PB-:  GENERAL  INDUSTRIAL
                                                                SURFACE  COATING
                                                                AP-42
                                                                Section  4.2.2.1
                                                                Reference  Number
                                                                      1
Controlling Pollution from the Manufacturing and- Coating of
Metal Products. 1. Metal Coating Air Pollution Control
(U.S.) Environmental Protection Agency, Cincinnati,  Ohio
Prepared for

Foster D. Snell, Inc., Cleveland,  Ohio
May 1977
                   U.S. DEPARTMENT OF COMMERCE
                National Technical Information Service

-------
                NOTICE





THIS DOCUMENT HAS BEEN REPRODUCED



FROM THE BEST COPY  FURNISHED US BY



THE  SPONSORING AGENCY.  ALTHOUGH IT



IS RECOGNIZED  THAT CERTAIN PORTIONS



ARE  ILLEGIBLE, IT IS'BEING  RELEASED



IN THE  INTEREST  OF MAKING  AVAILABLE



AS MUCH  INFORMATION AS POSSIBLE.

-------
EPA-625/3-77-009
                         CONTROLLING POLLUTION
                       FROM THE MANUFACTURING
                   & COATING OF METAL PRODUCTS
                                      METAL COATING
                             AIR POLLUTION CONTROL-I
    U.S. ENVIRONMENTAL PROTECTION AGENCY
    Environmental Research Information Center * Technology Transfer
                        MAY 1977

-------
              ACKNOWLEDGEMENTS
    This seminar publication contains materials prepared for the U.S. Envi-
ronmental  Protection  Agency Technology Transfer Program  and has been
presented at Technology Transfer  design seminars throughout the United
States,
    The technical information in this publication was  prepared by Burton J.
Sutker of Foster D. Snell, Inc., and Uday Potankar of JACA Corporation.
                              NOTICE
    The mention of trade names or commercial products in this publication
is for illustration purposes, and does  not constitute  endorsement or recom-
mendation for use by the U.S. Environmental Protection Agency.

-------
                             INTRODUCTION
    Recent findings of high levels of hydrocarbons in the nation's air have spurred
renewed activity by the U.S. Environmental Protection Agency (EPA) toward their
control.  One result will be more emphasis on reducing hydrocarbon emissions from
industrial activity.  EPA's Office of Air Quality Planning and Standards will shortly
publish guidelines to the states for control of these emissions in several industries,
including metal coating.  In the next 2 years, federal standards for new plant construc-
tion in these industries are also expected.

    In the metal painting and coating industry, most hydrocarbon emissions are trace-
able to the solvent in the coating  material, all of which eventually evaporates.  Our
purpose is to acquaint supervisory and management personnel in the industry with
methods of reducing the emission of organic solvents to the atmosphere and to help
them assess the costs.  We will be as practical as possible and will present a number
of realistic options.

    The logical sequence of steps toward achieving compliance suggests a division of
this publication into two parts.

    Part A is concerned with reducing and controlling hydrocarbon emissions at their
in-plant sources.  It includes background material on the nature of hydrocarbon emis-
sions and step-by-step information on plant surveys and emission control procedures.

    Part B details the techniques available for end-of-line treatment of these emis-
sions. Because these techniques often involve the use of heat energy, methods for re-
covery of this heat will also be described.

    This handbook is part of the  U.S. Environmental Protection Agency Technology
Transfer Seminar Series for the  Machinery and Mechanical Products Industry.  A
companion publication discusses  control of air pollution from metal cleaning processes,
and a third volume delineates water pollution control in the metal industry.

    This publication, like the others in the series, provides practical, realistic op-
tions , based on the current literature and on the experience and know-how of people
throughout the industry.
                                       111

-------
REDUCING HYDROCARBON EMISSIONS
THROUGH (JM-PLANT PROCESS CHANGE
               Part B
   TREATMENT OF HYDROCARBON
   EMISSIONS AND HEAT RECOVERY
              May 1977
       U.S. Environmental Protection Agency
          Technology Transfer Office

-------
                            CONTENTS

                                                                   Page

INTBODUCTION	   iii

PART A — REDUCING HYDROCARBON EMISSIONS THROUGH
    IN-PLANT PROCESS CHANGE   ..... 	    1

Chapter I -'PHYSIOCHEMICAL FACTORS AFFECTING  HYDROCARBON
    EMISSIONS IN COATING OPERATIONS	    2

    Definition of Organic Solvents	    2

    Evaporation Rates  	    3

    Atmospheric Concentrations	    3

    TLV and  LEL	    5

    Reactive  Hydrocarbons	    6

    Calculations for Determining Reactivity of a Coating  Formula  ......   10

Chapter II - PLANT OPERATING FACTORS AFFECTING
    HYDROCARBON EMISSIONS	   12
                   #
    Non-processing Factors Affecting Emissions	  .   12

    Processing Factors Affecting Emissions  ...............   ^

Chapter IE — PLANT  SURVEYS OF HYDROCARBON EMISSIONS  ......   26

    Obtaining the Regulatory Requirements	   26

    Determining Coating Operations To Be Regulated	   26

    Identifying Major  Emitters	   27

    Estimating Amount and Type of Emissions	   27

    Measuring Emission Levels	<	  .   27

    Planning  for Compliance	   28

-------
Chapter IV - EMISSION REDUCTION BY IK-PLANT PROCESS
    CHANGE: OPPORTUNITIES AND PROBLEMS  ............   29

    Emission Control Through Formula Changes   .  ,	   29

    Emission Control Through Process Changes   .  .  ,  .	,   45
SUMMARY
PART B —  TREATMENT OF HYDROCARBON EMISSIONS
    AND HEAT RECOVERY   	,	   50

Chapter I — DISPOSAL OF SOLVENT VAPORS ..............   51

    Combustion ".......................	   51

    Vapor Adsorption   ,  .	   81

Chapter n - HEAT  RECOVERY	   82

Chapter IH — COST OF COMBUSTION AND HEAT
    RECOVERY SYSTEMS	   92

SUMMARY	   96
                                   VI

-------
                                 FIGURES
Figure                                                                   Page
                                   PART A
   1     Typical Evaporation Rates	     4
   2     Comparison of TLV and LEL for Some Solvents	     6
   3     ARB Reactivity Classification of Organic Compounds  .......     8
   4     Solvent Emissions from Various Coating Formulas	    13
   5     Solvent and Resin Emissions from Typical Coating Formulas ....    14
   6     Emissions from Coated Metal Parts or Assemblies   	    16
   7     Potential Solvent Vapor Emissions from Coating Operations  ....    17
   8     Evaporation Rates of Various Formulas	    21
   9     Relative Costs of Coatings	    30
  10     Examples of Surface Coating and Added Thinner Formulas
         on an As-Purchased Basis Having Conforming Solvent Systems ...    31
  11     Energy Requirements for Comparable Operations	    33
  12     Solids vs. Viscosity for Caprolactone, Acrylic, and
         Polyester Polyols	    34
  13     Examples of Modern Formulas for High-Solids Systems	    35
                    *
  14     Relative Emissions of a Hypothetical Waterborne System
         Containing 20% Solvent and of a Conventional Solvent
         Base System	    36
  15     Comparison of the Amount of Organic Volatile Material
         Contained in High-Solids, Water-Soluble, and
         Conventional Paints	    37
  16     Heat Requirements for the Baking of Equivalent Solvent-Borne
         and Waterborne Coatings	    39
  17     Comparative Economics of High-Speed Curing Units   .......    42
  18     Comparative Costs of U.V. Curing and Infrared Curing	    43

                                   PART B
   I     Coupled Effects of Temperature and Time on Rate of
         Pollutant Oxidation	    53
   2     Typical Effect of Operating Temperature on Effectiveness of
         Thermal Afterburner for Destruction of Hydrocarbons and CO  ...    54
   3     Available Heats for Some Typical Fuels	    58
                                     Vll

-------
Figure                                                                    Page
   4      Generalized Available Heat Chart for All Fuels at Various
          Flue Gas Temperatures and
          Various  Excess Combustion Air .... ............  .    59
   5      Maxon Combustible Burner   .  . . ,   ..............    69
   6      Hirt Multijet Gas Burner   .......... ...  ......    70
   7      Afterburner Using a Discrete Burner   .............  ,    71
   8      Schematic of Catalytic Afterburner System    ...........    72
   9      Combustion Efficiency as a Function of Catalyst-Volume/
          Flow Ratio  .  , ..... ,  ......... .  ........    73
  10      Typical Temperature-Performance Curves for Various Molecular
          Species Being Oxidiaed Over Pt/Al2O3 Catalysts ,  .  .....  .  .    74
  11      Effect of Solvent. Concentration on Required
          Preheat Temperature  ........
  12      Typical Shell and Tube Heat Exchangers ...........  .  .   85
  13      Rotary Regenerative Heat Exchanger   ..............   86
  14      Heat Pipe   ............... . . ....... '  .  .   87
  15      Process Heat Recoverable from Afterburner   ..........   88
  16      Capital Cost of Incineration   .  ............  .....   93
                                                         «r
  17      Annual Variable Cost of Incineration   . .....  .  .......   94
Table                                                                     Page
   I      Combustion Constants	   57
  II      Enthalpies of Gases Expressed in
         Btu/scf of Gas	   65
 in      Combustion Characteristics of Natural Gas	   66
 IV      Combustion Data Based on 1 Pound of Fuel Oil	   67
  V      Comparison of Heat Recovery Techniques	   83
                                      Vlll

-------
             Part A

REDUCING HYDROCARBON EMISSIONS
THROUGH IN-PLANT PROCESS CHANGE

-------
                                   CHAPTER  I
                                           IN C
    The following topics are basic to the understanding and discussion of hydrocarbon
emissions and their control,  They include;

    *  Definition of Organic Solvents
    *  Evaporation Rates

    •  Atmospheric Concentrations

    *  TLV and LEL
    •  Reactive Hydrocarbons

    •  Calculations for Determining Reactivity of a Coating Formula


                        DEFINITION OF ORGANIC SOLVENTS
                                                           fS
    Volatile organic substances have been defined in the  Federal Register (CFR
52.1596, subsections (a) (i) and (k)) as follows:

        "Organic materials mean chemical compounds of carbon excluding
        carbon monoxide, carbon dioxide,  carbonic acid, metallic carbides,
        metallic carbonates and ammonium carbonates and having a vapor
        pressure of 0.02 pounds per square ineh absolute or greater at
        standard conditions, including but not limited to petroleum fractions,
        petrochemicals and solvents.

        For the purposes of this section, organic solvents include diluents
        and thirmers which are liquids at standard conditions and which are
        used as dissolvers, viscosity reducers, or cleaning agents."*

    Although this definition was evolved for a specific region,** it is EPA's most re-
cent designation of the hydrocarbons to be controlled in maintaining acceptable atmo-
spheric burdens.
 "FederalRegister, Volume 38 - No. 218. November 13. 1973, pp. 31398-31399.
F*For "New Jersey portions of the New Jersey. New York, Connecticut Interstate and Metropolitan Philadelphia
  Interstate Air Quality Control  Regions."

-------
                               EVAPORATION RATES

     The vapor pressure of a chemical compound, central to its evaporation rate, re-
lates only to its temperature.  It has been measured for most of the pure compounds;
tables of value are readily available for most normal plant circumstances. *

     There are several empirical relationships between volatility (or relative evapora-
tion rate) and vapor pressure. According to Gaynes,  "The simplest form for calcula-
ting relative evaporation rates is multiplying the molecular weights in question by the
vapor pressures." **

     For example, ethyl acetate has about the same vapor pressure as ethyl alcohol,
yet it evaporates twice as fast because  its molecular weight is twice that of the alcohol.
This comparison is also true for butyl alcohol and butyl  acetate or ethyl alcohol and
toluol.

     Tysall and Wheeler state that "the  rate at which a solvent evaporates from a film
is a technological measurement—combining the effect  of a number of basic physical
properties such as vapor pressure, latent heat of evaporation and density of solvent
vapor ,t

     Doolittle presents typical evaporation rates for a  series of solvent compounds by
comparing them to n-butyl acetate, which is arbitrarily  given a rate value of l.tt
Figure 1 shows that the rates range from 1 to 3,000.

     Finally, evaporation rate is influenced by air circulation over the surface of the
solvent coating; the higher the volume of air, the faster  the evaporation.
                         ATMOSPHERIC CONCENTRATIONS

    In discussing air pollution, we are naturally concerned with concentration levels
of pollutants.  There are three main methods for calculating concentrations of solvent
vapors:  partial pressure,  volume, and milligrams per cubic meter.

PARTIAL PRESSURE

    There is a general relationship between the amount of a solvent in the air and the
vapor pressure of this solvent.  At any temperature a solvent will continue to evaporate
until the air becomes saturated, much as water evaporates to cause humidity. At a
normal atmospheric pressure of 760mm of mercury, the mixture of air and solvent
vapor will behave as if each exerted a pressure the total of which would be  760 mm.
The solvent will exert its own vapor pressure, called partial pressure.
 *For instance, see Chemical Engineers'Handbook. Perry et al., New York:  McGraw-Hill.
**Gaynes, N. l.,MetalFinishing, Volume 74, No. 1, Jan., 1976, p. 29.
 tTysall and Wheeler, The Science of Surface Coating, New York: Van Nostrand, 1962.
ft Doolittle, The Technology of Solvent and Plasticizers, 1954.

-------
                                                    Relative Evaporation Rate
                                                   with n-Butyl Acetate at 1 .0
       Methyl chloride                                        27.5
                                            I
       Isopropyi alcohol                      I                  3.0

       n-Butyl  acetate                                         1 .0

       Xyiene                                                 0.63
       Dipentene
0.08
       di-Ethylene glyco! monerhyl ether       i                  0.01
Source: The Technology of Solvent and Plasticizers. Doolittle, 1954.
                             Figure 1, Typical Evaporation Rates


    In a fixed volume of airs enough solvent will evaporate to create that pressure.
For instance, at temperature T, if a solvent has a vapor pressure of 200mm of mer-
cury there will be in that mixture of air and solvent vapor enough solvent to represent

rrr of the total  volume.  This also means that in a mole of the mixture there will be
t bu

rrr mole of the solvent and rr-  mole of air.  If the solvent has a molecular weight of
/bu                        7bu

92 and. the air a molecular weight of 29, there will be 24.2 Ibis.  (r-rr x 921 of solvent
                                                             \760     /

and 21.41bs. ( r^r x 291 of air, for a total weight of 45.6 or a weight concentration of

53 percent solvent and 47 percent air.  This is useful  for calculating emissions meas-
ured in pounds of solvent per hour.

VOLUME

    Another way of expressing the concentration of solvent vapor is by volume, stated
either in percent by  volume or in parts per million (ppm).  In the case mentioned above,

the concentration of  the solvent vapors in percent by volume would be (•=—- x 100
                                                                  \760
which is 26.S percent.  Expressed in ppm, the concentration would be 263,000 ppm.

MILLIGRAMS PER CUBIC METER

    Finally,  there is away,  increasingly used, to exprass concentrations in milli-
grams per cubic meter (mg per m3).  Using the above example,  the concentration
would be 24.2gm (26.3%, x 92) of solvent in 22, 4 liters, which converts to 24,200mg

x 5— =  1.08 x 106  mg/m3.

-------
    The solvent used in all these calculations is highly volatile.  In the case of a less
volatile solvent, xyiene, the figures at 20°C would be approximately;

    *  Concentration by weight = 2.3%.
    •  Concentration by volume = 0,66% or 6,600ppm.
                           .  ,        1000            ,  -5
    *  Concentration by mg/m3 = 693 x  r^—j = 31,OOOmg/mJ.
                                      £^2* 4

    These calculations assume conditions  under which solvents can evaporate into fixed
volumes of static air and give the maximum concentrations under circumstances of
ideal equilibrium. In most industrial operations, air is moving so that equilibrium is
not achieved and actual concentrations are lower.
                                  TLV AND LEL

    There are two values of the air concentrations of a given solvent vapor that are of
considerable importance to industry: the threshold limit value (TLV) and lower explo-
sive limit (LEL).

    •  TLV relates to toxicity expressed in ppm and is an arbitrary value based on
       physiological considerations. It represents the conditions under which it is be-
       lieved that nearly all workers may be repeatedly exposed, day after day, with-
       out adverse effects.*
                        j>
    •  LEL, the lower explosive limit, represents a property of the vapor.  It is the
       lowest solvent concentration at which the mixture does not sustain combustion.
       For insurance and for other obvious reasons, it is industry practice to provide
       enough ventilation to maintain a solvent concentration well below this  limit.
       The usual value is set at 25 percent of the LEL.  Explosive limits are usually
       given in percent by volume; one percent  is equal to 10,000ppm.

    TLV, LEL, and 25-percent LEL values for some typical classes of solvents are
given in Figure 2. You will note mat TLV's are much lower than even the 25-percent
LEL.  The practical importance of this fact will be discussed later.

    In the above discussion of concentrations, the equilibrium concentration  of xyiene
by volume was shown to be 6,600ppm at 20°C.  As shown in Figure 2, the TLV is
lOOppm and the  LEL is 10, OOOppm. This means that about 60 times more air than is
necessary for evaporation has to be supplied to  comply with the TLV and about 2.5
times more to comply with the 25-percent LEL.

    In general,  TLV and LEL requirements demand much larger volumes of exhaust
air than are necessary from a strictly operational point of view.
*N. Irving Sax, Dangerous Properties of Industrial Materials, Fourth Ed., New York: Van Nostrand Reinhold
 Company, 1975,

-------

Toluene
Xyiene
Isopropyl olcohol
Methyl-ethyl-ketone
n-Bufyl acetate
Methylene chloride
TLV
200
100
40CS
200
150
500
LEL
13,000
10,000
25,000
18,100
17,000
None
25% LEL
3,300
2,500
6,300
4,500
4,300
None
Source:  N. Irving Sax, Dangerous Properties of Industrial Materials. Fourth Ed., New York: Von Nostrand
       Reinholci Company, 1975,

                Figure 2. Comparison of TLV and Lit for Some Solwents (ppm/woiumej
                             REACTIVE HYDROCARBONS

     Because of the national scope of coating operations, it is expected that a federal
policy will be proposed through the  Environmental Protection Agency.  Major consum-
ers of hydrocarbon-based coatings, such as the automotive,  coil, and can coating seg-
ments of the metal coating industry, will be a prime initial target for emission
guidelines.   This could lead to guidelines  for other high-volume repetitive coating op-
erations such as those for paper, textiles, wood, and adhesive laminations.  To date,
there are no federal guidelines for hydrocarbon emissions from coating operations.

     Meanwhile,  many states  and  other political subdivisions have either proposed or
actually enacted legislation to limit atmospheric contamination by hydrocarbon emis-
sions.  For instance,  California, particularly the Los Angeles basin area, has pro-
mulgated Rules 66, 102,  and  442, to be discussed shortly.

     Industry personnel should become familiar with the terminology in existing regula-
tions so they can determine whether the solvents they use are likely to be affected by
later guidelines.

     Solvents used in coating processes are classified according to their photochemical
reactivity.  Briefly,, photochemical reactivity, sometimes shortened to "reactivity,"
is "the tendency of an atmospheric system containing the organic compound in question
and nitrogen oxides to undergo, under the influence of ultraviolet radiation (sunlight)
and appropriate meteorological conditions, a series of  chemical reactions that result
in the various manifestations associated with photochemical air pollution.  These in-
clude eye irritation, vegetation damage and visibility reduction. "*
*Control Techniques for Hydrocarbon and Organic Solvent Emissions from Stationary Sources, AP.68, EPA.

-------
    As a result of these definitions and because of severe local smog conditions, a
special regulation was issued to cover use in the Los Angeles area of materials  con-
taining reactive hydrocarbons.  This is "Rule 66, " which has become a byword for
legislation on hydrocarbon emissions.  In part, it reads:

    "For the purposes of this rule, a photochemically reactive solvent is any solvent
    with an aggregate of more than 20 percent of its total volume composed of the
    chemical compounds classified below or which exceeds any of the following indi-
    vidual percentage composition limitations,  referred to the total volume of solvent:

    (1) A combination of hydrocarbons,  alcohols, aldehydes, esters, ethers, or
        ketones having an olefinic or cyclo-olefinic type of unsaturation:  5 percent;
    (2) A combination of aromatic compounds  with eight or more carbon atoms to the
        molecule except ethyl-benzene:  8 percent;
    (3) A combination of ethyl-benzene, ketones having branched hydrocarbon struc-
        tures, trichloroethylene,  or toluene: 20 percent."

    There has been considerable controversy,  however, about the facts on which Rule
66 was based and especially about its applicability to areas other than the Los Angeles
basin.

    The rule was eventually amended by  two other rules of the Southern California Air
Pollution Control District:

    1. Rule 102 changed the listing of solvents in Rule  66 by the following additions or
subtractions:

       Type (1) Solvents — perchloroethylene is excluded.
       Type (2) Solvents — methyl benzoate and phenyi acetate are excluded.

       Type (3) Solvents — no  change.

    To aid industry in determining if specific coating formulas were in compliance, an
expanded tabulation was issued  by the Air Resources Board (ARB) in Resolution  76-12
of February 19, 1976. It states:

       "Now,  Therefore, Be It Resolved, the Air Resources Board hereby adopts for
       the purposes of inventory and planning, the classification of organic compounds
       according to  photochemical reactivity  as set  forth in Appendix V attached
       hereto,."

    The Appendix V referred to in the resolution is presented here as Figure 3.

    2. Rule 442, the second rule amending Rule 66, imposed specific limitations on
emissions.  Note that these are not clear as to  definition of a coating line or entire
coating plant, althougtr subsection (b) uses the word "collectively. "

-------
       Class I
  (Low Reactivity)

  Cj-Cn Paraffins

  Acetylene

  Benzene

  Benzaldehyde

  Acetone

  Methanol

  Tert-alkyl alcohols

  Phenyl acetate

  Methyl benzoate

  Ethyl Amines

  Dimethyl formamide

  Perhalogenated
  Hydrocarbons

  Partially halogenated
  paraffins

  Phthalic Anhydride (2)
  Phthalic Acids
                (2)
  Acetonitrile' '

  Acetic Acid

  Aromatic Amines

  Hydroxyl Amines

  Naphthalene C1^

  Chlorobenzenes *  '
  Nitrobenzenes
                (1)
  Phenol
         (1)
       Class II
 (Moderate Reactivity)

 Mono- tert- alky 1-benzenes

Cyclic Ketones

Alkyl acetates

2-Nitropropane

£3+ Paraffins

Cycloparaffins

n-alkyl Ketones

N-melhyl pyrrolidone

N,N-dimethyl acetatnide

Alkyl Phenols(1)

Methyl phthalates f2'
     Class III
[High Reactivity)

All other aromatic hydro-
carbons
(including partially halo-
genated)

Aliphatic aldehydes

Branched alkyl Ketones

Cellosolve acetate

Unsaturated Ketones

Primary 6  secondary C2+
alcohols

Diacetone alcohol

Ethers

Cellosolves

Glycolsd)

C2+Alkyl  phthalates(2)

Other Esters W

Alcohol Amines(2)

Cg+ Organic acids  + di acid f

C%+ di acid anhydrides^2'

Formin »2^
 (Hexa methylene-tetramine)

Terpenic hydrocarbons

Olefin oxides ^2'
(1) Reactivity data are either non-existent or inconclusive, but conclusive date from similar compounds are available;
   therefore, rating is uncertain but reasonable.
(2) Reactivity data are uncertain.

Source; Communication from the State of California Air Resources Board, Appendix V, Resolution 76-12,

                      Figure 3.  ARB Reactivity Classification of Organic Compounds

-------
   "(a) A person shall not discharge organic materials into the atmosphere from
        equipment in which organic solvents or materials containing organic solvents
        are used, unless such emissions have been reduced by at least 85% or to the
        following:

        (1)  Organic materials that come into contact with flame or are baked, heat
             cured or heat polymerized, are limited to 1.4 kilograms (3.1 pounds)
             per hour not to exceed 6.5 kilograms (14.3 pounds) per day.

        (2)  Organic materials emitted into  the atmosphere  from the use of photo-
             chemically reactive solvents are limited to 3.6 kilograms (T.9 pounds)
             per hour,  not to exceed 18 kilograms (39.6 pounds) per day, except as
             provided in subsection (a) (1).   All organic materials emitted for a dry-
             ing period of 12 hours following their application shall be included in this
             limit.

        (3)  Organic materials emitted into the atmosphere from the use of non-
             photochemically  reactive solvents are  limited to 180 kilograms  (396
             pounds) per hour not to exceed  1350 kilograms  (2970 pounds) per day,
             except as provided in subsection (a) (1).  All organic materials emitted
             for a drying period of 12 hours  following their application shall be in-
             cluded in this limit.

    (b) Equipment designed for processing a continuous  web,  strip or wire which
        emit organic materials shall be collectively subject to the limitations stated
        in subsection (a).

    (c) Emissions of organic materials into the atmosphere required to be controlled
        by subsection (a) shall be reduced by:

        (1)  Incineration, provided that 90 percent or more of the carbon in the or-
             ganic material being incinerated is oxidized to non-organic materials,
             or

        (2)  Incineration, provided that the concentration of organic material follow-
             ing incineration is less than SOppm,  calculated as carbon and with no
             dilution, or

        (3)  Adsorption, or

        (4)  Processing in a manner determined by the Air Pollution Control Officer
             to be not less effective than (1)  or (3) above." *
•Communications from State of California Air Resources Board, July 26, 1976.

-------
     CALCULATIONS FOR DETERMINING REACTIVITY OF A COATING FORMULA
    Each coating formula containing solvents may have to be revised according to Rule
66 or Rule 102,  For determining if revision is necessary to achieve conformity, use
the calculations that follow.

    For evaluating solvents  m connection with Rule 66;

         Given:  A coating solvent with the following composition:
             Toluene

             Xylene

             Methyl isobut}7! ketone

             Isophorone

             Saturated aliphatic solvents

                                  Total
 15.0%

  2,0%

  7.0%

 10.0%

 66.0%

100.0% by volume
         Problem:  To determine if this solvent system is photocheinically reactive as
         defined by Rule 66.

         Solution; Tabulate the materials in the solvent that may be photocheinically
         reactive.  Columns (1), (2), and (3)  refer to the  photochemieally reactive
         groupings listed on page 8.*
Chemical Name
Toluene

Xylene

Methyl isobutyl
ketone
Isophorone
Aliphatic
solvents
Total
Limit
Classification
Name
Aromatic
hydrocarbon
Aromatic
hydrocarbon
Branched alkyl
ketone
Cyclic ketone
C, + paraffins



(1)
0 %

0

0

10.0
0

10.0%
5 . 0%
(2)
0 %

2.0

0

0
0

2.0%
8.0%
(3)
15.0%

0

7.0

0
o

22.0%
20.0%
*Rcaders may need a chemical handbook to relate these compounds—and those used in their plants—to the
 classifications in Columns (1 I. (2). and (3) of Rules 66 and 102,
                                          10

-------
    This system is photochemically reactive on three counts:

    *  The group (1) total exceeds the 5 percent allowance.

    •  The group (3) total exceeds the 20 percent allowance.

    *  The total of all groups (34 percent) exceeds the 20 percent total allowance.

    Utilizing the expanded ARB tables and definition of Rule 102, the positioning of
various solvents changes as follows:
Chemical Name
Toluene
Xylene
Methyl isobutyl ketone
Isophorone
Aliphatic solvents
Total
(1)





0.0%
(2)



10.0%
66.0%
76.0%
(3)
15.0%
2.0%
7.0%


24.0%
    The significant differences between Rule 66 and Rule 102 (plus the ARB tables) are
the movement of some solvents into higher reactivity categories and the inclusion of
aliphatic solvents as part of the reactivity calculation.
                                       11

-------
                                  CHAPTER  II
    The first part of this chapter will be a discussion of emissions from various
coating formulas and thicknesses.

    In the  second part, we will examine the steps of the coating process to see what
each contributes to the total emission picture.
                 NON-PROCESSING  FACTORS AFFECTING EMISSIONS

    The amount of organic emission is related to:

    *  Composition of the coating;

    •  Amount of coating applied;
    *  Post-application chemical changes; and
    •  Non-solvent contaminants.

COMPOSITION OF THE COATING

    As we have seen,  the amount of solvent emitted depends on the composition of the
coating material.

    Figure 4 shows in ver}' general terms the amounts of solvent emitted under the
same conditions from various coating compounds.  In general, low-solids lacquers
will produce more emissions than high-solids urethanes, and significantly more than
waterborne systems.

    Some typical values of solvent emissions in grams per square meter for different
coating systems are given in Figure 5.  The five enamels shown in the figure, which
contain from 29 to 5? percent solvent, will emit 52 to 79 grams per square meter.
Note that there will also be  some emissions from unreacted resin  components and de-
composition products that volatilize during baking.*
* Resin emissions generally come from thermosetting coatings that require polymerization or crosslinking of low
 molecular weight fractions. These components gradually build in molecular weight (with decreased volatility) as
 the exposure time and temperature increase. Hence, the emissions contain both solvent and polymer fractions.
                                        12

-------
                                            DISPERSION
                                             LACQUER
                                   NONAQUEOUS
                               DISPERSION LACQUER
                       NONAQUEOUS
                    DISPERSION ENAMEL
              HIGH SOLIDS URETHANE
             32 percent SOLIDS, WATER-BORNE SPRAY
           40 percent SOLIDS, WATER-BORNE SPRAY
           HIGH SOLIDS (80 percent SOLIDS)

           SO percent SOLIDS, WATER-BORNE SPRAY  (80% water/20% solvent)
           ELECTRODEPOSITION {PRIMING ONLY)
           POWDER COATING
             5     10    15    20     25    30     35    40     45     50    55

                POUNDS OF ORGANIC SOLVENT EMITTED PER GALLON OF SOLIDS APPLIED
Source:  Private Communications.

                Figure 4. Solvent Emissions from Various Coating Formulas
                                         13

-------
                               Solvent:
                                                                                      Primer                         Primer
                                                                                       Vinyl               Primer     Zinc
                                         Enamel,   Enamel,  Enamel,  Acrylic   Alkyd    Zinc     Primer     Zinc      Vinyl     Epoxy
                                         Air Dry   Baking  Dipping   Enamel  Enamel  Chromate  Surfncer  Chromate  Chromatc  Polymide   Vinyl
           Solvent [-"missions
Aliphatic hydrocarhons   Minetal spirits
Aromatic hydrocarbons   Toluene
Salural«d alcohols       n Propyl  alcohol
Ketones                 Methyl ethyl
                         ketone
Saturated nsters         n-Butyl acetate
Saturated estors         Ethylene glycol
                         mono methyl
                         el her
           Resin  Emissions
Alkyd
Vinyl
Acrylic
Epoxy
lire! ha ne
CeUulusic
Amino
Resin ester
Styrene
Phenolic
Hydrocarbon
 *At typical coating thicknesses.

 Source: Private Communications.
                                                                                                                        Vinyl
                                                                                                                       Acrylic Polyurethane
.5703 .4323 .2912
.5242 .4683 .3872
.2150 ,4343
.5473
.4029
Emission Factor g/tn *
6B.2 56 fl 2B.4
4.4 7.5 3.4
.4 13-1

1 7

56 . B 11
5.5 2.1 5
11
GO. 2 9
in. 2 9
12
.0
.3
.n
. n
.9
.4
16.7 59.3
!i . 6 5.0
1.0 B 4

10.3
3.1
13.
R

42

12
»
a

.2

.9

It. 4
17.5
22 9
11.0

Phthalic anhydride    3.8       4.6
Vinyl chloride
Methyl melhacrylate
Epichlorohydrin
Toluene diisocyanate
Methyl ethyl kotone
Elhanolamine
Maleic anhydride
StyretiR
Phenol
Turpentine

Total:               74.4      68.5
                                                                                                                                        .5811   .7588
                                                                                4.0
15.S
                                                             51.9
                                                                      78.8
                                                                              62.7
                                                                                       61.11
                                                                                                  52.2
                                                                                                            74 .5
                                                                                                                       76.3
                                        1 .5
                                                1 .2
                               5.0
                                                                                                                                                 93.
                                                                                                                                                          .5071
                                                         4.6
                                                                                                                                                           79.9
                                               Figure 5.  Solvent and Resin Emissions from Typical Coating Formulas

-------
     Emissions from thermoplastic coatings are almost totally solvent.  These poly-
mers are applied at high molecular weight,  so that little change occurs between appli-
cation and baking.

     Emissions can also be linked to the nature of the part to be finished.  Expected
levels of emissions per-unit-produced are given in Figure 6 for some segments of the
metal coating industry. Although total emissions vary significantly from a beverage
can to a washer,  due to the area covered, the net emissions per square meter are
about the  same.

     Obviously, two factors must be considered in  selecting coating formulas:  the type
of material to be coated and the characteristics of the desired finish.  In a later sec-
tion, we will discuss how recent developments in formulation have increased the choice
of formulas,  with particular significance for overall emission reduction.

     In addition, an economic choice sometimes has to be made in meeting emission
standards: whether to invest in emission  control equipment or switch to a more expen-
sive coating.  This will be discussed later.

AMOUNT  OF  COATING APPLIED

     The total emissions during a coating process are affected by:

     *  The area to be  coated;

     •  The thickness of the coat;

     •  The efficiency with which it is applied; and

     »  The percent of  solvent in the coating.

     Area and thickness can be controlled  to some  extent by design of the part and the
application technique.

     Application efficiency—for instance, avoiding  overspray, overcoating, and excess
widths and using a minimum of passes to achieve the thickness desired—can profoundly
influence total emissions and is directly controllable by operating personnel.

     Figure 7 gives a simple equation for predicting the total amount of solvent emis-
sion from any operation involving non-waterborne  coating.

POST-APPLICATION CHEMICAL CHANGES

     Evaporation rates  of individual solvents in the coating can vary to such an extent
that  if significant air drying occurs before baking—generally the case—the solvent mix-
ture remaining in the coating at the beginning of the baking operation is much richer in
the high-boiling solvents.  For example, a solvent mixture initially consisting of equal
parts of isopropyl alcohol and xylene will  tend to lose isopropyl alcohol faster than
xylene. Thus, after air drying, which would remove most of the isopropyl alcohol, the
solvent to be  removed in baking could be mostly, if not entirely, xylene.


                                        15

-------

Type Of
Hydror.ar hon
Reproduced from t^Pil
besl available copy.

Total Sizn Oiam
Aliphatic Arnm^tir S.-tluralnd SnlursU'd S;oU Kntonp-s l->lprs l-'thcrs Alfcyd Vinyl Aiiylit (U Hhfthp Ammo l'^r Unit M ,/|vi2
Metal Cans. Exr.I. Bnvrragn -CSfi 785 .106 2R57 ;'HP 0(13 1171 1155 .077 0115 047 'iOI7 niHVmn 77 fl
Hcvprapp Cans 95P
Ductwork 3112 n
Canopies <*nd A%"nln(*R 155R.O
RRfrrgerators $23 2
Screening ^ ^^
Fencing 14*2.0
Enamcli-d Plumbing Fixtures !J7.7

Drynrs 33R fl

Washnrs 75" »
Mela 1 Poors ICxnl . nnrnge 8-1 "
785 lf.fi 2 B57 >B9 0(12 .071 055 077 005 042 5 017 Ofi45/ran 77 R
255R.O Sin 7 BP5».n 875 ^ e 111 2"1?.1 Iflll K 21 111 !'• Kf> I iK Mi IK.ni fl J.ll/lonnp 70 7
l??H.n 2HH , R 4125. fl 117 '! 1 2fi 1 IS fi nci 3 n f, n 4 F:» 2 Rlnn « In5/!isn 77,fl
111 2 79 .7 120. n ID9 P HO 7 ',1 In ?.(t III4I .1 1 !ll/unit 105 25
1.13 .2-1 3.M an mil in .fie in .n»7 r, 7 nnn i/m2 7 !i
1217.0 257. n 1IIH.O 41; (1 1 11 111 n Rf. . fl 11 I! ft fl PS . II 777S 1) Illl/lonnr 70.7
20 7 15. B - 17 S \f..n 2.fl .'!ii! r- 1 17/uriU 1(11 -•
1 pox y
HIT 8 S2.2 45.2 74 ) 54 7 l 'i i ) ) i t.i r, PB7 7 r, . IB/unit 107. n
Kpoxy
RO 5 1R <1 13 7 i;', .4 40 B 14 H II 1 II 2 Ml 1 4 7r,/iinil HB 1
51.7 1(1.9 I7ri I 17 7 1 4 H 17 5 . 1 2 B 'i:tll i; 4 32/miil 7(! 5
Gultnrs 3112. n 255C.O 53fl.() SB5II (1 :t75.7 (11 217 .1 1RI1 B 21.1 IP.R 1 IB S Ifi.l.H.O 211/tonnp 7(1,7
NOTI'l: In Metric Tons.

Source:  l.nvironmcntiil Protection Agency, Soun-e Assessment:  rriaritizatiox of Air Palltttton from Industrial Surface Coating Operations.
        I-.PA-650/2-7 5-019-9.1 cbruary 1975,
                                                        Figure 6. Emissions from Coated Metal Parts or Assemblies

-------
              where:
    W =  0-0623 A n  (1-0.0.IP)
    yy __        P       C          *^




W   =   weight of solvent vapors in Ib.

A   =   area coated (sq . ft.)

n    =   dry mils.

P    =   percent solids by volume

f    =   efficiency factor (dimensionless)
         empirically determined (f 
-------
OTHER CONTAMINANTS

    Coating operations produce other contaminants besides the solvent vapors.  A ma-
jor source of these is the spraying operation. We will discuss over spray and ways of
reducing it later.  For now, the relevant point is that part of the aerosol from the
spray gun may dry before it reaches either the target article and/or whatever device
(baffle or water curtain) has been  set up to catch the wet overspray.

    This dry material, in the form of particulates, will be part of the vapor exhaust.
In general, the amounts thus generated  are not significant.  However, if adsorption
devices are used for controlling the vapor emissions, the particulates will have to be
filtered out because they tend to foui the activated carbon beds.
                   PROCESSING FACTORS AFFECTING EMISSIONS

     There are a number ot commonly used coating processes.  A brief survey of these
will be followed by discussion of emissions associated with the various processing steps.

COMMONLY USED COATING PROCESSES

     The basic processes used for coating include spraying, dip coating, flow coating,
coil coating, and masking.

Spraying

     Typical spraying operations are performed in a booth, with a draft fan to prevent
explosive or toxic  concentrations of solvent vapors.  Essentially, there are three
spraying techniques:  air atomization, airless atomization, and electrostatic.

     Air atomization uses its own air  source, which may be heated, filtered and/or
humidified, or treated in some other  fashion.  Airless spraying, on the other hand,
atomizes without air by forcing  the liquid material  through specially designed nozzles
under a pressure of 1, 000-2, 000psi.   On release to atmospheric pressure,  some
of the solvents in the surface coating  vaporize and join with the straight hydraulic
forces at the nozzle as atomizing agents. In general, with airless spraying less sol-
vent will be volatilized in the spray booth than with air spraying, meaning that more
solids may have to be removed later during air drying or baking.

     During airless atomization, total volatilization of a portion of the solvent will
probably occur, and emissions from this type of booth will be similar to the solvent
formula.  Air atomization, which is based on partial volatilization of the  solvent blend,
is likely to produce emissions high in low-boilers.

     Electrostatic  spraying projects charged coating particles into an electrostatic field
created by a potential difference of ^bout 100,000 volts between the articles sprayed
and spray grids 12 inches away.  The particles of wet paint from.the spray gun enter
this  field with the  same potential as the grids and are thus repelled by them and at-
tracted to the article being sprayed.


                                        18

-------
 Dip Coating

     In dip coating operations, the object is immersed for the required time in a tank
 of the coating. When the object is removed, excess coating drains back into the tank,
 either directly or via a drain ramp.

 Flow Coating

     Articles that cannot be dipped due to their buoyancy, such as pressure bottles, are
 subjected to flow coating.  Material is fed through overhead nozzles in a steady stream
 over the article.  Excess coating drains by gravity from the coated object and is re-
 circulated.  Removal of excess coating material and solvents is aided by jets of heated
 air.

 Coil  Coating

     Coil coating is a technique for applying finishes to long flat strips or coils of metal,
 on one side or both, by means of rollers similar to those in a printing press.  Three
 power-driven rollers are normally used.   One of the rollers is partially immersed in
 the coating material.  The coating is then transferred by direct contact to a second
 parallel roller.  The object to be coated is run between the second and third rollers
 and is coated by the second roller.

 Masking

     Masking is a technique for applying coatings where sharp, clean edges are needed;
 for instance, for lettering, stripping, and two-color finishes.  The areas to be left un-
 coated are masked with cloth, plastic sheeting, tape,  or a special mask derived by
 photography from an artwork pattern (silk screening).

     The coating may then be applied by stencil or rubber squeegee.  Masking is usually
 removed while the coating is still wet to prevent frayed edges and to ensure sharpness.

 EMISSIONS FROM THE VARIOUS PROCESSING STEPS

     Emissions of solvent vapor vary not only with the coating formula but also with
 each individual processing step.

Spraying

     Paint-spray  booths generally have one side open to the rest of the plant; ventilation
 of the booth is necessary to ensure both operator and plant safety. Normally, spray-
booth ventilation  velocities of from 100 to 150 feet per minute per square foot of booth
 opening are adequate for manual operations.*  Insurance standards require that the
* Air Pollution Engineering Manual. AP-40.2nd Ed., U.S. EPA, May 1973.
                                         19

-------
average velocity over the open face of the booth be not less than about 1.5 feet per sec-
ond.  All fumes should be vented through a fume hood instead of into the plant.

    Discharge from a paint-spray booth consists of paniculate matter and organic sol-
vent vapors.  The particulate matter consists of entrained coating material that did not
adhere to the object being painted or to the inside  surfaces of the booth.  The organic
vapors are generated from the evaporation of solvents, resins, diluent, or thinner.

    Generally,  emission levels are increased by  overspray; that is, material that
misses the surface to be coated. The table below gives overspray percentages for the
various spray techniques,

                      Overspray Percentages as a Function of
                      Spraying Methods and Surfaces Sprayed*
Method of Spraying
Air atomization
Airless
Electrostatic
Disc
Airless
Air atomized
Flat
Surfaces
(%)
50
20 to 25

5
20
25
Table Leg
Surface
{%)
85
90

5 to 10 ,
30
35
Bird Cage
Surface
(%)
90
90

5 to 10
30
35
*Air Pollution Engineering Manual, AP-40,2nd Ed., EPA, May 1973.

    Solvent concentrations in spray booth effluents vary from 100 to 200 ppm for man-
ual operations.  Solvent emissions from spray booth stacks vary with the extent of the
operation, from less than 1 pound to more than 3,000 pounds per day.  No definitive
data is available for automatic spray booths.

    Virtually all solvents evaporate in the course of the coating sequence, each at its
own rate. For measuring purposes, this evaporation is viewed in terms of "flash-off,"
defined as the quantity of  solvent evaporated under either ambient or forced  conditions
from the surface of a coated object during a specific time.  The graph in Figure 8
shows flash-off times for various coating types applied b" spraying  and  is useful for
determining potential emissions from different coating  systems. The total emission
load, however, is significantly affected by the size, shape, and number of pieces being
coated and other factors.
                                        20

-------
   100

    90

    80

 -  70
 e
 01
 I  80

    40

    30

    20

    10
1.  LACQUER: CLEAR, SEMI-GLOSS. FLAT, PI6MENTEO. PRIMERS, PUTTIES, SEALERS
   VINYL ORGANISOLS. STRIPPA8LES. SOLVATEO POLYESTERS
2.  SQLVA1 ED VINYL PLASTISOLS
3.  STAINS: SPIRIT, OIL
4.  VARNISH: CLEAR AND PIGMENTED
5.  ALKYDS, ACRYLICS, PQLYURETHANES
6.  EPOXIES
                      I
                          i  i  i i
                                                    I
                                                            I
                                                                I
                                                                       J_
                                                                                I
                  3  4  S  6 7 8910
                  minutes
         20
40
60
1 hr
                             2 hr   3 hr 4 hr 6 hr 8 hr  12hr16 hr
                                           TIME
Source: Air Pollution Engineering Manual. AP40, 2nd Ed., EPA, 1973.

                         Figure 8. Evaporation Rates of Various Formulas

     Solvent emissions, then,  vary with types of spray operations. However, partieu-
late matter, the other type of emission, can be effectively removed {50 - 98 percent)
by techniques to control the particulate emissions.  These include:

     »  Dry Baffle.  In this method, the wet overspray collects on large panels called
       baffle plates, which catch 50 - 90 percent of the particulates produced by spray
       ing a high-solids enamel.  With low-solids lacquers containing highly volatile
       solvents, efficiencies may be much lower due to the rapid drying of the lacquer
       and poor adhesion of dry particles to the baffle.
     *  Paint Arrester.  Filter pads used in this method can remove up to 98 percent
       of paint particulates.  Filtering velocities should be less than 1.3 m/sec.

     *  Water Wash.  Water curtains and sprays are 95 percent effective in removing
       paint particulates.  A water circulation rate of 1 - 5 liters per cubic meter of
       exhaust air is usually recommended.  Surfactants may be added to the water
       aid in removing paint from  the circulating tank.

     In order of effectiveness, the paint arrester would be considered the best technique
for removing particulates when downstream solvent vapor processes such as catalytic
or other afterburners, heat exchangers, or carbon absorption beds are used. Water
washing  to remove particulates would be a  second choice, assuming  that the  solvent
vapor processes can tolerate some water in the vapor stream.  Baffle plates would be
considered the third and least effective method, although by far the cheapest.
                                          21

-------
     Concentrations of water-soluble solvent vapors are sometimes reduced, particu-
larly in non-re circulating sprays.  However, this creates a water  contamination prob-
lem necessitating treatment.  Solvents tend to increase the BOD (biochemical oxygen
demand) level of wastewater to a considerable degree (several hundred ppm).

     The following table shows the effectiveness of a water curtain  in reducing solvent
vapors in a sample  spraying operation:

               Emissions from Automatic Airless Spraying Operations
                          (Alkyd Coating with Xylol Solvent)

Operation

Spray (no water curtain)
Spray (water curtain on)
Emissions in Ib./hr,*

Particulate
0.5
0

Organic
4.0
3.5
These emissions total about 60 percent of the organic emissions from this particular operation. Typically, spraying
 accounts for 40 - 60 percent of the total emissions from a coating operation.
Source:  Foster D. Snell, Inc.

     Note that in the case above the water spray reduced organic vapors discharged to
the atmosphere by about 10 percent.  The contaminated water was collected and the
xylene recovered by  separation, with the balance discarded. "With highly soluble sol-
vents, for instance methyl-ethyl-ke tone, distillation may be necessary to recover the
solvent and minimize sewer disposal.

     Although air pollution was significantly reduced in this case, the disposal of  sol-
vent or parti culate-laden water to the sewer had to be carefully monitored to keep it
within water pollution guidelines.  It is important,  of course, to avoid  substituting one
set of pollution problems for another.

     Flash-off occurring after the spray operation but before baking is treated later in
this book as a separate category of emissions; Rule 66, however, includes pre-baking
emissions as part of spraying.

Other Application Techniques

     Emissions from other application techniques such as flow coating, dip coating, or
coil coating differ from spray coating emissions to the extent that these methods  re-
quire less coating material.  However, the expected solvent emission load from these
techniques can vary widely.

     In fact, flow coating may not be much better from an emission standpoint than
spray coating.   For flow coating, the proper percentage of solids and correct viscosity
must be  maintained.  Further, so much solvent is  lost during recirculation and air
                                        22

-------
    In general, electrical heating costs more than direct-firec1 go.;;, but total emission
loads are reduced.
    Baking or curing ovens can produce a variety of pollutants in addition to pure
"emissions" from the coating including (a) smoke and other products of incomplete
combustion resulting from improper operation of a gas- or oil-fired combustion
heating system, which can interfere with stack sampling procedures by fouling test
elements; and (b) aerosols arising from the partial oxidation of organic solvents ex-
posed to flame and/or high temperatures and from chemical  reactions that occur in the
resins (these can be deposited on heat exchangers, adsorption beds, and related hard-
ware-,  reducing their effectiveness).

    Emissions from ovens, therefore, vary significantly with the oven type (batch or
continuous), method of heating, condition of the part before it enters the oven (pre-
dried), and oven-operating parameters such as the allowable LEL,

Emissions from the Overall Coating Process

    In most coating operations ,40-80 percent of the solvent evaporates at the time
of application and/or during subsequent air drying.   The remaining 20 - 60 percent
evaporates in the oven.

    The table below provides an overview of this chapter and gives general emission
ranges as a percentage of the total emission load from typical coating operations.

               Percent Of Total Emissions from Various Coating Steps

Coating Method

Spray Coat
Flow Coat
Dip Coat
Roller Coat
Coating Step


Application
30-50
30-50
5-10
0-5

Pre/Air Dry
10-30
20-40
10-30
10-20

Bake
20-40
10-30
50-70
60-80
Source: Foster D. Snell, Inc.
    In a specific example, 30 percent of the emissions occurred during the spray
process itself and another 8 percent occurred in the conveyor between the spray booth
and the continuous curing oven.

-------
                                CHAPTER III

   PLANT                OF  HYDROCARBON

    An important early step in controlling emissions is to determine the volumes gen-
erated and their sources. Plant operators need to know which operations are most
responsible for solvent pollution.   Identifying emissions from each process is essential
to developing a central plan for complying in a cost-effective and practical manner,

    The plant survey gives this information and provides a basis for determining if the
plant is in violation of regulations and to what degree.  To carry out an effective plant.
survey the following steps are necessary:

    * Obtain the latest regulatory requirements;

    « Determine which coating operations are affected by the regulations;

    * Determine which coating operations are major emitters;

    * Estimate  the emissions from the sources  identified;

    * Measure the level of emissions; and

    * Develop a plan to minimize emissions and improve the plant's compliance
      position,


                 OBTAINING THE REGULATORY REQUIREMENTS

    Most cities and states have Air Pollution or Air Control Offices.  Contact the one
in your area for the latest regulations that may affect the operation of your plant.  Dis-
trict offices of the U.S. Environmental Protection Agency should also be asked for any
relevant information.

    There is always a tendency to let sleeping dogs lie and avoid involving local regu-
latory agencies if at all possible.   It is better,  however, to be aware of existing and
potential regulations and guidelines as they are promulgated.  One reason is that con-
struction and operating permits are required for  any equipment causing emissions, and
states keep records of these in order to later implement air pollution control plans.


             DETERMINING COATING OPERATIONS TO BE REGULATED

    Since emission standards  vary from one area to another, a coating line may be in
compliance in one state and  in violation in another. Opportunity may thus exist, over
the short term, for a company to  increase production in a plant bound by  less stringent
                                      26

-------
 blow-off  of excess  coating that flow  coating  is often done in a "tunnel" to keep the
 solvent-laden air in a fixed area.  The result is that a well-run flow coating operation
 using 60, 000 gallons of coating per year may use as much as 54,000 gallons of  makeup
 solvent to compensate for "tunnel solvent" losses.   This is much more wasteful than
 an air-atomized spray operation with  50 percent over spray.

     Dip coating solvent losses are generally under  10 percent, depending on time  of
 year and temperature in the plant.  This usually represents much less solvent loss
 than that occurring with spraying or flow coating and does not normally require much
 makeup solvent.

     From the standpoint of overall emissions, the  single most efficient coating method
 is  roller or coil coating, a process in which extraneous evaporation is practically neg-
 ligible, since all coating supplied to the coating head is placed onto the web to be coated.

 Pre-Drying Processes

     Enough solvent must evaporate before the coated part enters the finishing or curing
 oven to avoid bubbling, uneven coating thickness, and other adverse effects.
jvith_solvent evaporation, thp>-pfe-drymg protife&s allows time for The coating to level
 itself if it has been unevenly applied.  The skilled coating formulator can  often  vary
 solvent balances to minimize these problems, as well as to reduce  emissions from the
 pre-drying operation.

     Pre-drying is usually carried out on conveyors, which are often open to the atmo-
 sphere.   As will be discussed later, it may be advantageous to enclose these conveyors
 to  maintain the highest permissible vapor  concentration in the air surrounding the dry-
 ing parts. This allows a gentler drying of the coating to help prevent blisters or bub-
 bles in the curing oven.   Care must be taken,  however, to ensure that the atmosphere
 in  the oven is in keeping with LEL determinations.

     Emissions from pre-dryers will, in general, contain higher  concentrations of the
 low-boiling components of the solvent blend.

 Ovens

     The last step in  coating operations is  the final conditioning of the coating.  While
 certain coatings can  be totally air dried, this is usually too slow  for industrial  proc-
 esses.  In general, heat must be applied to speed the curing rate.

     A distinction can be made between drying and baking.  Drying generally refers to
 removal of volatiles  such as solvents. Baking is the process by which a coating cures
 or  otherwise changes to develop its film integrity.  However,  this distinction has less
 effect on  emissions than the methods used and the type of oven.

     There are two basic types of ovens:  continuous and batch.

     From an  emission standpoint, the difference is important only  insofar as the at-
 mosphere of a batch  oven is easier to control than that of a continuous oven.  However,


                                         23

-------
solvent evolution in a batch oven is a function of time and temperature, meaning the
coated part generally reaches the temperature required for baking the finish, making
subsequent handling difficult.

    In a continuous oven, the evolution of solvent vapors varies in different zones of
the ovsn.  This may enable more control, depending on the configuration of the oven's
exhaust system. In general, emissions in continuous ovens are more diluted than
those in batch ovens, reducing problems with LELs.  However, this dilution can make
emission control in exhaust  gases from continuous ovens more difficult and expensive.

    Ovens also differ in the way they provide heat.   Oven design should allow for:

    •  Sufficient time before contact with heat for the coated surface to level and for
       highly volatile solvents to evaporate slowly, inhibiting bubble formation;

    *  An initial low-temperature zone for continued slow evaporation of solvents, to
       further inhibit bubbles;

    *  Sufficient time and temperature for a full cure of the coating;

    *  Termination of the heating process before the coating is damaged;

    *  A cool-down zone to  set the coating and enable handling;

    *  Removal of emissions to prevent their interference with the curing process;
       and

    *  Enough air flow  to keep the atmosphere at approximately* 25 percent of LEL,
       well below the explosive limit,  to be maintained by control of coating formula-
       tion, air flow rate, and other variables.

    Along with the basic design of a curing oven, a choice of heat  source must be
made. This may be dictated by both the fuel or energy available and the emissions
expected.  Types of oven heating include:

    *  Direct-fired gas heat, in which the products of combustion combine directly
       with the process air. Oven burners may use either fresh makeup air or re-
       circulated oven gases containing evaporated solvents and other volatiles.
       Flame contact with recirculated gases may cause molecular cracking or con-
       version,  which may render the  effluent gases photochemically active.

    *  Indirect-fired gas heat, in which combustion products pass through one side of
       a heat exchanger and discharge directly into the atmosphere. Process air,
       heated before being circulated to the oven, passes through the other side.

    •  Electrical heat,  in which fresh  makeup air or oven gases are passed over elec-
       trical resistance or infrared heaters.  This is  similar to direct gas-fired heat,
       but it  eliminates combustion products.   However,  some solvent modification
       can result from  contact with the heating elements when resistance heaters are
       used.
                                        24

-------
emission standards,  allowing time for bringing all coating lines into compliance
without a loss in production.
                          IDENTIFYING MAJOR EMITTERS

     Coating operations should be assessed in terms of their overall contribution to total
plant emissions.  Large volumes of solvents used to clean applicator rolls contribute to
solvent emissions,  yet are not generally included in coating-line solvent calculations.
Spray booths used intensively but for short periods of  time must be considered.  "Tun-
nel losses" from flow coating lines contribute significantly to emissions.  A noncoating
operation, such as panel decreasing, may be an important emitter.  Finally, if a plant
makes its own coating, this operation can also be a major emitter.
                  ESTIMATING AMOUNT AND TYPE OF EMISSIONS

    Once the plant has determined the major sources of emissions, an overall tabula-
tion should be made of the amounts and types. This tabulation must include solvents
used for makeup, dilution, and cleaning.

    Coating suppliers should be contacted to find out the percentage of solids and types
of solvents in their products.  This also serves notice to the supplier that the plant is
interested in compliance-type solvents.

    Ideally, each article»would be coated and then weighed immediately after both air
drying and baking to determine how much weight loss (emission) takes place at each step.
Based on these weights, and on the temperatures of drying and baking and the formula-
tion supplied by the coating manufacturer, an estimate could be made of the type of
emissions from each  stage of the  process.   Obviously, this would not be practical for
auto bodies, refrigerator paneling,  and other large items.  Sample coupons or small
panels might be interjected into the coating line to obtain the information for large
pieces, however.

    As a first approximation; the daily consumption of coating multiplied by the per-
cent of solvent would produce a total solvent emission load.  This total load would then
be factored according to the breakdown in the table on page 26, presented to show per-
centage of emissions from individual process steps.  As stack testing is expensive,
some states accept the results of  such material-balance calculations.
                          MEASURING EMISSION LEVELS

    The only reliable method for determining actual emissions is to measure them in
the effluent streams.  The major effluent stream for gaseous emissions is the stack,
which transports the final emissions after the stream has passed through paint arres-
tors, water wash towers, adsorption devices, catalytic afterburners,  etc.
                                        27

-------
    However, the validity of this method, relied on more and more by regulatory
agencies, is impaired because of the following factors:

    »  Measurements are based on volumetric quantities, which are significantly af-
       fected by temperature;

    »  Many of the analytical techniques common^ used do not give a real value for
       the amount of material in a given volume of gas, and empirical factors have to
       be applied;
    •  Variations in air flow and/or concentrations are difficult to compensate for
       with current equipment; and
    «  Some of the emissions may be compounds for which no standard analyses are
       available,

    Stack sampling results are also affected by the point at which the sample is taken.
Although continuous operations would tend to produce a uniform level of emissions,
batch operations can produce constantly changing emission loads.  This means that for
a total picture of a given plant's operation, continuous monitoring is probably required.

    A further problem with  stack sampling is that,  in general, emissions from a plant
or coating  line are discharged through more than one stack.  Therefore,  each has to
be monitored, unless the exhaust can be combined before sampling.
                           PLANNING FOR COMPLIANCE

    Once it has been determined that a certain coating operation is the major emitter,
steps should be taken to reduce its emission load by formula changes, process modifi-
cations,  or other means.

    This should be followed by effective policing to ensure that the changes are, in
fact, producing the desired emission reduction.  The second major emitter should then
be approached in the same manner.

    In any comprehensive survey and action program,  the services of outside experts
may be worthwhile.  Experienced consultants have an up-to-date awareness of current
regulatory thinking, without preconceived biases as to how the regulations should be
approached or applied.  They have access to the latest technology in stack sampling
procedures, which can shorten the training period for plant personnel.   Finally, con-
sultants, using the plant's stack analyses and their familiarity with the regulations,
can advise plant managers how compliance may best br achieved.
                                       28

-------
                              CHAPTER  IV

     EMISSION  REDUCTION  BY  IN-PLANT  PROCESS
       CHANGE:  OPPORTUNITIES AND  PROBLEMS


    Earlier in this publication, the various approaches to emission reduction were
broadly presented.  In this chapter, we will discuss technical and economic aspects of
formula changes and the potential impact of process changes.


               EMISSION CONTROL THROUGH FORMULA CHANGES

    The problem associated with formula changes can be both technical and economic.
Before discussing these problems, we assume that experimentation has been or will be
performed to ensure that the new coating meets predetermined specifications, that ad-
equate supplies of the coating are available, and that plant personnel are fully trained
in its application.  Finally, the revised coating must be checked at the outset against
internal cost standards, a point illustrated by the sharply varying costs of the polymer
systems in Figure 9.  The data, although 8 years old, also illustrates the wide variety
of coating systems available.  Note that silicone and fluorochemical polymers are still
the most expensive.
                     *
    The main ways of varying formulas are discussed below, in terms of both advan-
tages and problems.

SOLVENT CHANGES

    As a result of regulations affecting the use of photochemically reactive solvents,
practically all the conventional formulas are now available with "conforming" solvents.
This means that the new formulations meet the requirements and limitations of old
Rule 66, discussed earlier.

    Figure 10 shows some of the types of systems that meet the requirements and the
compositions of their solvents.

    The solvent-changing approach, however, has two main limitations: (a) emissions
of non-photochemically reactive solvents are still limited by Rule 442, discussed
earlier, to 396 pounds per hour and a maximum of 2,970 pounds per day; and (b) re-
formulations generally result in higher costs. For example, a 100-percent xylene
thinner costs about 60  cents per gallon. The cost of the complying substitute formula
in Figure 10 would be 90 cents per gallon.
                                    29

-------
             Dominant resin-type in coating
  Re I cost/ft2 of
a 2-mil thickness
   of coating
       Oxidizing aikyd


       Oxidizing alkyd and meiamine and/or urea


       non-Oxidizing alkyd and meiamine


       non-Oxidizing alkyd and urea


       Vinyl chloride-acetate copolymer


       Acrylic-type copolymers


       Styrenated aikyds  (oxidizing)
       Phenolic
       Epoxy


       Epoxy and meiamine


       Meiamine and ethylcellulose


       Polyurethane and alkyd
       Silicone
       Silicone and alkyd


       Allyl ester copolymers


       Polyamide  (nylon) 10 mils, flame spray


       Polyterrafluoroethylene (Teflon) flame spray


       Poly (chlorofluoroethylene) (Kel-F)
      1.00

      1.30

      1 .50

      1.35

      1.50

      4.00

      1.10

      1.70

      2.00

      1.50

      1.50

      1.60

     10.00

      7.00

      6.00

      5.00

     13.00

     11.00
Source:  Kirk Othmer, Encyclopedia of Chemical Technology, 2nd Ed., Volume 5. Interscience. 1968.

                              Figure 9. Relative Costs of Coatings
                                          30

-------
Type of surface
coating
Enamel, air dry
Enamel, baking
Enamel, dipping
Acrylic enamel
Alkyd enamel
Primer lurfacer
Primer, epoxy
Primer, zinc
chr ornate
Primer, vinyl zinc
chromate
Epoxy -polyamide
Varnish, baking
Lacquer, (praying
Lacquer, hot ipray
Lacquer, acrylic
Vinyl, roller coat
Vinyl
Vinyl acrylic
Polyur ethane
Stain
Glaze
Wash coat
Sealer
Toluene replace-
ment thinner
Xylene replacement
thinner
Weight,
Ib/gal
7.6 i-
9. 1
9.9
8.9
8.0
9.4
10.5
10. 3

8.4

10.5
6.6
7.9
8. 4 ,
8.4
7.7
8.9
7.5
9.3
7. 3
7.8
7. 1
7.0
6.7

6.5

Composition of surface coatings, % vol
Nonvolatile
portion
- 39.6 i^-
42.8
59.0
30.3
47.2
49.0
57.2
37.8

34. 0

34.7
35. 3
26.1
16.5
38.2
12
22.00
15.2
31.7
21.6
40. 9
12.4
11.7




Hydrocarbon
Aliphatic
saturated
93.5
82.1
58.2

92.5
18. 0
44. 8
80. 0

17.5



7.0
16.4
1
-------
INCREASING SOLIDS IN  EXISTING COATING  FORMULATIONS
    An obvious step in reducing solvent emissions is to increase the solids content of
existing coating systems.

Advantages

    In addition to reduced solvent emissions, particularly during application and air
drying,  the benefits include:

    «  Reduced •inventory  space for drums.  Drums of solvent-based coating typically
       weigh 400  pounds.  The  following chart shows the effect of reducing solids in a
       coating formula in  a plant that consumes 1.200 dry pounds of coating per day.
Wet V.' eight
Per Drum
400
400
j
> Solids
1
! 30
1 60
1
	 "' " 	 '" 	 -J.™«.n«— in _ _ _ ._
Dry Weight
Per Drum
120
240
Drums Per Day
10
5
       Thus, a 100-percent increase in solids made possible a 50-percent reduction in
       drum storage area.

    *  Reduced drum handling by operators.  Increased solids per drum would also
       reduce the number of drum changes at the coater, freeing operators for other
       tasks.
    »  Reduced energy for  removing solvents.  Changing from a 30- to 60-percent
       solids system reduces by almost half the total  solvent load that must be re-
       moved.  Normally,  however, to achieve such a high percent of solids more
       polar or higher-potency solvents must be used.  These would typically have
       slightly higher heats of vaporization than hydrocarbon solvents.  Using typical
       values, we see the effect of a change in solids  on heat required to remove the
       solvent:
f/t Solids by
Weight
30
60
':c Solvent by
Weight
70
40
Avg. Heat of
Vaporization
of Solvent
Btu/lb .
160
200
Heat Required to
Volatilize Solvent
from l,2001bs.
of Dry Coating
448,000
160.000
       The chang,- h
       Addition;}! fia
     ea in a potential energy savings of almost 300,000 Btu.
•n entTg>. savings are given in Figure 11.
                                        32

-------
Operation
Spray booth
Heat metal
Other heat losses
Oven exhaust
Total Btu required
for processing
Solvent Base
Thermoset Acrylic
(Baking temper-
ature - 350°F)
233,280
466,100
231,840
879,984
1,811,204
High Solids
Polyester
(Baking temper-
ature - 350° F)
233,280
466,100
231,840
94,651
1,025,871
80% Solids
Urethane
(Baking temper-
ature - 180°F)
233,280
183,110
91,080
37,184
544,654
NOTE: For these calculations it was assumed that the average yearly temperature was 52° F and that .018 Btu will
      raise one cubic foot of air 1°F at 100% efficiency.
Source: Modem Paint and Coatings, March 1975.

              Figure 11. Energy Requirements for Comparable Operations (Btu per hour)


    *  Increased potential for compliance with emission guidelines.  The lower the
       emissions from any part of the coating operation, the more likely that the plant
       will be in compliance with emission restrictions. Care must be used in making
       formula  changes, however, to use solvents with emissions that are less photo-
       chemically active.

    *  Reduced freight costs. Freight costs can easily be 2 cents per pound of gross
       weight, with empty drums themselves weighing about 50 pounds.  In the exam-
       ple below, the freight cost for 1,200 pounds of dry coating would be reduced as
       follows by a 100-percent increase in solids.
Coating
Solids
30%
60%
Pounds of
Coating Purchased
4,000
2,000
Total
Drum Wt.
400
250
Total Wt.
4,400
2,250
Freight Costs
$88
$45
Problems
    There are, however, certain drawbacks to high-solids systems,  including:

    »  Higher viscosity of the coating system.  As solids are increased, so is the vis-
       cosity of the formula.  Typical increases in viscosities as a function of the
       solid  content for prepolymer coatings are given in Figure 12.  Higher applica-
       tion viscosity may be handled by either equipment or operational changes.  An
       increased coating temperature, for example, may reduce viscosities enough so
       that the higher-solids  system can be run on the same  equipment.
                                        33

-------
                1000 -
                 800 -
                 600
                 400
                 200
                            Acrylic and Polyester Polyols
                                               PCP-Q6QQX
                                                     PCP-0601X
                                  PCP-0300
                   40
SO        60        70       80

       Polyol Solids. Weight Per Cent
90
Source: Modern Paint and Coatings, March 1i75

              Figure 12. Solids »s. Viscosity for Caprolaetone, Acrylic, and Polyester Polyols

    *  Reduced storage stability.  The higher the percentage of solids, the harder it
       becomes to maintain a stable system.  Skinning-over becomes more of a prob-
       lem with higher solids, with redispersal more difficult. The tendency to thicken
       or gel with time can often be counteracted by additives, but these may have del-
       eterious effects on other coating properties,

    *  Less latitude with in-plant formula modifications.  Because of the instability of
       high-solids systems described above, it is usually difficult to modify  them
       in-plant.

    Some typical formulas for high solid coatings are presented in Figure IS.

SWITCHING TO WATERBORNE SYSTEMS

    Use of water-based coating  systems is still a further choice of formula variations
for emission control.

Advantages

    Differences between emissions from waterborne systems and  solvent-based sys-
tems are shown in Figure  14.  For instance, at 30-percent coating solids, a  waterborne
system containing 20 percent solvent and 80 percent water would have one-quarter of
the solvent emissions of a 100-percent solvent system.
                                          34

-------
                                 High-Solich Cooting (Whire» (80°.. Solids)
                                               Ui ethane
                         Materials
   Component I
      Mulfron R-22J1
      2-Ethyl-l,3 hexanedtal2
      R-966, TiO23
      Modaflow,  10% in ethylglycal acetate4
      FC 430, 10% in ethylglycal acetate5
      EAB 381-1/10, 10% in ethyiglycal acetate8
      Dibutyltin dilayrafe, 10% in ethylglycal acetate7
      Ethylglycal acetate
   Component II
      Oesmodur N-1001
                 Total Weight

                      94.5
                      94.5
                     280.0
                      10.4
                      15.6
                      10.4
                       0.5
                     163.1

                     331.0
                                                                   ,000.0
                                     High-Solids Acrylic Hard fnamel
                               Materials
   Disperse on roller mill
         Titanium dioxide
         Experimental Resin QR-542 (80% in Ektasolve EE acetate)
   Letdown
         Mill paste (above)
         Experimental Resin QR-542 (80% in Ektasolve EE acetate)
         Cymel 301
         p-TSA (30% in sopropanol)
         n-Butyl acetate
         n-Butonol
                         Formulation  Constants
   Solids content
      Titanium dioxide (45%)
      Binder (55%)
         Experimental Resin QR-542 (70%)
         Cymel 301  (30%)
   Volatile* content
   Catalyst, p-TSA (on binder)
   Spray viscosity, *4 Ford cup (sec) 35
       Solids Weight

            94.5
            94.5
           280.0


             1.0
  j         33U)
  I         801,0

 P'.iiK By Woigl.l

      60.0
      40.0

     100.0
      24.2
      22.0
       0.5
      14.4
      11.9
     173.0
Percent of Formula
      77
                                      23
                                       0.2
1.  Mobay Chemical Corp.
2,  Union Carbide Corp.
3.  E, I. duPont de Nemours & Co.
4,  Monsanto Co.
5.  3M Co.
6.  Eastman Chemical Products. Inc.
7.  M & T Chemicals, Inc.
Source: Modem Paint and Coatings, March 1975.
                       Figure 13. Examples of Modern Formulas for High-Solids Systems
                                                   35

-------
                                High Solids Coaling (100% solvent]
                                      tOO-%Solids
                                   Water Borne Coating (20% solvent)
                                          100- %Solids
                            10    20    30   40    50   60   70
                               Application Solids, Per Cent by Weight
Source:  Modern Paint and Coatings, March 1975

         Figure 14. Relative Emissions of a Hypothetical Waterborne System Containing 20% Solvent
                           and of a Conventional Solvent Base System
     Figure 15 further illustrates (he reduction in emissions from the substitution of
walerborne coatings for conventional or hig'b-solids systems.

     Additional advantages of switching to a waterborne system include:

     *  Reduction of flammabiljty levels.  While many waterborne formulations include
       "co-solvents," these often evaporate before heat treatment, considerably re-
       ducing problems in the ovens.  Much lower dilutions are required due to the
       lower percentage of  solvent and also to the "quenching" effects of the water
       vapor.

     *  Increase in usable polymers,  In solvent-based systems, relatively few mono-
       mers or prepolymers can be used because of solubility, viscosity, and related
       factors.  In particular, the molecular weights  are severely  restricted.  This
       affects the ultimate properties of the coating,  to waterborne coatings, the
       choice of monomers and/or prupolymers is  much wider,

     *  Higher-solids content at .quivalent viscosity.  In solvent polymerizations, as
       the molecular weight increases so does the viscosity.  Waterborne systems  are
       not as sensitive to viscosity from increased molecular weight.  Thus, to obtain
                                          3 A

-------
Paint System
1 . High-solids polyester
2. Coil-coating polyester
3. High-solids alkyd
4. Short-oil alkyd
5. Water-reducible polyester
6. Water-reducible alkyd
7. High-solids water-
reducible conversion varnish
8. High-solids water-
reducible conversion varnish
9. High-solids water-
reducible conversion varnish
% N.V.V.*
80
51
80
34
48
29
80
73.5
67
Volume Ratio:
Total Non-Volatile/
Organic Volatile
80/20
51/49
80/20
34/66
82/18
67/33
90/10
90/10
90/10
Ml of Organic
Volatile Liberated
per Sq Ft per Mi!
of Dry Film Coating
0.59
2.30
0.59
4.75
0.51
1.16
0.24
0.24
0.24
*Non-vo!atile by volume.
Source: Modem Paint and Coatings, March 1975.
             Figure 15. Comparison of the Amount of Organic Volatile Material Contained in
                       High-Solids, Water-Soluble, and Conventional Paints
       similar molecular weights,  a solvent system must be used with a much higher
       viscosity than that of a waterborne system.  In addition, waterborne system
       viscosities are less sensitive to solid contents than are those of solvent sys-
       tems. Thus, waterborne systems permit the use of higher solids with higher
       molecular weight for the same  required viscosity.

       Lower raw material cost.  The cost of solvent coatings includes the price of
       the solvent, whereas in aqueous-based coatings very little solvent is used and
       the water is free.  A typical example follows of raw material costs for equiva-
       lent solids systems, in which the solids cost 50 cents per pound and the solvent
       an average of 75 cents per gallon, or 10 cents per pound.

40% Solids ,
solvent-based
40% Solids ,
water-based
Cost of Solids
per 100 Ib.
$20.00
$20.00
Cost of
Solvent
$6.00
$1.50
Total Raw
Material Cost
$26.00
$21.50
                                         37

-------
       The cost of the solvent will be reflected in the selling price of the coating.  This
       was a prime factor in the significant cost increases of solvent-based coatings
       during the recent petrochemical shortage.
       Ease of clean-up.  Water-based systems can be readily cleaned up with water,
       whereas solvent systems require solvents.
Problems
    Drawbacks associated with switching to waterborne coatings include:

    *  Use of a comparatively untried technology. The traditional reliance on solvent-
       based  systems for metal coating has resulted  in a  lower level of interest in
       waterborne systems.  However,  many authors claim that dry-film properties
       have been developed that are equal or superior in every respect to those
       achieved by conventional solvent systems.
    *  Higher total-system energy requirements to remove water.  Water has a higher
       latent heat of vaporization(1,000 Btu/lb.)than most solvents (100-200 Btu/lb.).
       Thus,  it takes more heat energy to evaporate or remove a pound of water than
       a pound of solvent.  A comparison follows of two systems, one a 70-percent
       solids  solvent coating and the other a 70-percent solids aqueous coating.
Coating
Type
Solvent
Aqueous
Volatiles
Solvents
Water
Latent Heat of
Vaporization
Btu/lb.
200
1,000
Heat Required to
Volatilize l.OOOlbs.
Volatiles
200,000 Btu
1,000,000 Btu
       As a rule of thumb, at $1.25 per 1,000 cubic feet of gas and 1,000 Btu per
       cubic foot,  the cost of natural gas is $1.25 per 1,000,000 Btu.  Thus, evapo-
       rating the solvent costs 25 cents and the water $1.25.

    There may be compensating factors for the high cost of water removal, however,
in that some of the solvent that evaporates from waterborne coatings may be used for
heating requirements through burning of the oven exhaust gases. This depends  on in-
dividual plant operations and will be discussed again later.

    The higher energy requirement for evaporating the water is usually mitigated by
the fact that this constitutes only part of the heat loss of the oven; the exhaust gases
also carry away a portion of the heat requirement.  Figure 16 compares  the energy
balance in an oven curing a conventional solvent  system and an equivalent waterborne
coating that has a solvent component representing 20 percent of the volatile load. In
this instance the heat requirements are quite similar, with a 10-percent  edge in favor
of the waterborne svstem.
                                         38

-------
 EXHAUST RATES

 Solvent Coating

      /17.5 gallons of solvent /10 .000 cubic feet at 70°F A /  1 hour  \
      \        hour         / \         gallon          / \60 minutes/

       =     2925 cubic feet of air at 70°F, per minute.

       The exhaust rate for the solvent  system is 2,925 cubic feet of air at
       70°F per minute.

 Waterborne Coating

 A.    Solvent Requirement

      /17.5 gallons^          /10.000  cubic feet of air at 70°F\   I  1 hour  \
      \   hour    /   (0  '  \           gallon            /   \60 minutes/

       =     583 cubic feet of air at  70°F. per minute.

 B.    Humidity Requirement

      /17.5 gallonsN   (Q g.   /5.000  cubic feet  of air at 70°F .\   /   1 hour  \
      \    hour   /   l        \          gallon            /   ^60 minutes j

       = 1167 cubic feet of air  at 70°F.  per minute.

       The total exhaust requirement  is 1,750  cubic feet of air at 70°F. per minute.

 Solvent System

 Parts  and Conveyor Load

       n     _    /ll.OOO pounds\  / 0.12Btu \
       Qm   '    \     h^S	j  ( pound °F.j     (3500F.-700F.)

             370,000 Btu/hour.

 Panel  Loss Load


       Qp    =     (10.000  square feet)   / °'3 Btu   \     (350°F.-70°F )
                                         \ sq. ft.  °F./

       =     840,000 Btu/hour.
Figure 16. Heat Requirements for the Baking of Equivalent Solvent-Borne and Waterborne Coatings
                                      39

-------
        Exhaust Load
                         /292S cubic feet at 70°F A,  /0.075 pounds\  /  0.24 Btu
              60 minutes
        minute


    (3500F.-70°F.)
                 hour

              Total 2,095,000 Btu/hour.
cubic foot  I  I pound °F . /


855,000 Btu/hour.
              The total heat lost, i.e.,  the total heat input, is approximately 2,100,000
              Btu per hour, which can be supplied by burning 2,100 cubic feet of natural
              gas per 'hour.

        Waterborne System = 350° Bake

        Parts and Conveyor Load
                          11.000 pounds\   / 0.12 Btu
                              hour     /   \ pound °F.
                                   (350°F.-70°F.)
              =     370,000 Btu/hour,

        Panel Loss Load
                           (10,000 square feet)
                         0.3 Btu
                        sq. ft. PP.
                (350°F.-70°F.)
              =     840,000 Btu/hour.

        Water Evaporation Load
                          14 gallons of water \
                                  hour      /
                        B.33 pounds\    [ 1,178 Btu
                           gallon   I    y pound °F.
                    138,000 Btu/hour.
        Exhaust Load


              Qa    =

              60 minutes
                  hour
              Total:
1,750 cubic feet at 70°F.\  /0.075 pounds\
         minute        I  \  cubic foot    1


    (350°F.-70°F.)
      530,000 Btu/hour
    1,878,000 Btu/hour
                     0.24 Btu
                    pound °F.
              The total heat lost, i.e ,  the total heat input, is approximately 1,900,000
              Btu per hour, which can be supplied by burning 1,900 cubic feet of natural
              gas per hour.
Source: Metal Finishing, December 1975.
                                     Figure 16 (continued)
                                             40

-------
                          EPA-GAQPS Libraix
                          Mutual Plaza
                          RTF, H.C.  277H
TECHNICAL REPORT DATA
(Pietae fead fmstructtoiu on the reverse before complt
1. REPORT NO. 2.
EPA-625/3-77-Q09-V I
4. TITLE ANO SUBTITLE
Controlling Pollution from the Manufacturing and
Coating of Metal Products
7. AUTHORiSI
9. PERFORMING ORGANIZATION NAME ANO ADDRESS
Foster B. Snail, Inc.
and
JACA Corporation
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Environmental Research Information Center
Cincinnati, OH 45 268
PB 29967
i. REPOR-
Publication Date: May IS
6. PERPORMING ORGANIZATION CODE
S. PERFORMING ORGANIZATION R6POF
10. PROGRAM ELEMENT NO.
1HD622
11. CONTRACTi'GRANT NO.
13. TYPE OF REPORT ANO PERIOD COV
Technology Transfer
14. SPONSORING AGENCY CODE
EPA/ 600 /OO
1i. SUPPLEMENTARY NOTtS
-*>
1«. ABSTRACT
  fh« first volume acquaints supervisory and management personnel in the industry
  with methods of reducing the emission of organic  solvents to the atmosphere and
  helps them assess the costs.  Volume 2 outlines for plant operators practical and
  proven techniques for controlling hydrocarbon emissions  from metal cleaning
  operations, along with appropriate cost data.  Volume 3  addresses managers, engi-
  neers and other industry personnel responsible for  resolving the water pollution
  problems of a manufacturing facility.  It covers  regulations, in-plant controls,
  three methods for wastewater treatment, establishment of a working relationship
  with a municipality and a case history.
                          'N
17.
KEY WORDS ANO DOCUMENT ANALYSIS
1. DESCRIPTORS
Air Pollution
Water Pollution
ttWOOUCEO IT
NATION
INFORM
DAMN
SPIN
IS. DISTRIBUTION STATEMENT
Release to Public
b.lDENTIFIBRS/OPEN ENDED TERMS
Metal Products
Metal Coating
IAL TECHNICAL
ATION SERVICE
«t«NI Of COMMERCE
I6HEID, V*. 22161
IS. SECURITY CLASS 
-------
    *  Rust and corrosion potential.  Coating applicators, tunnels, and ovens would
       now be subject to water vapor, which could condense and drip down onto moving
       parts.  Ovens made of galvanized steel may be subject to water corrosion.

    *  Increased treatment of metal parts before coating.  Most metal parts have
       films of grease and oil that must be removed to achieve proper coating films.
       Solvent-based coatings have an ability to "self-clean" some of the surface if it
       has not been completely treated. Water-based coatings, however, would re-
       quire a completely oil-free surface, which might increase pretreatment costs.

    •  Slow drying at high humidity.  Coating operations that depend primarily on re-
       moval of volatiles during application and air drying will be slowed down on days
       of high humidity and slow water evaporation.

SWITCHING TO ULTRAVIOLET OR ELECTRON BEAM CURE SYSTEMS

    Ultraviolet (U.V.) or electron beam systems rely on the rapid uptake of high-
intensity energy from an external  source to polymerize the low-molecular-weight com-
ponents of the coating.   The materials are supplied at close to 100-percent solids so
that, except for extraneous matter, all that is applied in the first place remains in the
coating.

Advantages

    Benefits of using these systems include:

    *  Substantial emission reduction.  The systems  are inherently 100-percent solids,
       or 100-percent active.  Emissions are only incidental and can be as little as 5
       percent by weight.  There  is ozone from the U.V. process, but this can be  min-
       imized by proper controls.

    «  High-speed reactions.  Relative typical cure times for total-solids coatings
       would be:

                   Curing System             Time

                   Electron beam       1 second  or less

                     Ultraviolet               seconds

                       Oven                 minutes

    *  Low operating costs.  Figure  17  is a  synopsis of operating-cost comparisons
       for conventional, U.V. and electron beam curing. Figure 18 compares the
       costs of U.V. curing vs. infrared curing.  This is of particular interest, since
       infrared ovens can be readily  converted to U.V. units.

    *  Reduced floor space for coater.  Ovens normally take up much of the floor
       space in coating  lines.  A system with U.V. of electronbeam curing that re-
       quires minimum oven capacity will use less floor space.
                                       41

-------
Line Speed  (Fpm)

Beam Power

Machine Power (kW)

Power Costs ($/hr)

Maintenance ($/hr)

Nitrogen Gas

Water

Total Costs  ($/hr)

60
--
2900*
13.13°
.80
__
--
13.93
Heat
120 180 80
3.7*
5800 8800 100
26.25 39.38 2,00*
1.00 1.20 1.00
--
--
27.25 40.58 3.00
LI.V.
120
7.4
200
4.00
2.00
--
__
B.OO

180
11.1
300
e.oo
3.00
--
_-
9.00

60
1.25**
B
.16*
2.25
1.00
.20
3.61
Electron Beam
120
2.5
10
.20
2.50
1.50
.20
4.40

180
3.75
12
.24
2.75
2.00
.20
5.19
      Based upon an ultraviolet cure requirement of 1 j/cm^.

      Based upon an electron beam dose to cure of 2.5 megarads (0.25 j/cm^ for a 1 mil coating) .

      Based upon natural gas at 1000 BTU/cubic foot and converted directly using 1055 j/BTU.
                                                              *
      Based upon gas costs at $1.25/1000 cubic feet.

      Based upon power costs at 2t/kWhr.
Source; "Status of Electron Beam Curing," Paptr presented at the National Coil Coaters Association Meeting, tas Vegas. May 1971.
      (Cost figures updated)


                                        Figure 17. Comparative Economics of High-Speed Curing Units

-------

I
Oven length (ft)
Line speed (fpm)
Vehicle
Nonvolatile (%)
Film thickness (mils)
Coverage (wet) (sq ft/gal)
Coats
Cure time (sec)
Exit temp (°F)
Cool
Cost of system ($/gal)
Cost per sq ft ($)
Per coat
Total
Power (kW)
Power per sq ft ($)
Typical U.V.
10
60
Polyester
90-100
2
700-800
1
10
100
No
5.00-6.00

0.7-0.9
0.7-0.9
100
Less than 0.1
Typical 1R
90
60
Urea-olkyd
35-65
2
500
2
90
130
Yes
2.00-3.00

0.9-1 .3
1.8-2.6
250
App. 0.2 (2 coats)
Source: Journal of Paint Technology, Vol. 44, No. 571, Aug. 1972.

                  Figure 18.  Comparative Costs of U.V. Curing and Infrared Curing


Problems
     The disadvantages of switching to U.V.  or electron beam curable coatings systems
are:
    •  High formula costs.  As can be seen in Figure 18, formulas based on the types
       of polymers that can be cured by U.V. or electron beam cost several times as
       much as conventional coatings, necessitating tight control on overcoating and
       waste.

    *  Limited selection of polymers. Since this is a relatively new technology, the
       range of polymers available is still limited, although some can coatings, var-
       nishes, and inks have been developed.
                                         43

-------
    •  Special precautions for high intensity energy sources.  U.V.  and electron beam
       energy sources can cause injuries to workers if not carefully shielded and
       operated.
    •  High costs of coating and curing equipment.  Initial capital expenditures for
       coaters,  curing chambers, and protective shielding tend to be higher than
       equipment for conventional coatings.

SWITCHING TO 100-PERCENT SOLIDS COATINGS

    Total-solids systems represent an entirely different technology  and for most ap-
plications require new equipment that is generally not compatible with existing lines.
Because of the potential advantages, however, the automotive industry has begun using
some of these coatings for auto bodies. Further, chain-link fencing is processed with
a "green" coating, and many houseware items have protective plastic coatings.

Advantages

    Benefits of using 100-percent solids coating systems  include:

    •  Freedom from emissions.  There is no solvent vapor generated in the  curing
       process for total-solids coatings.  Emissions are  therefore negligible  and are
       limited to solid particles that can be trapped by relatively cheap systems like
       dust collectors.

    •  Reduced  energy consumption.   Since the coating is 100-percent solids, no heat
       is  required to volatilize solvent or water.  The only heat needed for thermo-
       plastic coatings is that necessary to melt or flux the material so that it will
       bond to the surfaces.

    Heats of fusion or melting tend to  be  lower than heats of vaporization, so  that the
net heat required per 100 pounds of dry coating would be less than that for either the
high-solids or aqueous systems.  Additional heat will be needed to cure the coating if
it is a thermosetting type; however, since no solvent or water has to be removed, the
total heat will still be lower than for an equivalent waterborne or high-solids solvent
system.

Problems

    Disadvantages of 100-percent solids coating systems are:

    •  Higher costs. On a relatively equivalent basis, solvent-based paints were ap-
       proximately 1-1.3 cents/ft2/mil of thickness, whereas fluidized-bed powder
       coatings  were 1.6 - 4.1 cents/ft2/mil,  depending  on the system.

    •  Limited selection of systems.   Only certain polymers are available in a  form
       that will  flux and  fuse (polyamides, polyesters, and some epoxies),  limiting
       total-solids coating formulations.
                                        44

-------
    •  Variable adhesion.  Adhesion, a direct function of the fusion process, may be
       adversely affected by any irregularity in temperature in either a 100-percent
       solids coating or the surface to be coated.
    *  Incompatibility with existing coating lines.  As mentioned, special new equip-
       ment is required for application and curing of 100-percent solids systems.
    •  Difficulty in applying uniform thin coatings.  The total-solids coating technique
       lends itself to thicker coatings.  Applications under Imil are difficult; 3 or
       more mils is more typical.
    •  Color changes in-process.  In 100-percent solids coating lines, large amounts
       of colored particles must be moved and  cleaned up before each color change or
       the next batch of articles may have off-specification colors or shades.
                 EMISSION CONTROL THROUGH PROCESS CHANGES

    Operating changes that a plant can consider in setting up its emission control pro-
gram include:

    •  Controlling emissions by incineration;

    •  Controlling emissions by adsorption;

    •  Improving spraying efficiency;

    •  Improving dip coating, flow coating, and coil coating efficiency;

    •  Purchasing prefinished roll stock;

    *  Increasing vapor concentration; and

    •  Educating plant personnel for process changes.

    The first options, controlling emissions by incineration and adsorption, will be
covered in Part B of this manual, which deals with treatment of hydrocarbon emis-
sions and heat recovery.  Discussion of the remaining process changes follows.
                                                           **"'\s
IMPROVING  SPRAYING EFFICIENCY

    The most commonly used air-spraying method, as explained earlier, is the most
inefficient coating method. Overspray (and thus emissions)  can often be reduced by
ganging spray nozzles of different spray patterns or by rotating the article to be
sprayed.  Prefinishing the article so only a touch-up is required may also cut spraying
losses.

    Other techniques for improving efficiency include minimizing manual spraying and
color changeovers through production control.
                                        45

-------
     The effect of inefficient spraying on emissions is obvious.  More air is required
to maintain the TLV,  necessitating more fan capacity.  If emissions have to be con-
trolled, added cfm are very expensive, as is shown below:

                          Effect of Over spray Reduction
Coating Formulation
20/80 Coating system

TLV
200ppm
2QOppm
Over spray
50%
10%
scfm
1,400
1,030
In this case, the savings in fuel for the afterburner would amount to about $4,000 per
year, which would be in addition to a savings of about 27 percent on the cost of the coat-
ing system used.

    A gross measurement of overspray can be obtained by a material balance between
the coating actually used and the coating that is on the articles after spraying.

IMPROVING DIP COATING, FLOW COATING, AND COIL COATING EFFICIENCY

    As  in all coating operations, control of coating weight or thickness is of primary
importance.  There are many devices that can be installed on production lines for
sampling on a random basis and for weighing the article if it has, for example, been
dip coated and can be weighed.  Off-weights will trigger either a manual or automatic
response to correct the situation.  In the case of dip coating, this corrective response
might be shorter immersion time,  reduced immersion depth, or increased air blow-off
pressure.

    Beta ray, gamma ray, and x-ray devices have been used in many areas of industry
for determining coating thickness on moving webs.  Their use in monitoring high-speed
coil-coating applications should be considered.

PURCHASING PREFINISHED ROLL STOCK

    Some items lend themselves to prefinishing and use of raw materials that come in
coil/machinable form.   For example, license plate stock is prime-coated at one loca-
tion and stamped and painted at another.  This not only places the prefinishing step in
a more efficient setting,  but also shifts some of the solvent emission load.  Since a
final product is still the responsibility of the ultimate finisher, however, precise con-
trol must be maintained over the prefinisher,

INCREASING VAPOR CONCENTRATION

    The cost of moving and heating air is proportional to the amount of air being
moved.  There is, therefore, a considerable operating-cost advantage in having vapors
as concentrated as possible.
                                       46

-------
    In our initial discussion of basic factors that affect emissions, we pointed out that
to maintain a safe TLV, 30 times more air must be supplied to the spray booths or
air-drying tunnels than is strictly necessary for a normal rate of evaporation.  If
booth and tunnel areas can be kept free of operating personnel, however, the TLV con-
centration requirements can be replaced by the much less demanding 25-percent LEL.
Further, operators  can continue  to work in the areas by using protective devices.

    Substitution of automatic spraying wherever possible will eliminate the need to
maintain TLV, with important economic advantages.  For instance, without TLV stric-
tures the ventilation requirements can be reduced by a factor of about 10, lessening
energy needed to move the air or to control its temperature.  Moreover, emission
control becomes much cheaper because (a) equipment size is drastically reduced, with
savings of 40 percent; and (b) fuel costs are also greatly reduced because less air has
to be heated and much less fuel is required per scfm.

    An important consequence of increasing vapor concentrations is that all equipment
conveying wet parts must be enclosed.  However, the economic advantages of increased
concentrations may  pay for the substantial modifications that enclosure requires.

    Vapor concentrations cannot be raised beyond safe limits or the limits placed on
recovery incineration equipment. For example, if emissions are controlled by com-
bustion with either an afterburner or catalytic converter,  there is no point going above
40 percent LEL; because of the considerable heat value of most solvent vapors, partic-
ularly hydrocarbons, severe overheating and equipment damage may result from excess
vapor combustion.  Indeed, this  is reported to be one of the most frequent problems
with afterburners, especially with catalytic units.
                       *•

EDUCATING PLANT  PERSONNEL  FOR PROCESS CHANGES

    The main problem in switching to a more efficient application method (without sig-
nificant change in system formula) may be human resistance to change.  This is par-
ticularly true where hand operations are replaced by automated methods. A change
that may even temporarily affect quality or production rate may be resisted by super-
visory personnel who pride themselves on high-efficiency/low-downtime  operations.
Therefore, any test of new equipment (for instance,  airless or electrostatic)  should be
closely  supervised by management-level staff.

    A second problem is that a new process often involves new materials with higher
costs per pound.  Economic advantage can be achieved only if the product is used at the
prescribed rate, restricting the  operating personnel's latitude in applying the coating.
Previously, if the coating was within 10 - 20 percent of desired weight or thickness,
there was little cost effect.  Higher costs per pound necessitate more precise control.

    Management must make it clear to employees that the changes are in everyone's
best interest.
                                        47

-------
                                SUMMARY
    In Part A of this volume, we have covered the basic terms connected with pollution
control, the properties of solvents that can cause pollution, and the ways that materials
and processes influence the amount of emissions.

    We have  seen that the factors most affecting emissions are;

    «  Total  volume or weight of coating used;

    *  Efficiency of application; and

    »  Composition of the formulas used.

    Many routine opportunities for pollution reduction will become evident in a simple
but thorough survey of the plant.  Some possibilities may be beyond one department's
direct control, but cooperative effort with other sections may enable  considerable re-
ductions in pollution and costs.
                       PRODUCT DESIGN CONSIDERATIONS

    Certain variables in design should be studied as possible aids to pollution control.
Managers should consider:

    •  Choosing material that will serve the intended use without painting, for instance,
       anodized aluminum, plastic,  or plated components;

    •  Standardizing and reducing the number of colors to minimize solvent needed for
       clean-up between color changes and to reduce inventory;

    •  Tightening specifications on coating thicknesses or number of coats required;
       and

    *  Eliminating pockets, rough coatings, or other features that require large
       amounts of paint for adequate coverage.
                         FABRICATION CONSIDERATIONS

    Manufacturing variables that may aid in pollution control include:

    *  Using precoated stock and limiting painting operations to touch-up of damage
       occurring during fabrication;

    »  Buying primed components and applying only a top coat;
                                       48

-------
    •  Increasing protection,  and saving paint, by using conversion coatings such as
       anodizing and phosphatizing;

    *  Assembling first and saving painting for the final step, to avoid two paint appli-
       cations ; and
    •  Fabricating the articles and then subcontracting the coating operation to a fin-
       isher who already has emission controls in place.
                            PROCESS CONSIDERATIONS

    We have discussed in some detail the process changes that can be made to reduce
emission levels in coating operations.  Those that will have a high impact on lowering
emissions at the source include:

    *  Replacing manually operated, air-atomized spray methods with, preferably, a
       combination of airless and electrostatic spraying,  to reduce overspray and help
       reduce ventilation needs;
    •  Converting to formulas with as high a solids content as possible;

    *  Switching wherever possible to waterborne coating;
    *  Reducing excessive ventilation;
    •  Using powder coating and U.V. curing, where feasible,  when new facilities are
       installed.

    If control devices prove necessary even after all possible design, manufacturing,
and process changes have been made, plant management should carefully examine the
total air balances in the facility and should study all unavoidable sources of emission,
with a view to increasing concentrations in waste streams.

    The end-of-line treatment of these waste streams will be the subject of Part B,
which follows.
                                       49

-------
 TREATMENT OF HYDROCARBON
EMISSIONS AND HEAT RECOVERY
            50

-------
                                 CHAPTER  I

               DISPOSAL OF SOLVENT VAPORS

    Most plants have hydrocarbon emissions that cannot be eliminated by the source
control methods discussed in Part A. There are a number of techniques for treating
these emissions, of which the most widely used are based on combustion that breaks
down organic pollutants into other,  nonpollutant substances.


                                 COMBUSTION

    Combustion of organic compounds is a widely used technique for air pollution
emissions control. At a high  enough temperature, carbon and hydrogen will combine
with oxygen to produce carbon dioxide and water.  Although there is some concern
over accumulation of carbon dioxide in the atmosphere, both CO2 and water will prob-
ably remain classified as nonpollutants for the foreseeable future.

    Elements other than carbon and hydrogen that may be present in the organic com-
pound will also be released (though not necessarily in oxygenated form) in the combus-
tion process.  Halogenated hydrocarbons like chlorine and fluorine are generally con-
verted to the acids, though in  certain cases phosgene may  result.  Sulfur is burned
to sulfur dioxide,  while nitrogen is converted to nitric oxide.

    Combustion is used for control of odorous sulfur and nitrogen compounds where
the amounts of SO2 or NO formed are too small to cause significant air pollution.
However, halogenated compounds are not normally burned, because of the extremely
corrosive and hazardous nature of the gases formed.  Even trace quantities of HC1 or
HP would force the use of exotic and expensive corrosion-resistant materials in the
control equipment. Greater-than-trace quantities would require additional controls
for the removal of acid gases. Thus, combustion for the control of halogenated hydro-
carbons is  impractical.

    In a recent review of solvent metal cleaning practices in industry,* Dow Chemical
Company found that halogenated hydrocarbons are used almost exclusively in vapor de-
greasing and in about half of the cold degreasing operations. From,a practical stand-
point, therefore, solvent combustion as an air pollution control  technique is limited
largely to the metal coating industry.
•"Study to Support New Source Performance Standards for Solvent Metal Cleaning Operations." Report to EPA by
 The Dow Chemical Company, June 1976.
                                       51

-------
    Combustion of solvent vapors can be accomplished in one of three ways:

    *  Direct-flame incineration
    •  Catalytic incineration

    »  Process boilers

DIRECT-FLAME INCINERATION

    Direct-flame or thermal incineration involves raising the waste gas temperature
and sustaining it long enough for any hydrocarbons present to combine with available
oxygen.  A direct-flame incinerator usually consists of a burner fueled with auxiliary
fuel and a mixing chamber,  The efficiency of the unit depends on the temperature and
residence-time characteristics of the unit and, to a  lesser degree, the  solvent burned
and the design details.

    Eighty-five percent combustion of most contaminants is easily obtainable at tem-
peratures of 1200°F - 1400°F and 95 percent at approximately 1500°F.   For direct-
flame units without  heat recovery, the principal expense is fuel.  The addition of heat-
reeovery equipment will increase capital costs but reduce those for fuel.

Gas Conditioning

    Any non-combustible material, such as particulate matter in the waste gas, will
simply pass through the incinerator at normal temperatures. Since the gas velocities
are generally lower in  the combustion chamber than in the incoming ductwork, the
combustion chamber will act as a settling chamber and dust will tend to accumulate
there.  This does not normally affect the performance of the unit unless the buildup
significantly reduces the combustion chamber volume or alters the flow pattern.  Where
the incinerator exhaust is  circulated back into the oven, the presence of particulate
matter may affect the quality of the coating.

    In most metal coating operations, the carryover of particulate matter is insignifi-
cant and no prior conditioning or precleaning is necessary.  Where large amounts of
paint are likely in the exhaust gas, a dry-type collector is preferred to avoid cooling
of the gases and increased incinerator fuel consumption.

Combustion Conditions

    To achieve efficient combustion of hydrocarbons to carbon dioxide  and water, the
solvent must be mixed with sufficient oxygen held at a uniform temperature of between
1200°F and 1500°F  for 0.3 - 0.5 seconds. Time and temperature are interrelated,  so
that a relatively short contact period and high temperature  can produce an efficiency
(i.e., degree of pollutant destruction) equivalent to a time/temperature unit with long
contact and low temperature.  This effect is illustrated in  Figure 1.  For normal
straight-chain solvents, operating temperatures of from 1200°F to 1300°F at a resi-
dence time of 0.3 -  0.5 seconds are generally used to achieve greater than 90 percent
control.  Methane,  cellosolve, or benzene-based compounds, however,  may require a
temperature of 1400°F  - 1500°F at conventional contact periods of 0.3  - 0.5 seconds.

                                       52

-------
en
Co
                                                                                Increasing Temperature
                                                                                                                                1800
2000
                Source: Afterburner Systems Study, Shell Development Company, 1972.



                                            Figure 1. Coupled Effects of Temperature and Time on Rate of Pollutant Oxidation

-------
     In cases where carbon monoxide formation in the incinerator is deducted from the
 unit's efficiency, such as under Rule 66 of the Southern California Air Pollution Con-
 trol District referred to earlier,  significantly higher time/temperature units are re-
 quired to achieve a given efficiency. This principle is illustrated in Figure 2.  The
 combustion of organic  carbon to carbon dioxide is a two-stage reaction:  the  first stage
 of oxidation to CO involves a relatively high-heat release and proceeds rapidly,  The
 second stage,  further oxidation to CO,, gives off less heat and is therefore an inher-
 ently slower reaction.

     The zone of combustion consists of a region of rising temperature followed by a
 dwell region with an essentially constant temperature.  The  design residence time of
 0.3 or more seconds should apply to the reaction zone only,  with additional volume
 provided for initial combustion and mixing.  Insufficient combustion chamber volume
 is probably the most significant design flaw  in units that fail to meet performance
 expectations.

    100
    90
    80
                                       Hydrocarbon + CO,
                                       per LAAPCD, Rule 66
 £
 c
 o
 u
    70
    60
    SO
              1200
1300
1400
1500
                                         Temperature, °f
Source; Afterburner Systems Study, Shell Development Company, 1972.

                  Figure 2. Typical Effect of Operating Temperature on Effectiveness
                   of Thermal Afterburner for Destruction of Hydrocarbons and CO
                                          54

-------
    Thermal design must take into account the following factors:

    •  Efficiency increases with operating temperature.
    •  Efficiency increases with detention time of up to approximately one second.
    *  Efficiency increases with initial hydrocarbon concentration.
    *  Efficiency decreases if the waste gas is preheated to a point approaching the
       combustion temperature.
    *  Efficiency increases with the degree of contact between the flame and the sol-
       vent vapors.
    *  Poor mixing yields low efficiency even if the temperature and residence time
       are sufficient,
    *  Carbon monoxide removal requires a minimum temperature of 1300"F  regard-
       less of retention time.

Process Design Principles

    The process design of a thermal incinerator involves selecting the general charac-
teristics for the unit,  establishing design values for temperatures and gas volumes,
and determining the fuel-firing rate and combustion chamber volume.  Once the proc-
ess has been fully described, the physical facilities for meeting process requirements
can be determined.

    The information required for the process design  calculations is:
                     *
    *  Inlet gas flow rate,  scfm;

    •  Met gas temperature,  °F; and

    *  Solvent type and vapor concentration range, % or ppm.

    Where a heat exchanger is used to preheat the gas, the temperature at the inciner-
ator inlet will be greater than the temperature at which the waste gas leaves the proc-
ess.  Heat exchanger design considerations are further described below under Heat
Recovery.

    The desired gas temperature at the incinerator must be specified.  Frequently,
air pollution regulations require the gas temperature to be above a certain minimum.
This may vary from about 1250° F for easily oxidized solvents to 1500° F for resistant
vapors.  Where carbon monoxide formation must be prevented,  a minimum design
temperature of 1400° F is recommended.  The desired gas temperature at the  inciner-
ator should be slightly in excess of the required minimum.

    The residence times of 0.3 - 0.5 seconds mentioned earlier should be considered
as minimum values for systems burning hydrocarbon solvents without  significant ob-
jectionable impurities.  Many units are operating satisfactorily at 0.3 seconds resi-
dence time, but only where extremely good mixing  is achieved.  For carbon monoxide
removal, the higher residence time of 0.5 seconds should generally be used.

                                       55

-------
    Prior to design of the incinerator, a burner type and fuel should be selected that
will be compatible with the source of oxygen for combustion, to the extent that this is
known.  If the contaminated air stream will be used to provide the oxygen, the size of
the incinerator and the heat requirements will be lower, since it will not  be necessary
to accommodate outside air in the system.  Burner types will be discussed presently.
Natural gas and propane-fired units use contaminated air almost exclusively. Oil-
fired burners may be set up to use contaminated air, but frequently use outside  air to
avoid fouling of the primary air blower and burner gun.

    Hydrocarbon solvents used in metal coating have a high fuel value (Btu per pound
of solvent) and will contribute to the heat required  for incineration.  At concentrations
of 100-200ppm, the fuel saved will be almost negligible, but the savings will be signif-
icant beyond 25-percent LEL concentrations. Heat available from the solvent is nor-
mally included in the heat balance.

    Fuel Requirements

    The first step in the design of a thermal incinerator is to determine the amount of
fuel required to heat the waste gas stream to the design temperature.  The amount of
fuel will depend on the flow rate,  composition,  and temperature of the incoming waste
gas, the type of fuel,  and whether the oxygen required will be derived from the waste
gas or from external air.  Any heat value from the solvent vapors in the waste gas
stream will reduce the fuel needs.

    When a fuel is burned, carbon and hydrogen in the fuel combine with  air to produce
carbon dioxide and water.  The heat energy released raises this,carbon dioxide and
water to a very high temperature. When air is used as an oxygen source, the nitrogen
present must also be heated, lowering the temperature in the gas mixture (approxi-
mately 3400°F).  If the combustion products are further mixed with a waste gas stream,
the temperature of the resultant mixture will be still lower.  In a normal design situa-
tion, the final desired temperature of the gas mixture is known and the problem be-
comes one of finding the proper fuel-addition rate.

    The potential heat energy released by various  organic materials burned at 60° F is
termed the gross or higher heat value.  Since water is a combustion product of most
fuels, the energy available to heat combustion products must be reduced by the heat of
vaporization of the water formed.  The resultant heat is termed the aet or lower heat
value.  Gross and net heat values, along with combustion air requirements for a num-
ber of fuels used in incineration,  are shown in Table I.  In heating waste  gas  streams,
the net heat released is distributed among the combustion products and the waste gases.
Since the ratio of combustion air to fuel is known,  calculations are simplified by work-
ing with the concept of available heat.  This is excess heat remaining for other pur-
poses after the combustion products have been heated to a specified temperature. As
may be seen in Figure 3, the available heat is reduced as the design temperature in-
creases, since more heat is consuraed in heating the combustion products.  If excess
air is included with the combustion products, the available heat for other purposes is
reduced still further,  as is shown in  Figure 4.  It  should be noted that Figure 4 is ap-
proximate , in that it is assumed that the ratio of combustion air to heat content is con-
stant for all fuels.
                                        56

-------
        Table  I

Combustion Constants



No . Substance

1 . Carbon*
2 . Hydrogen
3 , Oxygen
4. Nitrogen (aim)
5. Carbon monoxide
6, Carbon dioxide
Poroffin series
7 . Methane
8 , El bane
9 . Propane
10, n-Butane
1 1 . Isobutane
12. n-Pentane
13. Isopentane
14. Neopenrane
15. n-Hexane
Olefin series
16, Erbylene
17. Propylene
18. n-Burene
19. bobuterie
20. n-Pentene
Aromatic series
21. Benzene
22. Toluene
23. Xylene
Miscellaneous gases
24. Acetylene
25. Naphthalene
26. Methyl alcohol
27. Erbyt alcohol
23 . Ammonia
29. Sulfur*
30. Hydrogen sulfide
H C If.., *4>*.v!,4a
Jl . gutter aioxioe
32 Water Vapor
33! Air



Formula

C
H2
0?
N?
CO
CO-,

CH<
C2H6
C3H8
C^H^j
C^HHJ
C6H,?
*-tjH|2
C6HW
C6H,"

CjH,,
C-,Hn
CIH«
C4H8
Cr,H,u

Cf,Htl
C,H8
CuHio

C2H?
CirjHg
CH3OH
C?H6OH
NH3
s
H2S
C(™JT
3\J2
H,O




Molecular
Weight

12.01
2.016
32.000
28.016
28.01
44.01

16.041
30.067
44.092
58.118
58.118
72.144
72.144
72.144
86.169

28.051
42.077
56.102
56.102
70.128
V,
78.!07
92,132
106,158

. 26,036
128.162
32,041
46.067
17.031
32.06
34.076
A4 fVA
O4 ,UQ
18.016
28"°



Lbper
Cu Ft

_
0.0053
0.0846
0.0744
0.0740
0.1170

0.0424
0.0603
0.1196
0.1582
0.1582
0.1904
0.1904
0.1904
0.2274

0.0746
0.1IIO
0.1480
0.1480
0.1852

0.2060
0.2431
0.2803

0.0697
0.3384
0.0846
0.1216
0.0456

0.0911
01711
• I/ O»]
0.0476
0,0766



Cu Ft
per Lb

__
187.723
11.819
13.443
13.506
8.548

23 .565
12.455
8.365
6.321
6.321
5.252
5.252
5.252
4.398

13.412
9.007
6.756
6.756
5.400

4.852
4.113
3.567

14.344
2.955
1 1 .820
8.221
21.914

10.979
s 77f|
•J ./ ' U
21 .017
13.063


SpGr
Air
1.0000

^ -
0.0696
1.1053
0.9718
0.9672
1.5282

0.5543
1.0486
1 .5617
2.0665
2.0665
2.4872
2.4872
2.4872
2.9704

0.9740
1.4504
1 .9336
1 .9336
2.4190

2 .6920
3,1760
3.6618

0.9107
4.4208
1.1052
1.5890
0.5961

1.1898
27A4A
.*OrU
0.6215
1.0000

Heat of Combustion


Bru per Co Ft
Gross
(High)
_
325


_
322



1013
1792
2590
3370
3363
40)6
4008
3993
4762

1614
2336
3084
3068
3836

3751
4484
5230

1499
5854
868
1600
44)

647

-
Net
(low)
_
275


_
322



913
1641
2385
3113
3105
3709
3716
3693
4412

1513
2186
2885
2869
3586

3601
4284
4980

1448
5654
768
1451
365

596

-
Btu per Lb
Gross
(High)
14,093
6J,100


_
4,347



23,879
22,320
21,661
21,308
21,257
21,091
21,052
20,970
20,940

21,644
21,041
20,840
20,730
20,712

18,210
18,440
18,650

21,500
17,298
10,259
13,161
9,668
3,983
7,100
_
-
Net
(Low)
14,093
51,623


—
4,347



21,520
20,432
19,944
19,680
19,629
19,517
19,478
19,396
19,403

20,295
19,691
19,496
19,382
19,363

17,480
17,620
17,760

20,776
16,708
9,078
11,929
8,001
3,983
6,545

-
For 100% Total Air
Moles per Mole of Combustible
or
CuFI par Cu Ft of Combustible

Required for Combustion
02
1.0
0.5




0.5



2.0
3.5
5.0
6.5
6.5
8.0
8.0
8.0
9.5

3.0
4.5
6.0
6.0
7.5

7.5
9,0
10.5

2.5
12.0
1.5
3.0
0.75
1.0
1.5



Hi
3.76
1.88




1.88



7.53
13.18
18.82
24.47
24.47
30.11
30.11
30.11
35.76

11.29
16.94
22,59
22.59
28.23

28.23
33.88
39.52

9.41
45.17
5.65
11.29
2.82
3.76
5.65



Air
4.76
2.38




2.38



9.53
16.68
23.82
30.97
30.97
38.11
38.11
38.11
45.26

14.29
21.44
28.59
28.59
35.73

35.73
42.88
50.02

11.91
57.17
7.15
14.29
3.57
4.76
7.15



Flue Products
CO,
1.0
_




1.0



1.0
2.0
3.0
4.0
4.0
5.0
5.0
5.0
6.0

2.0
3.0
4.0
4.0
5.0

6.0
7.0
8.0

2.0
10.0
1.0
2.0
—
SO,
1.0
1.0



H2O
_
1.0




_



2.0
3.0
4.0
5.0
5.0
6.0
6.0
6.0
7.0

2.0
3.0
4.0
4.0
5.0

3,0
4.0
5.0

1.0
4.0
2.0
3.0
1.5

1.0



N2
3.76
1.88




1.88



7.53
13.18
18.82
24,47
24.47
30.11
30.11
30.11
35.76

11.29
16.94
22,59
22.59
28.23

28.23
33.68
39.52

9.41
45.17
5.65
11.29
3.32
3.76
5.65




For 100% Total Ai»
Lb per Lb of Combustible


Required for Combustion
Oj
2.66
7.94




0.57



3.99
3.73
3.63
3.58
3.58
3.55
3.55
3,55
3.53

3.42
3.42
3.42
3.42
3.42

3.07
3.13
3.17

3.07
3.00
1.50
2.08
1.41
1.00
1.41



NS.
8.86
26.41




1.90



13.28
12.39
12.07
11,91
11.91
11.81
11.81
11.81
11.74

11.39
11.39
11.39
11.39
11.39

10.22
10.40
10,53

10.22
9.97
4.98
6.93
4.69
3.29
4.69



Air
11.53
34.34




2.47



17.27
16.12
15.70
15.49
15.49
15.35
15.35
15.35
15.27

14.81
14.81
14.81
14.81
14.81

13.30
13.53
13.70

13.30
12.96
6.48
9.02
6.10
4.29
6.10



Flue Products
C02
3.66
_




1.57



2.74
2.93
2.99
3.03
3.03
3.05
3.05
3.05
3.06

3.14
3.14
3.14
3.14
3.14

3.38
3.34
3.32

3.38
3.43
1.37
1.92
_
$02
2,00
1.88



H2O
™
8.94




—



2.25
.80
.68
.55
.55
.50
.50
.50
.46

.29
.29
.29
.29
.29

0.69
0.78
0.85

0.69
0.56
1,13
1.17
1.59
nr_,
0.53



N2
8.86
26.41




1.90



13.28
12.39
12.07
11.91
11.91
11.81
11. 8r
11.81
11.74

11.39
11.39
11.39
11.39
11 .39

10.22
10.40
10.53

10.22
9.97
4.98
6.93
5.51
3.29
4.69



'Carbon and sulfur are considered as gases for molal calculations only.
NOTE: This table is reprinted from Fuel Flue Gases, courtesy of American Gas Association.
                        All gas volumes corrected to 60°F and 30 in. Hg dry.

-------
       140.000
    _  120.000
       100,000
        80.000
        60.000
        40,000
         3,000
         2,400
         1.800
         1.200
    OQ
    «5
          600
              Haavyfuel oil 14° API
                 152,000 Btu/gal.
                              Light fuel oil 36.5° API
                                 138,000 Btu/gal
                                 Commercial butane
                                  3210Btu/cu.ft
                         Commercial propane
                         2558 Btu/cu.ft.
                          Natural gas
                          1059 Btu/cu.ft
                                                                                1
             300     600
900     1200    1500     1800    2100
           Flue gas exit temperature °F
2400    2700    3000
Source: Control of Gaseous Emissions.  EPA Air Pollution Training Institute, 1973.

               Figure 3. Available Heats for Some Typical Fuels (Referred to 60°F)
                                               58

-------
90

80

70

60

SO
^\S
           200  I  600
              400
excess air
                                                               This chart is only applicable
                                                               to casts in which there is no
                                                               unturned fuel in the products
                                                               of combustion.

                                                               The average temperatuie of the
                                                               hot mixture just beyond the end
                                                               of the flame may be read at the
                                                               point where the appropriate %
                                                               excess air curve intersects the
                                                               zero available heat line.
                 1000  I  1400  1  1800   2200   2600   3000
              800   1200   1600   2000    2400   2800   3200

                      Flue gas temperature f
Source: Control of Gaseous Pollutants, EPA Air Pollution Training Institute, 1973.

        Figure 4. Generalized Available Heat Chart for All Fuels at Various Flu* Gas Temperatures and
                        Various Excess Combustion Air (Referred to 60°F)

                          *

    When the available heat of combustion is distributed in the waste gas, its temper-
ature will rise.  However, this rise is not uniform throughout the combustion range.
For this reason, it is common practice to work in terms  of enthalpy (heat content) of
the gases at various temperatures.  Enthalpies of common gases are shown in Table n.
As this table shows,  raising  the temperature of one standard  cubic foot of CO2 from
200°F to 1200°F requires 33.55-3.39, or 30.16  Btu.


    When the combustion air is drawn from outside the waste gas stream, calculations
of fuel requirements are relatively simple; the use of waste gas for combustion  re-
quires further data on  combustion air requirements.  Data needed for computing natu-
ral gas or propane requirements are given in Table ITJ; data for oil are presented in
Table IV.


    Sample computations follow for determining fuel requirements for various design
conditions.


                         Computations for Gas Requirements

Example 1—Given:  3000 acfm of air containing SOOppm of toluene.  The air temper-
                    ature is  300°F.


            Find:   The amount of natural gas required to heat the gas stream to
                    1400°F,  assuming that combustion air is drawn from the gas
                    stream and ignoring the fuel value of the solvent.
                                         59

-------
Since the combustion air is to be drawn from the waste gas stream, it will be neces-

saxy to write an equation balancing heat input and consumption in terms of an unknown

gas quantity:



    1 .   Heat Input



        = (Available heat at  1400 °F at 0% excess air) x G +



          (Credit for preheat of combustion air from 60° F to 300° F)



        where G = scfm of natural gas required.



        •  From Table in,



           - Available heat at 1400" F = 668.6 Btu/scf gas



           - Amount of combustion air required at 0% excess ai~



           = 10.36 scf/scf  gas



        »  From Table II, for air, enthalpy difference (300 °F - 60 °F)



           = 4.42 Btu/scf air



    2.   Heat Consumption
                                                          #


        =  (scfm waste gas - scfm required for combustion) x



           enthalpy difference (1400 °F - 300 °F)


                                      460 + 60
        *  scfm waste gas = 300 acfm x -T^-T — r-rr = 2053 scfm
                                          ' O UU
        *  scfm required for combustion = 10.36 x G scfm



        •  From Table II, for Air, enthalpy difference (MOOT - 300°F)



           = 26.13 - 4.42



           = 21.71 Btu/scf



        *  Thus, heat consumption



           = (2053 - 10.36 x G) x 21.71 Btu/min.



    3.   Heat Balance:  Heat Input = Heat Consumption



        •  Thus, 668 x G +4.42 x 10.36 xG = (2053 - 10.36xG) x 21.71



        •  Solving for G , G = 47 . 5  scfm natural gas




                                      60

-------
Example 2—Using the data from Example 1, compute the gas consumption considering
the heat available from the combustion of toluene.

Toluene combustion will  enter the heat balance by providing a heat input and by reducing
the unburned air that must be heated to the design temperature.

    1.  Toluene Flow Rate

        = 2053 scfm waste gas x —g - 1.027 scfm

    2.  Gross Heat for Toluene

        *  From Table I, for toluene, gross heat

           = 4484 Btu/scf x 1.027 scfm

           = 4605 Btu/min

    3.  Available Heat from Toluene

        •  Using Figure 4, at a flue gas temperature of 1400°F and at 0% excess air,
           available heat from toluene

           = 0.61 x Gross Heat

           = 0.61 x 4605 Btu/min

    4.  Credit for Preheat

        *  From Table I, combustion air  required

           = 42.88 scf air/scf toluene

        *  Total combustion air

           = 1.027 x 42.88 scfm air

           = 44.04 scfm air

        *  Credit for preheat

           = 44.04 scf x 4.42  BtU
                            scfm
          = 44.04 x 4,42 Btu/min
                                      61

-------
    5.  Total Available Heat From Toluene

        =  available heat + credit for preheat

        -  0.61 x 4605 -i- 44.04 x 4.42 Btu/min

           300:1.7 Btu/min

    6.  Heat Balance: Heat Input   Heat Consumption

        668 x G \ -4.42 x 10.30 xG i  3003.7

        --  (2053 - 10.36 x G - 44.04) x 21.71

        »  Solving for G ,  G ~ 43 . 3 scfm natural gas

    Comparing the two examples,  it may be seen that the natural gas savings is 4.2
sefm, or 8.8% if the heat value of the solvent is considered.

                       Computations for Oil Requirements

    To illustrate the effects of oil  firing and the use of external combustion air on the
calculations, the situation in Example 2 may be reworked for oil firing.

Example 3— Determine the quantity of #2 fuel oil  (PS 100, Table IV)  required to incin-
erate 3000 sefm of air conditioning 500ppm of toluene,  where the air temperature is
300° F.  The oil burner is supplied with  120% theoretical air taken from outside the
waste gas stream.

    1.  120% theoretical air - 20% excess air

        From Figure 4, the available heat at MOOT and 20% excess air is approxi-
        mately 55% of the gross heat value of fuel oil,  which is 130,000 Btu/gal. from
        Table IV.

    2 .  Heat Input

        =  the sum of the available heats of oil and toluene plus rrodit for preheat
           combustion air:

        -  0.55 x 136,000 x Q + 3,003.7
           .197.3^ X6.83 ~x4.42 f? x Q

        where Q  quantity of oil burned,  gallons/minute
                                       62

-------
        = 74,800 x Q + 3,007 + 5,596 x Q

        = 80,756 x Q + 3,003.7 Btu/min

    3.   Heat Consumption is the waste gas flow less the air consumed in the combus-
        tion of toluene

        - (scfm waste gas - scfm required for combustion of toluene)

          x enthalpy difference (1400° F - 300°F)

        • Thus, heat consumption
                                             TJ 4, -
          = (2053 sefm - 44.04  scfm) x 21.71 —-

          = 43,615 Btu/min

    4.   Heat Balance:  Heat Input  = Heat Consumption

        * Thus

          80,756 Q f 3,003.7 ~ 43,615

        » where

          Q = 0.5 gallons/minute

    A comparison of gross heat inputs lor Examples 2 and 3 shows:

        gross heat input (Example 2) --43.3 scfm x 1100 Btu/scf

                                   - 47,630 Btu/min

        gross heat input (Example 3) - 0.5 gal/min x 136,000 Btu/gal

                                   = 08,000 Btu/min

    Avoiding the use of outside air for fuel combustion (as in Example 2) results in a
significant savings of heat  input and thus of operating costs.

    Combustion Chamber Size

    The size of the combustion chamber will be determined by both the volumetric How
rate of the waste gas stream and combustion products at the design temperature and
the design retention time.  Since the combustion chamber should be considered as only
that zone in which the design combustion temperature is attained,  some burner types
may necessitate a mixing zone before the  combustion zone. Calculations follow for the
combustion chamber volume.
                                       63

-------
Example 4 — For the conditions in Example 2,  find the combustion chamber volume re-

quired for combustion at. 1400° F using natural gas fuel with internal combustion air;

the desired retention  time is 0.50 seconds.



    1.   From Example 2, waste gas required for the combustion of natural gas



         =  10,36 scf/scf x 43.3 scfm



         •--  448.fi scfm



         Combustion products from natural gas (from Table TH)



            1 1.45:! sel'/scf x  43.:: selm



         -  4 !>!}.«) scfm



    2-   From Example 2, waste gas required for the combustion of toluene



         =  42.88 scf/scf x 1.027 scfm



         =  44 . 04 scfm



         Combustion products from toluene (from Table I)



         --;  (7 . U -(•• 3 . 0 -i- 33 . *8) set/He! \ 1 . 027 scfm



            Ki.OiJ scfm



    3.   Fltnv through tht1 c'omlni.-.Siun rhambrr



         ;-  (2053 - 44S.C5 - 44.04) * (l!>5.9 i 4(5. Oi>)



         =  2102 scfm


           „   n   460 + 1400
         rr  2.02 x — — — jTT- acfm at 1400° F
                   460 + GO


         -  7520 acfm at 1400° F



    4.   Volume of combustion chamber needed  for 0.5 second retention time


           -,-on   r     , _       ,    1   min
         ~  <>  second


           !i2.G7 cubic feet
                                       64

-------
                                         Table II
           Enthalpies of Gases Expressed in Btu/sef of Gas, Reference 60°F
"F
60
77
100
200
300
400
son
600
700
800
900
1,000
1, 100
J , 200
!, 300
i , 400
1 , 500
1 , 600
1,700
l.BOO
1 , 900
2, 000
2 , 1 00
2, 200
2, 300
2,400
2, SOO
3, 000
3,500
4,000
4, SOO
5,000
5,500
6,000
6, 500
N2
.
0. 31
0.74
2. SK
4. 42
6.27
8. 14
1 0. 02
11. 93
13. 85
15.80
17. 77
19.78
21.79
23. 84
25, 90
27.98
30. 10
32.21
34. 34
56. 48
38.65
40. H4
4 3 . 00
45. 24
47. 46
49. 67
60. 91
72. 31
83. 79
95. 57
107. 04
1 18. 78
132. 54
142. 57
°2
.
0. 31
0, 74
2.61
4. SO
6. 4 3
H. 40
1 0. 40
12.43
14. 49
16. 59
IB. 71
20, 85
23. 02
25,20
27. 40
29. r, 2
51.85
34. 10
36. 34
38.61
40. 90
43. 17
45.47
47. 79
50. 11
52.43
64. 18
76. 1 5
88. 29
100. 64
1 i 5.. 20
125. 89
139,. 74
151. 72
Air
-
0. 32
0. 74
) r j>
4.42
(>. 2't
H, 17
1(1. 07
12, 00
13.95
IS. '?2
17. 92
19, 94
2 1 . 98
24. 0^
2d. 13
2H. .M
30. 58
52.50
34. 66
36. 82
38. <>9
41. 1 H
4 5. (9
•is. (> 1
47. 83
50. 07
61. 39
72. 87
84. 42
96. i 1
107. VI
J 19. 78 .
151.75
143. 76
!12
-
0. 31
0. 73
2, 55
4. 40
*,. 24
H. ()'»
". H'.i
11.77
H. 61
is. 47
17. 56
19. 20
21.08
22, ''5
24. 87
26. 80
28. 70
50. 62
32. 52
34.45
56. 45
58. 49
10. ''7
42. (.6
•14. 71
46. 82
c<7. 22
68. 14
79, 58
90. 6S
102. 42
1 14. 21
126. 16
1 58. 5-
CO
-
0. 32
0.74
2. 58
4. 4 <
6. 2'i
H. IK
10. 08
12. 01
1 5. <>(,
15. 94
17. 94
19. 97
22. 02
2-1. 10
.'(,. 19
28. U
50. 44
32. 58
34. 74
36. 1H
5<)
SO. 23
6 1 . s i
7 3 . 00
B9. Mi
<»6. 2 1
107. V5
i 19. 70
1 51.5,'
14 5. 57
Enthalpies arc for a gaseous system, and do not include latent hr.tt o
  Lv -  1, 059. 9 Btu/lb or 50. 34 Btu/scl' nf IbO vapor at <';0°F ,ind  1-4. »
CO2
0. 39
0. 94
5. 3')
'.. 'IK
S, ii'i
I I . ' • .'
14. 14
17. 45
20. 54
23. 71)
26. 92
50.21
5 5. S'.
5(,. M-,
40. «>
4 i. 8^
47. 5^
50. 8"
54. 48
58, 07
('..!. 71
(.'.. <'••
(,'), 02
r ^ , "M
H>. 4 .
80. 15
98. 96
118. l^
137. t',2
157. 20
17t,. 'H
!'•«.. 7?
2 1 !>. 77
236. 88
>) vaporlKai
96 psia.
1UO"
0. 36
0. 85
2. 98
r-. 14
* . ' "*
'*. '' <'
I 1, HI
14. 1 1
)!». 45
18. K4
21.27
2 i. 74
2»». .',(,
28. H2
51. 42
54. OK
it.. 77
V>. 49
42. 26
4'-. 06
47. '»!
50. 7H
'• i. tiH
;>t<. (.4
Ml. S'-i
62. (.0
77. 'IS
'M, 'i;
t 10. 28
1 ,'J:. 'Ml
14 '.. •'.'
161.1) 7
178. 41
19', 82
r*n .

Source: Air Pollution Engineering Manual, EPA AP-40, 2nd Ed., 1973,
                                             65

-------
                                                Table III
                           Combustion Characteristics of Natural Gas
                                         Average analytic, volume '
                                            CO,           0
                                            02
                                            Cli,
                                            n-C4H10
 0
HI. 1 1
 9.665
 !. 505
 0. 19
 0.24
 0.09
 0.05
100. 00
                                         Average gross heat, i, 100 Btu/ft

                                         Air required (or combustion

                                         Theoretical - 10. 360 ft  /ft  gas
                                         id0",, excess air - II. 432 ft*/ft' Has

                                         Products of comfauation/ft  of gas
                         Theoretical air
Vol
CO2 1. 134 ft'
Hj,O 2.083
N, 8.2)6
U
Tol«! I FT-iTl It *
s
Available heat, Bui/It gas,
Temp, * F
100
150
ZOO
250
300
350
400
450
500
550
600
700
800
900
1,000
t. 100
1, 200
1, (flu
1, 400
1, SOU
1,600
1, 700
1,800
1,900
2,000
2, 100
2,200
2, 300
2,400
2, 500
3,000
3, 500
Wt
0. 112 Ib
0, 099
0. 609

0. K40 Ih

"* baser! on latent he.it
Vol
I. IJ4 ft1
1 . 08 »
9.H7)
u.-m
M.ViMl1

of vapor izat son ol water at
Available heat, tttu, . , . , .
Available heat,
with theoretical air
988.
976.
963.
952.
941-
928.
917.
906.
894.
882.
870.
846.
820.
797.
772.
747.
/21.
f.'li.
66K.
M2.
614.
589.
562.
534.
507,
478.
450.
421.
393,
364.
219,
70.
6
1
7
1
0
B
8
2
6
7
9
2
7
7
6
t.
1
i)
1,
7
Ij
8
)
8
5
7
7
9
0
6
1
4
Wt
0. 1 J2 Ib
0. 099
0, 711
0. (H7
0. 'H>9 Ih

60 "F


992.2
973.0
958. 5
949.9
912.0
917.8
905. 1
891. 5
878.0
864. 1
850,4
821. ft
792. 3
765. 3
736.2
70h, 6
1,76. ',
li'l t. I,
61'.. 4
'.HI. ',
'.S2. '(
521. 7
491.7
459.9
4?S, I
394. 9
362. 5
329. 1
295,6
262.6
94.2
--
            'Average of two «»mple8 analyzed by Southern Calif. C»« Co. , 1956.
Source: Air Pollution Engineering Manual, EPAAP-40, 2nd Ed.. 1973.
                                                    66

-------
en
              Reproduced from
              best available copy.
                       Table IV

Combustion Data Based on 1  Pound of Fuel Oila'b
Const i tuerit
Common Name

Density, Ibs./gal.
Thcorcticnl air
(401 sat'd at 60°F)
Flue gas
constitu-
ents with
theoret-
ical air

CO
so:;
'V
H70 formed
ItO (fuel)
up (air)
Total
Amount of
flue gas
with %
excess
air as
indicated






0
7.5
10
12.5
15
17.5
20
30
40
50
75
100
S0_ % by vol. and wt.
with theoretical air
Approximate Btu/gal.
PS No. 100
Kerosine/
Distillate
6.83
ft3
197.3
26.73
0.002
154.8
28.76
-
1.367
211.659
211.7
226.5
231.4
236.4
241.3
246.2
251.2
270.9
290.6
310.4
359.7
409.0

0.0011
Ib
15.04
3.104
0.004
11.44
1.368
_
0.0662
15.9786
15.98
17.11
17.48
17.86
18.24
18.61
18.99
20.49
22.00
23.50
27.26
31.02

0.0025
136,000
PS No. 200
Straight-run
fuel oil
7.50
ft3
;85.i
27.08
0.142
145.2
22.75
-
1.283
196.455
196,5
210.4
215.0
219.6
224.3
228.9
233.5
252.0
270.5
289.1
335.3
381.6

0.072
Ib
14.11
3.144
0.0240
10.74
1.082
-
0.0621
15.0521
15.05
16.11
16.46
16.81
17.17
17,52
•17.87
19.28
20.69
22.11
25.63
29.16

0.16
142,000
PS No. 300
Low-crack
fuel oil
8
ft3
179.1
27.61
0.130
140.5
19.18
0.011
1.242
188.673
188.7
202.1
206.6
211.1
215.6
220.0
224.5
242.4
260.3
278.3
323.0
367.8

0.069
Ib
13.66
3.207
0.0220
10.39
0.9118
0.0005
0.0601
14.5914
14.59
15.62
15.96
16.30
16.64
16.98
17.32
18.69
20.05
21.42
24.84
28.25

0.15
146. 000
PS No. 400
Heavy -crack
fuel oil
8.33
ft3
177.2
27.86
0.142
139.0
17.86
0.011
1.228
186.101
186.0
199.4
203.8
208.3
212,7
217.1
221.5
239.3
257.0
274.7
319.0
363.3

0.076
Ib
13.51
3.236
0.0240
10.28
0.8491
0.0005
0, 0595
14.4491
14.45
15.46
15.80
16. 14
16.48
16.81
17.15
18.50
19.85
21.21
24.58
27.96

0.17
152,000
                  a. Combustion products calculated for
                     fuel included where indicated.
                  b. Maximum accuracy of calculations:
  combustion with air 40% saturated at 60°F. All volumes measured as gases at 60"F. Moisture in

   1:1,000,
         Source: Air Pollution Engineering Manual, EPA AP-40,2nd Ed., 1973.

-------
    Burner Types

    A great deal of ingenuity has gone into the design of commercial afterburners.
Experience shows that economic and performance advantages accrue to systems that
incorporate uniform and short flame zones, maximum contact between the fumes and
the flame, and intensive mixing. Beyond the initial combustion and mixing zone, the
design features are less critical, so that simple cylindrical or rectalinear sections
tend to be used.  Cooling may have to be provided beyond the combustion zone to pro-
tect the blower and  stack.  This may be accomplished by heat recovery or by the in-
troduction of outside air,

    There are two types of burner  designs, based on arrangement: distributive and
discrete.

    Natural gas and propane have commonly been used for afterburners, since gaseous
fuels  are adaptable  to uniform and short combustion zones.  Distributive burners, shown
in Figure 5A, allow the use of the waste gas stream for combustion air and are compact
and efficient.

    The line-type burner,  shown in Figure 5B, uses a gas manifold or multiport con-
struction which injects gas into a network of divergent openings from a metal plate.
The waste gas enters through perforations in the sidewalls of the plate and is mixed
with the gas by the jet action, forming a short flame zone beyond the plate.  Flame
contact is extended  by this  design.   The metal plate must be highly temperature resis-
tant to avoid damage, and the amount of preheat is limited to approximately 1000°F for
the same reason.

    A variant of the same general type, the multijet burner, is shown in Figure 6.  It
differs in that a' ceramic burner grid may be used. Gas and air enter the conical mix-
ing and combustion  zone from the upstream face.  This type of construction is less
subject to heat damage, but part of the waste gas stream must be bypassed if a stable
flame is to be obtained.  Flame contact is lost with the portion of fume bypassed, and
the combustion chamber must be extended to allow for mixing of the gases beyond the
flame.  Figure 6 shows baffles used to provide mixing.

    Where oil is burned, or only interruptible gas is available, a single premix burner
is commonly used.  The air supplied to the burner may be outside air,  or if the waste
gas stream is reasonably clean it may be withdrawn from the ducting upstream. An
arrangement of this type is shown in Figure 7.  Since flame contact in this type of  sys-
tem is relatively poor, some arrangement is needed to obtain rapid mixing of the flame
with the waste gas.  Baffles may be used for this purpose, or turning vanes in the air
inlet zone. Occasional!}', a tangential inlet for the gas stream is provided in a cylin-
drical combustion chamber.  The artificially induced swirl provides mixing throughout
the combustion zone.

    Afterburners have been constructed in both horizontal and vertical configurations.
The horizontal type is more compatible with heat recovery systems, while the vertical
type (under forced draft) can reduce the cost of the stack.
                                       68

-------
                                                                    To Stack
   Fume
     \E£ZZZZZZZZiZ^2CS.
                                                          ** -rf * * .-lf.lCjf.il'
    X

    X
       u
     Natural
      Gas
                                                                   1 C-f-fTi-^ *"J J '
                     A.  Afterburner with Distributed Burner
BURNER
PROFILE
OPENING
 PROFILE
 PLATE
 INCINERATOR
   HOUSING

                           COMBUSTIFUME BURNER
                        B. Maxon Combustifume Burner
    Source:  Afterburner Systems Study, Shell Development Company, 1972.

                      Figure S. Maxon Combustible Burner
                                      69

-------
Natural
 Gas
                             Natural
                              Gas
                                    A. Hirt Multijet Gas Burner
                                            \\\\\\\\\\ \\X\\\\\\ \\\\\\\\\
  \\\\\\\

Exhaust
                                                        Baffles
                                                           \\X\\\\ \\\\\\\\\\\\\\\\\\\\\
                           Fume
                           B. Afterburner System Employing Multijet Burner
Source; Afterburner Systems Study, Shell Development Company, 1972.


                                   Figure 6. Hirt Multijet Gas Burner
                                                70

-------
               Fume
     Fuel-
         Combustion
            Air
          (Fums)
» Exhaust
Source: Afterburner Systems Study, Shell Development Company, 1972

                        Figure 7. Afterburner Using a Discrete Burner
    Accessories and Controls

    Direct-flame incinerators will require thermocouple temperature sensors, used
to monitor;

    *  Inlet waste gas'temperature.  If the waste gas is preheated, a preheat bypass
       control is sometimes used to prevent the temperature from going above safe
       maximum inlet values.

    *  Temperature in the combustion area. Temperatures much in excess of the re-
       quired minimum result in higher fuel and maintenance costs; for this reason,
       the fuel injection rate is controlled through a burner control that incorporates
       a sensor.
                                   "X
    »  Temperatures downstream from the combustion area, especially  if heat re-
       covery is a part of the system.

    In addition, sensors and controls are frequently provided to monitor or control
pressure drops across the incinerator and combustion fires.  The safeguards used
against combustion fires include flame detectors, automatic shutdown provision in
case of flame-out, and pressure switches.  These devices are based on approval by
safety and fire protection organizations such as the Underwriters Laboratory, toe.,
the Factory Mutual System, the Factory Insurance Association, and the National Fire
Protection Association,
                                       71

-------
  CATALYTIC INCINERATION

       The catalytic incinerator differs from the direct-fired unit in that an active cata-
  lyst is used to reduce the ignition temperature of the solvent in the waste gas stream.
  Since less heating of the incoming gases is required,  fuel consumption is reduced.
  Further benefits include the smaller combustion chamber and the reduced equipment
  maintenance associated with the lower operating temperatures.

       Structurally, a  catalytic incinerator differs from a direct-fired unit in that a
  burner  system in a preheat chamber is used to raise the temperature of the incoming
  gases to 600°F - 900°F.  The hot gases are then passed through a catalyst bed where
  the fume is burned,  releasing further heat and elevating the gas temperature to 800° F-
  1100°F.  Since no flame contact is involved, the preheat section is frequently a dis-
  crete burner followed by a simple mixing zone, although a distributive burner may also
  be used.  Little or no combustion occurs beyond the burner, so the residence time at
  peak temperature can be quite short.  The net result is a  somewhat smaller physical
  system than the direct-fired unit.  Figure 8 is a diagram of atypical catalytic incinerator.

  Gas Conditioning

       The catalysts used in catalytic incineration, normally platinum or palladium.,  are
  extremely sensitive to contamination and catalyst failures are common. Heavy metals
  such as mercury, arsenic,  lead, and zinc will inactivate the catalytic surface. Plastic
  resins and tar like materials may coat the catalyst, as may otherwise-inert materials
  such as dusts and metallic oxides. As a result,  the Incoming waste gas stream must
  be completely free of materials of this type.  Where doubt exists as to the contaminants
  in the gas stream, it is wise to use thermal incineration.

       Where the advantages of catalytic incineration appear to warrant the expense of
  preeleaning gases, fabric filters or electrostatic precipitators may be used for dust
  and scrubbers may be used for either particles or heavy liquid droplets. Water vapor
Fume Stream
  704QO°F"
Preheat Catalyst
Burner Element
'=f^isjjz~ 600-900°F
I
800-1 00°F

                   Clean Gas
                   to Stack
                     Combustion/Mixing
                         Chamber
 Optional Heat
  Recovery
(Regenerative or
Recycle System)
  Source: Afterburner Systems Study, Shell Development Company, 1972.

                        Figure 8. Schematic of Catalytic Afterburner System
                                          72

-------
will not affect  the  incinerator,  although the temperature loss associated with  water
scrubbing will  increase fuel requirements.

Catalytic Combustion

     The degree of solvent vapor oxidation that can be expected on a catalyst is affected
by the vapor composition, the reaction temperature, the surface area of the catalyst,
and the degree of contact between the solvent and catalyst. The last two variables will
depend on the commercial design of the catalyst and its support structures.  A gener-
alized curve showing the relative effect of catalyst volume/flow ratio on combustion
efficiency is given in Figure 9.  It is normal practice to follow the manufacturer's
recommendation concerning catalyst volume for a given application, since much of the
applications technology is based on field experience.
 100
  so
5


I
  40
  20
                   0.5               1.0               >.S              2.0

                         Volumt of Catalyst/Volumitric How Rate of Wasti Strum (Relative I
2.5
Source: Afterburner Systems Study, Shell Development Company, 1972.

              Figure 9, Combustion Efficiency as a Function of Catalyst-Volume/Flow Ratio
                                          73

-------
    Both the solvent used and its concentration influence the design preheat tempera-
ture for the catalyst.  Figure 10 shows catalyst temperatures required for combustion
of typical solvent hydrocarbons with some of the common combustible gases.  These
temperatures must be modified to reflect the concentration of solvent. As may be seen
from Figure 11, lower vapor concentrations are considerably more difficult to oxidize
because of chemical kinetics and the lack of additional heat gain from the catalyst.

    Commercial catalysts consist of small quantities of platinum or platinum/palladium
alloys  deposited on metallic or  ceramic support structures. One manufacturer electro-
plates  the active metal onto  the surface of fine nichrome ribbon that has been geometri-
cally packed to obtain a high surface area and void ratio.   The nichrome structure is
quite inert and will withstand temperatures up to 1500"F without damage,   Several
                              400
   600
Temptrature, °f
800
1000
                                          1200
Source: A fterburner Systems Study, ShelI  Development Company, 1972.

            Figure 10. Typical Temperature-Performance Curves for Various Molecular Species
                          Being Oxidized Over Pt/AI2O3 Catalysts
                                          74

-------
-a
01
                                                                            Solvent continuation

                                                                            in wast* slrtam
                                                                               600               700

                                                                           Preheat Temperature, op
1000
            Source:  Afterburner Systems Study, Shell Development Company, 1972.



                                               Figure 11.  Effect of Solvent Concentration on Required Preheat Temperature

-------
other manufacturers use a ceramic base coated with a porous wash coat of aluminum
oxide. The catalyst is deposited on or in the porous surface coat in the form of small
crystals.  The base structure may be palletized or, more commonly, formed into a
honeycomb structure.  The aluminum-oxide-based form provides somewhat greater
catalyst activity than the all-metal form.

Process Design Principles

    For catalytic incinerator process design, it will be necessary to determine the
degree of preheat required for the solvent fume, emission concentration,  and the ef-
ficiency desired. The subsequent heat and volume calculations are similar to those
for thermal incinerators, except mat the heat is released in two stages.  The cross-
sectional area of catalyst required is based on manufacturers' recommended face
velocities, which range from about 5 to 35 feet per second.  A typical calculation of
the fuel requirements and temperature profile follows:

Example 5—Given:   A waste gas stream of 4500 acfm at 200°F cc-itains SOOppm of
                    hexane.  The gas stream is to be catalytically incinerated with a
                    desired efficiency of 90%.  Natural gas is available as a fuel.

           Find;    The preheat fuel required, and the temperature and rate of gas
                    flow at the exit of the catalyst.

    1.  Since natural gas is available, a distributive preheat burner will be used to
        avoid the introduction of  outside air.               „

    2.  From Figure 10,  it may  be  seen that a catalyst temperature of approximately
        900°F should be sufficient.  Figure 11 confirms this temperature, at hexane
        concentrations of less than 10% LEL.   Since the temperature rise over the
        catalyst will be small with only SOOppm of hexane, the conservative assump-
        tion of 900°F preheat temperature will be used to size the burner. The natu-
        ral gas flow will actually be slightly less than the calculated value.

    3.  Waste gas flow in scfm

           ._ftn   460 + 60
        =  4500 x ———r—
                  460 + 200

        =  3545 scfm

    4.  Heat Input (neglecting contribution from hexane)

        =  (Available heat at 900°F, 0% excess air) x G

           + credit for initial heat of combustion air to 200°F

        where G = scfm natural gas required
                                       76

-------
    *  From Table HI, available heat at 900°F


       = 797.7 Btu/scf


    *  Credit for initial heat

                  (2 f*y Q \ T*
       = G x 10.36    .     x enthalpy difference (200°-60°)
                  scfgas         w

    *  From Table H, enthalpy difference (200*- 60")


       = 2.58 Btu/scf air


    *  Thus, heat input


       = 797.7 xG + Gx 10.36x2.58


       » 824.4 xG Btu/min


5.  Heat Consumption


    =  (scfm waste gas - scfm needed for combustion) x

       enthalpy change (900° F - 200" F)


    *  From Table E, enthalpy difference (900°F-200°F)


       = 2.58
                  »

    •  Thus, heat consumption


       = (3545 - 10.36 x G) (15.92 - 2.58) (from Table 11)


       = 47290- 138.2 xG


6.  Heat Balance: Heat Input = Heat Consumption


    *  Thus, 824.4 G = 47290 - 138.2 G


    •  Solving for G,  G = 49.1 scfm natural gas


7.  The combustion gas at this point will consist of:


    (i)  air  = 3545 - 10.36x49.1 = 3036 scfm


    From Table in, products of natural gas combustion


    (ii) CO, = 1.134 scf/gas x 49.1 = 55.7 scfm
                                   77

-------
     (iii) H,O = 2.083 scf/scf gas x 49.1 = 102.3 scfm

     (iv) N2 = 8.236 scf/scf gas x 49.1 = 404.4 scfm

 8.  Heat available from hexane burning over catalyst

     *  Hexane flow rate = 3545 x -~r- = 1.064 scfm

     •  From Figure 4, at 9QO°F  and 0% excess air, approximately 72% of gross
        heat from hexane is available.  Gross heat value for hexane is 4762 Btu/scf
        (from Table I)

     •  Heat available from hexane

        = 0.72 x 4762 x  1.064 Btu/min

        = 3648 Btu/min

 9.  Combustion air required to burn hexane (from Table I)

     =  45.26 scf air/scf hexane x 1.064 scf

     =  48.2 scfm

     Products of hexane combustion (from  Table I)
                                                      *
     CO,  = 6.0 x 1.064 = 6.4 scfm

     H2O = 7.0 x 1.064 = 7.5 scfm

     N2 = 35.76 x 1.064 = 38.0 scfm

10.  Combustion gases after the catalyst will consist of (from steps 7 and 9):

     (i)  air = 3036 - 48.2 = 2987.8 scfm

     (ii) CO2 = 55.7 + 6.4 = 62.1 scfm

     (iii) H2O = 102.3 + 7.5 = 109.8 scfm

     (iv) N2 = 404.4 + 38.0 = 442.4 scfm

11.  Temperature at the end of the catalyst zone can be calculated by equating heat
     available from hexane (step 8) to the heat consumed in temperature increase
     in the combustion gases from step 10.

     •  Assuming a linear enthalpy change between 900°F and 1000"F for air, CO,,
        H2O, and N2, the heat consumed by combustion bases (from Table II)
                                    78

-------
           3 2         x 2987.8 scfm air at AT
               100 F
              3.22    ;o~  x 62.1 scfm CO, x AT
                    luu F
             + 2.43 ~~' .1  x 109.8 scfmH,O x AT
                    100 F

             + 1.97 ^'ff'f x 442.4 scfm N, x AT
                    100 F

           = 73.14 x AT Btu/min

        *  Equating this to heat available from combustion of hexane (step 8)

           73.14X AT = 3648

                    T = 50°F

        *  Thus the final gas temperature

           = 900 + AT

           = 950°F

   12.  The final now rate, from step 10

        =  2987.8 + 62M + 109.8 + 442.4

        =  3602 scfm
        =  9767 acfm at 950°F

    These are typical calculations for sizing a catalytic afterburner.

Accessories and Controls

    The accessories and controls for a catalytic incinerator are similar to those re-
quired for a direct-fired unit and include a temperature-regulated preheat burner sys-
tem and a flame sensor for emergency shutdown.  The burner control may be activated
by the post-catalyst temperature to smooth out fluctuations in final temperature due to
solvent variations.

    Catalysts tend to become less effective with time, even with a clean waste-gas
stream.  The normal life will vary from a few months to 2 or 3  years, depending on
the nature of the waste load.  As the  incinerator becomes less efficient,  operating
temperatures must be increased to offset the decreasing catalyst activity.  Frequent
                                       79

-------
or continuous effluent monitoring for hydrocarbons or carbon monoxide is the only way
to be sure that emission regulations are being met unless the unit is operated at exces-
sive temperatures.  However,  continuous monitoring is expensive, as is overheating.

PROCESS BOILERS

     The use of boilers—existing or planned—for fume combustion ma}7 enable significant
reductions in both capital and operating costs.  The initial saving of the cost of a fume
incinerator will be  somewhat offset by the cost of boiler modifications and ducting.
However, if the distance between the fume source and the boiler is not great, long-
term savings can be substantial.  An important factor is circumventing the cost of
additional fuel for air pollution control only. In incineration, fuel is the major cost;
further, the lighter grades used (gas or distillate oil) may be difficult to obtain.

     There are a number of conditions that must be met for a boiler to function satis-
factorily for fume incineration:

     *  The fume should be almost completely combustible. If not, the solids present
        will either foul the heat exchange surfaces or  cause the boiler emissions to ex-
        ceed applicable particulate emission regulations.  If there are significant quan-
        tities of solids in the waste gas, the costs of increased maintenance  of the
        boiler and/or control of the particulates may well exceed  the purchase price of
        a fume incinerator.

     *  The contaminated gas stream should, preferably,  constitute only a small frac-
        tion of the air requirements of the boiler.  If the volume of the gas stream is
        large, special attention must be paid to the oxygen balance, mixing,  and con-
        tinuation of the air flow when the fume-emitting process is shut down.

     •  The oxygen concentration of the contaminated  gas  stream  should be close to that
        of air to avoid incomplete combustion, which can produce tars that coat heat
        exchanger surfaces.

     •  The boiler must operate at all times when fume incineration is required.

     *  The fumes must be free of compounds,  such as halogenated hydrocarbons, that
        accelerate corrosion of the boiler.

    In addition:

     »  Baffling  may be  required in the combustion chamber to ensure adequate mixing
        and combustion of the fumes without bypassing.

     * If the boiler-firing rate varies greatly, it may be worthwhile to install a small
       auxiliary boiler  that will operate under steady load conditions to produce a base
       quantity  of steam and serve as a fume incinerator.

     Before a process boiler is used for fume incineration, a careful analysis should
be made of the operations involved.  In a new facility, it is generally possible to plan
the layout and characteristics of individual boiler units to ensure that thev will be
                                        80

-------
economical and effective.  In existing facilities some problems are inevitable, and an
objective analysis is necessary to make certain that projected economies are realistic.
                              VAPOR ADSORPTION

    Although adsorption of organic vapors is a proven technique for their control, its
use in the metal coating industry for solvent vapor control is not common, for the fol-
lowing reasons;

    *  The low concentrations of solvent vapors in gas streams from metal coating
       operations make this technique very expensive.

    •  Flow rates associated with metal coating are large and the large adsorption
       units necessary for vapor removal are not cost-effective.

    *  Vapors from metal coating operations contain mixtures of solvents, and the
       solvents used frequently change, depending on coating needs.  This makes
       solvent recovery by adsorbers impractical because of high costs of solvent
       separation, unless the solvent mixture can be used elsewhere as a secondary
       fuel.

    *  Process gas must be cooled to less than 100°F for adsorption, requiring addi-
       tional cooling equipment.

    *  Adsorption beds are extremely susceptible to fouling from particulate matter
       in the process gas. The effective life of the adsorption medium can thus be
      prohibitively low.

    *  Under certain  circumstances, low-temperature adsorption systems may re-
       quire corrosion-resistant construction materials, increasing the initial outlay.
                                       81

-------
                                 CHAPTER II

                           HEAT  RECOVERY


    Since fume incineration involves heating significant flows of air to temperatures as
high as 1500°F, the potential for heat recovery is obvious.  For large flows in thermal
incinerators, heat recovery is nearly always economically justified.  Catalytic incin-
erators operating at lower temperatures offer less opportunity for heat recovery, as
do small incinerators where the increased capital requirement may exceed the value of
the heat recovered over a reasonable period of time.

    Heat may be recovered and reused in a number of ways.  The most common is pre-
heating the incoming contaminated gas stream to directly reduce the fuel needed for the
desired operating temperature in the incinerator (primary heat recovery). Where the
fume source is an oven, part of the incinerator exhaust gas may be used to heat the
oven, either directly or indirectly. Heat may also be recovered as either hot air or
steam for use in plant operations unrelated to the fume source (secondary heat
recovery).

    The choice of whether to use heat  recovery, and in what form, depends on a num-
ber of process and economic factors.  These are summarized in Table V. In this
table, "effectiveness ratio" is the percentage of actual heat reoovery relative to that
theoretically obtainable.

    Primary heat recovery for reducing fuel required for incineration is almost uni-
versally accomplished by using either  cross-flow tubular or regenerative heat ex-
changers.  The tubular type provides relatively low-cost cross-flow configurations for
the low-pressure differentials encountered in incineration processes.  Single- and
double-stage units are illustrated in Figure  12.  Since the cost is almost directly pro-
portional to the number of stages, tubular exchangers are most commonly used in
single-stage applications,  where 40-50 percent heat recovery is considered adequate.
The units  are sensitive to fouling of the tube surfaces and are difficult to clean because
of the complex arrangement of tubes.  They should, therefore, be avoided where soot,
tars, or possible polymerization products are present in the stack gases.

    For high-efficiency heat recovery, regeneration by rotary heat exchangers is most
commonly used. An example of this type is shown in Figure 13. A wheel with large
surface area and sufficient bulk for a large heat capacity is rotated between the hot and
cold gas streams.  Heat is captured by the portions of the wheel exposed to the hot gases
and lost as the wheel is rotated into the cool gas stream. Since the wheel is constantly
exposed to heating and cooling, warping and thermal stress are potential problems in
metal wheels and sophisticated construction is needed to  overcome them. In recent
years, however, ceramic materials with low thermal expansion (and freedom from
corrosion) have been successfully used for the wheels.
                                        82

-------
                                    Table V
                     Comparison of Heat Recovery Techniques
     Type
Effectiveness
   Ratio, E
Additional
Auxiliary
Equipment
Limitations and
   Problems
Common Use
 Tubular
 exchanger
 gas/gas
1 sta-50% max
2 sta-62% max
3 sta-85% max
I Regenerative
I(rotary)
 exchanger
  up to 85%
            1. May be easily
               fouled; frequent
               cleaning and
               maintenance.

            2. Failures,  differ-
               ential thermal
               expansion.

            3. Hot surfaces may
               crack or polymerize
               fume components,
               lay combustible
               deposit, initiate a
               fire.

            4. Bulky, heavy, added
               roof load and/or
               floor space.

            5. Corrosion if cools
               below dew point of
               flue gas.
                   Primary,
                   secondary
                   heat recovery
            1. Easily fouled.  Use
               only on relatively
               clean streams.

            2. Burnout if failure on
               rotary drive motor.

            3. Requires attention
               to pressure balance
               to control leakage
               at seals.

            4. Avoid cooling flue '
               gas to dew point,
               but  otherwise is
               relatively insensi-
               tive to corrosion.

            5. Ignition if overheat
               fuel-rich stream.
                    .Secondary heat
                    recovery, to
                   |heat air enter-
                    ing oven
                                       83

-------
                                Table V (continued)
f"""""- 	 -• 	 —- 	 	 — 	 _— 	 	 	 	 	 	
Type
; 	 	 	 	 	
* 	
Flue gas
recycle
to oven










Steam,
generators ,
boilers,
water
h (inters






Heat pipe


Effectiveness
Ratio, E

30%












to 75%









Lo (>0%



Additional
Auxiliary
Equipment

Safety
Controls











Extra burn-
ers and con-
trols; safety
controls
Extra duct-
ing, blowers
controls







Limitations and
Problems

1 . Process must be
compatible with flue
gas (condensation?
sulfur in fuel? CO
or CO , ? reduced
oxygon? unburned
fuel?)

2. Usefulness depends
on temperature and
heat requirements
of fume generating
process.
1. Ties steam genera-
tion to fume process
and vice versa.
2. Match steam heating
load to afterburning
heat release.

3 . Dew point and con-
densation on cold-
water coils.
1. Can only be used for
hot side tempera-
ture up to 800° F

Common Use














Secondary heat
recovery








To preheat oven
air, secondary
heat recovery
     jd from; Afterburner Systems Study, Shell Development Company, 1972.
    Rotary heat exchangers are capable of very high efficiency and offer significant
cost and space savings over other high-performance heat exchangers.  Fouling can be
a problem  if  stick materials are handled, but inert dusts are tolerated at reasonable
concentration. Cleaning is easier than with tubular-type exchangers.

    The principal operating problem with rotary exchangers has been leakage from the
seals separating the hot and cold gas streams.  Since these seals must tolerate motion
while subjected to temperature extremes, finding a substance and a design that would
give reasonable life expectancy was  technically difficult.  However, in recent years
manufacturers have managed to overcome this problem, at least for  relatively clean
gas streams.

-------
           A. Shell and Tube Type Exchanger Cross-Flow Type
                                                      Cooltd
                                                      Flue
                                                      Gas
  B. 2-Pass, Cross-Flow Exchangers (Arranged to Place Units Counter-Flow)
Source: Afterburner Systems Study, Shell Development Company, 1972.

            Figure 12.  Typical Shell and Tube Heat Exchangers
                                 85

-------
                 Rotating
                  Matrix
Source: Afterburner Systems Study, Shell Development Company, 1972.

                        Figure 13. Rotary Regenerative Heat Exchanger


    A heat recovery technique known as heat pipo is also commonly used.  Here a re-
 rigerant inside a series of tubes circulates between the hot and co.'d sides (Figure 14).
\s the liquid refrigerant enters the hot side of a tube, it absorbs heat front the hot
jases and evaporates. The evaporated refrigerant then gives up its heat to the cold
ur stream and condenses.  In this way, heat can be transferred from hot gases to eold
Air.  Heat pipes are used for  secondary heat recovery purposes like comfort heating.
\ limitation of the heat pipe technique is that the maximum hot-side temperature can-
lot be much in excess of 800°F.  For this  reason,  the technique is used in conjunction
vith ovens on coating lines that have water-based or powder-deposition paint technolo-
gies, where incineration of the oven exhaust is not required.

    In cases where additional process steam or heat is required in a plant, it may be
more economical to  let the fume incinerator serve the dual function of controlling air
                                         86

-------
                hot side
cold side
           hot gases
            evaporation
  condansation
tube tilled
with refrig-
erant
                                                          cold air
                                  Figure 14, Heat Pipe
pollution and providing process heat. If separate boilers are constructed for process
heat and for heat recovery to reduce fuel consumption in the  incinerator»  the total cost
will be higher than for a single burner and heat recovery unit.  For true economy,
plant operations must be such that incineration of waste gases would not be necessary
at times when the heat recovered from them could not be used in processes.

    The process heat or steam available from afterburners of even moderate capacity
can be substantial, as is shown in Figure 15.  Recovery may be accomplished by burn-
ing the contaminated gas stream in a boiler, by using an afterburner followed by only
the heat exchange portion of a boiler, or by using a conventional gas-liquid heat ex-
changer to produce a hot-fluid stream for process use.

    Where the fume source is an oven, it is common to circulate part of the exhaust
gases from the afterburner back to the oven to provide either part or all of the heat
requirement.  Care must be taken to prevent the exhaust gases from harming the prod-
uct being processed in the oven.  Temperature control may require sophisticated ex-
haust and outside air blending, but safety is enhanced by a low oxygen atmosphere in
the oven. The oxygen needs for fuel combustion prevent use of a totally closed system
and result in more exhaust gas than can be used in the oven.  Thus, the total heat re-
covery efficiency is low.  However, exhaust-gas recirculation can be used with heat
recovery for other purposes to obtain a high overall recovery efficiency.
                                        87

-------
    20
    16
  - 12
  9
  &
                                                                       Flue Gas Temperature
                                                                                                  15000F
                                                                                                 14000F
                                                                                                 130QOF
                                                                                                 12QO°F
                                                             Note:
                                                             Finit Temperature of Flue Gas = 350°F
                                                             Fume Strewn Used for Oxygen Supply
                                                               (No Outside Air Used in Combustion)
                             4,000
        8,000

Afterburner Capacity, scfm
12,000
16,000
Source: Afterburner System Study, Shell Development Company, 1972.

                             Figure 15. Process Heat Recoverable from Afterburner
                                                      88

-------
    The following calculation shows the effect of primary heat recovery on afterburner
fuel requirements in a sample case:

Example 6—Calculate the fuel savings expected by the application of a heat exchanger
           of 50% effectiveness, as shown in the schematic:

                             1400°F               Qt  Oven

                                                 A:  Afterburner

                        T                        P:  Primary Heat Exchanger

using data described in Example 2 (oven exhaust =  3000 acfm at 300°F, containing
500ppm toluene).

    1.  Heat exchange effectiveness

0
300° F

,

P




A
          _
where:  C    = specif ic heat of gas , Btu/°F

        TIO  = Oven exhaust temperature entering the heat exchanger,  °F

        TQO = Oven exhaust temperature leaving the heat exchanger, "F
        TIA  = Afterburner exhaust temperature entering the heat exchanger, °F

        TQA = Afterburner exhaust temperature leaving the heat exchanger, °F

To avoid using Cp values between unknown temperatures, the above expression can be
converted in terms of enthalpies, H:

            Hoo~Hio
        e = -
            HIA ~HIO

Where subscripts have  the same meaning as above.

        •  In this example, e  = 0.50

    2.  Oven exhaust temperature leaving the heat exchanger, TQO

        *  Assuming all gases are air, from Table II

               HOC, -4. 42
           e = _°2 - _ =  o.50
               26.13-4.42

           HOQ = 15.28 Btu/scf
                                       89

-------
        *  From Table n, for air, the corresponding temperature

           TOO * 868 °F (by linear interpolation)

    3.  Heat input, as in Example 1 and 2

        = (Available heat at 1400°F at 0% excess air) x G

          + (Credit for preheat of combustion air from 60° F to 868° F)

          + (Available heat from toluene)
        =  668 G + 15.28      x 10.36  SC* &ir x G
                         scf          scf gas

           + 0.61 x 4605 + 44.04 x 15.28

        =  826 G + 3482 Btu/min

    4 .  Heat consumption at the afterburner

        =  (2053 -  10.36 x G - 44.04) x (26.13 - 15.28)

        =  21798 - 112.4 x G Btu/min

    5.  Heat Balance: Heat Input = Heat Consumption
                                                          *

           826G + 3482 = 21798 - 112.4 x G

        *  Solving for G ,

           G = 19.5 scfm natural gas

A reduction in fuel consumption from 43 . 3 scfm (Example 2) to 19 . 5 scfm is achieved
by the introduction of a heat exchanger of 50% effectiveness.

    The final afterburner exhaust temperature (leaving the heat exchanger), TQA> can
be calculated as follows;

    6.  Heat Balance: Heat absorbed by oven gases in the heat exchanger - heat lost
        by afterburner exhaust, or

        Tl    "tt   — W   W
          OO    IO ~  1A    OA

        15.28 - 4.42 = 26.13 - HCA

        HOA  =15.28
                                       90

-------
        •  From Table E,  for air, the corresponding temperature

           TOA« 868°P

The heat contained in the incinerator exhaust at 868°F can be utilized for oven heating
as well as to meet a portion of total plant steam needs.  If the incinerator exhaust is
used for steam production in a waste heat boiler, and if we assume the final incinerator
exhaust temperature at 350°F, the heat available for steam production can be calculated
as follows:

    7.  Heat available for  steam production

        =  2053 scfm x (H868 -H350 )

        =  2053 x (15.28 -  5.36) from Table H

        =  20,366 Btu/min

        -  1.22x 106 Btu/hr

    This would be roughly equivalent to l,0001bs/hr of low pressure steam.

    The choice of a heat recovery system should be made after analyzing the inciner-
ator heat capacities and the total heat needs of the plant, using the steps below as a
guide:

    1.  Determine the-heat recoverable from the incinerator exhaust. This heat can
        be used for preheating the oven or for generating steam, heating water, or
        comfort heating.

    2.  Determine the preheat that can be applied to the oven exhaust, that is, the
        highest amount compatible with the safe  maximum oven exhaust temperature.

    3.  Determine the heat that can be recycled  to the oven as incinerator exhaust by
        using assumed exhaust to outside-air ratios and the known oven temperature
        requirement.  In considering this as an option, the effect of incinerator ex-
        haust on product quality must also be evaluated.

    4.  Determine the purposes for which heat recovered from the incinerator exhaust
        could immediately be used.  During periods of incinerator operation when there
        is no  hot water or  steam demand in the plant, the value of heat recovery is
        limited. Heat recovery for comfort heating can be used only in the winter
        months.

    5.  Determine the heat load from steam, hot water, and comfort heating, based
        on past history of  fuel burned for these purposes.

    If the total heat demand from steps 2,3, and 5 is comparable to the amount of heat
recoverable, then the plant should consider heat recovery units with high effectiveness
ratios (Table V). If the heat demand is small compared to the recoverable heat, heat
exchangers with lower effectiveness can be used,  thereby reducing capital costs.

                                       91

-------
                                CHAPTER III

                            OF                       AND

                   HEAT  RECOVERY  SYSTEMS

    The combustion of dilute organic fumes in air streams is one of the most expensive
forms of air pollution control,  In smaller units, installed costs may range up to $25
per cfm capacity.  Operating costs are also high because of the fuel consumed in heat-
ing air streams to high temperatures.  Heat recovery will reduce fuel consumption but
entails higher capital investment.  Selecting a combustion and heat recovery system
for a specific use requires careful consideration of capital and operating-cost tradeoffs
to insure a minimum overall cost,

    Installation costs for thermal incinerators range from 75 to 200 percent of the cost
of the basic equipment, averaging about the same as equipment costs.  Installation costs
will normally consist of installation labor and materials, plus auxiliary equipment that
consists of:

    *  Ducting;
    *  Blower motor controls and instrumentation;
                                                         *
    •  Insulation;
    *  Blower motor housing for noise control;

    *  Instrumentation,  including air pressure regulator, temperature monitoring and
       recording, flame safety controls, and fuel rate monitoring; and
    *  Foundations and structural steel.

    The installed costs of thermal and catalytic incinerators, with and without heat ex-
changers , are shown in Figure 16.  Estimates were based on roof-top location with al-
lowance for structural steel and assuming custom-designed units.  The cost, therefore,
majr be considered  as in the upper limits in the less-than-10,000 cfm range.  Pre-
engineered units with a few thousand cfm capacity may be installed at a total cost of
approximate!}' two-thirds that of a custom-designed and fabricated unit.

    Costs of operation will depend on the number of shifts,  the temperature of the con-
taminated air stream, and the incineration temperature.  Variable cost factors are
fuel and electricity consumed and labor for operation. Maintenance, taxes, insurance,
and overhead charges on space are commonly taken as a fraction of capital costs.  Fig-
ure 17  shows estimated operating costs for various incinerator configurations.  All
costs are based on  an 8-hour single shift operation,  inlet fume temperature of 70°F,
electricity at $0.03/kWh, direct operating labor of 0.5 hr/shift, and miscellaneous
                                       92

-------
    275-
    250
A. Thermal without heal exchanger
8. Thermal with primary heat exchanger
C. Thermal with primary and secondary heat exchanger
0. Catalytic without heat exchanger
E. Catalytic with primary heat exchanger
    225
    200
    t7S

I
•5
     125
    100
      75
     50
                                           10
                                         15
                                   Flow, sctm x  1fl3
20
25
30
                                      Figure 16.  Capital Cost of Incineration
                                                         93

-------
245
220
A.  Thermal without heat exchanger
6  Thermal with primary htat exchanger
C.  Thermal with primary and secondary heat exchanger
D.  Catalytic without heat exchanger
E  Catalytic with primary heat exchanger
195
170
145
120
 95
 70
 45
 20
                                     10
                                        15

                                 How, scfm x 103
20
25
30
                           Figure 17. Annual Variable Cost of Incineration
                                                  94

-------
costs of 9 percent of capital.  The operating cost would increase by slightly less than
factors of 2 or 3 for two- or three-shift operations.

    The total annual cost of ownership may be derived from Figures 16 and 17 by com-
bining a suitable fraction of the capital cost with the annual operating cost.  At current
interest levels, the annual cost of capital is commonly taken as 14-18 percent of total
investment.  The data in Figures 16 and  17 have been left uncombined to facilitate the
approximation of total annual costs for multishift operations.
                                        95

-------
    The first part of this publication was concerned with reducing hydrocarbon emis-
sions at  the source through  changes in  coating formulas  and plant equipment and
processes.

    In this part we have discussed, in some detail, the end-of-line treatment of emis-
sions that must be dealt with after all other practicable control measures have been
taken.

    Throughout the entire volume the emphasis has been on presenting pollution con-
trol techniques as sets of options,  with enough information on each method—pro and
con, descriptions, diagrams, and  simple calculations—for assessing its suitability for
a given plant.

    In both sections of this publication, plant managers,  engineers, and operating per-
sonnel should find the basic information they will need to plan, in logical sequence,
for reduction and treatment of hydrocarbon emissions from metal coating processes.
                                       96

-------