemicals, U.S. Production and Sales.
  U.S. International Trade Commission  (formerly U.S. Tariff
  Commission).  Washington.  1963-1974.
                              120

-------
                                        ?
            TJfV r*                        s-*
         (N)  ||   N0 +  (2N)  HOROH  +  (N)  ^^ X~  Heat
            HC—C-7
               O
         Maleic anhydride  Glycol
Phthalic anhydride
H-
      O     O    0  0
      ||     II    II  II
   5RO-C-CH=CH-C-ORO-C  C-
        Prepolymer
-OH
N
Heat
Catalyst
Styrene
O O 0
II II II
OT?O C* f"*H I^*H~ f^— ^P/^— ^*
wr\w ^. v^n L.n L. wi\^ v-
CH2 £
0 00
ii II II
-ORO-C-CH-CH-C-ORO-C
^
O
II
/-»
"A
0
II
A J
                                        Unsaturated polyester resin
In  1973,  78% of unsaturated polyester  resins were used for
the production of glass fiber reinforced plastics (FRP) and
the remainder for non-reinforced applications.64

The average annual growth of unsaturated polyester resin
production from 1963 to 1973 was 15%.65   After a slight
decline  in 1970,  growth in the 1971-1973 period averaged
22%.65/6S   These large increases in polyester resin consump-
tion were  the result of a large increase in the pleasure
boat business,  the large-scale use of  synthetic marble and
FRP tub/shower units in construction,  and the first large-
scale applications of low profile, low shrink resins in the
transportation and equipment markets.
6£f1972 Annual  Statistical Report.  The  Society of the Plastics
  Industry.  New York.  November 16, 1973.
65Anderson,  E.  V.  Growth Slows in Top  50 Chemicals' Output.
  Chemical  and Engineering News.  52i10-13,  May 6, 1974.

                                121

-------
The years 1973 and 1974, especially the latter, were charac-
terized by widespread shortages of raw materials, including
styrene, maleic anhydride, and most glycols.  As of late
1974, raw material availability was generally adequate,
although some glycols were still in short supply.  As a
result of the severe raw material shortages in the first nine
months of 1974, reported unsaturated polyester resin produc-
tion for this period was down by 10% compared to the same
period in 1973.66  Production for the first quarter of 1975
was down 24% compared to the first quarter of 1974.4  How-
ever, growth is expected to resume after the first half of
1975.

The average annual growth rate for unsaturated polyester
resins is expected to be 11% to 13% in the period 1974 to
1979.  This rate of growth will result in an increase in
production from an estimated 3.85 x 105 to 4.4 x 105 metric
tons in 1974 to 6.6 x 105 to 8.21 x 105 metric tons in 1979.
The demand for phthalic anhydride in this area is expected
to grow at a level somewhat below that of total unsaturated
polyester resin production.  This is due to the substitution
(some of which is believed to be permanent) of isophthalic
resins for some phthalic-based resins caused by the recent
raw materials shortage.

The third major outlet for phthalic anhydride is the produc-
tion of alkyd resins, which accounts for 19% of domestic
phthalic anhydride production.3  Alkyd resins are a type of
polyester resin,  being the reaction product of a polybasic
acid and a polyol.  Alkyds, however, contain an unsaturated
monocarboxylic acid,  such as those from the drying oils, as
66Anderson, E. V.  Recession Stifles Output of Top 50
  Chemicals.  Chemical and Engineering News.  53:30-33,
  May 5, 1975.
                             122

-------
a modifying agent.  Thus, after coating on a surface, the
alkyd resin can undergo further oxidative polymerization
like a drying oil to yield a very tough, elastic, weather-
resistant film.  These synthetic enamels have been used
extensively for the finishing of automobiles and household
appliances.

Until 1961, alkyd resin manufacture was the largest single
user of phthalic anhydride.  The rapid rise in plasticizer
demand and, later, in unsaturated polyester resins, combined
with the slow growth of alkyd resins  (due primarily to
competition from other surface coating resins), resulted in
alkyds dropping to third on the list of phthalic anhydride
consumers.  Phthalic anhydride consumption for alkyd resin
manufacture has remained nearly static in the period 1967
to 1974 (9.34 x 101* metric tons in 1967 versus 8.89 x 104
metric tons in 1974).63

Total alkyd surface coatings consumption is expected to
decline at an average annual rate of -5% to -1% in the
period 1975-1978, mostly as the result of increased compe-
tition from other coating systems and stricter anti-pollution
laws governing the use of solvent-based paints.

Exports and various miscellaneous products account for 7%
of the phthalic anhydride produced in the U.S.3  Included
in this category are anthraquinone and its derivatives
(intermediates for the synthesis of anthraquinone dyes), lead
phthalate (a PVC stabilizer),  tetrachloro- and tetrabromo-
phthalic anhydride  (fire retardants in polyester resins),
diallyl phthalate  (a crosslinking agent for polyesters),
dibutyl phthalate and phthalonitrile  (insecticides), phenol-
phthalein  (a pH indicator and laxative), phthalein  and
xanthene dyes, phthalimide  (used in various organic  syntheses
and in the manufacture of perfume and indigo dye),  and  eosin
(tetrabromofluorescein) inks and dyes.
                              123

-------
Exports have exhibited a sharp increase during the past
four years, as is shown below:4'63

                             Phthalic anhydride
             Year         exports/ 103 metric tons
             1971                    3.48
             1972                    6.26
             1973                   10.08
             1974                   15.42

During the first quarter of 1975, however, exports ran at
an annual rate of only 1.1 x 103 metric tons.  The reason
is that foreign phthalic prices had dropped below the domes-
tic price.4

Domestic phthalic anhydride production increased at an
average annual rate of 8.4% during the period 1963 to 1973.3
In 1974, production totaled 4.692 x 105 metric tons.  During
the first quarter of 1975, however, production totaled only
5.62 x 104 metric tons.4  This corresponds to an annual rate
of 2.248 x 105 metric tons, which is less than half the
nominal plant capacity in the U.S.  The sharp drop in
phthalic production reflects the declines in phthalic's
major markets, polyvinyl chloride and unsaturated polyester
resins.

Phthalic production is expected to recover in the second
half of 1975 to reach a total output of approximately
3.18 x 105 metric tons for the year.  Annual production is
projected to reach 4.99 x 105 metric tons in 1978 and
6.35 x 105 metric tons in 1980.4
                              124

-------
                         SECTION VII

                         APPENDIXES
A.   Storage Tank Calculations

B.   Rationale for Not Considering All Species Listed in
     Table 21
                              125

-------
                        APPENDIX A

                 STORAGE TANK CALCULATIONS


The procedure for calculating the emissions from storage
tanks is outlined in this section.  The equations given
below were derived in References  39-43.

Step 1.  Calculate the equivalent gasoline breathing loss;

      24  /   p  \0.68
--_
1000  i4.7-p
                       rjl.73/tiM0.51  ( Am\ 0 . 50 p  p      (A
                       D     (H  }      (AT)        c      (A
where  L  = equivalent gasoline breathing loss, bbl/yr
        P = vapor pressure of material stored at bulk
            temperature, psia
        D = tank diameter, ft
       H1 = average tank outage, ft
       AT = average daily ambient temperature change,  °F
       F  = paint factor
        C = diameter factor

The bulk temperatures of the material stored were taken to
be 200°F for naphthalene tanks, 300°F for phthalic anhydride
tanks, and 74°F for o-xylene tanks.  The latter temperature
was obtained by adding 5°F to the ambient temperature, as
recommended in References 39-43 for tanks held at ambient
temperature.  The ambient temperature was assumed to be 69°F,
the national mean ambient temperature.

                             126

-------
Tank diameters were computed by assuming a height of 50 ft
for all tanks.  The average tank outage, i.e., freeboard,
was taken as one-half the tank height, or 25 ft.

The average daily ambient temperature change, AT, was taken
as 20°F, which is the national average value.  The paint
factor, Fp, was assumed equal to unity, the value for white
paint in good condition.  This factor can be as high as
14.6 for gray surfaces.  The diameter factor, C, is equal
to unity for tanks 30 ft or larger in diameter.  For smaller
tanks, the value is obtained from a graph given in Reference 41,
and is between 0.25 and 1.0.

Step 2.  Calculate the equivalent gasoline working loss:
where  F  =
        V =
        N =
                        Pg=  157000  PVNKT
equivalent gasoline working loss, bbl/yr
tank capacity, bbl
number of turnovers per year
turnover factor =1.0 for N £ 36
                  180 + N
                                          (A-2)
                                 6N
                                      for N > 36
Step 3.  Compute total equivalent gasoline loss, L :
                         L  = L  + F
                          9    Y
Step 4.  Compute petrochemical losses:
                    L = o.os (|)LC
                                           (A-3)
                                           (A-4)
where  L = total petrochemical loss, bbl/yr
       M = molecular weight of chemical stored
       W = liquid density of chemical stored, Ib/gal
                             127

-------
Step 5.  Calculate emission factor:

                    Lj = L  (42) (W)                     (A-5)

                    E1 = •=—-                           (A-6)
                     E = f-                            (A-7)

where   Lj = petrochemical loss, Ib/yr
       Cap = production capacity, ton/yr
        E1 = emission factor, Ib/ton
         E = emission factor, g/kg

The necessary input data for the above calculations are
tabulated in Tables A-l and A-2, while the results are
summarized in Tables A-3 and A-4.  The tank numbers in
these tables correspond to those in Tables 7 and 14 for o-
xylene and naphthalene based plants, respectively.
                             128

-------
                 Table A-l.  STORAGE TANK  INPUT  DATA FOR o-XYLENE BASED PLANT
Tank number
1
2
3 & 4
5 & 6
7
8
                                           Input  data
Average ambient temp. , °F
Average ambient temp, change, °F
Material molecular weight
Liquid density, Ib/gal
Vapor pressure at bulk temp., psia
Bulk temp. , °F
Tank diameter, ft
Tank outage, ft
Paint factor
Diameter factor
Turnover factor
No. turnovers per year
Tank capacity, bbl
69
20
106
7.37
0.194
74
17.3
25
1.0
0.81
0.297
230
2,070
69
20
106
7.37
0.194
74
54.8
25
1.0
1.0
1.0
23
21,000
69
20
148
8.35
0.33
300
17.9
25
1.0
0.83
0.63
65
2,260
69
20
148
8.35
0.33
300
17.5
25
1.0
0.82
0.58
72
2,140
69
20
148
8.35
0.33
300
18.5
25
1.0
0.84
0.85
44
2,380
69
20
148
8.35
0.33
300
26.1
25
1.0
0.98
0.85
44
4,760
to
ID

-------
                           Table A-2.  STORAGE TANK INPUT DATA FOR NAPHTHALENE BASED PLANT
Tank number
1
2
3
4
5
6 & 7
8
9
                                                     Input data
Average ambient temp. ,
Average ambient temp.
change, °F
Material molecular
weight
Liquid density, Ib/gal
Vapor pressure at bulk
temp. , psia
Bulk temp. , °F
Tank diameter, ft
Tank outage, ft
Paint factor
Diameter factor
Turnover factor
No. turnovers per year
Tank capacity, bbl
69
20
128
8.0
0.31
200
58.4
25
1.0
1.0
1.0
1
23,810
69
20
128
8.0
0.31
200
58.4
25
1.0
1.0
1.0
13
23,180
69
20
128
8.0
0.31
200
41.3
25
1.0
1.0
1.0
26
11,905
69
20
148
8.35
0.33
300
17.3
25
1.0
0.81
0.37
148
2,095
69
20
148
8.35
0.33
300
37.0
25
1.0
1.0
1.0
2
9,524
69
20
148
8.35
0.33
300
12.8
25
1.0
0.65
0.40
130
1,143
69
20
148
8.35
0.33
300
22.6
25
1.0
0.93
1.0
2
3,571
69
20
148
8.35
0.33
300
22.6
25
1.0
0.93
0.92
40
3,571
U)
o

-------
              Table A-3.  STORAGE TANK CALCULATION SUMMARY FOR o-XYLENE BASED PLANT
Tank
No.
1
2
3
4
5
6
7
8
Material stored
o-Xylene
o-Xylene
Crude product
Crude product
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Capacity,
gal
87,000
880,000
95,000
95,000
90,000
90,000
100,000
200,000
Losses
gal/yr
546
2,770
858
858
836
836
872
1,747
Ib/yr
4,027
20,417
7,165
7,165
6,977
6,977
7,282
14,584
Emission
factor, g/kg
0.031
0.157
0.055
0.055
0.054
0.054
0.056
0.112
U)

-------
           Table A-4.   STORAGE TANK CALCULATION SUMMARY FOR NAPHTHALENE BASED PLANT
Tank
No.
1
2
3
4
5
6
7
8
9
Material stored
Naphthalene
Naphthalene
Naphthalene
Crude product
Crude product
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Phthalic anhydride
Capacity,
gal
1,000,000
1,000,000
500,000
88,000
400,000
48,000
48,000
150,000
150,000
Losses
gal/yr
2,582
3,970
2,899
953
1,422
490
490
577
1,310
Ib/yr
20,658
31,759
23,194
7,954
11,872
4,090
4,090
4,819
10,937
Emission
factor, g/kg
0.165
0.254
0.186
0.064
0.095
0.033
0.033
0.039
0.088
U)
ro

-------
                         APPENDIX B
RATIONALE FOR NOT CONSIDERING ALL SPECIES LISTED IN TABLE 21

A number of the possible reaction products listed in Table 21
were not studied explicitly in this work since they are emitted
in, at most, trace amounts.  The compounds are not emitted in
greater than trace amounts for the following reasons:

     They are not formed in greater than trace amounts, due
     either to unfavorable conditions in a commercial reactor,
     or to the fact that they are reaction products of
     feedstock impurities which are themselves present in
     sma11 amounts.
     They are too reactive to be stable under the process
     conditions.

An exception is citraconic anhydride, which may be emitted
in greater than trace amounts.  However, it is undoubtedly
included in the reported data as maleic anhydride, since the
two are chemically very similar.  This is not considered to
be a drawback because it is reasonable to assign citraconic
anhydride the same TLV as that of maleic anhydride (in lieu
of other information) and to treat the two together for the
purpose of pollution control.

Each of the compounds listed in Table 21 but  not explicitly
included in the study is listed in Table B-l, together with
the reason for its exclusion.
                               133

-------
                           Table B-l.  COMPOUNDS NOT INCLUDED IN STUDY
                   Compound
                                                    Reason for exclusion
CO
o-Methylbenzyl alcohol
o-Toluic acid
p_-Toluic acid
Toluene

o-Hydroxymethylbenzoic acid
Citraconic anhydride
Phthalide
Phthalaldehydic acid
Phthalic acid
Terephthalic acid
p_-Benzoquinone
Isophthalic acid
o'-Carboxylphenyl-o-methylphenyl acetate
o_' -Methylphenyl-o-methylphenyl acetate
bis(o-Methylphenyl)methyl ether
1,2-bis(o-Methylphenyl)ethane
Reactive intermediate
Reactive intermediate
Precursor is feed impurity
Reactive; formation not favored under com-
  mercial reactor conditions
Reactive intermediate
Reported as maleic anhydride
Reactive intermediate
Reactive intermediate
Emitted as phthalic anhydride
Precursor is feed impurity
Precursor is feed impurity
Precursor is feed impurity
                                               Formation not favored under commercial reactor
                                                 conditions; would be decomposed by incinera-
                                                 tion

-------
As a further justification for not considering the above
compounds, upper limits were estimated for the source severity
due to these compounds.  All of the above compounds are in-
cluded in the category "miscellaneous organics" listed in
Tables 18 and 19.   (See also Table 33, footnote 1, and
Table 34, footnote  p.)  For the purpose of this calculation,
it is assumed that  the miscellaneous organics in the reactor
product stream are  partitioned between the crude product and
switch-condenser off-gas in the same ratio as phthalic anhy-
dride.  Using the data in Table 18 for o-xylene based plants,
this procedure yields 33.3 kg/hr miscellaneous organics in
the crude product stream, and 0.8 kg/hr in the switch-con-
denser off-gas.  The former amount is assumed to be removed
during product purification and sent to the secondary in-
cinerator.  The latter amount is sent to the main process
incinerator.  Removal efficiencies of 96.5% in the main
process incinerator and 99% in the secondary incinerator are
assumed.  These are the values for total organics obtained
from Reference 13.  The resulting emission rates are
0.028 kg/hr from the main process incinerator and 0.333 kg/hr
from the secondary  incinerator.  The corresponding source
severities are:

            1.15 x  10~5 f                .
        S = 	==-rr	 for main process incinerator
                1 J_iV
            5 5 x lO"1*
        S = —	=jrr=	  for secondary incinerator
                IJjV
Assuming an average TLV for the miscellaneous organics of
0.001 g/m3  (equivalent to maleic anhydride), the source
severities are 0.01 for the main process incinerator and
0.55 for the secondary incinerator.
                              135

-------
A TLV of 0.001 g/m3 represents a very stringent assumption,
since maleic anhydride has the lowest TLV of the major species
which are emitted.  If the TLV for phthalic anhydride
(0.006 g/m3) is used in the calculation, the severity for the
secondary incinerator is less than 0.1.  In addition, the
estimated severities are for the total of all species included
in the miscellaneous organics category.  The severity for any
individual compound should be much smaller.

A similar calculation based on the data in Table 19 for a
naphthalene based plant results in the following source
severities for miscellaneous organics:
        S = 1 03 x
            — ' — ==-r= - for main process incinerator
            C QQ y 10""1*
        S = ——==rr=	 for secondary incinerator
Assuming a TLV of 0.001 g/m3 yields source severities of 0.01
for the main process incinerator and 0.61 for the secondary
incinerator.
                              136

-------
                        SECTION VIII

                      GLOSSARY OF TERMS
ATMOSPHERIC STABILITY CLASS - Class used to designate
degree of turbulent mixing in the atmosphere.

BADGER-SHERWIN-WILLIAMS PROCESS - Fluid-bed process developed
by Sherwin-Williams Co. and the Badger Company for the oxi-
dation of naphthalene to phthalic anhydride.

BASF PROCESS - Fixed-bed process developed by Badische Anilin
and  Soda Fabrik for the oxidation of o-xylene to phthalic
anhydride.

CRITERIA POLLUTANT - Emission species for which an ambient
air quality standard has been established.

EMISSION FACTOR - Weight of material emitted to the atmos-
phere per unit weight of phthalic anhydride produced.

FLAKER - Device which solidifies liquid phthalic anhydride
in the form of flakes.

MAIN PROCESS INCINERATOR - Incinerator which burns the
off-gas from the switch condenser.

NON-CRITERIA POLLUTANT - Emission species for which no
ambient air quality standard has been established.

PARTIAL CONDENSER - Condenser used to condense phthalic
anhydride as a liquid in the fluid-bed process.

SECONDARY INCINERATOR - Incinerator which burns the
waste streams from the product purification section of
the phthalic anhydride plant.

SWITCH CONDENSER - Condenser used to condense phthalic
anhydride as a solid.

TANK OUTAGE - Distance from liquid surface to top of
storage tank.


                              137

-------
                         SECTION  IX

          CONVERSION FACTORS AND  METRIC PREFIXES67
To convert from
degree Celsius
joule  (J)
kelvin
kilogram  (kg)
   CONVERSION FACTORS
         to
degree Fahrenheit
British thermal unit
degree Celsius
pound-mass (Ib mass
  Multiply by
t« = 1.8 t£ + 32
9.479 x lO'4
t£ = tj - 273.15
2.204

kilogram (kg)
meter (m)
meter (m)
meter (m)
meter (m)
meter3 (m3)
meter3 (m3)
pascal (Pa)

Prefix Symbol
mega M
kilo k
milli m
micro y
nano n
avoirdupois)
ton (short, 2,000 Ib
mass)
angstrom
foot
micron
mile
barrels (42 gal)
gallon (U.S. liquid)
pound- force/inch2
(psi)
PREFIXES
Multiplication
factor
106 5
103 5
10~3 5
10~6 5
10~9 5

1.
1.
3.
1.
6.
6.
2.
1.


102
000
281
000
215
293
642
450


X
X

X
X

X
X


io-3
1010

IO6
io-*

IO2
10-"











Example
MJ
kg
mm
ym
nm
= 5
= 5
= 5
= 5
= 5
X
X
X
X
X
IO6
IO3
io-3
10~6
io-9
joules
grams
meter
meter
meter
67Metric Practice Guide.  American Society for Testing and
  Materials.  Philadelphia.  ASTM Designation:  E 380-74.
  November 1974.  34 p.
                              138

-------
                          SECTION X

                         REFERENCES
 1.   Ockerbloom,  N.  E.   Xylenes and Higher Aromatics.  Part
     3:   Phthalic Anhydride.   Hydrocarbon Processing.
     2^:162-166,  September 1971.

 2.   Landau,  R.,  and H.  Harper.  Phthalic Anhydride.  Chem-
     istry and Industry (London).  July 29, 1961, p. 1143-
     1152.

 3.   Phthalic Anhydride.   Chemical Marketing Reporter.
     205:9  March 4, 1974.

 4.   Anderson, E. V.  Phthalic Anhydride Makers Foresee
     Shortage.  Chemical and Engineering News.  53:10-11,
     June 30, 1975.

 5.   Bernardini,  F., and M. Ramacci.  Oxidation Mechanism
     of o-Xylene to Phthalic Anhydride.  Chimica e 1'Industrie
     (Milan).  48^9-17,  January 1966.

 6.   Friedrichsen, W.,  et al.  Production of Phthalic
     Anhydride.   British Patent No. 1,082,326 (to BASF),
     September 6, 1967.

 7.   Battacharyya, S. K., and I.  B. Gulati.  Catalytic
     Vapor-Phase Oxidation of Xylenes.  Industrial and
     Engineering Chemistry.  5_£: 1719-1726, December 1958.

 8.   Spitz, P. H.  Phthalic Anhydride Revisited.  Hydro-
     carbon Processing.   4^7:162-168, November 1968.

 9.   Schwab,  R.  F.,  and W. H. Doyle.  Hazards in Phthalic
     Anhydride Plants.   Chemical Engineering Progress.
     ££:49-53, September 1970.

10.  i Phthalic Anhydride:   Return to Overcapacity.  Chemical
    i and Engineering News.  £9_: 18-19, March 15, 1971.
                             139

-------
11.  Phthalic Anhydride by Vapor-Phase Oxidation.  The Oil
     and Gas Journal.  Tl:92, March 12, 1973.

12.  Riley, H. L.  How to Design and Operate Fluidized-
     Catalyst PA Plants.  Hydrocarbon Processing and
     Petroleum Refiner.  £2:167-172, June 1963.

13.  Personal communication, H. M. Lacy, Monsanto Company.
     April 7, 1975.

14.  Fawcett, R. L.  Air Pollution Potential of Phthalic
     Anhydride Manufacture.  Journal of the Air Pollution
     Control Association.  20^:461-465, July 1970.

15.  Uhl, V. W., and H. P. Voznick.  Molten Salt as a Heat
     Transfer Medium.  Chemical Engineering Progress.
     59^:33-35, May 1963.

16.  Fried, J. R.  Heat-Transfer Agents for High-Temperature
     Systems.  Chemical Engineering.  80^:89-98, May 28, 1973.

17.  Seifert, W. F., L. L. Jackson, and C. E. Sech.  Organic
     Fluids for High-Temperature Heat-Transfer Systems.
     Chemical Engineering.  7_9^ 96-104, October 30, 1972.

18.  Purdy, R. B., et al.  Indirect Heating with Aromatic
     Oils.  Chemical Engineering Progress.  59:43-46,
     May 1963.

19.  Conant, A. R., and W. F. Seifert.  Dowtherm Heat Transfer
     Medium.  Chemical Engineering Progress.  59:46-49,
     May 1963.

20.  Petersen, D. E., and R. K. Bedell.  UCON Heat Transfer
     Fluid.  Chemical Engineering Progress.  59:36-39,
     May 1963.

21.  Davis, W. J., and P. G. Benignus.  Therminol FR-2 Heat
     Transfer Systems.  Chemical Engineering Progress.
     59^:39-42, May 1963.

22.  Frikken, D. R., K. S. Rosenberg, and D. E. Steinmeyer.
     Understanding Vapor-Phase Heat-Transfer Media.  Chemical
     Engineering.  £2:86-90, June 9, 1975.

23.  Simard, G. L., et al.  Vanadium Oxides as Oxidation
     Catalysts.  Industrial and Engineering Chemistry.
     47^1424-1430, July 1955.

24.  Cullis, C. F.  Heterogeneous Catalytic Oxidation  of
     Hydrocarbons.  Industrial and Engineering Chemistry.
     5$^: 18-27, December 1967.

                               140

-------
25.  Glueck, A. R. , and C. N. Kenney.  The Kinetics of the
     Oxidation of Sulphur Dioxide over Molten Salts.
     Chemical Engineering Science.  23^1257-1265, 1968.

26.  Schwartz, W. A., et al.  Engineering and Cost Study of
     Air Pollution Control for the Petrochemical Industry,
     Volume 7:  Phthalic Anhydride Manufacture from Ortho-
     Xylene.  Air Products and Chemicals, Inc., Houdry
     Division.  U.S. Environmental Protection Agency.  Research
     Triangle Park.  Publication EPA-450/3-73-006-g, July 1975.
     108 p.

27.  loffe, I. I., et al.  Kinetics and Vapor Phase Oxida-
     tion Mechanism of Aromatic Hydrocarbons.  VI.  Zhurnal
     Fizicheskoi Khimii  (Journal of Physical Chemistry).
     29^:692-698, 1955.

28.  Shelstad, K. A., J. Downie, and W. F. Graydon.  Kinetics
     of the Vapor-Phase Oxidation of Napthalene Over a
     Vanadium Catalyst.  Canadian Journal of Chemical Engin-
     eering.  38^102-107, August 1960.

29.  Graham, J. J., and P. F. Way.  Phthalic Anhydride by
     Fluid Bed Process.  Chemical Engineering Progress.
     _5_8:96-100, January 1962.

30.  Personal communication, L. B. Evans, U.S. Environmental
     Protection Agency  (Data originally supplied by U.S.
     Steel Corp.).  June 12, 1975.

31.  Chopey, N. P.  Fluid-Bed Phthalic Anhydride.  Chemical
     Engineering.  69_: 104-106, January 22, 1962.

32.  Riley, H. L.  Design of Fluidized Reactors for Naphtha-
     lene Oxidation:  A Review of Patent Literature.  Trans-
     actions of the Institution of Chemical Engineers  (London).
     37^:305-313, 1959.

33.  Improvements Relating to the Oxidation of Aromatic
     Hydrocarbons.  American Cynamid Company.  British
     Patent No. 850,817, January 2, 1957-

34.  Graham, J. J.  The Fluidized Bed Phthalic Anhydride
     Process.  Chemical Engineering Progress.  66:54-58,
     September 1970.

35.  Phthalic Anhydride  (Sherwin-Williams/Badger).  The
     Badger Company, Inc.  Hydrocarbon Processing.  46:215,
     November 1967-
   i
   \
36.\  Rousseau, W. P. C.  Production of Phthalic Anhydride.
     U.S. Patent No. 3,080,382  (to the Badger Co.), March 5,
     1963.
                                141

-------
37.  Personal communication, L. B. Evans, U. S. Environmental
     Protection Agency  (Data originally supplied by Union
     Carbide Corporation).  June 12, 1975.

38.  Ferrari, D. C., and C. G. Bertram.  Method and Apparatus
     for the Removal of Organics from Chemical Waste Gases.
     U.S. Patent No. 3,624,984 (to the Badger Co.), December
     7, 1971.

39.  Evaporation Loss from Fixed Roof Tanks.  American
     Petroleum Institute.  New York.  API Bulletin No. 2518.
     1962.  38 p.

40.  Use of Variable Vapor Space Systems to Reduce Evapor-
     ation Loss.  American Petroleum Institute.  New York.
     API Bulletin No. 2520.  1964.  14 p.

41.  Petrochemical Evaporation Loss from Storage Tanks.
     American Petroleum Institute.  New York.  API Bulletin
     No. 2523.  1969.  14 p.

42.  Evaporation Loss from Floating Roof Tanks.  American
     Petroleum Institute.  New York.  API Bulletin No. 2517.
     1962.  13 p.

43.  Evaporation Loss in the Petroleum Industry - Causes and
     Control.  American Petroleum Institute.  New York.  API
     Bulletin No. 2513.  1959.  57 p.

44.  Personal communication, R. G. Lunche and A. B. Netzley,
     Los Angeles County Air Pollution Control District.
     March 28, 1975.

45.  Air Pollution Engineering Manual, 2nd Edition.  Danielson,
     J. A. (ed.).  Environmental Protection Agency.  Research
     Triangle Park.  Publication No. AP-40.  May 1973.  987 p.

46.  Personal communication, W. Fitzgibbons, Standard Oil of
     Ohio (SOHIO).  June 23, 1975.

47.  Threshold Limit Values for Chemical Substances and
     Physical Agents in the Workroom Environment with Intended
     Changes for 1975.  American Conference of Governmental
     Industrial Hygienists.  Cincinnati.  1975.  97 p.

48.  Nonhebel, G.  Recommendations on Heights for New
     Industrial Chimneys.  Journal of the Institute of Fuel.
     33_:479-511, July 1960.

49.  Turner,  D. B.  Workbook of Atmospheric Dispersion Esti-
     mates.   U.S. Department of Health, Education, and Welfare,
     Public Health Service.  Cincinnati.  Publication No.
     999-AP-26.  1969.
                               142

-------
50.  Personal communication, L. B. Evans, U.S. Environmental
     Protection Agency (Data originally supplied by Stepan
     Chemical Company).  June 12, 1975.

51.  1972 National Emissions Report.  U.S. Environmental
     Protection Agency.  Research Triangle Park.  Publication
     No. EPA-450/2-74-012.  June 1974.

52.  Eimutis, E. C., and M. G. Konicek.  Derivations of
     Continuous Functions for the Lateral and Vertical
     Atmospheric Dispersion Coefficients.  Atmospheric
     Environment.  £: 859-863, 1972.

53.  Naphthalene Feedstock Outlook Mixed.  Chemical and
     Enginering News.  5_2^ 14-17, July 8, 1974.

54.  Chi, C. T., and T. W. Hughes.  Technical and Economic
     Evaluation of Phthalic Anhydride Plant Air Pollution
     Control.  Monsanto Research Corporation.  Dayton.
     Preliminary Draft Report.  U.S. Environmental Protection
     Agency, Contract 68-02-1320, Task 25.  August 1976.

55.  Zimmer, J. C.  New Phthalic Anhydride Process.  Hydro-
     carbon Processing.  53:111-112, February 1974.

56.  Weyens, E.  Recover Maleic Anhydride.  Hydrocarbon
     Processing.  J33_: 132-134, November 1974.

57.  Wirth, F.  Recover MA from PA Scrubber Water.  Hydro-
     carbon Processing.  5_4_: 107-108, August 1975.

58.  Wu, W. H., and J. R. Maa.  Make Phthalates Direct.
     Hydrocarbon Processing.  53:117-118,  April 1974.

59.  Austin, G. T.  Industrially Significant Organic
     Chemicals, Part 8.  Chemical Engineering.  24:107,
     July 22, 1974.

60.  Chemical Origins and Markets, 4th Edition.  Menlo Park,
     Stanford Research Institute, Chemical Information
     Services, 1967.  99 p.

61.  Edwards, P-  Chemicals '75/Aliphatics.  Chemical
     Marketing Reporter.   207;40-42, January 6, 1975.

62.  Burke, D. P.  Forecast '75 Riding it Out.  Chemical
     Week.  116:17-24, January 8, 1975.

63.  Synthetic Organic Chemicals, U.S. Production and Sales.
     U.S. International Trade Commission (formerly U.S.
     Tariff Commission).   Washington.  1963-1974.
                              143

-------
64.  1972 Annual Statistical Report.   The Society of the
     Plastics Industry.   New York.  November 16,  1973.

65.  Anderson, E. V.  Growth Slows in Top 50 Chemials'  Out-
     put.  Chemical and Engineering News.  52:10-13,
     May 6, 1974.

66.  Anderson, E. V.  Recession Stifles Output of Top 50
     Chemicals.  Chemical and Engineering News.  53:30-33,
     May 5, 1975.

67.  Metric Practice Guide.  American Society for Testing
     and Materials.  Philadelphia.  ASTM Designation:
     E 380-74.  November 1974.  34 p.
                             144

-------
                                 TECHNICAL REPORT DATA
                          (Please read Instructions on the reverse before completing)
 1. REPORT NO.
 EPA-600/2-76-032d
2.
                            3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Source Assessment: Phthalic Anhydride (Air Emissions)
                            5. REPORT DATE
                             December 1976
                                                       6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                            8. PERFORMING ORGANIZATION REPORT NO.
 R. W. Serth and T.W. Hughes
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Monsanto Research Corporation
1515 Nicholas Road
Dayton, Ohio  45407
                            10. PROGRAM ELEMENT NO.
                            1AB015;  ROAP 21AXM-071
                            11. CONTRACT/GRANT NO.
                             68-02-1874
 12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC  27711
                            13. TYPE OF REPORT ANC
                            Final; 1/75-4/76
  NO PERIOD COVERED
                            14. SPONSORING AGENCY CODE
                             EPA-ORD
15.SUPPLEMENTARY NOTES Project officer for this report is Dale A.  Denny, Mail Drop 62,
919/549-8411 Ext. 2547.
16. ABSTRACT .
          The report gives results of an analysis of atmospheric  (air) emissions from
 ortho-xylene- and naphthalene-based phthalic anhydride manufacturing plants. Uncon-
 trolled and controlled emission factors are given for each species emitted to the
 atmosphere from each source within a typical plant, based on the latest data available.
 Emissions  data are used to calculate three factors designed to quantify the hazard
 potential of the emissions:  (1) source severity (the ratio of maximum mean ground-
 level concentration of a pollutant to the concentration which constitutes an incipient
 health hazard),  (2) the industry contribution to total atmospheric emissions of cri-
 teria pollutants, and (3) the population exposed to high contaminant levels from a
 representative plant.   Detailed process descriptions and flow sheets are presented
 for the BASF fixed-bed ortho-xylene process and the Badger-Sherwin-Williams fluid-
 bed naphthalene process.   Present and future aspects of pollution control technology
 in the industry are discussed, including a number of possible process modifications.
 Economic and production trends in the phthalic anhydride industry and in each of the
 industries that are major consumers of phthalic anhydride are analyzed.  Water-
 related emissions are  to be discussed in a future,  separate  report.
17.
                              KEY WORDS AND DOCUMENT ANALYSIS
                 DESCRIPTORS
               b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Phthalic Anhydride
Industrial Processes
Xylenes
Naphthalene
                Air Pollution Control
                Stationary Sources
                Source Assessment
                Ortho-xylene
13B
07C
13H
18. DISTRIBUTION STATEMENT

 Unlimited
               19. SECURITY CLASS (ThisRepon)
               Unclassified
21. NO. OF PAGES

     154
               20. SECURITY CLASS (This page)
               Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
                                         145

-------