&EPA
         United States
         Environmental Protection
         Agency
          Industrial Environmental Research EPA-600/7-78-167
          Laboratory         August 1978
          Research Triangle Park NC 27711
Combustion
Research on
Characterization
of Particulate
Organic  Matter
from Flames

Interagency
Energy/Environment
R&D Program Report

-------
                  RESEARCH REPORTING SERIES


Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application  of en-
vironmental technology. Elimination of traditional  grouping was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

    1.  Environmental Health Effects Research

    2.  Environmental Protection Technology

    3.  Ecological Research

    4.  Environmental Monitoring

    5.  Socioeconomic Environmental Studies

    6.  Scientific and Technical Assessment Reports  (STAR)

    7.  Interagency Energy-Environment Research and Development

    8.  "Special" Reports

    9.  Miscellaneous Reports

This report has been assigned  to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded  under the  17-agency  Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development  of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include  analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments  of,  and development of, control technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
                        EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for  publication. Approval does not signify that the contents necessarily reflect
the  views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                     EPA-600/7-78-167
                                            August 1978
         Combustion Research
on  Characterization  of  Participate
    Organic Matter from Flames
                        by

                 R.A. Hites and J.B. Howard

             Massachusetts Institute of Technology
              Department of Chemical Engineering
               Cambridge, Massachusetts 02139
                   Grant No. R803242
                      Task No. 1
               Program Element No. EHE624A
              EPA Project Officer: John H. Wasser

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

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

-------
                                  ABSTRACT

     The formation and emission of soot and polycyclic aromatic hydrocarbons
(PAH) from both laminar flames and a turbulent continuous flow combustor were
studied.  The amount of soot and PAH produced in the latter system was dimin-
ished by a decrease in the atomizing air pressure.  Benzene fuel gave more
soot and PAH than did kerosene, and the PAH from benzene were more substituted.
Nitrogen and sulfur doped fuels did not affect total soot production.  The
distribution of soot and PAH within the flame was consistent with the concept
that certain PAH may serve as intermediates in soot formation.

     The PAH formed during combustion of coal, wood, and kerosene were sepa-
rated and identified by capillary column gas chromatography and mass spectro-
metry.  The PAH from coal combustion were found to be similar to airborne
PAH from Indianapolis, a high coal consumption area, while the PAH from kero-
sene combustion were similar to airborne PAH from Boston, an area of low coal
consumption and high consumption of petroleum derived fuels.
                                                        9
     Mixed charge exchange chemical ionization mass spectrometry allows the
differentiation of PAH isomers on the basis of ionization potential.  This
method may be able to predict the mass spectra of presently unidentifiable
isomers.

     GC-MS analysis of carbon blacks found a wide range of PAH including
sulfur containing aromatics which had not previously been identified in carbon
blacks.   The quantity of PAH was somewhat dependent on conditions of forma-
tion.

     Nitrogen containing fuels primarily produce polycyclic aromatic compounds
in which the nitrogen is in a cyano substituent (approximately 70%).  This
contrasts with the observation that nitrogen heterocyclic compounds are the
major nitrogen containing compounds in airborne particulate matter.

     This report was submitted in fulfillment of Grant No.  R803242 by the
Massachusetts Institute of Technology under the sponsorship of the U.S.
Environmental Protection Agency.  This report covers the period August 1,
1974 to July 31, 1977, and work was completed as of September 30, 1978.
                                       ii

-------
                                  CONTENTS

Abstract
Overview  ............................. •  1

Appendices

     A.  Soot and Hydrocarbon Formation in a Turbulent Diffusion
         Flame  ................. ..........  5

     B.  Formation of Soot and Polycyclic Aromatic Hydrocarbons in
         Combustion Systems •. .................... 19

     C.  Soot Concentration Measurements in Fuels Doped with
         Nitrogen and Sulfur Containing Compounds .......... 41

     D.  Source Identification of Urban Airborne Polycyclic
         Aromatic Hydrocarbons by Gas Chromatographic Mass
         Spectrometry and High Resolution Mass Spectrometry ..... 64

     E.  Mixed Charge Exchange-Chemical lonization Mass
         Spectrometry of Polycyclic Aromatic Hydrocarbons ...... 70

     F.  Charge Exchange-Chemical lonization of Polycyclic
         Aromatic Compounds ..................... 72

     6.  Characterization of Sulfur-Containing Polycyclic
         Aromatic Compounds in Carbon Blacks  ............ 76

     H.  Cyano-Aromatic Compounds Produced by the Combustion of
         Nitrogen Containing Fuels  ................. 81
                                     iii

-------
                               FIGURES
Number

   A-l Water-flushed soot sampling probe                           7
   A-2 Influence of atomizing-air pressure on axial profile of
       soot mass loading (cold gas  velocity = 0. 96 m/s)              9
   A-3 Influence of atomizing-air pressure on axial profile of
       soot mass loading (cold gas  velocity = 2. 67 m/s)              9
   A-4 Influence of atomizing-air pressure on axial profile of
       PCAH mass loading                                        10
   A-5 Influence of the type of fuel on axial profile of soot mass
       loading                                                    10
   A-6 Influence of the type of fuel on axial profile of PCAH
       mass loading                                              10
   B-l Summary of proposed mechanisms of soot formation         23
   B-2 Molecular beam mass spectrometer system                 25
   B-3 Schematic of mass spectrometer instrumentation            27
   B-4 Schematic of gas  introduction system for studying
       mass discrimination                                       29
   B-5 Profiles of signal intensities relative to argon for major
       stable species in  an acetylene-oxygen flame near the
       sooting limit                                              34
   B-6 Profiles of signal intensities relative to argon for minor
       species in an acetylene-oxygen flame near  the sooting limit  35
   C-l Overall experimental arrangements                         44
   C-2 Burner details                                             45
   C-3 Water-cooled sampling probe                               47
   C-4 Variation of soot  concentration with height above burner
       for each of four additives in benzene                         48
   C-5 Variation of soot  concentration with height above burner
       for pyridine additive in benzene                             51
   C-6 Variation of soot  concentration with height above burner
       for pyrrole additive in benzene                             52
   C-7 Variation of soot  concentration with height above burner
       for thiophene additive  in benzene                            53
   C-8 Variation of soot  concentration with height above burner
       for carbon disulfide additive in benzene                     54
   C-9 Variation of soot  concentration with height above burner
       for pure methane fuel with no additive                       56
                                  IV

-------
Figures (Continued)

Number                                                        Page

   C-10 Variation of soot concentration with height above burner
        for thiophene and pyrrole additive in hexane                57
   C-ll Variation in soot concentration with height above burner
        for each of three additives in benzene and a lower
        equivalence ratio                                         58
   D-l  Capillary-column gas chromatogram of the PAH fraction
        of coal combustion products                               67
   D-2  Capillary-column gas chromatogram of the PAH fraction
        of wood combustion products                              67
   D-3  Capillary-column gas chromatogram of the PAH fraction
        of kerosene combustion products                           68
   D-4  Alkyl homolog distribution plots for several PAH series
        in coal combustion products                               69
   D-5  Alkyl homolog distribution plots for the pyrene-type series
        in the combustion products of coal, wood, and kerosene, and
        in air particulate matter from Indianapolis  and Boston       69
   D-6  Composite alkyl homolog distribution plots  for PAH in the
        combustion products of coal, wood, and kerosene, and in
        air particulate matter from Indianapolis and Boston         69
   E-l  Plot of the abundance ratio obtained by CH4-Ar chemical
        ionizatiqn mass spectrometry as a function of ionization
        potential for a series of four tetracyclic polycyclic aro-
        matic hydrocarbons                                       71
   G-l  Compounds identified in carbon blacks by GC/MS and HRMS  77
   G-2  Packed-column gas chromatograms of furnace black 1 and
        furnace black 3                                          79
   G-3  Glass capillary-column gas chromatogram of the extract
        of furnace black 1                                        80
   H-l  Gas chromatograms of fraction 3,  isolated from organic
        compounds produced by combustion of solution of 32%
        pyridine in o-xylene                                      84

-------
                              TABLES

Number
 A-I  Effect of Injected Water on Gas Composition                 8
 A-n  Computed and Experimental Soot Mass Loading             11
 A-HI Polycyclic Aromatic Hydrocarbons  in the Combustion
       Products of Kerosene and Benzene                         12
 A-IV Distribution of PCAH for Kerosene  and Benzene Fuel
       Along the Burner Axis                                    13
 A-V  Arithmetic Mean Particle Diameter (Kerosene and
       Benzene; Cold Gas Velocity = 0.96  m/s)                    14
 A-VI Arithmetic Mean Particle Diameter (Kerosene; Cold Gas
       Velocity = 2.67 m/s)                                     14
 B-I  Masses at Which Signals Were Observed but Positive
       Identifications Were Not Made                            37
 B-n  Additional Species Near the End of  the Reaction Zone in
       a C2H2/C-2 Flame                                       39
 C-l  Information on Benzene and Additives                      50
 D-l  PAH Identified by G. C. M. S.                              68
 E-I  Abundance Ratios for Selected PAH Obtained by CH4-Ar
       Chemical lonization Mass Spectrometry                    71
 F-l  Some of the Polycyclic Aromatic Compounds Investigated    74
 G-I  Carbon Black Characteristics                             79
 H-I  Fractionation of Soot Extracts                            83
 H-H  Mass Spectra of Peaks 1-7 in El and CE/CI Modes           84
                                 VI

-------
                                  OVERVIEW

     Polycyclic aromatic hydrocarbons (PAH) are ubiquitous in the environment.
Many specific PAH have been found to be carcinogenic in animals.   Since PAH
are present in small particulates, they are potentially extremely dangerous
to human health since they can easily be inhaled deep in the respiratory
system.  Consequently PAH are highly suspect in the relationship  of lung
cancer and atmospheric.pollution.

     Since different !PAH vary widely in their toxicity and carcinogenic
effects, from harmless to extremely hazardous, it is essential to be able
to identify the specific PAH in a mixture.  Since common sources  of PAH
contain only minute quantities of many different species, the combination
of gas chromatography and mass spectrometry (GC-MS) is extremely  well suited
to resolving PAH.  The mass spectrometer provides identification  through
empirical formulae and the partial elucidation of major structures.  Gas
chromatography, since it is very sensitive to shape, allows the separation
and quantification of isomers.  Unfortunately, one difficulty encountered
in exact structure determination is the inability of mass spectrometry to
discriminate between isomeric PAH.

     Since the major source of airborne PAH is combustion products, variables
affecting PAH and soot production in combustion systems may have  great
effects on airborne pollution.  The increased use of aromatic fuels (from
coal) and additional controls on NO  production may result in increased
                                   /\
amounts and different distributions of PAH.  Knowledge of the variation of
PAH production within a flame as well as the mechanism operating  in its
formation will aid in attempts to limit emission of these compounds.

     The general goal of this project has been an improved understanding of
the production of soot and PAH in combustion systems.  Toward this end,
studies on a variety of combustion systems were carried out.  The types of
combustor were varied as were the fuels; samples were collected and analyzed
for PAH.  In some cases, improvements in the state-of-the-art analytical
tools were required.  All of our work has been published in the open litera-
ture and these reprints from the body of this report.  The following para-
graphs are a summary of our work and a guide to the appended reprints.

     Soot and PAH formation in a turbulent flame were observed, while para-
meters such as mixing intensity, fuel equivalence ratio and type  of fuel were
varied to determine the effects on total mass, composition and size distribu-
tion of PAH produced.  Lowering the atomizing air pressure resulted in a
10,000 fold increase in soot emission.  (Appendix A, Fig. 2).  A  similar
effect was observed in the profile of PAH (Fig. 4).  This effect  may be attri-
buted to improved mixing of fuel and air with increased pressure.  In general,
but with some exceptions, more intense mixing and hence increased Op consump-

                                      1

-------
tion in the first section of the flame and decreased combustion in the
later portions, results in decreased soot formation.

     The presence of anthracene in the fuel has little effect on the amount
of soot or PAH formed in the turbulent diffusion flame.  (Appendix A, Figs. 5
and 6).  However, benzene combustion produced significantly less PAH and soot
than the combustion of kerosene.  This surprising result was attributed to
the flame temperature, and hence the rate of soot burnout, being higher in
the benzene case.

     The distribution of specific PAH was also studied (Appendix A, Table 3).
Both kerosene and benzene showed enrichment of higher molecular weight PAH
further from the nozzle.  Kerosene produced markedly less PAH containing
methyl and phenyl groups.  These substituted PAH decreased with increasing
distance from the nozzle at a faster rate than did the unsubstituted PAH.

     Benzene produced slightly smaller particles of soot than did kerosene.
Increased cold gas velocity also produced smaller particles.  Both these
effects may be caused by a higher flame temperature:  the adiabatic flame
temperature of benzene is higher than that of kerosene, and higher gas
velocities reduce heat loss and raise flame temperature.

     These results indicate that poorly maintained combustion sources such
as home heating burners may operate under conditions resulting in high soot
formation.  The use of staged combustion for NO  emission control may also
dramatically increase PAH and soot production,  if the first stage is too
fuel rich.

     A molecular mass beam spectrometer system has been developed for study-
ing molecular distributions within a flame (Appendix B).  Aromatic hydro-
carbons have been observed under conditions of high soot formation.  However,
the present sensitivity is too low owing to a high background level of hydro-
carbons.  This should be improved by the addition of cryogenic pumping in
the system to decrease aliphatic hydrocarbons.

     Studies were also conducted burning nitrogen and sulfur containing com-
pounds in a combustion system comparable to that used for PAH studies (Appen-
dix C).  Fuels containing one percent nitrogen or sulfur were obtained by
doping benzene with pyridine, pyrrole, thiophene, or carbon disulfide.  Soot
production was found to be independent of additives.

     Since PAH from different combustion sources differ considerably in type
and concentration, a correlation between fuel utilization in an area and
composition of air particulate matter would be expected.  The airborne PAH
of high and low coal burning areas, Indianapolis and Boston, were compared.

     PAH emissions from combustion or coal, wood, and kerosene were analyzed
by capillary column GC-MS to identify and roughly to quantitate the components
(Appendix D).  There were several significant differences:  (1) there is a
greater relative abundance of alkylated PAH in the coal combustion products;
(2) higher molecular weight species are relatively more abundant in wood and
kerosene combustion products; (3) sulfur containing compounds are present only
in the coal soot PAH fraction.

-------
     Alkyl homolog plots were constructed for seven groups of isomers
(Appendix D, Figure 4).  These plots represent the distribution of the alkyl-
ated species within a PAH mixture.  The curves for the pyrene series show a
correlation between the curves for the coal PAH and Indianapolis air particu-
late PAH, as well as between the kerosene PAH and the Boston sample.

     The major differences between coal PAH and Indianapolis PAH are that the
concentrations of high molecular weight compounds in soot are higher than
those of airborne PAH.  This may be due to the sensitivity of these compounds
to combustion and collection conditions.

     Since the spectra of isomeric PAH are almost identical, GC-MS cannot
unequivocally assign structures to many compounds.  In charge exchange
chemical ionization mass spectrometry, the mass spectra are observed in the
presence of argon and methane.  The molecule of interest.reacts either with
an argon ion (Ar+) or with an ethane carbonium ion (C9H(-+).  The ratio of
                 +                                      +
molecular ions (M ) and protonated molecular ions (M + 1 ) differs for each
isomer and can be correlated with the ionization potential (Appendix E).
Variations in the ionization potential only affect the rate of ionization by
Ar+ ions, and do not affect the rate of protonation by CpHr+.  Since ioniza-
tion potentials can be calculated for a particular isomer from molecular
orbital theory, this approach holds promise to identify presently unknown
PAH.  The correlation also holds for several aromatic nitrogen containing
compounds as well as the two sulfur compounds that were available (Appendix
F>-

     Carbon black, a major component of automobile tires, has a wide environ-
mental distribution.  GC-MS analysis of four different carbon blacks shows a
broad distribution of PAH, somewhat dependent on conditions of formation
(Appendix G).  The higher temperature of formation of one of the carbon blacks
reduced the total PAH by a factor of ten.  The distribution of compounds was
not affected markedly by conditions of formation, although sulfur analogues
of PAH were detected for the first time in carbon blacks produced from sulfur
containing petroleum fractions.
                                                               )
     Nitrogen containing aromatic compounds (aza-arenes) are produced in
combustion of fuels with significant nitrogen contant (Appendix H).  Up to
this time these compounds have not been widely studied, although they may
become important pollutants with the increased use of nitrogen containing
fuels.  Little is known about the distribution of neutral aza-arenes since
the standard analysis procedure only extracts bases.  We have used charge
exchange chemical ionization mass spectrometry to analyze the neutral aza-
arene fraction of compounds produced by the combustion of high nitrogen
fuels (Appendix H).  The spectra of 1- and 2-cyanonaphthalenes were identified,
and confirmed by the use of authentic samples.  Other peaks were tentatively
identified as isomers of cyanoacenaphthylenes and cyanophenanthrenes.  These
three classes of compounds accounted for 80% of the aza-arenes in the fraction
studied.  Nitrogen heterocyclic compounds formed only a minor component (less
than 30%).  Neutral nitrogen compounds have not been evaluated for biological
activity or for environmental effects.

     The knowledge gained about the distribution and production of polycyclic

-------
aromatic compounds can promote the effort to control their emissions as well
as help evaluate their potential harm.

-------
                 APPENDIX A
      Soot and Hydrocarbon Formation in a
            Turbulent Diffusion Flame
From:   Sixteenth Symposium (International) on
       Combustion, The Combustion Institute,
       Pittsburgh, 1977.  Reprinted with the
       permission of the copyright holder.

-------
       SOOT AND HYDROCARBON FORMATION IN A TURBULENT
                                DIFFUSION FLAME
          G. P. PRADO", M. L. LEE, R. A. HITES, D. P. HOULT, AND J. B. HOWARD00
     Departments of Chemical and Mechanical Engineering, Massachusetts Institute of Technology,
                                 Cambridge, Massachusetts.
       Formation and emission of soot and  polycyclic aromatic  hydrocarbons (PCAH) from
      a turbulent continuous flow combustor have been studied. Measurements included mass
      concentration of both soot and PCAH, composition of individual PCAH, and size distribution
      of soot particles, as a function of mixing intensity, fuel equivalence ratio and type of fuel
      (kerosene or benzene). Both soot and PCAH concentrations reach maxima early in the flame,
      after which PCAH decays rapidly and soot decays much  slower. The maximum  PCAH
      concentration always preceded that of soot in agreement with the concept that certain PCAH
      may serve  as intermediates  in soot formation. An approximate calculation based on the
      assumptions of local equilibrium with respect to soot formation and a Gaussian distribution
      of air-fuel mixedness gave satisfactory correlations of the data on soot formation at the higher
      cold gas velocity. At the lower velocity, the amount of soot and PCAH formed was drastically
      increased and strongly dependent on fuel atomization.
               1. Introduction

  In  gas turbine and most other practical
combustors, fuel and air are fed separately and
the combustion occurs primarily within a rela-
tively confined region. Turbulent mixing rates
in  such systems have important  effects  on
emission of pollutants such as  nitric oxide,
carbon monoxide, soot and polycyclic aromatic
hydrocarbons (PCAH).  The present investiga-
tion was focussed on soot and PCAH formation
in a  turbulent diffusion flame. The under-
standing of the formation of these  species is
important in relation both to flame radiation
and to  air pollution. Knowledge of the con-
centration  of soot  particles  is necessary  to
predict flame emissivities for radiative heat
transfer problems.  In addition, some PCAH
  "On  leave from:  Centre de  Recherches sur la
Physico-Chimie des Surfaces Solides, Centre  Na-
tional de la Recherche Scientifique, 24 Avenue du
President Kennedy 68200-MULHOUSE, FRANCE.
  00 To whome correspondence should be sent.
Room 66-454, Department of Chemical  Engineering,
Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139.
known  for their carcinogenic properties  are
associated with soot particles, and because
these small particles are easily injested deep
into the respiratory  system, they may be one
of the types of particulates most hazardous
to human health. Also, the NC^ control strate-
gies such as lower-temperature and/or staged
combustion presently being considered tend
to increase PCAH and soot formation. Use of
highly aromatic fuels made from coal  would
focus increased attention on emissions of par-
ticulate organic matter.
   In  turbulent flames, the nonuniform  dis-
tribution  of fuel in  air can be represented as
pockets of gas or  eddies of different fuel-air
ratios, some of which are sufficiently fuel rich
to form soot and PCAH while others are fuel
lean, compared to the mean. Although the air
and fuel  eventually may  become mixed by
molecular diffusion, soot and  possibly PCAH
emissions will still occur if the rate of mixing
is slow enough to allow the amount of such
materials  formed to  exceed that which can be
burned out in the  residence time available.
  In recent studies at M.I.T.1-2 of the role of
turbulent mixing  on nitric oxide and carbon
monoxide  formation, nonunifonnities  asso-
ciated with mixing  were  allowed by  either

-------
assuming that burnt gas eddies have a normal
distribution of equivalence ratio  about the
mean value, or by representing the mixing by
a large number of fluid "particle" interactions
simulated by a random number generator.
  The objective  of the present work was to
study  the  role of turbulent mixing  in the
formation of soot  and PCAH in the same
combustor used previously to characterize ni-
tric oxide and carbon monoxide emissions.
             2. Experimental

  The burner used in this study is described
elsewhere.2 It is basically a cylindrical burner
in which  the combustion air is supplied at
atmospheric pressure at  one end through  a
45-degree blade angle swirler. Kerosene or
benzene was supplied through an air assist
atomizer (De La van model 3060-9-11) on the
burner axis.  The atomizer nozzle air-pressure
drop was controlled independently of the fuel
flow rate.  The atomizer air flow amounted to
about 5%  of the total air flow.
  Gas samples  containing soot and  PCAH
were collected at different flame positions with
a stainless steel  water-cooled probe  (Fig. 1)
in which about 1% of the cooling water was
injected into the sampled gas at the probe tip
in order to quench the reactions and to prevent
deposition of soot along the walls. The mova-
ble tube permitted control of the amount of
water injected into the probe. This  design
allowed quantitative collection of the soot  and
condensibles without tedious scraping of the
probe after each sample is  taken. A similar
probe was previously designed and used else-
where.4 The sample  stream,  comprised  of
water, gases  and participates, passed  through
a glass fiber filter where the soot and essential-
ly all the  PCAH were deposited. The water
and gas then passed through a condenser for
     further cooling,  after which  the water was
     collected in a large flask and the gas passed
     through a wet test meter after flowing through
     an ice bath condensate trap and vacuum pump.
     For the preliminary runs, the amount of soot
     and  associated  PCAH  was  determined by
     weighing  the dried  filter  before  and   after
     collection. The filter and soot deposit was then
     extracted with methylene chloride and  the
     filter was again dried and weighed. The soot
     was thus classified as carbonaceous residue
     (referred to as soot  in the text) and methylene
     chloride soluble  material. The carbonaceous
     residue was removed from the filter, dispersed
     in an ultrasonic  bath and deposited on per-
     forated carbon electron microscope grids for
     particle size distribution analysis.  A Philips
     E.M. 300 microscope,  containing  an   anti-
     contamination device was used. The size dis-
     tributions  were  determined  by  measuring
     about 2,000 particles per sample, using a  semi-
     automatic Zeiss particle counter. The reprodu-
     cibility was about 5% with the same operator,
     20% with different  operators.
       The methylene chloride extract of flame soot
     was concentrated to a volume of  from 0.1 ml
     to 10 ml, depending on  the concentration of
     organics,  and analyzed  by  gas chromato-
     graphy. A Perkin-Elmer 3920 gas chromato-
     graph  with  flame   ionization detector and
     model-l computing integrator was  used for
     quantitative analysis  of individual  PCAH. A
     standard solution consisting of phenanthrene
     (representative of 3 ring compounds), pyrene
     (4 rings), perylene  (5 rings) and  coronene (7
     rings) was used for calibration of the integrator,
     and each chromatographic peak in a sample
     run  was compared  to the standard compound
     with the closest retention time. A  1.83  m (6
     ft) x 3.175mm (1/8 in) O.D. stainless steel
     column packed with 3% OV-17 on 100/120
     mesh Gas Chrom Q was used throughout these
     experiments. After  injection of a  sample, the
Sample
  In
	___  Water  Out
ESS^SSSSSSSS
                     Water  In
             x\\xxxw\\x^-—Movable  Tube
                      Sample + Water
                      to  Filter
                          Movable  Tube
                         Fie. 1. Water-flushed soot sampling probe.

-------
gas chromatograph  was programmed  from
70°C to 320°C at the rate of 12°C/min.
  Identification of individual compounds was
accomplished with a Hewlett-Packard 5982 A
gas chromatograph/mass spectrometer which
is interfaced with an HP 5933 A data system.
The same gas chromatographic conditions
were used as previously described for quan-
titative measurements. Mass spectral data were
automatically acquired during a gas  chroma-
tographic run and stored  on disc for future
interpretation. Structural assignments of indi-
vidual PCAH were made from mass spectral
data and relative gas chromatographic reten-
tion as compared with standard compounds.
  The separation of phenanthrene and anthra-
cene isomers was accomplished by high reso-
lution gas chromatography using a  19  m  X
0.26mm, i.d., glass capillary column coated
with SE-52  rriethylphenyl silicone stationary
phase. The  column was  programmed  from
70°C to 170°C at 2°C/min after sample injec-
tion.
  The reliability  of the sampling technique
was tested by performing reproducibility tests
on the amount of soot collected. The  reprodu-
cibility was  better than  ±10% when  the sam-
pling time was long enough (10 minutes  to
one hour depending on conditions) to collect
at least 20 mg of material.
  Another concern was the possible oxidation
of PCAH and soot particles by the injected
water. Such  oxidation would cause a  decrease
of the weight of products collected, and possi-
bly a change of  the size of the particles  if
they do not burn  internally. The effect on the
amount of products  collected  could not  be
determined, due to sticking of material within
the probe when no -water  was  introduced.
However, gas analysis with and without water
injection  (Table  I)  indicated no significant
reaction between the water and the gas.  In
general, the  addition of about 3 1 per hour
of water inside the probe gave rise to slightly
                 TABLE I
   Effect of injected water on gas composition
(volume%). Kerosene; fuel equivalence ratio = 1.0;
distance from nozzle = 10 cm; cold gas velocity =
 0.96 m/s; atomizing-air pressure = 184 kPa (12
                   psig)

no water
3 1 water /hour
CO
2.8
3.0
C02
12.0
11.3
02
1.25
1.60
higher CO and O2 concentrations and slightly
lower CO2 concentration. This was probably
due to better quenching of the CO/CO2 reac-
tion with  water than without.  In light of this
observation and  the  short  quenching time
(~-10~4s),  reaction between water and the solid
materials  collected should not be important.
                3. Results

  Three types of experiments were carried out,
using aviation kerosene as  fuel. The total  air
flow was always held constant at  56.7 kg/hr.
The mean  fuel equivalence ratio was varied
by  changing the fuel  flow rate.  Then, at a
constant mean fuel equivalence ratio  of 1
(stoichiometric), the atomizing air pressure
was varied in order to change the intensity
of fuel-air  mixing within the burner. Three
atomizing air pressures were studied: 184 kPa
(12 psig), 205 kPa (15 psig) and 239 kPa (20
psig). The variation of atomizing air pressure
was carried out for two cold gas velocities:
0.96 m/s and 2.67  m/s. The cold gas velocity
was varied at constant total flow rate by using
two different burner diameters, 12.70 cm and
7.62 cm. Finally, at a fuel equivalence  ratio
of 1, a cold gas velocity of 0.96 m/s and  an
air  atomizing  pressure of 184 kPa, kerosene
was replaced  by  benzene and  by benzene
doped with 0.5% by weight anthracene. These
fuels were chosen in  order  to compare the
behavior of a pure aromatic fuel with that of
aviation kerosene which contains about 17%
by weight monoaromatic compounds and 1%
PCAH.

3.1 Soot and Total PCAH Mass Loading

  Soot concentrations, along  the  burner axis
between 4cm from the nozzle and the exhaust
(53 cm from the nozzle) for the case of kerosene
fuel are shown in Fig. 2 for a cold  gas velocity
of 0.96 m/s, and in Fig. 3 for a cold gas velocity
of 2.67 m/s.  For the lower velocity, the total
PCAH concentrations are shown in Fig. 4. It
was not  possible  to obtain reliable data  on
total PCAH concentration for tha runs at higher
velocity, due to the obscuring of the chroma-
tographic  PCAH   profile  by   appreciable
amounts of unburned  kerosene all  along the
axis of the burner.
  For a cold gas velocity of 0.96 m / s, it appears
that soot formation occurs very early in the
flame, well within  the recirculation zone. The
soot concentration  profile  exhibits a strong
increase between 1/2 and  1 burner diameter

-------
 

 2

     10
        0   10   2O   30   40   50   60   70
           DISTANCE  FROM  NOZZLE (cm)

  Fie. 2.  Influence  of  atomizing-air  pressure  on
axial profile of soot mass loading. Kerosene/air; fuel
equivalence ratio = 1.0; cold gas velocity = 0.96
m/s; atomizing-air pressure = (o) 184 kPa (12 psig),
(D) 205 kPa (15 psig), and (A) 239 kPa (20 psig).
from the nozzle, followed by a slow decrease
to the exhaust for the 205 and 239 kPa cases.
  At 184  kPa the profile does not exhibit a
maximum, the strong increase being followed
by a smaller one. The profiles of PCAH also
exhibit a maximum, just upstream of the soot
maximum. However, the decrease  in  PCAH
concentration is  more drastic than  for  the
carbonaceous residue, indicating  that either
PCAH are intermediates of soot formation, or
their burnout is more rapid than that of soot.
For  this  cold  gas  velocity (0.96 m/s),  the
atomizing air pressure was found  to have a
strong effect on soot loading all along the axis
of the burner, particularly at the exhaust. For
example,  Fig. 2 indicates that a decrease in
atomizing air pressure from 239 to 184 kPa
results in  a 20 fold increase of the  maximum
of soot formation and a 104 fold increase in
soot emission.
  From these mass loading profiles, the flame
can be divided into two distinct zones, in the
first of which (first 10 cm  from the nozzle)
the soot formed and in the second of which
soot combustion occurred. For  a cold  gas
velocity of  2.67 m/s, these two  zones were
 found to have rather different characteristics
 than in  the 0.96 m/s case. The maximum of
 soot formation was lower, much less depen-
 dent  on  atomizing  pressure,  and  located
 downstream of the recirculation  zone (about
 15 cm from the burner). Also,  more burnout
 of soot occurred at lower rather than at higher
 air atomizing pressure.
   The profiles of soot and PCAH when burn-
 ing pure aromatic fuels (benzene and anthra-
 cene-doped benzene) are shown in Figs. 5 and
 6 respectively. The profiles  obtained  with
 kerosene are  also  shown for comparison. It
 appears, within the limit of reproducibility,
 that  anthracene has very little effect on the
 amount  of soot or PCAH formed. The dif-
 ference  at 4, 6 and 8 cm between benzene
 and anthracene-doped benzene results mainly
 from the presence  of  unburned  anthracene,
 as shown by the gas chromatographic analyses
 described below.  The  comparison  between
 kerosene and  benzene  shows  also  that  the
 combustion of benzene generates PCAH and
 soot  particles i) earlier in the process and  ii)
 in larger amounts (about three times as much)
     250
                                                  200 -
     150 -
 O
 O
o
o
z
6
o
     10O -
              10    20     30     00     50

             DISTANCE  FROM NOZZLE  (CM)
  Fic. 3. Influence of atomizing-air pressure on
axial profile of soot mass loading. Kerosene/air; fuel
equivalence ratio =  1.0; cold gas velocity = 2.67
m/s; atomizing-air pressure = (o) 184 kPa (15 psig),
(D) 205 kPa (15 psig) and (A) 239 kPa (20 psig).

-------
 at the maximum of formation. On the other
 hand, the amount of PCAH formed with ben-
 zene decreases at a faster rate than with kero-
 sene to  reach a  lower total  amount near  the
 exhaust. This may be due to the fact that the
 adiabatic flame  temperature for benzene is
 higher  than for  kerosene, resulting in more
 severe pyrolysis  and post-combustion  of  the
 PCAH. For example, the adiabatic flame tem-
 perature for benzene and kerosene are respec-
 tively: 2330 K and 2281 K at a fuel equivalence
 ratio of 1, and 1363 K and  1107 K at a fuel
 equivalence ratio of 3.
   Finally, increasing the fuel equivalence ratio
 resulted in  an increase of the amount of soot
 formed  as  shown in Table II.  It was  not
 possible to measure the total amount of PCAH
 for these conditions, due to obscurring of the
 gas chromatogram by unburned kerosene, as
 pointed out above. However, it was possible
 to measure  the higher mass compounds. For
 example, the  concentration  of  cyclopen-
 ta[cd]pyrene  (C18H10)  had  a larger depen-
 dence  than  soot  on  variations  of  fuel-
 equivalence ratio.
                                          10"
      01
      E
      t
      (j
      a.
      U.
      o
      o
      z.
      a
      <
      O
10 -I-
           0    10
           DISTANCE
            20   30   40   5O
            FROM NOZZLE (cm)
  FIG. 4. Influence of atomizing-air  pressure  on
axial profile of PCAH mass loading. Kerosene/air;
fuel equivalence ratio  = 1.0; cold  gas velocity =
0.96 rn/s; atomizing-air pressure = (o) 184 kPa  (12
psig),'(D) 205 kPa (15 psig) and (A)  239  kPa  (20
                                      O
                                      O
                                      in
                                      Z
                                      Q

                                      O   10"
                                      _i
                                          1O
          Anthracene
          Doped  Benzene
                                             0   10   20   30  40  50   60  7O
                                               DISTANCE  FROM  NOZZLE (cm)

                                       FIG. 5. Influence of the type of fuel on axial profile
                                     of soot mass loading. Fuel equivalence ratio = 1.0;
                                     air atomizing pressure = 184 kPa (12 psig); cold
                                     gas velocity = 0.96 m/s.
                                     en
                                     E
                                     (J
                                     Q.
                                              -z.
                                              O
                                              I/I
                                              I/)
Anthracene
Doped Benzene
Benzene
                                                  10  --
                                                  10
                                                 10   20
                                               DISTANCE
     .30   40   50   6O   7O
     FROM  NOZZLE ( cm)
                                       Fie. 6. Influenceof thetypeof fuel on axial profile
                                     of PCAH mass loading. Fuel equivalence ratio =
                                     1.0; air atomizing pressure =  184  kPa  (12 psig);
                                     cold gas velocity = 0.96 m/s.
                                            10

-------
                 TABLE II
  Computed and experimental soot mass loading.
     Kerosene; cold gas velocity = 2.67 m/s;
   air-atomizing pressure = 184 kPa (12 psig);
     distance from nozzle = 15 cm;  = fuel
   equivalence ratio; a = standard deviation of
     Gaussian distribution of fuel/air ratios.

       <1>        0.8   0.9   1.0   1.1   1.2

experimental
  values (mg/sm3) 96   125  230  255   410
computed values
  (mg/sm3) a
  = 0.8
60    80  105   190   280
computed values
  (mg/sm3) a/
  = 0.8
 6    30  105   400   800
kerosene at these short distances from  the
burner nozzle.
  These tables form the data base for the total
amount of PCAH plotted in Fig. 4 and 6 and
described above. Some additional conclusions
can be drawn:
  a) With kerosene as well as with benzene,
     there is an enrichment of heavier PCAH
     as samples are collected at greater dis-
     tances  from the burner nozzle.
  b) There  is a gross reduction  and in some
     cases  an absence  of  PCAH  containing
    . methyl and phenyl groups for kerosene
     fuel as compared to benzene.
  c) In both cases the disappearance of PCAH
     containing methyl and phenyl groups and
     those containing a saturated carbon (such
     as fluorene)  with  increasing  distance
     from the burner  nozzle is faster than the
     disappearance of more stable PCAH.
3.2 Individual PCAH Analysis

  Although the total mass of PCAH found in
the combustion  products is useful in  under-
standing general trends in PCAH  production
for variovis systems,  this information is  not
particularly useful in determining the health
effects   of  these   products.   It   is  well
documented 5i6|T that the majority of PCAH are
noncarcinogenic; however, those that are car-
cinogenic range in activity from  very weak
to extremely potent. Because of this fact, recent
emphasis has been placed on the determination
of the exact  identities and concentrations of
these potential carcinogens in exhausts repre-
sentative of those that may be of environmental
importance.
  Gas chromatographic-mass  spectrometric
analysis of the methylene chloride extracts of
the soot  samples,  and subsequent peak area
integration of the chromatographic peaks have
provided detailed quantitative and qualitative
coTTipositional information on the polycyclic
aromatic  hydrocarbons  formed within  the
flame. Table  III lists the identities and struc-
tures of  these compounds as determined  by
gas chromatography-mass spectrometry. Quan-
titative values for 26  PCAH  ranging from
two- to seven-ring structures  measured at  dif-
ferent points along the burner axis at  an ato-
mizing air pressure  of 184 kPa are reported
for kerosene and  benzene in Table IV. The
lack of data  at 4, 6 and  8 cm  is  due to the
obscurring of the  chromatographic  PCAH
profile by appreciable amounts of unburned
                            3.3 Particle Size Analysis

                              Selected samples were subjected to particle
                            size analysis  by  electron  microscopy  at  a
                            magnification of 81,000, which was brought
                            to 200,000 by further photographic magnifica-
                            tion. Two types of carbonaceous material were
                            observed. Close to the nozzle,  in the zone of
                            soot formation, the particles looked like agglo-
                            merates of partially coalesced spherical units.
                            This type of material was observed previously
                            in different systems, namely a flow reactor
                            for the  thermal decomposition of  benzene,8
                            a propane-oxygen premixed flame,9 and a die-
                            sel  engine.10 It has been  attributed to  the
                            coalescence of viscous carbonaceous particles
                            through  collisions  occurring possibly in the
                            sampling system.8'11 It has been suggested that
                            such material may indicate the evolution of
                            soot particles through a liquid state to a solid
                            material  by dehydrogenation and carboniza-
                            tion reactions.8 Alternatively,  rapid surface
                            growth  during particle agglomeration might
                            contribute to the observed shape. The individ-
                            ual spherical units in this  material were very
                            poorly defined  and it was not possible to
                            measure their size. Farther from the nozzle,
                            the material had the classical  appearance of
                            necklace-like units of spherical  particles, indi-
                            cating that the collisions occurred mainly be-
                            tween already solidified particles. The size of
                            these well defined spherical units are reported
                            in Tables V and VI. In these tables, the poorly
                            defined  material refers either to the above
                            described coalesced material or to a strongly
                            oxidized material (Table VI, 239 kPa, exhaust)
                            having a different structure than the coalesced
                                      11

-------
  10
  12
 13
                                            TABLE III
          Polycyclic aromatic hydrocarbons in the combustion products of kerosene and benzene.
                    The structure of number 11  is unknown, its mass is 176 (C14H8).
Number           Mame

  1          Indene


  2          Naphthalene


  3          Hethylnaphthalene
            Biphenyl
            Biphenylene


            Acenaphthylene


            Acenaphthene
Methylbiphenyl


Fluorene


Hethylacenaphthylene


Phenanthrene
     +
Anthracene


Methylphenanthrene
                         Structure
                            "  cl^"
                                 "CH,
                                              CH.
 14   4H-cyc1openta[def]phenanthrene

 15        2-phenylnaphthalene


 16        Fluoranthene
Number

  17


  18


  19


  20
     Name
                                                                         Structure
Benzacenaphthylene


Pyrene


Benzofluorene


Methylfluoranthrene    &
                                                     Methylpyrene


                                           21    Benzo[
-------
                                         TABLE IV
  Distribution of PCAH (nig/sni') lor kerosene ami benzene fuel alonji the burner ;i\is. X—Unable to
              measure Ix-cause of unburned kerosene peaks; K—kerosene: B—Ix'ir/.ene
Distance from burner nozzle (CMII)
Compound
Number"
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18°
19
20
21
22
23
24
25
26
27
28
29
Total
4
K
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
2.4
X
6.9
X
X
0.9
4.4
X
X
X
X
X
X
X
75b
B
8.9
117
55
135
38
121
4.9
8.1
32
24
20
52
10
20
9.0
26
17
24
12
11
11
21
5.8
3.6
4.3
1.8
4.2
4.0
1.5
802
6
K
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
7.5
X
12
X
X
4.9
11
X
X
X
X
X
X
X
130b
B
12
151
88
160
57
162
6.8
16
62
39
43
108
19
38
19
51
35
49
25
26
24
45
15
14
9.4
8.7
8.5
6.3
2.6
1302
8
K
0
X
X
X
X
X
X
X
X
X
X
X
X
X
X
8.7
X
16
X
X
6.7
16
X
X
X
X
X
X
X
179b
B
3.8
6.3
28
107
25
131
3.0
5.9
30
15
14
62
5.8
18
8.3
31
20
31
9.8
9.8
13
29
5.2
8.0
5.7
4.8
9.0
9.2
3.0
709
K
0
X
9.0
7.5
22
185
10
0
27
7.0
22
53
0
19
0
35
21
55
2.4
6.8
17
53
4.0
11
6.6
3.0
5.4
17
6.4
603
10
B
0
42
12
84
18
130
3.0
7.3
30
5.0
10
73
4.6
16
6.7
50
20
43
9.1
5.0
21
11
3.9
15
13
6.2
9.2
14
5.8
665
20
K
0
0.8
0
0
0.3
18
0.8
0
0.3
0
2.2
2.1
0
1.2
0
4.5
2.1
16
0
0.5
3.7
17
0
1.3
1.2
0.8
0.8
5.4
2.6
81
B
0
11
0.4
1.0
0.8
28
1.9
0
0.6
0
2.0
1.8
0
1.2
0.5
4.9
2.0
10
0.1
0.4
3.6
9.9
0.7
0.9
0.9
1.3
1.0
2.6
2.1
90
30
K

1.2
0
0
0
29
0.5 .
0
0.4
0
2.5
2.5
0
1.3
0
6.1
2.5
19
0
0.5
4.5
25
0
1.8
1.5
1.7
1.1
12
4.3
117
B
0
2.7
0.5
1.0
0.4
3.5
0.3
0
0.2
0
0.6
0.4
0
0.3
0.3
1.0
0.4
2.5
0.1
0.2
0.8
2.7
0.2
0.4
0.3
0.1
0
0.1
0.1
19
K
0
1.9
0
0
0
19
0.6
0
0.2
0
0.9
1.3
0
0.6
0
3.2
1.2
9.2
0
0.3
2.2
16
0
1.1
0.8
1.2
0.7
5.8
3.9
68
40
B
0
3.1
0.3
0.4
0.2
3.7
0.2-
0
0.1
0
0.3
0.1
0
0.1
0.2
0.5
0.2
1.5
0
0
0.3
1.1
0
0.1
0
0
0
0
0
12
  w Calculated 95% confidence limits for measurement of pyrene concentration = ±18%.
  (b)These values have been derived by assuming that the relative concentration of pyrene (compound
no. 18) to total  PCAH is the same at  4, 6 and 8 cm as  at 10 cm. In fact, there  is an enrichment of
heavier PCAH as  samples  are collected at greater distances from the burner nozzle. As a result, the
actual values at 4, 6 and 8 cm are probably higher than the values in the table.
  M See Table 3 for key.
particles but so strongly agglomerated that it
was very difficult to distinguish the individual
particles. The arithmetic mean particle diame-
ter  is  defined as
                               n, being the
number of particles  contained in a diameter
interval centered on d{.
  From these tables, several conclusions can
be drawn:

  a) The  atomizing-air  pressure, even when
drastically changing the total amount of soot
emitted, does not affect the particle size.
  b) Burnout of soot tends to increase slightly
the average  diameter  of  the population of
particles.  This  is  an  indication  that  small
particles burn out first and possibly also  that
particles are oxidized internally and then  col-
lapse, as  observed by  others.9 Indeed, even
if  burnout does  not reduce  the diameter of
the particles,  it  reduces strongly  their mass
(at least for 205 and 239 kPa). If particles were
not burning  internally, one would expect an
eventual decrease in their average  size.
   c) Benzene produces slightly smaller parti-
                                         13

-------
                                        TABLE V
 Arithmetic mean particle diameter (A). Cold gas velocity = 0.96 m/s; mean fuel equivalence ratio = 1.0.

                                                   atomizer air pressure, kPa

Distance from nozzle, cm
6
10
53 (exhaust)

184
191
211
249
kerosene
205
0
220
298

239
O
254
o
benzene
184
189
184
223
  "poorly defined carbonaceous material
 cles than kerosene, even though the total mass
 is much larger. The same  effect, with even
 a larger impact, was observed when increasing
 the cold gas velocity (Table VI). This effect
 in both cases may result from an increase in
 temperature. The adiabatic flame temperature
 for benzene is indeed higher than for kerosene,
 as pointed above, and consequently so is the
 final temperature. Increasing the cold gas ve-
 locity results in a  higher temperature through
 the reduction of heat loss per  unit of energy
 input. It has been previously observed11 that
 an increase in temperature results in a decrease
 in the size of particles through an  increase
 in the number  of particles formed per unit
 volume, other parameters being constant. For
 a given amount of carbonaceous material, the
 final size of the particles  will be smaller  if
 a  larger  number of  nuclei  is generated.
 However, smaller particles  will be produced
 also if for a given number  of  nuclei there  is
 less  carbonaceous material to provide  surface
 growth. Since, the total amount of soot asso-
 ciated with the smaller particles is equal to
 (effect of cold gas velocity at 239 kPa) or larger
 than  (benzene as  compared to kerosene) the
 amount associated with larger particles, the
 effect is one of nucleation, probably related
 to the change in temperature.  For the effect
 of cold gas velocity at  184 kPa and 205 kPa,
                 TABLE VI
Arithmetic mean particle diameter. Kerosene. Cold
 gas velocity = 2.67 m/s; atomizer air pressure =
 184 kPa (12 psig); distance from nozzle = 20 cm.
       mean fuel
      equivalence
         ratio
0.8   0.9   1.0   1.2
diameters (A)
     146   145   139
  "poorly defined-carlxmaceous material
                       both the amount of soot and the particle size
                       decrease, and it is not possible to distinguish
                       between nucleation and surface growth effects.
               4.  Discussion

  Any model of soot formation in a turbulent
fuel spray combustor has to combine both the
mixing of fuel and air and the chemistry of
soot formation and burnout. Neither of these
two processes is well understood and only very
approximate modelling of soot formation in
this system  is  now possible.
  An approximate mixing model previously
usedl-2 assumed that the nonuniform distribu-
tion of fuel and air can be represented by a
Gaussian distribution  of uniformly premixed
pockets of gas, or  eddies, of different fuel-air
ratios, centered on the mean fuel-equivalence
ratio. Such a model may be used with reasona-
ble accuracy i) downstream of the recirculation
zone  and  ii)  after all the  fuel  has been
evaporated.
  The recirculation zone in our  burner is
assumed to  extend 1.5 to 2 burner diameters
from  the nozzle. Accordingly, the  maximum
of soot formation (Fig. 2 and 3) occurred inside
the recirculation zone at the lower velocity
and outside at the higher velocity. Since the
fuel is most probably entirely evaporated out-
side the recirculation  zone, the above two-
conditions are presumably obeyed for the data
obtained at  the higher  air velocity.
  By  combining the Gaussian distribution of
fuel/air ratios  and the amount of soot pro-
duced in each eddy, the total soot concentra-
tion could in principle be computed. However,
the chemistry of soot  formation is  not well
known  and few data  are  available on  the
amount of soot emitted for a given fuel equiva-
lence ratio. Several reviews of this subject are
available.9'10-11  Studies of  premixed  flames
indicate that soot begins to form at lower C/O
                                           14

-------
ratios than would be predicted from an equi-
librium calculation (roughly,  for C/O — 1).
To assess the departure of soot emission from
equilibrium in premixed flames, we employed
the equilibrium program  developed recently
by Svehla and McBride12 which uses a  free
energy minimization technique to compute the
equilibrium composition  of 56 molecular or
radical species in the C/H/O/N/Ar system.
Soot was assumed to have the thermodynamic
properties of graphite. Comparison of equilib-
rium values of soot formation and experimental
values available in the literature indicates that
experimental  values can  be as  high as  five
times the equilibrium value  (premixed ben-
zene/oxygen flame13 or  as low as one fifth
of the equilibrium  value (premixed acetyl-
ene/oxygen flame6. However, in the absence
of data  on  premixed  kerosene-air  flames,
and neglecting the fact that actual values are
probably  higher  due to  the  aromatic frac-
tion  contained in  the fuel,  we used equili-
brium values to approximate the concentration
of soot formed when burning kerosene.
  The values so computed, assuming a Gaus-
sian distribution of fuel/air  ratios  centered
on the mean  ratio , are reported in Table
II together with the experimental values. The
standard deviation a of the distribution was
assumed to be either independent of 4>, and
equal to 0.8,  or dependent  on  with cr/
being 0.8. The value of 0.8 agrees with values
reported by others2 for similar conditions in
the early zone of the burner.
  It is worth noting that,  in this picture, soot
will  be formed only in the small  fraction of
eddies having a fuel equivalence ratio larger
than about 2.8 and  contained in  the tail of
the distribution. As a result, the computed soot
concentrations are very sensitive to the values
selected for cr. In spite of the above | imitations,
it appears from Table II that this approximate
calculation predicts the right order or magni-
tude for both soot concentration and the  gen-
eral effect of <1> for the higher  velocity runs.
  As  indicated  before, at low velocity the
maximum of soot  formation appeared inside
the recirculation zone, and its magnitude was
extremely sensitive  to the fuel atomization
conditions. A possible reason for this behavior
is that the relative velocity between the larger
droplets and  the air remains smaller than the
extinction velocity  in the low velocity runs,
in contrast to  the high velocity runs where
the combustion air velocity is expected to be
high enough to extinguish the droplet flame.
Individual diffusion flame envelopes around
large droplets are known to  generate soot  in
fuel spray combustors.17 The extinction veloc-
ity is proportional to the square root  of the
droplet diameter17 and the size of the droplets
decreases with increasing atomizing air pres-
sure. These factors offer a plausible explana-
tion  for the effect of atomizing air pressure
on soot formation at a relative velocity lower
than the extinction  velocity and for the lack
of sensitivity at higher velocity.
  With regard to soot burnout, the soot con-
centration  in  the exhaust (50 cm from the
nozzle) decreases with  increasing atomizing
air pressure, and hence with increasing inten-
sity of mixing, at the lower gas velocity (Fig.
2) but increases  at the  higher velocity (Fig.
3). Although these trends in the two figures
may at first appear to be mutually inconsistent,
further inspection reveals substantial  agree-
ment. The mass of  soot  burned out, as mea-
sured by the decrease in soot loading between
the point of maximum concentration and the
exhaust, decreases consistently with increasing
atomizing air pressure in both figures. While
this  behavior  may appear to  be opposite  to
that  generally expected for improved mixing
of fuel and air, it should  be noted as observed
previously2 that an increase in atomizing air
pressure will  increase  the consumption  of
oxygen in the first zone  of the flame thereby
lowering the oxygen concentration in the sec-
ond zone. Accordingly, the observed decrease
in mass of soot burned with increased atomiz-.
ing air pressure can be attributed to decreased
oxygen availability  for  soot burnout  in the
second zone.  The amount of unburned soot
emitted in the exhaust can either decrease (Fig.
2) or increase (Fig. 3) with increased atomizing
air pressure depending on whether the amount
of soot formed in the primary zone is extremely
sensitive (Fig. 2) or relatively insensitive (Fig.
3) to changes  in the  atomizing air pressure.
              5. Conclusions

  The results of this study can be utilized in
the  consideration of  two general  types  of
practical  combustors.  The  first type corre-
sponds to a  system  with a  high combustion
air velocity typical of gas turbine combustors,
and  the second type corresponds to a system
with a low combustion air velocity typical of
domestic  home heating burners. The amount
of soot formed in the  second case is usually
higher than in the first and  is very sensitive
to fuel atomization characteristics. One practi-
cal consequence is that the common, unmain-
tained domestic home  heating burner may be
the  source of considerable  soot and PCAH,
some of which may have considerable effects
                                      15

-------
on human health. The identification of  indi-
vidual PCAH is therefore important in order
to assess the exact health hazard generated by
a particular combustion system.

                Acknowledgment

  We are grateful to C. Clampitt, J. Dillon, R. Flagan
and K. Shamsavari  for valuable help at different
steps of  this project. The work was initiated with
support of Honeywell, Inc. and continued under
Contract No. NSG3076 of the National Aeronautics
and Space Administration (Lewis Research Center)
and Grant No. R803242 of the  Environmental Pro-
tection Agency.  One of us (Gilles P. Prado) was
supported in part by the  Centre National  de la
Recherche  Scientifique, Ministere  de 1'Education
Nationale, France.
               BIBLIOGRAPHY

 1.  POMPEI, F( AND HEYWOOD, J. B.: Combustion and
    Flame 19, 407 (1972).
 2.  APPLETON, J. P. AND HEYWOOD, J. B.: Fourteenth
    Symposium (International) on Combustion, p.
    777,  The Combustion Institute,  Pittsburgh,
    1973.
 3.  FLACAN, R. C. AND APPLETON, J. P.: Combustion
    and Flame 23, 249 (1976).
 4.  LAHAYE.J. ANDLARACNE,J.: Centre de Recherches
    sur  la Physico-Chimie des Surfaces Solides,
    C.N.R.S., Mulhouse, France, Personal Commu-
    nication.
 5.  SCHOENTAL,  R.:  Carcinogenesis by  Polycyclic
    Aromatic Hydrocarbons and by Certain Other
    Carcinogens, Chapter in "Polycyclic Hydrocar-
    bons" (E. Clar, Ed.), p. 133, Academic Press,
    1964.
 6.  FISHBEIN, L., FLAMM, W. G. AND FALK, H. L.:
    Chemical Mutagens,  p. 275, Academic Press,
    1970.
 7.  MILLER,  E. C. AND  MILLER, J.  A.: Chemical
    Mutagens, Vol. 1 (A. Hollaender, Ed.), p. 105,
    Plenum Press, 1971.
 8.  PRADO, G.  P.: Mechanisme  de Formation des
    Particules  de Noir de  Carbone lors de la De-
    composition  Thermique de Benzene.  Ph.D.
    Thesis, C.U.H.R.  and  U.L.P. of Strasbourg,
    France, 1972.
 9.  WERSBORC, B. L.: Physical Mechanisms of Car-
    bon Formation in Flames. Sc.D. Thesis, M.I.T.,
    Cambridge, Mass., 1972.
10.  VUK,  C. T., JONES, M.  A.  AND JOHNSON, J. H.:
    The Measurement  and  Analysis of the  Physical
    Character  of Diesel  Particulate Emissions.
    Paper presented  at  Automotive Engineering
    Congress and Exposition,  Detroit,  Michigan,
    February 1976.
11.  LAHAYE, J., PRADO, G.  P.  AND DONNET, J. B.:
    Carbon 12, 27(1974).
12.  DONNET, J. B. AND  VOET,  A.: Carbon  Black,
    Physics,  Chemistry,  and  Elastomer  Rein-
    forcement, p. 96, Marcel Dekker  Inc., 1976.
13.  PALMER, H. B. AND CULLIS, C. F.: The Formation
    of Carbon from Gases, Chapter in "The Chemis-
    try and Physics of Carbon," Vol. 1 (P. L. Walker,
    Jr., Ed.), p. 265, Marcel Dekker,  1965.
14.  LAHAYE, J. AND PRADO,  G. P.: Mechanisms of
    Carbon  Black Formation,  Chapter   in "The
    Chemistry and Physics of Carbon," Vol. 13, (P.
    L. Walker, Jr., Ed.), Marcel Dekker, to be pub-
    lished.
15.  SVEHLA, R. A. AND McBfliDE, B.  J.: Fortran IV
    Computer Program Calculation  of Thermody-
    namic and Transport  Properties of  Complex
    Chemical Systems, NASA TN D. 7056, 1973.
16.  HOMANN, K.  H., MORGENEYER, W. AND  WAGNER,
    H.Gc.: Combustion Institute European Sympo-
    sium, p. 396, Academic Press, N.Y., 1973.
17.  SJOCHEN, A.:  Fourteenth Symposium  (Interna-
    tional) on Combustion,  p. 919, The Combustion
    Institute, Pittsburgh, 1973.
                                        COMMENTS
  P. J. Pagni,  University of California,  USA. Your
mass data are  very interesting. However, the size
data are at best qualitative because of the extensive
sample manipulation. Two techniques which might
be useful in this apparatus are: (1) Particle replica-
tion which permits  micrographs  to be obtained
without removing the particles from the filter on
which they are originally collected1  and (2)  air-as-
pirated, porous walled probes which nearly elimi-
nate wall effect difficulties.2
                REFERENCES

1.  KAY, D. H. (ED), "Techniques for Electron Micro-
   scopy," 2nd ed., p. 115, F. A. Davis Co., Philadel-
   phia (1967).
2.  MITCHELL, C. A. AND SILVERMAN, L.; "The Bounda-
   ry Layer Diluter—A  New  Gas  and Particulate
   Calibration Device," Air Pollution Control Asso-
   ciation Paper, presented to the New England
   Section, April  (1963).
                                               16

-------
  Authors' Reply. As indicated in the paper, the
soot appears  as aggregates of individual, spherical
particles.  Any collection  technique which  allows
collisions of  particles in the sampling train,  as is
the case in this  study, may affect the aggregate
characteristics. However, the sizes of the individual,
spherical units  are not affected by such collisions
as long as they are solid before entering the probe,
which is the case in this study for samples collected
farther than  10 cm from the burner. Because this
study focussed only on the size of these individual
units, we believe that the data reported are accurate
to within 10%. We are not familiar with the particle
replication technique, but we are not optimistic that
such  a technique would  permit the detection  of
particles as small as 20 Adiameter which are directly
measured with the electron microscope.
  Richard G. Cann, National Bureau of Standards,
USA. In your paper, you make the important as-
sumption that the combusting system is essentially
premixed. It is critical  that this be verified since
variation in residual fuel droplet size and concentra-
tion or gaseous fuel-rich pockets could easily domi-
nate  the  soot formation characteristics. Would you
expand  on  the theoretical  or experimental  ap-
proaches you need to establish the homogeneity of
the system and the confidence limits.

  Authors' Reply. It is correct that fuel-rich pockets
as well as possible fuel droplet combustion dominate
the soot formation characteristics in turbulent diffu-
sion  fuel spray systems. Here, for the high velocity
runs, we believe that droplets evaporate  prior to
combustion, but that  might not be the case for the
low  velocity runs, as indicated in the paper.  We
do not assume that the system is premixed; on the
contrary, we represent the mixing of fuel and air
by a  gaussian distribution of pockets (or eddies)
of different  fuel-air ratio, in which only very fuel-
rich pockets from soot. We are presently measuring
the CO/CO2 ratio in axial as well as traverse direc-
tions in the  flame  in order to determine more
precisely the  distribution of eddies (see ref.  1,  2
and  3 in the paper).  In parallel, computations are
being developed to determine the trajectory of fuel
droplets  and their evaporation characteristics.
  S. C. Graham,  Shell Research Ltd., England. In
your presentation, you state that benzene flames with
relatively high adiabatic  flame temperatures, give
smaller soot  particles  (within  the agglomerated
chains) than the lower temperature kerosene flames.
You then attribute this to the effect of temperature
on the particle nucleation rate—i.e. the higher the
temperature, the greater the rate, — on the assumption
that the final number density of individual particles
is  strongly dependent  on, if not identically  equal
to, the total number of nuclei that are formed within
a unit volume  of  gas  throughout the combustion
process. I should like to point out that your  inter-
pretation ignors the collisions that .must occur be-
tween the (very small)  soot nuclei,  and I suggest
that the particle number densities involved are far
too high for such collisions to be ignored,1'2 whether
they lead  to  chain formation or to  the  formation
of bigger particles  via coalescence.
  An  alternative and very different model, in which
coalescent collisions between particles (partly soot,
partly PCAH) dominate the  early stages of particle
growth, also  leads to the result  that the ultimate
size of the particles within the agglomerated chains
decreases  with  increasing temperature. In its sim-
plest  form3  this model leads to the equation for
free-molecule coagulation
       Vs'6 (t)
where  V  is the  particle  volume,  k'  and  k"  are
constants  and   is the participate volume fraction.
For simplicity consider the  situation  where <)> is
constant. This gives

                V= fc
where At is the interval between the onset of nuclea-
tion and the time at which particle collisions cease
to result in the coalescence of individual particles.
However it is inevitable that coalescent collisions
will give way to chain-forming or possibly to non-
sticky collisions either.
i)  when the gas is quenched to room temperature
   so that the condensed phase (partly soot, partly
   hydrocarbon) is wholly  solid, or
ii) when the chemical processes of dehydrogena-
   tion/cracking/pyrolysis of the condensed phase
   have  proceeded sufficiently for the condensed
   phase to have turned completely into  soot.
Whenever ii above is the limiting factor, an increase
in temperature will result in a shorter  At giving
smaller particles within the agglomerated chains.
                 REFERENCES

1. GRAHAM, S. C. AND HOMER, J. B.: Faraday Sympo-
   sium; Chem. Soc., 7, 86 (1973).
2. GRAHAM, S.  C., HOMER, J. B.  AND ROSENFELD, J.
   L. J.: Proc. Roy. Soc. (London). A344, 259 (1975).
3. GRAHAM, S.  C., HOMER, J. B.  AND ROSENFELD, J.
   L. J.: Second  Symposium (European)  on Com-
   bustion, p. 374 (1975).
                                           17

-------
  Authors' Reply. The qualitative interpretation that
the increase  of  particle  number density may  be
attributed to increased  flame temperature does not
ignore particle collisions, but the validity of the
interpretation clearly depends upon the relative rates
of nucleation  and  coagulation with the latter rate
in turn depending on  the rate of particle  growth
by reaction with gaseous  species.
  Grahm's coagulation equations are interesting but
their application to soot formation would appear
to require a suitable treatment of nucleation and
surface growth,  both of  which  processes  signifi-
cantly overlap the coagulation process.1'2
              REFERENCES
HOWARD, J. B.,  WEHSBORC, B. L., AND WILLIAMS,
G. C.: "Coagulation of Carbon Particles in Pre-
mixed  Flame," Faraday  Symposium  of  the
Chemical Society (London), No. 7, pp. 109-119,
1973.
WERSBORG,  B. L., HOWARD, ]. B., AND WILLIAMS,
G. C.: "Physical Mechanisms in Carbon Forma-
tion  in  Flames,"  Fourteenth Symposium (In-
ternational) on Combustion, p. 229, The Combus-
tion Institute, Pittsburgh, 1973.
                                               18

-------
                 APPENDIX B
  Formation of Soot and Polycyclic Aromatic
     Hydrocarbons in Combustion Systems
From:   Proceeding of the Stationary Source
       Combustion Symp., Vol I:  Fundamental
       Research, EPA Report 600/2-76-152a,
       June 1976.
                        19

-------
FORMATION OF SOOT AND POLYCYCLIC AROMATIC HYDROCARBONS
              IN COMBUSTION SYSTEMS
  Development of a Molecular Beam Mass Spectrometer
                        By

                  James D.  Bittner
          Department of Chemical  Engineering
         Massachusetts Institute  of Technology
            Cambridge, Massachusetts 02139
                        20

-------
                                  ABSTRACT

   The overall objective of this research program has been to study the pro-
duction of soot and polycyclic aromatic hydrocarbons (PCAH) in combustion
systems.  The program has two phases.  In one phase the production of soot and
PCAH in a turbulent diffusion flame was studied.   The results have been reported
elsewhere.  In the phase to be discussed in this  paper, a low pressure flat
laminar premixed flame is being used to study the kinetic relationships
between soot, PCAH and other hydrocarbon species  that may be important as soot
nuclei or surface growth species.  The molecular beam-mass spectrometer system
developed to study the gas phase species will be  described.  Preliminary mass
spectral data on an acetylene-oxygen flame will  be reported.
                                      21

-------
                                  INTRODUCTION

   The formation of soot in combustion systems has been studied for many years
with the objective of attaining a working knowledge of the process to aid in
predicting flame emissivities for radiative heat transfer problems.  Recently,
concern has been expressed that soot and some of the organic compounds assoc-
iated with incomplete combustion are pollution problems that may have to be
controlled.  It is known that some polycyclic aromatic hydrocarbons (PCAH)
produced in flames and adsorbed on soot particles are carcinogenic.  Soot
and other organic particulate matter may not be the most abundant particulates
emitted, but because the small particles (<0.2 ym) are easily injected deep
into the respiratory system, they may be one of the types of particulates
most hazardous to human health.  The effects of two expected changes in the
operation of combustion systems threaten to increase the emissions of parti-
culate organic matter.  First, the staged combustion strategy for controlling
NOX requires primary combustion in a fuel rich atmosphere that leads to
increased soot and PCAH formation.  Second, the planned increased use of coal
and coal derived liquids will also lead to increased soot and PCAH formation
since aromatic fuels are known to be troublesome in this respect (1,2).  The
necessity for learning how to burn low H/C fuels (aromatics) without consi-
derable soot and PCAH formation is especially important for small systems such
as gas turbines and internal combustion engines where residence times available
for burnout are limited (3).

     The mechanism of soot formation in combustion systems is not well under-
stood.   Some of the mechanisms proposed over the years and the confusion
surrounding the role of PCAH in the process are illustrated in Figure 1.
Since soot particles contain many more carbon atoms and a much lower H/C ratio
than the fuel molecules, soot formation must involve processes of aggregation
(top to bottom) and dehydrogenation (left to right).  The extreme routes, the
C2 route and the saturated polymer route, are unlikely to occur under typical
combustion conditions.  The C2 route may be important at temperatures around
3000°K.  At lower temperatures, around 700°K, and long reaction times the
saturated polymer route may occur.  But, in the range 1200-2200°K most mechan-
isms can be represented by the central portion of Figure 1.   From this figure
it is seen that PCAH species have been postulated to serve as nuclei for soot
growth by surface decomposition of any number of hydrocarbon species (including
themselves).   They have been postulated to have sufficiently low vapor pres-
sures to physically condense to liquid droplets which then form soot.   They
have also been postulated to be stable by-products of the soot formation process,

     In light of the concern over particulate organic matter as a pollutant
and the confusion surrounding soot formation mechanisms and  the role of PCAH,

                                     22

-------
                SUMMARY OF PROPOSED MECHANISMS OF SOOT FORMATION
ro
oo
                                  DEHYDROGENATION
                       AROMATIC
                       FUEL
                  PARAFFINIC
                  FUEL
                    \
                                                            C2 MOLECULES
Cg ^2-^
                     HYDROCARBON
                         IONS
POLYMERIZATION
     AND
OEHYDROGENATION
  \
SOOT
NUCLEUS
                                             CONDENSATION
                                 PCAH
                          STABLE
                          PCAH
                          BY-PRODUCTS
                         POLYACETYLENES
                             N   ^^SURFACE
                                       GROWTH BY
                                       DECOMPOSITION
                                    PHYSICAL
                                    CONDENSATION
                                  DEHYDROGENATION
                                      AND
                                  POLYMERIZATION
                  SATURATED  OR
                  UNSATURATED
                  POLYMERS
                                               POLYMERIZATION
                  LIQUID  DROPLETS      CARBON PARTICLES
                  CONTAINING      —"* WITH GRAPHITE-
                  POLYCYCLIC COMPOUNDS LIKE STRUCTURE
                      FIGURE 1.  SUMMARY OF PROPOSED MECHANISMS OF SOOT FORMATION

-------
the specific objective of this research is to study the chemistry of soot and
PCAH formation.  Emphasis is on the identification and measurement of relative
concentration profiles of the gas phase hydrocarbon species and the measure-
ment of soot particle concentrations, size distributions and total mass of
soot at different stages of combustion in a flat premixed flame.


                            EXPERIMENTAL APPARATUS

DESCRIPTION

     The molecular beam mass spectrometer system that has been developed
during this project to study premixed low-pressure flat flames is schemati-
cally represented in Figure 2.  The flame is stabilized on a flat water-cooled
drilled copper plate burner that is 7.1 cm in diameter.  The burner chamber
is pumped by a 60 cfm Stokes mechanical pump.  The flame is sampled along the
centerline of the burner by moving the burner relative to the rest of the
apparatus.  The sample is withdrawn supersonically through a quartz nozzle.
The nozzle is 
-------
       ELECTRON
       MULTIPLIER

   OUAORUPOLE MASS
        FILTER
           IONIZER


       LN2- COOLED
        WALLS

    TUNING FORK
      CHOPPER

CALIBRATION GAS
     EFFUSIVE
     SOURCE
        6 INCH
        DIFFUSION
        PUMP
   4 INCH
   DIFFUSION
   PUMP
   COLLIMATOR
   6 INCH
   DIFFUSION
   PUMP
                                          SKIMMER
                                          QUARTZ NOZZLE

                                          BURNER
MECHANICAL
VACUUM PUMP
                            !=<•— PREMIXED GASES
     FIGURE 2,   MOLECULAR BEAM MASS  SPECTROMETER SYSTEM
                          25

-------
     For studying the participate phase the mass spectrometer is removed and
the top flange is replaced by a flange that has a mechanism for mounting elec-
tron microscope grids and moving them in and out of the beam.  The grids are
exposed to the beam for a measured time interval with the use of a mechanical
shutter.  The particles collected on the grid are analyzed by making electron
micrographs using either shadowing or transmission procedures.  Number densi-
ties and size distributions are obtained with the aid of a Zeiss semi-automatic
particle counter.  Computation techniques developed by Wersborg will be used
to calculate profiles of average soot particle size, soot particle number
concentration, rates of nucleation, surface growth and coagulation (5).

     A schematic of the mass spectrometer instrumentation is shown in Figure 3.
The molecular beam is chopped by the fixed-frequency tuning fork operating
at 220 Hz.  The signal from the electron multiplier at any particular mass
has a DC component corresponding to ions originating from the background gas
molecules and an AC component at 220 Hz corresponding to ions originating
from the beam molecules.  The signal is amplified with an Extranuclear fast
electrometer preamplifier.  The amplified signal is DC coupled to an Extra-
nuclear electrometer to obtain a signal corresponding to the background compo-
nent.  A reference signal from the chopper and the AC component of the signal
from the preamplifier are introduced into a Princeton Applied Research HR-8
lock-in amplifier to obtain a signal proportional to the beam component.
Signals corresponding to beam and background components are then displayed
either on a dual-beam storage oscilloscope or a chart recorder.

                                  CALIBRATION

Considerations

     Calibration of supersonic molecular beam systems for flame sampling is
not a straight forward task due to the many effects that can influence the
composition of the beam prior to reaching the mass spectrometer ionizer.  Shock
formation in front of the skimmer orifice, species condensations, pressure
diffusion in the free jet, Mach number focussing downstream of the skimmer,
scattering of the beam by the background gas in any or all of the chambers,
and effusion of the background gas into the beam are effects that can distort
the beam composition and are discussed extensively in a review article by
Knuth (6).  Skimmer interference problems can be avoided by proper design of
external,and internal angles of the skimmer and the selection of the proper
distance between the tip and the supporting wall.  Species condensations in
free jet expansion are not believed to be important at flame temperatures and
sub-atmospheric pressures for most flame constituents (6), however little is
known about the behavior of the high molecular weight hydrocarbons and care
must be taken to look for condensation effects in the interpretation of the
data.  Pressure diffusion, Mach number focussing and background scattering
and effusion are more difficult effects to avoid altogether and all can be
functions of source conditions.  A study of their importance in this system
and techniques for calibration of these effects is now underway.  The limited
data indicate that background scattering and effusion are negligible under the
flame conditions presented here.  Pressure diffusion and Mach number focus-
sing are lumped together as "mass discrimination" and can be studied in this
                                      26

-------
                                              ELECTRON
                                              MULTIPLIER
                                  SIGNAL
rsi
        ELECTROMETER
      STORAGE
      ^	^

    OSCILLOSCOPE
   PAR HR-8
   LOCK  IN
   AMPLIFIER
OSCILLOGRAPHIC
  RECORDER
                                    MASS
                                   FILTER
                                           IONIZER
                                          REFERENCE SIGNAL
                  EXTRANUCLEAR

                      MASS

                      SPEC.

                     CONTROLS
                                                                              TUNING
                                                                              FORK
                                                                              CHOPPER
                   FIGURE 3,  SCHEMATIC OF MASS SPECTROMETER INSTRUMENTATION

-------
 system with  the  aid  of the  effusive  source  placed  in  the  second  stage  upstream
 of the collimator.

      The system  for  studying  mass  discrimination is shown schematically  in
 Figure 4.  Mixtures  of the  major stable  species are made  by metering through
 critical orifice flow meters  and mixing  dynamically.  These mixtures may be
 introduced into  the  mass  spectrometer ionizer  either  as an effusive beam in
 which no mass  discrimination  effects occur  or  as a supersonic  beam  (through
 the burner and quartz nozzle)  in which high molecular weight species are usually
 preferentially concentrated relative to  low molecular weight species.

      To introduce the mixture effusively without mass discrimination,  a  stain-
 less steel sintered  disc  with a  nominal  pore size  of  0.5  ym is used to leak
 into the effusive source  in the  second stage about 1% of  the total gas flow by
 the disc.  The pressure on  the high  pressure side  of  the  sintered disc is main-
gained in the  range  0.1 to  1  torr.   Tubing  size and pressures  are designed to
 give viscous flow (no mass  discrimination)  everywhere upstream of the  sintered
 disc and effusive flow through the disc  and downstream from it.  Under these
 conditions the(flow  rate  of a component  through the disc  is inversely  propor-
 tional  to the  square root of  its molecular  weight  but its density anywhere
 downstream of  the disc is proportional to its  partial pressure upstream.  Since
 the electron impact  ionizer is a density detector, ratios of ion signals  are
 proportional to  the  ratios  of partial  pressures in the gas mixture upstream of
 the disc.  The effusive source inlet system has been  tested with commercially
 prepared standard gas mixtures and reproduces  the  suppliers analysis to  within
 2% on any component.

      By comparing signal  ratios  in the effusive beam  to signal ratios  in  the sonic
 beam the mass  discrimination  between two species in a mixture  can be characterized
 by an enrichment factor,  ex.,,:
                        aAB=  rE =     SX
                         MB    ig i   lg t    Ag b  Ag t


where    a.B  = enrichment  factor of  species A relative to species B, dimensionless


         I.   = beam  signal intensity of species A, amperes


         ID   = beam  signal intensity of species B, amperes


         S    = sonic introduction

         E    = effusive introduction

         X.   = mole  fraction of species A

         Xg   = mole  fraction of species B


                                      28

-------
                 MECHANICAL
                 VACUUM
                 PUMP
                   l
       STAINLESS
       STEEL
       SINTERED
       DISC
                       ALUMINA
                       TRAP
 DIFFUSION
 PUMP
     MKS BARATRON
     PRESSURE
     TRANSDUCER
T   T   T   T   T
 CO   C02   H2    C2H2 02
    EFFUSIVE
    SOURCE^
 M.S.
STAGE
                                          2ND
                                         STAGE
                 1ST
                STAGE
               BURNER
               CHAMBER
	 rvvv
MIXING
VOLUME
1


— — — wr-
J
 |   CRITICAL
E^  ORIFICE
 |   FLOW METERS
 Ar
   FIGURE 4,   SCHEMATIC OF GAS INTRODUCTION SYSTEM FOR STUDYING
            MASS DISCRIMINATION
                         29

-------
     The effusive source system is also used to obtain relative mass spectro-
meter sensitivity factors for stable species, S.,,:
                                  s   .-
                                   AB   kB

where   k. = sensitivity for species A, amperes/torr

        kg = sensitivity for species B, amperes/torr

     Mass discrimination effects have been studied most extensivity by Sharma
et al.  (7).  The enrichment factor may or may not be a function of the source
conditions (flame conditions at the sampling point in this case) depending
upon how the system geometry and source conditions dictate the point of transi-
tion from continuum to free molecular flow.  Although the study of these effects
in this system is not complete, preliminary experiments ar room temperature
with a mixture approximating the burned gas composition of the acetylene-
oxygen  flames studied here indicate that enrichment factors for the major
species except that for hydrogen are relatively insensitive to source condi-
tions.  An increase in source density by a factor of five resulted in a decrease
in aH2,co of 23% and changes in all other enrichment factors of less than 12%.
Further experimental work to assess the effect of gas composition and source
density on the enrichment factors is planned.
                                                          . .         i
     The calibration procedures for mass discrimination in the sonic beam and
for mass spectrometer sensitivities are carried out at room temperature.  The
effect of temperature on these two calibration steps must be considered.  The
mass spectrometer relative sensitivities in some cases might be functions of
temperature.   In conventional (residual gas analyzer) mass spectrometer ion-
izers variation with temperature of absolute ionizer sensitivities for total
ionization have been found to be due only to the effects of temperature on
gas density and the speed of the neutral molecule (which effects ion collection
efficiency), implying that actual total ionization cross-sections are not
temperature sensitive (8).  In the type of molecular beam ionizer used here
the neutral beam molecules are directed along the axis and toward the quadru-
poles, therefore the ion collection efficiency might be expected to be high
and less sensitive to translational temperatures perpendicular to the beam
axis.  Although total ionization cross-sections are temperature independent,
fragmentation patterns and therefore mass spectrometer sensitivity (when a
specific ion is used to follow concentration changes) is highly dependent upon
the vibrational energies of the molecule-ion (9).   As temperature increases
more fragmentation occurs.  With current knowledge of vibrational  relaxation
processes in supersonic expansions vibrational  energies are impossible to
predict and one must if possible avoid the use of high electron energies where
fragmentation, occurs.  The appearance potentials of fragment ions, like heats
of reaction,  seem to vary but little with temperature (10).   So, if fragmenta-
tion does not occur in room temperature calibrations it is 'not likely to
occur in. the case of the flame.   However, in spite of these arguments, care
must be taken in the interpretation of the data and simple experiments must
be done to check temperature effects on mass spectrometer sensitivies due to

                                      30

-------
fragmentation effects whereever possible.

     The effects of temperature on the mass discrimination enrichment factors
must be assessed using the models designed to handle the prediction of enrich-
ment factors for pressure diffusion and mach number focussing (5,6).  Analysis
of these models suggests that these mass discrimination effects depend upon
temperature in only the way in which it effects the source density.  Therefore
enrichment factors from room temperature calibrations carried out at a reduced
pressure to reproduce the source density conditions should apply to the higher
pressure and temperature conditions in the flame.

Major Stable Species

     In accordance with the considerations discussed above, the calibration
for most major stable species involves the measurement of enrichment factors
and relative mass spectrometer sensitivities.  With this calibration information,
ratios of mole fractions in the flame are calculated from ratios of signal
intensities.  In the flames studied here 5 mole % argon has been added to the
unburned gases as an aid in following density changes in the flame front.
Since calibration of this system for water is very uncertain the following
relationships are used to calculate mole fractions from ratios of mole frac-
tions in the tail of the flame where diffusion effects are negligible:
an oxygen balance
                         * XAr + XCO + XC02 + \
                     "l
                            T = V + XCO  + V + XH90
                           . 1     f.   ~n—       c.   ~~^-~
and argon balance



                     njT^Arh = XAr

and from experimental measurements

                     XH2   XC2H2    X02  XCO   XC02
                     v»  Y> ~v  ' ~Y   ' If
                     AAr   AAr      AAr  AAr   AAr

where nj/np is the ratio of initial number of moles to final number of moles
and (Xn)j refers to the mole fraction of species A in the unburned gas.  Carbon
and hydrogen balances can then be used to check the consistency of the data
and calibrations.  From the calculated ratio Xu0n  in the tail of the flame and
                                      31

-------
experimental !„ ~  ratio it is possible to obtain a calibration factor for
water:

           X
            H20
where I.  = signal intensity of species i, amperes

      X.  = mole fraction of species i

      CM Q .  = calibration factor for water relative to argon.


Now with this additional piece of experimental information the oxygen and
argon balances are not needed in the reaction zone where steep gradients
require the use of diffusional terms.  The assumption is made that C^O Ar is
constant throughout the flame (i.e. the enrichment factor is not a function
of gas density or composition and mass spectrometer relative sensitivity
factors are not functions of flame position.)

Higher Molecular Height Hydrocarbons

     Due to the difficulty of introducing compounds that are liquids of low
vapor pressure and solids at room temperature into the system to obtain mass
spectrometer sensitivities and mass discrimination enrichment factors, all
higher molecular weight hydrocarbon species measurements are made on a
relative basis.  To account for any changes in mass spectrometer relative
sensitivities that might occur, the fragmentation pattern at 20 eV electron
energy of perfluorotri butyl ami ne is recorded before and after each experiment.


                               EXPERIMENTAL PROCEDURE


     To check out the experimental technique and procedures several fuel-
rich acetylene-oxygen flames have been investigated.   All flames had a cold
gas velocity of 50 cm/sec and burned at a pressure of 20 torr.   Five mole
percent argon was added to the unburned gas.  Four fuel  equivalence ratios
were investigated, <|> = 1.5, <|> = 2.0,  = 2.4 (sooting limit) and <|> = 3.0.
Complete profiles were made for <|> = 1.5, 2.0 and 2.4.   Two points near the
end of the oxidation zone were sampled in the cj> = 3.0 flame.

     The mass spectrometer resolution controls were set to give unit resolu-
tion from 18 to 502 a.m.u.   The electron energy was set at a nominal  20eV
which corresponded to an actual  electron energy of 17.1eV according to argon
ionization potential  measurements.   At these conditions,  room temperature

                                     32

-------
fragmentation of 02, C02, CO and h^O does not occur.  For acetylene  peak 25
is less than 1% of peak 26.  Mass spectrometer sensitivity and beam enrichment
factor calibrations were made before and after each flame run.  Two sweeps
across the mass range 0 - 230 a.m.u. were made at each flame position at a
speed of ^ 1 a.m.u./sec.


                           RESULTS AND DISCUSSION


     Profiles of signal intensities relative to argon for the <|> = 2.4 flame
are shown in Figures 5 and 6.  The mole fractions of the major stable species
in the burned gas, as calculated from the procedure described above, are
included in Figure 5.

     The experimental points with the most scatter are the hydrogen and water
profiles.  The characteristics of quadrupole mass spectrometers make it diffi-
cult to have a high sensitivity for hydrogen and high molecular weight
hydrocarbon species simultaneously.  To improve upon this measurement different
tuning conditions can be used for the low molecular weight gases and the high
molecular weight hydrocarbons.  The noise in the water signal is due to a
large background peak at 18 a.m.u.  The additional cryogenic pumping in the
ionizer region should improve this measurement.

     The general features of the profiles are in agreement with the measure-
ments on a similar flame by Homann and colleagues (11,12).  About 5.5 mole %
of the acetylene is present in the burned gases.  The other hydrocarbons in
the burned gases are mostly polyacetylenes.  They increase through the early
part of the oxidation zone and maximize at about 1 cm.  Detectable amounts
of polyacetylenes up to CsH2 are still present in the burned gas.  Polyacety-
lene formation is preceded in the oxidation zone by vinyl acetylene (C^) and
a C5Hg compound that could be benzene or a straight chain hydrocarbon.  These
compounds have concentration maxima that are at least an order of magnitude
lower than the polyactylene of the same carbon number.

     The interaction between the major stable species and the hydrocarbon
species throughout the flame is complex and interesting.  Oxygen is completely
consumed by about 1.4 cm.  At this point CO and H2 have nearly reached their
final values.  C02 and H20 molefractions peak near 1 cm and decrease slightly
into the burned gas.  Acetylene continues to decrease until about 2.5 cm above
the burner.  These basic features can be explained qualitatively by the
following scheme of competing reactions early in the oxidation zone:
                                     33

-------
                C2H2/02/Ar           P= 20TORR

                0.469/0.485/0.05    VQ = 50CM /S
o
e>
cr
<

o

LU
                                     CO     Xco = 0.56
LU

(T
LU
                                                   =0.055
                   I            2           3

             DISTANCE  ABOVE BURNER , CM
FIGURE 5, PROFILES OF SIGNAL INTENSITIES RELATIVE TO ARGON FOR MAJOR STABLE

        SPECIES IN AN ACETYLENE-OXYGEN FLAME NEAR THE SOOTING LIMIT.

        *  = 2,4, P = TORR, V0 = 50 CM/S, 5M% ARGON


                              34 .

-------
                 C2H2/02/Ar           p= 20 TORR
                 0.469/0.485/0.05    V0 =  50 CM/S
               1234
         DISTANCE  ABOVE BURNER  , CM
FIGURE 6.  PROFILES OF SIGNAL INTENSITIES RELATIVE TO ARGON FOR MINOR
         SPECIES IN AN ACETYLENE-OXYGEN FLAME NEAR THE SOOTING LIMIT.
         * = 2.4, p = 20 TORR. VQ = 50 CM/S, SM% ARGON
                        35

-------
                                                    fast
Reaction leading to polyacetylenes and C02 and f^O are faster than reactions
involving the same radicals leading to CO and H2-  This ties up,oxygen in
the form of C02 and H20 and produces super equilibrium amounts of polyacety-
lenes.  Near the end of the oxidation zone where the temperature reaches its
maximum, reactions producing OH, CO and H2 from C02 and H20:

                     H20 + H -> OH + H2
                     C02 + H -* CO + OH

become important and the oxidation of polyacetylenesbecomes faster than their
production (13).  The oxidation rate decreases quickly because of the drop in
OH radical concentration leaving some polyacetylenes in the burned gas (14).

     The features of the flame with 4> = 2.0 were essentially the same but
with lower CO, H2, acetylene and polyacetylene concentrations as might be
expected.  At $ = 1.5 no acetylene or polyacetylenes were detectable in the
burned gas.

     Signals were observed at masses other than those shown in Figures 5 and
6.  Table I lists those masses at which positive identifications were not made
or contributions from several species were not sorted out.  Masses 15, 16 and
17 peak in the middle of the oxidation zone.  The most probable contributor
to peaks 15 and 16 is methane in this fuel rich flame.  Masses 29 and 30 are
maximum at the lowest sampling point, 0.17 cm and the contribution of H2CO to
peak 29 has not been determined at these mass spectrometer conditions.  Mass
34 appears only at the lowest sampling point.  Mass 39 maximizes at 0.62 cm
from the burner.  Mass 42, probably corresponding to propylene, maximizes at
0.35 cm.  Appearance potential measurements along with measurements at lower
electron energies to prevent contributions from fragmentation of hydrocarbons
must be made to make positive identifications.

                                     36

-------
TABLE I.  MASSES AT WHICH SIGNALS WERE OBSERVED BUT POSITIVE
        IDENTIFICATIONS WERE NOT MADE,  * = 1A, P = 20 TORR, VQ = 5Q CM/S


          MASS      POSSIBLE  CONTRIBUTING SPECIES
14
15
16
17
25
29
30
34
38
39
42
CH2
CH3 ,CI

H4
CH4 (MOST PROBABLE), 0
OH
C2H ,
HCO ,
H2CO,
H202
C3H2
C3H3 -
C3H6

C2H2
H2CO, CI30
C2H6 (UNLIKELY)


C3H3

                           37

-------
     Mass spectra were observed at two points near the end of the oxidation
zone of a sooting flame of fuel equivalence ratio   <|> = 3.0.  In addition
to the species observed in the 4> = 2.4 flame several higher molecular weight
aromatic hydrocarbons were observed.  Their masses, molecular formula and
possible structures are listed in Table II.  The sensitivity for higher mole-
cular weight hydrocarbons (100 a.m.u. and above) should be increased by an
order of magnitude when the cryogenic pumping in the ionizer region is added
since the present sensitivity is limited by noise in the beam spectrum caused
by high levels of hydrocarbons in the background.


                                  SUMMARY

     A molecular beam mass spectrometer system has been developed for studying
sooting flames.  Preliminary measurements suggest that molecular beam mass
discrimination effects may be relatively insensitive to source conditions and
that absolute concentration measurements of major stable species profiles
will be possible.

     Profiles of relative signal intensities of several non-sooting and barely
sooting acetylene-oxygen flames have been made.   The results are in agreement
with those of previous investigators.  Polyacetylenic hydrocarbons up to mass
146 were detected and are the major hydrocarbons present other than the un-
burned fuel.  Maxima in the carbon dioxide and water profiles are supportive
of the suggestion that reactions that store oxygen in these species are rapid
compared to reactions that lead to CO formation from the fuel in the early part
of the flame.  At the higher temperatures present at the end of the oxidation
zone this oxygen is released in the form of OH that attacks polyacetylene and
unburned fuel.

     Under strongly sooting conditions (4> = 3.0) aromatic hydrocarbons have been
detected.  The addition of cyrogenic pumping in the ionizer region is expected
to increase the sensitivity for these species by an order of magnitude.


                                 ACKNOWLEDGEMENTS

     This work has been done under EPA Grant No. R 803242.
                                     38

-------
  TABLE II,  ADDITIONAL SPECIES NEAP. THE END OF THE REACTION ZONE
         IN A C2H2/02 FLAME, * = 3,0, P = 20 TORR, VQ = 50 CM/S
MASS    MOLECULAR   POSSIBLE      STRUCTURE
        FORMULA     SPECIES
                                   CH.
                                     3
 92     C7H8   TOLUENE

102     C8H6   PHENYLACETYLENE
                                     HC=CH2
104     C8H8   STYRENE             (p)
                                    ^x.-x^ LJ/••>	r*LJ
                                        nVrf = ^rip

118     C9HIO   METHYL STYRENE        (A
                                          CH3
128     C|0H8   NAPHTHALENE
                                       H
130     C,0H|0  PHENYL BUTADIENE

               DIHYDRONAPHTHALENE
 42     C|,H|0   METHYL NAPHTHALENE
146     C|2H2   HEXACETYLENE
                                       Q
                     39

-------
                               REFERENCES

 1.  Butze, H.F., and R.C. Ehlers NASA Tech. Mem. NASA JMX-71789.  Paper pre-
     sented at Western States Section of the Combustion Institute, Palo Alto,
     CA. Oct. 20, 1975.

 2.  Schirmer, R.M. in "Emissions from Continuous Combustion Systems", W.
     Cornelius and W.G. Ahnew, ed., p. 189, Plenum Press, N.Y.  (1972).

 3.  Longwell, J.P., "Synthetic Fuels and Combustion" Plenary lecture presented
     to Sixteenth Symposium (International) on Combustion, Cambridge, Massa-
     chusetts, August, 1976.

 4.  Biordi, J.C., C.P. Lazzara, and J.F. Papp, Combust. Flame  23, 73 (1974).

 5.  Wersborg, B.L., J.B. Howard, and G.C.  Williams, Fourteenth'International
     Symposium on Combustion, p. 929, The Combustion Institute  (1973).

 6.  Knuth, E.L., "Direct Sampling Studies  of Combustion Processes" in "Engine
     Emissions, Pollutant Formation and Measurement", G.S. Springer and D.J.
     Patterson, ed., p. 319, Plenum Press,  N.Y. (1973).

 7.  Sharma, P.K., E.L. Knuth and W.S. Young, J.  Chem,  Phys.  6£,  4345 (1976).

 8.  Field, F.H. and O.L. Franklin, "Electron Impact Phenomena  and the Proper-
     ties of Gaseous Ions," pp.  202-203, Academic Press, N.Y.  (1970).

 9.  Ibid.  pp. 203-204.

10.  Ibid., p. 81.

11.  Homann, K.H. and H.Gg. Wagner, Ben'chte der  Bunsengesellschaft 69, 20
     (1965).

12.  Bonne, U., K.H.  Homann, and H.  Gg.  Wagner, Tenth Symposium (International)
     on Combustion,  p.  503, The  Combustion  Institute (1965).

13.  Homann, K.H. Combust,  and Flame 11, 265 (1967).

14.  Bonne, U., H. Gg.  Wagner, Berichte  der Bunsengesellschaft  69,  35 (1965).
                                     40

-------
           APPENDIX C
Soot Concentration Measurements
  in Fuels Doped with Nitrogen
and Sulfur Containing Compounds
         (Unpublished)
                41

-------
SOOT CONCENTRATION MEASUREMENTS IN FUELS DOPED WITH

     NITROGEN AND SULFUR CONTAINING COMPOUNDS
                       by

                William J. Kausch.Jr.
          Department of Chemical Engineering
        Massachusetts Institute of Technology
            Cambridge, Massachusetts 02139
                  September, 1977
                        42

-------
                                 INTRODUCTION

     The purpose of this experimental program was to study the formation of soot
and polycyclic aromatic hydrocarbons (PCAH) in fuels doped with nitrogen and
sulfur containing compounds.  The main fuel studied was benzene, but for pur-
poses of comparison a few experiments were also conducted with hexane and methane,
representing  (H/C) ratios of 1, 2.3, and 4, respectively.  The oxidant used was
oxygen rather than air to allow comparison with other flame studies conducted
with oxygen, and to assure that the origin of any nitrogen contained in poly-
cyclics formed in the inner cone of the flame would have to be the fuel rather
than atmospheric nitrogen.

     The use of pure oxygen rather than air led to a benzene flame speed greater
than could be stabilized at convenient gas flows on a one-inch diameter burner,
without resorting to a fuel equivalence ratio () greater than 7.  As a flame this
rich was undesirable, methane was added to the fuel in order to lower the flame
speed to an acceptable level.  A smaller diameter burner was not substituted
since the water-cooled sampling probe might create too much disturbance in a
smaller flame.
     The benzene and hexane studies were conducted with a CH^/CgHx (x = 6 or 14)
ratio of 1.5 and a <(> = 4.5.  For all studies the cold gas velocity (v) was kept
constant at 31.8 cm/sec.  Soot collected by the probe was trapped in glass wool
packed filters and the PCAH extracted and analyzed by gas chromatography.
                       APPARATUS AND EXPERIMENTAL DESIGN

     The arrangement of the apparatus is shown in Figure 1.  The burner used
was a modified Meker burner (see Fig. 2).   The four air inlets were replaced
with oxygen feeds, and the fuel inlet was attached to 0.25 inch stainless
steel tube through which the benzene/methane mixture was fed.  The fuel mixture
was preheated electrically with heating tape to vaporize the benzene.  The
oxygen feeds were also preheated to approximately the same temperature (300-
320°C) to prevent benzene condensation inside the burner.

     Temperatures were measured via thermocouples in the lines; CH^ and Q£
flows were monitored by critical orifices, and the benzene flow was measured
before vaporization by passing the liquid through a capillary tube with the
upstream and downstream pressures carefully measured.  The liquid flows used
corresponded to Reynolds numbers of between 800-1000 through the capillary
tube.  Thus the flow was always laminar and proportional to the pressure drop
through the tube.


                                      43

-------
                                                       Exhaust
                                                                                & soot out
                 Critical
                 Orifice
                 Flowmeter
Ar
      liquid fuel tank
                                                                                                                    exhaust
glass wool packed
        filter
                                                                  Heating
                                                               p   ape     water~collection
                                                                                                               ice  bath
                                               FIGURE 1.   Overall Experimental Arrangements

-------
fuel
                                                      Top View
                                                     1/8" square
                                                        grid
                                                      Ventrur i
                                                                          glass
                                                                          shield
                                                                             base  to support
                                                                             glass shield
                                                                             V
                                                          Side  View
                       heating  tape
                                     FIGURE  2.   Burner  Details

-------
                             EXPERIMENTAL PROCEDURE

     A water cooled probe (Figure 3) was used to collect the soot, which was
trapped in a glass wool packed tube.  One difference between the present sampling
technique and the one used previously in this laboratory was necessary.  The
extremely sooty benzene flame caused some clogging of the probe due to soot
build-up at the tip; consequently the probe was kept out of the flame until all
conditions were ready for the actual sample to be collected.  The bypass
filter, which previously collected "extraneous soot" before and after the actual
sampling run,- was totally eliminated.

     Four additives were used for the doping procedure.  Two compounds (pyrrole
or pyridine); were added to obtain 1% by weight nitrogen in the benzene, and
two other compounds (thiophene or carbon disulfide) were used to obtain 1% sulfur.
One percent of N or S was selected as being representative of the practical
interest for many fuels.

     The amount of gas collected through the probe was monitored using a wet
test meter.  As usual, the soot was extracted with acetone and methylene chloride
to obtain the yellow-orange colored PCAH.  The analysis of these solutions is
achieved with GC, and: as necessary, combined GC/MS techniques.  Soot concentrations
were calculated by weighing the soot collected (after extraction of PCAH) and
knowing the amount of gas sampled.

     The flame was sampled at different heights above burner (HAB) to obtain
the soot concentration profiles.


                             RESULTS AND DISCUSSION

     Figure 4 is a composite for all four additives to benzene.  It is assumed
that the soot concentration is approximately independent of the additive; thus
the composite figure should give a fairly good representation of the actual
soot concentration.  The validity of this approximation is based on the
relatively small amounts of additive required to obtain the desired 1% N and
S concentrations, and the similarity of the chemical structure of these com-
pounds to benzene.  The 1% concentrations were calculated based on molecular
weights and densities, as follows (using CS_ as an example):

                                 MW  CS2 = 76.14

                                 Atomic wt. S in CS  = 64.14

                                 MW benzene = 78.11

                                      46

-------
Sample
  In
                                  - Water   Out
                                    Water  In
           :^^^
         Water  Spray

Sample + Water
to  Filter
FIGURE 3.  Water-Cooled Sampling Probe

-------
   100
    80
    60
C
01
(J
C
o
(J
o
o
CO
                                                                     4.5
     x  =  CS
    (•) =  Pyrrole


    O =  Pyridine


    Q=  Thiophene
                                                      FIGURE  4.   Variation  of  Soot Concentration

                                                       with  Height  Above  Burner for Each of

                                                       Four  Additives  in  Benzene
              20
                        40        60         80        100


                             Height Above  Burner, mm
120
                                          48

-------
     Letting x = wt. fraction (benzene/CS ) ,
                                76.14 + x(78.11)
                                                 =
Solving, x = 81. 1A.  This implies that we need 1 g of CS2 per 81.14 g of benzene.
Using densities @ 20°C,

                                density C^ = 0.8786 g/ml

                                density CS2 = 1..2632 g/ml


one obtains 1.00 ml of CS0 per 116.67 ml of C,H, .
                         L                   a b

     This is admittedly the lowest ratio of the four additives, but for the
other three compounds the (C/H) ratio is approximately one, as is that of
benzene.  Furthermore the structures are all or mostly aromatic, benzene being
of course totally aromatic (see Table 1) .  In all cases the carbon atoms are
in aromatic or at least double-bonded positions.  Thus we confidently make, the
aforementioned approximation, and also consider only the €5^5 and CH4 in the
balanced chemical reaction to determine the 4.5 fuel equivalence ratio.

     The benzene/oxygen/ uethane flame had a very bright and distinct inner
cone of approximately 70 mm height, thus points shown in Fig. 4 below 70 mm
HAB can be assumed to be similar to a premixed flame, while those above 70 mm
HAB are from the diffusion-like tail of the flame.  Concentrations are reported
in ug/cm^ at 20°C and 1 atm.  on a water free basis.

     The curve in Fig. 4 has been drawn to favor the lower concentration points
(at a given HAB) especially at low HAB.  The reason for this is that the small
shell-like diffusion flame which formed around the edges of the burner was
observed to have much more soot than inside the flame (probably due to low
temperatures which prevented soot burnout), and due to the sampling probe
being moved quickly in and out of the flame at the beginning and ending of the
sample collection, the probe was exposed to this soot for a fraction of a
second.  This effect is negligible in the upper portions of the flame (> 70 mm)
where there was so much soot that 40-50 mg were usually collected; however, in
the lower portions of the flame ( < 40 mm), this effect might have some small
significance since usually less than 10 mg of soot was collected.  The curve
is thus drawn very low at HAB's less than 40 mm, despite some high points in
this region.

     Unfortunately, Fig. 4 does not represent the whole story, as some highly
suspect points have been removed.  These points are shown in Figs. 5-8, which
represent each of the additives separately.  The curves drawn through the. data
ignore the suspected bad points.  These curves tend to differ slightly from
the composite graph (Fig. 4), probably due only to an insufficient number of
points.
                                       49

-------
Table 1.  Information on Benzene and Additives
Compound
Benzene
Pyridine
Pyrrole


Thiophene
Carbon
Disulfide
Structure
0
0
o
T
H
Q
S-CoS

MW
78.11
79.1
67.04


84.14
76.14

Density
.8786
.9812
.9691


1.0882
1.2632

Volume ratio
ml additive
ml benzene
1% N or S
-
1.00/18.88
1.00/18.81


1.00/49.52
1.00/116.64

C/H ratio
1.0/1.0
1/1
4/5


1/1
1/0

% Aromatic
Atoms
(C,S, or N)
100%
100%
80%
(not N)

80%
(not S)
100%

                                     50

-------
  100
   80
   60
00
3.
                                                                     <(> = 4.5

                                                                  C,H/0,/CH
                                                                   662   4
                                                                 5H N (Pyridine)
                                                                      doped
                                        FIGURE 5.  Variation of Soot Concentration with
                                          Height Above Burner for Pyridine Additive in
                                           Benzene
              20
60       80        100      120

   Height Above Burner, mm
                                                                      140
                                               .51

-------
  100
oc
3.
   80
   60
0
   40
                                                                  <(> = 4.5



                                                                C6H6/02/CH4



                                                                  C^H N  (Pyrrole)

                                                                         doped
   20
                                         FIGURE 6. Variation of Soot Concentration with

                                           Height Above Burner for Pyrrole Additive in

                                           Benzene
             20        40       60        80        100

                                  Height Above Burner, mm
                                                             120
                                                                       140
                                          52

-------
   100
    80
t>o
3.
C
a)
o
c
o
o
    60
       4.5
                                                             C,H,/0,/CH;
                                                              662   4
                                                             C.H,S (Thiophene)
                                                              44,    ,
                                                                   doped
    20
                                         FIGURE 7.  Variation of Soot Concentration with

                                           Height Above Burner for Thiophene  Additive in

                                           Benzene
              20        40       60        80       100


                                 Height Above Burner, nun
120
                                       53

-------
   100
    80
00
3.
c
o
cj
    60
                                                                  =  4.5



                                                                C,H,/0,/CH.
                                                                662    4



                                                                  CS-  doped
                                       FIGURE 8.  Variation  of  Soot  Concentration with

                                         Height Above  Burner for  Carbon Disulfide Additive

                                         in Benzene
    20
                                  60       80       100


                                    Height Above Burner, mm
                                                              120
                                           54

-------
     A circle around a point indicates that it does not appear on the composite.
The reasons why the points are believed to be in error are listed below with
numbers, referring to the numbers in the figures.
(1) Sample was taken at a height greater than 110 mm, where, due to quivering
of the diffusion flame, it was difficult to tell when the probe was in the
flame and when it was not.
(2) A very large amount of soot was collected and some leaked through the
filter during the PCAH extraction.
(3) The probe appeared to clog slightly.
(4) The probe was exposed to the outer diffusion flame shell (discussed above)
for much longer than usual, typically due to the probe slipping during a run.
(5) Much lower than usual gas suction rate through the filter.
(6) Much higher than usual gas suction rate through the filter.
(7) Filter containing soot chipped after completion of sampling.  Pieces were
weighed, but a small piece of glass must have been lost.

     Two other fuels (methane and hexane) were also tested on this burner at
the same 4> and v.  Graphs of data from these two cases are shown in Fig. 9 and
10.  The methane/oxygen flame was not doped with any additive, but the hexane/
oxygen/methane flame contained both thiophene and pyrrole (both five member
rings) in proportions such that the final liquid mixture had 1% N and 1% S.
The (CH^/CgH^) ratio was also set at 1.5, the same as for the benzene studies.
Despite the same ()> and v, neither the hexane nor the methane flame exhibited
an inner cone; however the shape of the soot concentration curves are approxi-
matly the same as before.  The range of soot concentration curves are approxi-
mately the same as before.  The range of soot concentrations with hexane is
l/25th of that for benzene (from 0-100 yg/cm3 to 0-4 yg/cm3) and for methane
it is l/50th (from 0-100 yg/cm3 to 0-2 yg/cm3) .

     It was earlier stated that the presence of the additive was negligible
calculating the fuel equivalence ratio.  This approximation may become suspect
when one extends it to two additives, as was done with hexane.  However, cal-
culations show that the approximation is still very good.

     A few samples were also made at a <)> = 4.0, using once again a benzene
and methane mixture as fuel.  In order to attain this lower fy with the same v
on our one- inch diameter burner, the ratio (CH^/CfcHfc) unfortunately had to be
increased from 1.5 to 2.0.  The resultant flame had an inner cone height of
only 54 mm, rather than 70 as with the (CI^/CgHfc) ratio = 1.5 and  = 4.5.
     The presence of more methane and less benzene, as well as a lower $ would
lead one to expect lower soot concentrations at a given HAB, yet a comparison
of Figs. 4 and 11 show that this is not the case.  This phenomena can perhaps
best be explained by the difference in inner cone heights.  If one were to
plot the concentration versus the height, above this inner cone, the resultant
shift to the right of the  = 4.0 curve would bring it under the 
-------
  2.00
  1.60
E
u
eo
   1.2
o
w  .80
   .40
              20
                                                         Pure
                                                              =  4.5
                                             FIGURE 9.   Variation  of  Soot  Concentration with
                                               Height Above  Burner for  Pure Methane Fuel with
                                               No Additive
                                60        80       100

                                Height Above Burner, mm
                                                           120
                                             56

-------
 4.0
3.20
                                                                  <(> = A.5

                                                               C6'V°2/CH4
                                                            C.H,S  (Thiophene) and
                                                             4 4
                                                            C H N  (Pyrrole)
                                                                     doped
                                           FIGURE 10.  Variation of Soot Concentration
                                             with Height Above Burner for Thiophene ar
-------
   100
0)
u
c
o
u
o
o
in
    80
   60
                                                            C.H /O./CH.
                                                             662    4
                                                                    4.0
cs_
                                                          Q =  CHS
                                                                4  4
                                 FIGURE  11.   Variation  of  Soot  Concentration with Height

                                  Above Burner  for  Each of  Three Additives in Benzene

                                  and a Lower Equivalence Ratio
               20
                        40
                                 60
                                           80        ion


                                        Above  liurnor,  mm
                                                             120

-------
                               SUMMARY AND CONCLUSIONS

     The benzene/methane mixture used to obtain the majority of data in this
report was approximately 25 times more sooty than the hexane/methane mixture
and 50 times more sooty than the pure methane/oxygen flame.  The general shape
of the soot concentration profiles are about the same for all fuels, as they
should be since all of the cold gas velocities were the same.

     The data for each individual additive is somewhat sketchy, but the compo-
site seems to yield a fairly firm picture of the soot concentration profile.
This is probably due to many sources of error inherent in the techniques used,
necessitating a large number of data points.  The single most important source
of error is the deposition of soot at the inlet to the probe, sometimes making
the opening smaller or even clogging it.

     The data taken at a $ = 4.0 do not exhibit much, if any, evidence of repro-
ducibility.  Apparently many more runs would have been necessary at these condi-
tions to obtain a firm idea concerning the actual location of the concentration
profile.  Consequently these results are only included here for the sake of
completeness, not because they convey too much meaning.
                                       59

-------
                               APPENDIX - DATA TABLE
Notes for Table:
      (a)  A blank indicates an entry is unchanged from the last listed
      value, except in the Error Code Column, where a blank indicates  that
      the concentration reported is reasonably accurate.
      (b)  Fuel column indicates fuel in addition to CH^, which was present
      in all runs.
      (c)  All (|> = 4.0 runs have (CH^CeHg) = 2.0, whereas all  = 4.5
      runs have (CH/CH) =1.5
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Additive HAS
1 % N or S (mm)  Fuel
CS0 20.7 4.5 C,H,
i DO
a. 2
15.7
71.05
61.3
C.H_N 80.85
4 5
53.25
26.65
101.6
116.35
111.25 4.0
85.0
95.45
63.65
59.95
33.1
12.5
Soot
(mg)
11.0
2.6
11.2
24.8
59.1
25.3
32.8
4.2
65.95
40.1
25.76
200.7
55.9
89.3
73.3
14.6
2.1
Gas
U)
NA
13.5
15.5
1.05
1.53
0.592
1.59
1.23
0.943
0.943
0.298
2.22
0.732
1.06
1.09
.898
1.14
Cone . Error
(yg/cm3) Code
NA NA
.19
.72
23.76
38.56
42.73
20.7
3.42
68.87
42.57 1,3
86.27
90.19
76.48
84.53
67.28
16.28
1.88
                                                                  (continued)
                                       60

-------
APPENDIX - DATA TABLE (continued)
Run Additive
No. 1 % N or
18 C5H5N
19
20
21
22
23
24 C.H.S
4 4
25
26
27
28
29
30
31
32
33
34 CS2
35
36
37
38
HAB
S (mm) <(>
11.
31.
71.
109.
89.
136.
148.
116.
88.
63.
37.
17.
45.
89.
110.
91.
88.
112
38.
68.
84.
95 4.5
5
4
3
9
9
7
1
8
9
9
7
2
6
0
95 4.0
2 .4.5

1
0
2
Soot
Fuel (mg)
3
4
56
133
33
102
88
50
60
32
1
0
4
26
51
58
C,H, 38
o o
89
0
20
62
.37
.23
.67
.5
.2
.8
.3
.1
.9
.8
.8
.35
.4
.6
.2
.4
.5
.6
.73
.2
.6
Gas
1
1
0
1
0
1
1
0
1
0
1
0
0
0
0
0
0
0
0
0
0
.153
.18
.846
.585
.737
.339
.172
.758
.58
.876
.15
.547
.903
.383
.492
.547
.410
.986
.355
.465
.657
Cone. Error
(pg/cnH) Code
2
3
66
84
42
76
75
67
38
37
1
0
4
69
103
101
93
90
2
43
95
.92 4
.58
.96
.27
.05 5
.77 1,2
.42 1
.2 1,2,5
.51 2
.5
.55
.639
.85
.47
.95 6,1
,7
.83 4?
.89
.05
.4
.3 6
                                           (continued)
               61

-------
APPENDIX - DATA TABLE (continued)

Run
No.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
Additive HAB
1 % N or S (mm) $ Fuel
63.9 4.0
C4H5N 70.0 4.5
86.6
102.1
76.25 4.0
55.5
43.6
C5H5N 49.2 4.5
60.95
14.15
86.5
98.95
None 98.95 CH.
4
83.25
83.25
70.3
54.85
52.6
67.9
39.4
Soot
(rag)
41.2
28.1
46.6
51.7
22.6
25.9
7.1
22.7
23.2
0.93
55.57
44.3
2.38
0.19
1.39
0.62
0.13
0.51
5.89
0.07
Gas
0.542
0.542
0.488
0.484
0.456
0.491
0.573
0.607
0.496
0.685
0.712
0.493
1.367
2.94
1.097
1.59
1.426
13.715
.13.192
21.147
Cone.
(yg/cm )
76.07
51.77
95.6
106.71
41.45
52.85
12.4
37.38
46.65
1.36
78.0
89.88
1.78
0.065
1.266
.3897
0.091
0.037
0.446
0.0033
Error
Code


6
4?



4





7






        62
                                    (continued)

-------
APPENDIX - DATA TABLE (continued)
Run Additive HAB
No. 1 % N or S (mm) 4>
59 C.H.S 25.5
4 4
60 and 44.05
61
62
63
64
53.35
71.25
90.25
110.3
Soot
Fuel (mg)
C,H,. 1.01
6 14
1.02
1.68
6.06
6.53
10.45
Gas
U)
12.35
3.81
11.59
5.902
5.048
2.91
Cone . o
(yg/cm )
0.082
0.267
0.145
1.027
1.294
3.592
Error
Code






               63

-------
             APPENDIX D
       Source Identification
    of Urban Airborn Polycyclic
    Aromatic Hydrocarbons by Gas
  Chromatographic Mass Spectrometry
and High Resolution Mass Spectrometry

 From:   Biomedical Mass Spectrometry,
        4, 182 (,1977).  Reprinted
        with the permission of Heyden
        & Son Ltd.
                 64

-------
       Source Identification  of  Urban  Airborne
       Poiycydic Aromatic Hydrocarbons by Gas
                           raphic Mass  Specfrometry  and High
       Resolution  Mass Spectrometryt
       M. L. Lee, G. P. Prado, J. B. Howard and Ronald A. Hitest
       Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139. U.S.A.
       Poiycydic aromatic hydrocarbons formed during the combustion of three common fuels (coal, wood and
       kerosene) were separated and identified by capillary-column gas chromatographic mass spectrometry and were
       compared to airborne polycydic aromatic hydrocarbons from Indianapolis, a high coal consuming area, and
       Boston, a low coal consuming area. High resolution mass spectral data were utilized in the construction of alkyl
       homolog plots for the comparison of alkyl distribution within each sample.
INTRODUCTION
The suspected relationship between atmospheric pollu-
tion and lung cancer is not yet substantiated with good
epidemiological and analytical surveillance data to pro-
vide good quantitative correlations. It seems clear, how-
ever, that a relationship does exist. For example, the
incidence of lung cancer is twice as high among city
dwellers as among rural residents, and it is most common
in cities where general industrial  pollution  is the
heaviest.1 Polycyclic aromatic hydrocarbons (PAH),
well known constituents of air paniculate matter, have
been credited with being the major class of compounds
responsible  for  this observed carcinogenic activity.
Many specific PAH have been shown to be carcinogenic
in animals; and, although  inferences for man from
experimental studies involving animals must be  drawn
with  particular  caution,  there are good  reasons to
believe that airborne PAH participate in the induction
of human lung cancej.1
  The distribution of PAH in a particular atmospheric
environment, and hence, the carcinogenic potential of
that environment, is largely dependent on the sum of its
sources. PAH are generally produced from combustion
processes utilizing fossil fuels, and  their formation is
dependent on various combustion parameters such as
fuel type, temperature and fuel-to-air ratio. Distribu-
tions of PAH  in emissions from different combustion
sources differ considerably in type  and concentration of
species. Thus, the sum of contributing sources generally
produces an air particulate PAH  fraction which is
unique to the area sampled.2'3
  It  has been  determined  that  approximately 90%
of the PAH emissions in the United States is due to
coal combustion processes, including coal-fired furnaces
and coal-refuse bank burning, and to coke production.4
Most of these processes are concentrated in the mid-
western states. Coal consumption is highest in Ohio and
t Abbreviation: PAH = polycyclic aromatic hydrocarbons.
t To whom correspondence should be addressed.
Pennsylvania with each state consuming more than 60
million tons in 1970." Michigan, Indiana, and Illinois
each used between 26 and 50 million tons." Contrast
these figures with a total of 4.6 million tons consumed by
all of New England during the same period of time.4 It
seems reasonable to  expect, therefore, that the PAH
isolated from the combustion products of coal would be
quite well  represented in  midwestern air particulate
matter. On the other hand, New England airborne PAH
should  be similar to  the products obtained from the
combustion of fuel oil and gasoline since these are the
major sources of PAH emissions in New England.
  The objectives of this study were twofold. First, the
PAH fractions obtained from the combustion of three
common fuels (coal, wood and kerosene) were analyzed
to determine their detailed compositions. Capillary-
column gas chromatography combined with mass spec-
trometry was utilized for separation and identification of
the mixture constituents. Of particular interest was the
identification of the combustion products of coal, which
is the major source of atmospheric PAH. Since the usage
of coal is likely to increase because of the present energy
crisis, information on the PAH distribution in coal com-
bustion products is valuable for future studies concern-
ing the health effects of new modes of energy produc-
tion. Second, the distributions of PAH obtained from
the three different fuels were compared with those of air
particulate matter from a high coal consumption area
(Indiana) and a low coal consumption area (Mas-
sachusetts) by high resolution mass spectrometry to see
if correlations could  be made between fuel utilization
and air particulate matter.


EXPERIMENTAL
The apparatus used for burning coal and wood samples
was constructed by placing a stainless steel wire mesh
screen (no asbestos) on a tripod and placing a 16 cmx
10 cm i.d. metal cylinder endwise on the screen. About
ten pieces (approximately 20 g each) of  a Pittsburgh
seam bituminous coal or wood chips from a construction
                                               65

-------
grade fir were placed in the chimney (cylinder) and
ignited from  underneath with  a  Bunsen burner. The
burner was removed, and soot samples were collected by
repeatedly lowering a 250 ml Erlenmeyer flask, which
was cooled by a continuous flow  of water through the
flask, into the flame for about 30 min, at which time the
fuel was completely consumed. The soot collected on the
glass surface  was scraped with a spatula into a small
beaker. Each soot sample (250-400 mg) was then placed
in a Soxhlet  apparatus and extracted  with  250 ml of
methylene chloride for 18 h. The extract was evaporated
to dryness with a rotary evaporator, dissolved in 50 ml of
cyclohexane and transferrred to  a  separatory funnel.
The cyclohexane layer was then washed five times with
50 ml of nitromethane. All nitromethane portions were
combined and evaporated to dryness. The residue was
then transferred to a 1 cm i.d. glass column packed with
2 g of silicic acid and eluted with 300 ml of hexane. The
eluate was evaporated to  dryness  and dissolved in 0.5-
5 ml of methylene chloride prior to gas chromatographic
analysis.  This procedure was shown  previously to
be effective in separating and purifying the PAH from
the bulk of other organic  compounds in the extract.5 It
should be noted that nitrogen-containing polycyclics are
removed  by   this procedure  but  sulfur-containing
polycyclics are not.
  Kerosene was burned in a specially constructed tur-
bulent diffusion  flame   burner  using  an   air-assist
atomizer.6 The air-atomizing  pressure was  170 KPa
(101bin~2g),  the cold gas velocity was 0.96ms~' and
the fuel-air equivalence ratio was 1 (stoichiometry). This
burner and these conditions simulate  a poorly main-
tained home-heating furnace. Soot samples were col-
lected from this burner  using a novel  water-cooled
probe7 at the exhaust of the combustion chamber (50 cm
from  the burner  nozzle).  The  collected  soot was
extracted with methylene chloride and the extract con-
centrated for gas chromatographic analysis with a rotary
evaporator. All glassware was acid cleaned prior to use,
and all solvents were Nanograde (Mallinckrodt) quality.
Blank runs demonstrated the absence of contamination.
The total yield of PAH in each sample was 2-3% of the
soot extracted.
  The PAH samples isolated from air paniculate matter
from Indiana  and Massachusetts were utilized in previ-
ous studies and descriptions of each sample work-up
procedure can be found elsewhere.8'9
  Gas  chromatographic  mass  spectrometry of each
sample  was   performed  with  a  Hewlett-Packard
5982Ag.c.m.s. system interfaced  to a HP 5933A data
system. A 19 mx 0.26 mm i.d. glass capillary column
coated with SE-52 methyl-phenyl-silicone  stationary
phase was used  to separate mixtures prior to mass
spectral analysis.  The oven temperature  was  pro-
grammed  from 70 °C to  250 °C  at 2°Cmin^'  during
each chromatographic run. The mass range (from 50 to
350a.m.u.) was continuously scanned every  2.5 s. The
mass  spectrometer was  operated  at  70 eV ionizing
energy.
  High  resolution  mass spectral   information  was
obtained on each sample by introducing an aliquot of the
methylene chloride extract into a high resolution mass
spectrometer  through  a direct introduction probe sys-
tem and slowly vaporizing the sample at  a continually
increasing temperature while making several exposures
on a photographic plate. The developed plate was read
on a computerized comparator, and the exact masses
were converted to elemental composition. The corre-
sponding intensities were  then  arranged as  tables  of
carbon number vs. number of double bonds and  rings.
One such table was generated  for each exposure and
then a composite table was formed by adding the corre-
sponding entries  of  each table. This  technique has
been described previously.ia"  The results from this
high resolution m.s. technique have been validated by a
comparison with quantitative g.c. data obtained on the
Indianapolis air particulate extract.  The two methods
give results which are highly correlated (r = 0.93, N = 6)
and thus can be considered self-consistent.
  The high resolution m.s. system consists of a DuPont
Instruments 21-110B mass spectrometer and a D. W.
Mann comparator interfaced to an IBM 1802 computer.
The ionizing energy was 70 eV and the resolving power
was about 20 000. This high resolution  m.s. system and
its operation are described elsewhere.12
RESULTS AND DISCUSSION	

Considerable emphasis has been placed on the necessity
for improved resolution of components in complex PAH
mixtures because of the ongoing interest in exact iden-
tification  of  potential   carcinogens.13'1'?  Capillary-
column g.c. has proven to be the most efficient separa-
tion tool presently available for the analysis of PAH, and
its use has provided the means for extremely detailed
studies of  PAH mixtures.8-u-15 Figures 1, 2 and 3 show
high resolution gas chromatograms of the PAH fractions
obtained from the three fuels studied here. Table 1 lists
the compounds identified by g.c.m.S: and g.c. retention
data. Several qualitative  and semiquantitative observa-
tions  result from  the comparison  of  these chromato-
grams: (1) there is a  greater relative  concentration of
alkylated PAH in the coal combustion  products as com-
pared with either wood or kerosene; (2) there is a greater
relative concentration of high molecular weight species
in the wood and kerosene combustion  products as com-
pared with  coal;  (3)  the  coal  soot PAH  fraction
contains   significant   amounts  of  sulfur-containing
compounds, which are absent in the other combustion
products.
  The construction of alkyl homolog plots from high
resolution mass spectral data  is an  effective  way of
representing the distribution of alkylated species within
a PAH mixture.10'11 Figure  4 shows the alkyl homolog
distribution for seven major groups of isomers found in
the coal soot PAH fraction. In agreement with the gas
chromatogram, there tend to be considerable amounts
of alkylated species. The high  sensitivity of this tech-
nique shows the presence of alkyl groups containing up
to six carbons for the phenanthrene type series (Z = —
18) and  five carbons for  the pyrene type  series (Z = -
22). The comparison of the Z= -22 series for coal,
wood and kerosene combustion products with  air par-
ticulate matter from Indianapolis and Boston is shown in
Fig. 5. There is a  remarkable similarity between the
curves for coal-soot PAH and Indianapolis air particu-
lates as  well as between the kerosene-soot  PAH and
Boston air particulates.
                                                  66

-------
70
90
110
130
ISO
170
190
210
230
230
Temp. PC)


Time(min)   Q        10       20       30       40       50        60       70      80        90       110

 Figure 1. Capillary-column gas chromatogram of the PAH fraction of coal combustion products. Conditions: see text. Key: see Table 1.
Temp.(°C)
Tlme(mln)  0
70
90
110
130
150
170
190
210
230
250
                   10
                           20
                                    30
                                             40
                                                      50
                                                                60
                                                                         70
                                                                                 80
                                                                                           90
                                                                                                   110
Figure 2. Capillary-column gas chromatogram of the PAH fraction of wood combustion products. Conditions: see text. Key: see Table 1.
Early chromatographic peeks with less retention than acenaphthylene have been identified as substituted methoxyphenols and several
other nonpolycyclic compounds.
   Figure  6 represents composite alkyl homolog plots
.which were constructed by adding intensities of all par-
ent  compounds, all  Ci  alkylated  compounds,  all  C2
alkylated compounds, etc., for all series represented in
Fig. 4  and  plotting the resulting sums.  Again, the
similarities observed in Fig. 5 are evident in this compo-
site. The similarity in the  results obtained  from the
turbulent diffusion flame burner and Boston air particu-
lates is consistent with the fact  that fuel  oil (which is
chemically  similar to kerosene) is Boston's principal
heating and energy producing fuel, while coal burning is
practically nonexistent.
  The sulfur-containing polycyclic aromatic species in
the coal combustion products (see Table 1) are identical
to those found in the air particulates from Indianapolis.8
This observation, in addition to the alkyl homolog dis-
tributions, lends support to  the belief that  the major
contributors  to  Indianapolis  airborne PAH are coal
combustion processes. The relatively low concentration
of high molecular weight compounds in the coal soot
                                                      67

-------
70
90
no
130
150
170
190
210
Z30
250
Temp. (°C)


Time (min)  0        10       20        30       40       50        60       70       80        90       110

Figure 3.  Capillary-column gas chromatogram of the PAH fraction of kerosene combustion products. Conditions: see text. Key: Table 1.
PAH fraction is not consistent, however, with what has
been observed in the Indianapolis air participates. There
are three possible explanations for this difference. First,
it was found that by altering the combustion conditions
(especially  temperature) in the production of carbon
                        blacks15 the  relative proportion  of  higher molecular
                        weight to lower molecular weight PAH species could be
                        either considerably increased or decreased. The same
                        trends are certainly present in the combustion of coal.
                        Second, the practice  6 of sampling air paniculate matter
Table 1. PAH identified by g.c.mj.

  Peak No.         Compound
     1   Methylnaphthalene
     2   Biphenyl
     3   Ethylnaphthalene"
     4   Acenaphthylene*
     5   Methylbiphenyl
     6   Dibenzofuran
     7-  Propylnaphthalene"
     8   Fluorene
     9   Methyldibenzofuran
    10   C,4He"
    11   Methylfluorene
    12   Ethyldibenzofuran*
    13   Dibenzothiophene
    14   Phenanthrene
    15   Anthracene
    16   Ethylfluorene'

* Could be dimethyl.
b Could be trimethyl or ethylmethyl.
c Could be methylanthracene.
" Could be methylpyrene.
eak No.               Compound                 Peak No.
 17   Propyldibenzofuran"                       33
 18   Methylphenanthrene"                      34
 19   4H-cyclopentaIde/'lphenanthrene             35
 20   Methyl-4W-cyclopenta[der']phenanthrene      36
 21    Ethylphenanthrene*                        37
 22   Fluoranthene                             38
 23   Benz(e]acenaphthylene                     39
 24   Benzo[def1dibenzothiophene                40
 25   Pyrene                                  41
 26   Ethyl-4H-cyclopenta[derlphenanthrene°       42
 27   Methylfluoranthene"                       43
 28   Benzo|alfluorene                          44
 29   Benzo[b]fluorene                          45
 30   Benzo[g/i/)fluoranthene                     46
 31    C18H,0 (unknown)                          47
 32   Cyclopentalcdlpyrene                      48
         Compound
Benz[a|anthracene
Chrysene
Methylchrysene*
Methylcyclopenta(cdjpyrene'
Benzofluoranthene
Benzo(e]pyrene
Benzo[a)pyrene
Perylene
Methylbenzopyrene0
C21H12 (unknown)
C21H12 (unknown)
lndeno[1,2,3-cd]pyrene
Dibenz(a,/i]anthracene
Dibenz[a,c]anthracene
Benzo[g/7/]perylene
Anthanthrene
                        * Could be methylbenz(a]anthracene.
                        'Could be methylbenzo[0/i/]fluoranthene.
                        0 Could be methylbenzofluoranthene.
                        h Probably cyclopentlbc or fglacenaphthylene, see Ref. 17.
on glass fiber filters followed by drying at slightly ele-
vated temperatures for accurate weight measurements
tends to volatilize many of the lower molecular weight
compounds and enrich the sample in higher molecular
weight  species. The third explanation is drawn from
experiments with the turbulent diffusion flame burner in
which there was  an enrichment of  higher molecular
weight  PAH as samples were collected at greater dis-
                        tances from the burner nozzle.   Since the coal- and
                        wood-soot samples were collected from within the flame
                        instead of at greater distances, it is possible that these
                        samples were not as enriched with high molecular weight
                        species as the exhaust from a typical coal combustor.
                          The  complementary  use of  high  resolution gas
                        chromatographic mass spectrometry for detailed com-
                        pound identification and of high resolution mass spec-
                                                        50
                                                        o

-------
        0    14      16     18     20
                         Corbon number
Rgure 4. Alkyl homolog distribution plots for several PAH series
in coal combustion products. The lines are labeled by the molecu-
lar weight of the unsubstituted species. Example isomers are:
178   (Z=-18)phena'nthrene,   202   (Z=-22)pyrene,   226
(Z= -26)cyclopenta(co1pyrsne, 228  (Z= -24)chrysene,  252
(7=-28)benzo[s]pyrene, 276  (Z=-32)indeno[1,2,3-cd]pyrene
and 278 (Z = -30)dibenzanthracene. The value of Z is derived
from the general formula CnH2n+z.
trometry for alkyl distribution monitoring can greatly
aid in the characterization of airborne PAH and in the
identification of pollution sources. This  information is
important for the engineer to design those cleaner com-
bustion processes  which will be necessary in the light of
the expected increase in the consumption of coal.
Acknowledgements

The authors thank C. Clampitt and J. Dillon for technical assistance
during this project. This work was supported by Grant No. R803242
from the Environmental Protection Agency.
                                                                   20
                                                                       202
                                                                                    ~Air pnrticulote (IndionuDolis)
                         I8      19
                        Carbon number
                                                                                                    20
Rgure 5. Alkyl homolog distribution plots for the pyrene-type
series (Z= -22) in the combustion products of coal, wood and
kerosene, and in air paniculate matter from Indianapolis and
Boston. The abundance of the parent compound in each series
was normalized to 100.
     100 -
       0   C0     C,      C
                 Number of olkyl sidechoin carbons
Figure 6. Composite alkyl homolog distribution plots (see text)
for PAH in the combustion products of coal, wood and kerosene,
and in air paniculate  matter from Indianapolis and Boston. The
abundance of the parent composite in each series was nor-
malized to 100.
                                                 REFERENCES
 1. Paniculate Polycyclic Organic Matter. National Academy of
    Sciences, Washington, D.C. (1972).
 2. K. D. Bartle, M. L. Lee and M. Novotny. Int. J. Environ. Anal.
    Chem. 3, 349 (1974).
 3. J. L. Shultz, A. G. Sharkey and R. A. Friedel, Biomed. Mass
    Spectrom. 1, 137(1974).
 4. H. S. Stoker, S. L. Seager and R. L. Capener, Energy, p. 163.
    Scott Foresman, Glenyiew, Illinois (1975).
 5. M. L. Lee. Ph.D. Thesis, Indiana University (1975).
 6. J. P. Appleton and J. B.  Heywood, Fourteenth International
    Symposium on. Combustion, p. 777. The Combustion Insti-
    tute, Pittsburgh, Pennsylvania (1973).
 7. G. P. Prado, M. L. Lee, R. A. Hites, D. P. Hoult and J. B. Howard,
    Sixteenth   International  Symposium  on  Combustion.
    The Combustion Institute. In press.
 8. M. L. Lee, M. Novotny and K. 0. Bartle, Anal. Cham. 48 1566
    (1976).
 9. A. Hase and R. A.  Hites, in Identification and Analysis of
    Organic Pollutants in Water, edited by L. H. Keith, p. 205. Ann
    Arbor Science, Ann Arbor, Michigan (1976).
 10. R. A. Hites and W. G. Biemann. Adv. Chem. 147, 188 (1975).
 11. A. Hase and R. A. Hites, Geochim. Cosmochim. Acta 40,1141
    (1976).
 12. K. Biemann, Application of Computer Techniques in Chemi-
    cal Research, pp. 5-19. Institute of Petroleum, London (1972).
 13. 0. Hoffmann, W. E. Bondinell and 'i. L. Wynder, Science 183,
    215(1974).
 14. M. L.  Lee, M.  Novotny and K. 0. Bartle, Anal. Chem.  48, 405
    (1976).
 15. M. L. Lee and R. A. Hites, Anal. Chem. 48,1890 (1976).
 16. Paniculate Polycyclic Organic Matter. Appendix A. National
    Academy of Sciences, Washington, D. C. (1972).
 17. B. D. Crittenden and R. Long, Combust. FlameV), 359 (1973).
 Received 29 July 1976
                                                            69

-------
               APPENDIX E
Mixed Charge Exchange-Chemical lonization
Mass Spectrometry of Polycyclic Aromatic
              Hydrocarbons
    From:  Reprinted with permission
           from J. Amer. Chem. Soc.,
           99, 2008 (1977).  Copyright
           by the American Chemical
           Society.
                   70

-------
Mixed Charge Exchange-Chemical lonization Mass
Spectrometry of Polycyclic Aromatic Hydrocarbons
  The exact structural identification of polycyclic aromatic
hydrocarbons (PAH) and their alkylated derivatives is a dif-
ficult problem, particularly when they are encountered  as
complex mixtures. The analytical power of mass spectrometry,
which has had wide application in this Held,1'4 has been limited
because electron impact  mass spectra of isomeric PAH are
almost identical. The purpose  of this note is to report that
charge exchange-chemical ionization mass spectrometry, using
an argon-methane reagent gas,5 easily differentiates PAH
isomers.
  The mass spectra of a series of PAH were measured with a
Hewlett-Packard 5982A gas chromatographic-mass spec-
trometer system by injecting approximately 200  ng of each
compound (dissolved in methylcne chloride) on a  1 80 X 0.32
cm o.d. stainless steel column packed with 3% Dexsil 300 on
80/ 1 00 mesh Chromosorb W. The reagent gas mixture ( 1 0%
methane in argon) served as the carrier gas for the gas chro-
matographic column which was held isothermally at a tem-
perature appropriate to each sample being analyzed. The mass
spectrometer was continuously  scanned from 50 to 350 amu
at 81.2 amu/s. The ion source  pressure was 0.8 Torr and its
temperature was 1 70 °C. Data  were collected and processed
by a HP 5933A data system. Precautions were taken to assure
the absence of water vapor in  the ion source, since water is an
excellent proton donor and can greatly increase the abundance
of the protonated molecular ion. In these experiments, there
were no observable traces of water vapor (m/e 18 or  19).
  The resulting mass spectra showed considerable differences
in the  relative abundances of the molecular (M+) and pro-
tonated molecular (M + \+) ions when different PAH isomers
were analyzed. Table I lists the compounds analyzed in this
stu4y,  the resulting ratio of the abundance of the  protonated
molecular to molecular ion ((M + 1)/M), and the first ion-
 Table I. Abundance Ratios for Selected PAH Obtained by CH4-
 Ar Chemical lonization Mass Spectrometry

                                             Abundance
                             First ionization      ratio.
    Compound       Formula  potential (eV)°   (M + 1)/M*
Pentacene
Tetracene
Anthanthrene
Perylene
Benzo[a]pyrene
Anthracene
Benz[a]anthracene
Dibenz[a.A]anthra-
cene
Pyrene
Coronene
Benzo[f]pyrene
Acenaphlhene
Chrysene
Fluoranthene
Fluorene
Acenaphthylene
Phenanthrene
Triphenylene
Naphthalene
Benzene
C22H14
CI8H12
C22H,2
C20H12
C2rjH|2
C,4H,0
C18H,2
C22H|4
C,6HIO
C24H12
C20HI2
C12H,0
CI8H,2
C,6H,0
C,3H,0
CI2H8
CI4HIO
C]8H]2
CioH8
C6H6
6.42 (
6.88 (
7.02 (
7.03 (
7.17 (
7.42 (
7.47 (
7.55 (
7.56 (
7.58 (
7.58 (
7.70
7.74
7.76
7.86
8.02
8.02
8.11
8.14
9.29
).32
).45
).38
).32
).73
).82
1.83
J.95
J.73
).66
).82
.00
.26
.57
.66
.34
.59
.73
.68
>.79
ization potential of each compound. It is obvious from this table
that the (M + 1)/M ratio has a high positive correlation with
ionization potential (r = 0.877,/»« 0.01). This trend is con-
sistent with the  expectation that as the ionization  potential
increases, charge transfer processes will be less effective for
electron extraction while at the same time protonation becomes
more favorable.
  This technique should be quite useful for the elucidation of
         1.5
     M + I
      M
         1.0
             ocoo
   ° Values were averaged from experimental data found in ref 6-8;
 their variability was usually less than ±0.1 eV  * The reproducibilily
 of these measurements was ±4% over a 3-month period. The ratios
 have been corrected for the natural abundance of I3C.
                 7.0            75            8.0
                            I. R lev)
 Figure I. Plol of the abundance ratio ((M + I )/M) obtained by CH4-Ar
 chemical ionization mass spectrometry as a function or ionization potential
 (IP) for a series of four tetracyclic polycyclic aromatic hydrocarbons: I.
 tetracene; 2. benzja (anthracene: 3. chrysene; 4. Iriphcnylenc

 specific isomeric structures of PAH. By using a mixed charge
 exchange-chemical ionization reagent gas, such as described
 here, different mass spectra can be obtained  for most PAH
 isomers while conventional mass spectral techniques provide
 little differentiation. This fact is demonstrated by the series
 of tetracyclic compounds shown in Figure 1. The (M + 1 )/M
 ratio of each compound  is  plotted as a function of its first
 ionization potential. It is interesting to note that this abundance
 ratio increases from 0.45 to 1.73 as the isomer becomes more
 nonlinear, making differentiation quite easy.  If a standard
 PAH compound were not  available, it seems probable that the
 mass spectrum of that compound could be predicted from its
 ionization potential. The ability to calculate ionization po-
 tentials from molecular orbital  theory7-8 offers considerable
 promise for  the future identification of presently  unknown
 PAH.

   Acknowledgments. The authors thank D. P. Beggs (Hew-
 lett-Packard, Avondale, Pa.) and G. P. Arsenault (Chemistry
 Department, MIT, Cambridge, Mass.) for helpful suggestions
 concerning this work. This work was supported by grant
 R803242 from the U.S. Environmental Protection Agency.

 References and Notes

 (1)  R. C. Lao. R. S. Thomas. H. O|a. and L. Dubois, Anal. Cham.. 45, 908
    (1973).
 (2)  R. A. Kites and W. G. Blemann, Adv. Chem. Ser., No. 147, 188 (1975).
 (3)  M. L. Lee. K. D. Bartle. and M. Novotny. Anal. Chem.. 4B, 405 (1976).
 (4)  M. L. Lee and R. A. Hites. Anal. Chem.. 48, 1890 (1976).
 (5)  The ability of CH,-Ar mixtures to provide combined CE-CI spectra has been
    previously described by D. P. Beggs (Hewlett-Packard Applications Note
    No. 176-19). The methane acts as a tow energy proton donor which produces
    an Intense protonated molecular ion while the argon participates in charge
    exchange reactions to produce a fragmentation pattern normally found In
    electron impact mass spectra.
 (6)  H. Kuroda. nature (London).  201, 1214 (1964).
 (7)  M. J.  S. Dewar. F. R. S. Haselbach. and S. O. Worley. Proc. R See. London.
    Ser .A. 315, 431(1970).
 (8)  M. S. Sung. C. R. Acad. Sci.. Ser. C. 278, 37 (1974).

                                M. L. Lee, Ronald A. Hites*
                        Depart mem of Chemical Engineering
                       Massachusetts Institute of Technology
                           Cambridge. Massachusetts 02139
                                  Received August 17. 1976
                                                            71

-------
             APPENDIX F
Charge Exchange - Chemical lonization
  of Pplycyclic Aromatic Compounds
                  72

-------
                            Carcinogenesis, Vol. 3: Polynuclear Aromatic
                            Hydrocarbons, edited by P. W. -Jones and
                            R. I. Freudenthal. Raven Press, New York ©  1978.
       Charge Exchange-Chemical lonization Mass
    Spectrometry  of Polycyclic Aromatic Compounds

                Ronald A. Hites and George R. Dubay
     Massachusetts Institute of Technology, Department of Chemtciil Engineering,
                       Cambridge, Massachusetts 02139
   Certain polycyclic  aromatic  compounds (PAC)  have been  of interest
since the early 1800s because  of their carcinogenic behavior.  Since  only
specific PAC isomers  are carcinogenic,  it is important to have an analytical
tool which can differentiate  among  these  isomeric compounds. Although
GC-MS has had a wide application for the analysis of PAC, one problem
remains: isomeric  PAC give identical  electron  impact  mass spectra; thus,
the unambiguous identification of these  compounds is sometimes  impossible.
   We have found that charge exchange-chemical ionization mass spectrometry
can be used to distinguish many  isomeric PAC (1). We have established that
spectra of PAC have characteristic ratios of the protonated molecular ion to
the molecular ion when 5% to  10% methane in argon is employed as the
reagent gas. The two  most prominent ions in  the reagent plasma are C2HB*
and Ar*. The molecule of interest either reacts with the strongly acidic ion,
CoH5+, to give the protonated molecular ion, or it reacts with Ar* ions to
give the molecular ion. The relative rates of these two competing reactions
determine the intensity ratio of the  two ions. Variation in ionization potentials
will not significantly affect the rate of protonation by C2H5*  ions, but it will
affect the rate of ionization by Ar* ions.  Thus,  we have found that ionization
potentials are correlated with the  ratio  of the protonated molecular ion to
the molecular ion (M + 1/M).
   A wide variety of PAC have been studied, and the characteristic ratios '
for M 4- 1/M were determined.  A partial  listing of the PAC investigated is
given in Table 1. (All values have been corrected for the natural  abundance
of 13C.) Within any of the subgroups listed, it can be seen that the ionization
potential (IP) of isomeric PAC correlates  with the M + 1/M value.  Ap-
parently, an increase in the IP causes the rate of ionization by Ar+ to decrease
while the rate of protonation by C^Hs* is unaffected. Thus, there is an in-
crease in the relative rate of protonation as  evidenced by an increase in the
ratio of M + 1/M.
   Table  1 shows that this technique is applicable to the differentiation  of
                                   73

-------
               TABLE 1. Some of the pot/cyclic aromatic compounds investigated
Compound
Tetracene
Benzo[a]pyrene
Anthracene
Benz[a]anthracene
Pyrene
Benzo[e]pyrene
Chrysene
Fluoranthene
Phenanthrene
Triphenylene
4-Methylbiphenyl
3-Methylbiphenyl
2-Methylbiphenyl
9-Methylanthracene
2-Methylanthracene
2-Methylphenanthrene
Benz[b]indole
Benz[g]indole
Benzo[b]quinoline
Benzo[f]quinoline
.' Dibenzothiophene
Benzo[l ,2]dibenzothrophene
Formula
CigHi2
C2flHl2
CuHio
CuHt2
CJ^HIQ
C20'"12
Ci8Hi2
ds'MQ
CnHig
CiaH12
Ci3Hi2
CisH,2
Ci3H12
Ci5H,2
Ci&H12
CisHn
C12H9N
CI2H9N
C,3H9N
C,3H9N
C,2H8S
Ci6H,oS
IP(ev)
6.8Bd
7.\7*
7.42''
7.47d
7.S6d
7.56d
7.74d
7.76d
8.02"
8.1 \d
8.15«
8.30«
8.60C
7.36'
7.42'
7.90»
8.1 1*
8.25*
7.88^
8.14-
—
—
M + 1/M"
0.45
0.73
0.82
0.83
0.73
0.82
1.26
1.57
1.59
1.73
—
—
—
—
—
—
—
—
—
—
—
—
M+ 1/M6
	
—
—
0.70 ± 0.08 c
0.64 ± 0.02
—
—
—
—
1.24 ±0.03
1.79 ± 0.10
2.16 ± 0.10
2.45 ± 0.23
0.62 ± 0.07
0.64 ± 0.03
0.89 ± 0.01
1.03 ± 0.04
1.13 ± 0.07
1.68 ±0.09
2.20 ± 0.06
0.24 ± 0.09
1.25 ±0.08
     0 Reagent gai was 10% methane in argon (P-10) from Matheion Gas Products.
     6 Reagent gas was 5% methane in argon (P-5) from Matheion Gas Products.
     c 90% confidence limits on a minimum of 3 measurements.
     d Values are found in ref. 1.
     e Determined by photoelectronic spectroscopy, J. P. Maier and D. W. Turner, Faraday Disc.
    Chem. Soc., 54:149, 1972.
     / Determined using charge transfer spectra, O. B. Nagy, Tetrahedron, 31:2453, 1975.
     * Determined by appearance potentials, P. Nounou, J. Chem. Phyt., 63:994, 1966.
     * Calculated values, N. S. Hush, A. S. Cheung, and P. R. Hilton, J. Electron Spec. Rel. Phen.,
    7:385, 1975.
     1 Calculated values, M. J. S. Dewar, A. J. Hargot, N. Trinajstic, and S. D. Worley, Tetrahedron,
    26:4505, 1970.

simple PAC and to some of their methyl derivatives.  It  is also useful for
some  aromatic  heterocycles as  illustrated  by the nitrogen-containing com-
pounds. A  difference in the  ratios is observed  which  is dependent on the
basicity of the nitrogen in the ring. Lack of available standards has limited the
investigation of  sulfur-containing aromatic heterocycles;  the two we have
studied are shown.
   In conclusion, it should be noted that this work is in its initial stages. Sub-
stantial work is  needed to establish the capabilities  and  limitations  of the
technique.  This  charge exchange-chemical ionization technique is not  ex-
pected to replace electron impact GC-MS; but it should be  a useful  supple-
mental tool for differentiating isomeric PAC.  In addition, by using ionization
potentials calculated from MO theory, it is hoped that this technique  can be-
come  predictive  and be extended to molecules for which no standards  are
available.
                               74

-------
                      ACKNOWLEDGMENTS

  This work  has been  supported by  giants from the U.S. Environmental
Protection Agency (R803242)  and  the Energy Research and Development
Administration (EE-77-S-02-4267).


                           REFERENCE

1. Lee, M. L., and Hites, R. A. (1977):  Mixed charge exchange—chemical ionization
  mass spectrometry of polycyclic  aromatic hydrocarbons. A Am. Chem. Soc., 99:
  2008.
                                    75

-------
                  APPENDIX G
     Characterization of Sulfur-Containing
Polycyclic Aromatic Compounds in Carbon Blacks
                       76

-------
                       Reprinted fnmi ANALYTICAL CHEMISTRY. Vol. IS. \\\ft< ISJHI. Nmcmlvr l:'Ti:
               Copyright 197(i hy Ihe American Chemical Society iind reprinted liy permission of the i-npyri^hi owner
Characterization of Sulfur-Containing Polycyclic Aromatic
Compounds in  Carbon  Blacks

M. L. Lee and Ronald A. Kites'
Department of Chemical Engineering. Massachusetts Institute ol Technology. Cambridge, Mass. 02139
Computerized gas chromatographlc mass spectrometry and
high resolution mass spectrometry have been used to identity
sulfur-containing polycyclics and polycycllc aromatic hydro-
carbons In carbon blacks  obtained from sulfur-containing
petroleum feedstocks. Twenty-eight compounds have been
Identified, seven of which are sulfur-containing polycyclics.

  Carbon black is a material of considerable commercial im-
portance:  More than 1.5 billion pounds per year of domestic
carbon black are used in the manufacture of tires alone (/).
11 is also a material of potentially great environmental concern
because of (a) wide  environmental distribution of carbon
black, primarily in automobile tire dust, and (b) the potent
carcinogenicily of a number of compounds adsorbed on carbon
black such  as  certain  polycyclic aromatic  hydrocarbons
(PAH). These considerations have led to several studies of the
organic compounds associated with carbon black. For exam-
ple, two recent studies (2, 3) reported the identification of
cyclopenta|rd]pyreneasa major constituent of carbon black
extrncts; in addition, 11 other PAH (2) and several oxygenated
polycyclics (3) were also reported.
  This paper reports on the analysis of organic extracts of
several carbon blacks which were manufactured under varying
conditions. Of particular interest is the first reported identi-
fication of sulfur-containing polycyclics in carbon black. In
addition, the detection of high-boiling PAH has been extended
to include compounds of molecular weights up to 376 (C:i»H m).
Capillary column gas chromatography combined with mass
spectrometry (GO/MS) has allowed the positive identification
of'21 compounds and the tentative identification of 10 cithers
(see Figure 1). High-resolution mass spectromelry (HUMS)
of these same samples has verified the elemental composition
of individual compounds, especially for the sulfur polycy-
clics.

                  EXPERIMENTAL
  Samples of four different furnace blacks (seeTahle I) wereobuu'ned
from a commercial source (Cabot Corporation, Boston. Muss). The
aromatic feedstocks used in the production of three of these furnace
blacks were derived from refinery and naphtha-based ethytene type
tnrs. They were over 90% aromatic hydrocarbons, and had a consid-
erable amount of organic sulfur (1.2-3.1%). Appropriate amounts (see
Tahle I) of each furnace black were extracted wit h methylene chloride
for 18 h in a .Soxhlet apparatus. Soxhlet thimbles were extracted wilh
Nanoprade methylene chlnride (Mallinckrodl) for several hours prior
to each sample extraction to remove any organic contaminants in the
thimble or apparatus. The melhylene chloride extracts were then
evaporated to minimal volumes (I-10 ml) by rotary evaporntion under

Figure 1. Compounds identilied in carbon blacks by GC/MS and
HRMS
* Structure is presumed correct but has not been verified by comparison with
authentic compounds. " Exact position of benzo g>oup is not known. c The in-
crease in molecular weight of PAH species also increases the number ol possible
isomers: the lack of authentic compounds in this molecular weight range pre-
vents the elucidation of the exact structures lor these particular GC peaks. The
structures given are examples only; many other isomers are possible. a Detected
only by high resolution mass spectrometry. The structures given are examples
only: many other isomers are possible
                                                        77

-------
   naphthalene
  icenaphthylene
  dibenzothiophene
   phenonthrene
    anthracene
    H2,C
  4H -cyelopentafdef]-
  phenanthrene
   tluoranthene
 benzfejacenoph-
 thylene"
benzo[def] dibenzo-
thiophene0
                                   10
                                  pyrene
                              benzofaldibenzo-
                              thiophene
                                    12
                              benzo[ghj] fluoran-
                              thene
                              cyclopenlo[cd]py-
                              rene
ben z[o_] anthracene
 benzoQJf luoran-
 thene and benzo[k]
 fluoronthene
                             benzofluoronthene
                          benzofdejjnaphtho-
                          benzothiophene0'1"
                                                        benzo[e_] pyrene
                                 20

                          benzo[a]pyrene
      22
Indeno[l,2,3-cd]-
pyrene
                                                            23
                                                     benzo[£M]perylene
                                                            24
                                                        anthonthrene
                                                                                    30"
                                       7S

-------
                                                   10
                                                                    23*24
TEMP(
                                        I
                                       190
                                                              310
                                                                          370
                                        10
                                                              20
                                                                                     30
TIME(MIN)       0

Figure 2. Packed-column gas chromatograms ol the extract of (A) furnace black 1 and (0) furnace black 3 (see Table I). GC conditions: See text.
Key: See Figure 1
                                                               chromatographic run. High-temperature gas chromatograms were
                                                               run on a HP 5720A gas chromatograph with a 180 cm X 0.32 cm o.d.
                                                               stainless steel column packed with 3% Dexsil 300 on  80/100 mush
                                                               Chromosorb W which was programmed from 70 to 370 °C at 12 °C/
                                                               min with a carrier gas flow rate of 25 ml/min.
                                                                 High resolution mass spectral information was obtained on each
                                                               sample by introducing an aliquot of the methylene chloride extract
                                                               into a high-resolution mass spectrometer through a direct introduc-
                                                               tion probe system and slowly vaporizing the sample at  a continually
                                                               increasing temperature while several exposures  on a photographic
                                                               plate were made. After development, the plate was read on a com-
                                                               puterized comparator, and the exact masses were converted to ele-
                                                               mental compositions. The HRMS system consists of a DuPont In-
                                                               struments 21-110B mass spectrometer and a D.W. Mann comparator
                                                               interfaced to an IBM  1802 computer. This system and  its operation
                                                               have been previously described elsewhere (4). Both mass spectrom-
                                                               eters were operated at 70-eV ionizing energy.

                                                                            RESULTS  AND DISCUSSION

                                                                 Figure 2 compares packed-column gas chromatograms of
                                                               extracts from  furnace blacks 1 and 3 (see Table I).  Peak
                                                               numbers refer to compounds listed in Figure 1  which were
                                                               identified by gas chromalographic mass spectrometry and
                                                               retention data. In all cases where an exact identity is reported,
Table 1. Carbon Black Characteristics



No.
1
9
3
4




Feedstock
Elhylene tar
Refinery tar
Natural gas
Ethylene and
refinery tars

Furnace
temp,
K
1400-1600
1400-1600
1400-1600
1800-2000


Particle
size,
nm
260
75
75
30

Weight of
carbon
black
extracted,
g
8
2
2
34


Yield
of
PAH,
%
0.2
0.1
0.1
0.01

vacuum prior to gas chromatographic analysis. The total yield of PAH
from the tour carbon blacks is given in Table I.
  Gas chromatographic mass spectrometry of each sample was per-
formed on a Hewlett-Packard 5982A GC/MS system interfaced to a
H1J 591)3A data system. A 19 m X 0.2fi mm i.d. glass capillary column
coated with SE-52 methylphenylsilicone stationary phase was used
t,o sepnrate mixtures prior to mass spectral analysis. The oven tem-
perature was programmed from 70 to 250 °C at 2 °C/min during each
                                                            79

-------
                                                JJ	
                                                                                          I   lr
                                                                                           ,a
TIHF t*C1
70 00 110 130 150 170 190 210 230 250
 Tim WIN)   0         10        20        30        40        50         60        70        80         90        110

 Figure 3. Glass capillary-column gas chromatogram of the extract of furnace black 1 (see Table I). GC conditions: See text. Key: See Figure 1
 the mass spectrum and GC retention time were identical with
 those of authentic materials. The gas chromatograms of ex-
 tracts from carbon blacks 2 and 4 are very similar to that of
 carbon black 3 (Figure 26) and are, therefore, not shown.
   These GC/MS data, taken together with the PAH yields
 given in Table I, indicate that the amounts and structures of
 PAH associated with carbon blacks are quite dependent on
 the conditions of carbon black formation. Wallcave et al. (2)
 found that of eight carbon blacks examined, cyclopenta [cd] -
 pyrene was not detected in three of them and varied consid-
 erably  in concentration  in the others. Of the  four furnace
 blacks  analyzed in this  study, the  organic composition  of
 number 1 differed substantially from the other three, which
 were qualitatively quite similar to one another. Furnace black
 1 was manufactured by a different process than 2 and 3, al-
 though furnace temperatures were the same. In  addition, the
 higher furnace temperature used in the production of number
 4 seemed to reduce the total amount of PAH by a factor of ten
 as compared to the others, although the qualitative distribu-
 tion of PAH was still very similar to 2 and 3. The main dif-
 ference in the manufacture of 2 and 3 was the nature of the
 feedstock used.
   A high-resolution gas chromatogram of the extract of fur-
 nace  black 1 is shown in Figure 3. Again, numbers refer to
 compounds identified and listed in Figure  1. In addition to the
 identification of a number of previously unresolved isomers
 and trace compounds, four sulfur-containing polycyclic aro-
 matic compounds were detected. However, because of the
 unavailability of standard compounds, exact identification
 of  only dibenzothiophene and benzo[a]dibenzothiophene
 could be made. On the other hand, proposed structures (see
 Figure 1) seem to be reasonable when compared  to structures
 of PAH identified in the same mixture. For example, although
 there are a  number of possible structures for  CuHsS, the
 similarity in the structures of benzo[
-------
                APPENDIX H
Cyano-Aromatic Compounds by the Combustion
       of Nitrogen Containing Fuels
                     81

-------
                Reprinted from ENVIRONMENT SCIENCE & TECHNOLOGY, Vol. 12, Page 965, August 1978
              Copyright is) 1978 by the American Chemical Society and reprinted by permission of the copyright owner
Cyanc-arenes Produced by Combustion of Nitrogen-Containing Fuels

George R. Dubay and Ronald A. Hites*
Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge. Mass. 02139
D Cyanonaphthalenes (both  isomers)  and  cyanoacena-
phthylenes (four isomers) were identified in the sixil generated
by the combustion of aromatic hydrocarbon fuels doped with
6-!l()% pyridine. These wore by far the most abundant nitro-
gen-containing organic compounds in this combustion ef-
fluent; mulliring. nitrogen heterocyclic compounds, such as
those commonly observed in airborne particulale matter, were
a minor component. These identifications were made by gas
chromatographic mass spectrometry following a preliminary
separation by alumina column chromatography. The envi-
ronmental significance of  these findings is discussed.
  Certain organic compounds in scxit cause cancer in man (/).
Determining the structures of these compounds and under-
standing their biological activities have been the subjects of
intense research over the last 50years (2). It is now known that
the major class ol carcinogenic compounds associated with
soot is the polycyclic aromatic1 hydrocarbons (PAH) (,'/). Ni-
trogen-containing aromatic compounds (aza-arenes) are also
associated with soot I'M, and  some are known to be carcino-
genic (5). However, because they are much less abundant than
1'AH, these compounds have received proportionately less
attention. In the future, however, it is likely that the envi-
ronmental abundance of  a/.a-arcnes will increase as fuels
higher in organic nitrogen content are burned. We have,
therefore, undertaken the identification of the major aza-
arenes produced by the combustion of a model fuel containing
1-6% nitrogen.
  Several researchers have developed methods for the analysis
of aza-arenes in atmospheric particulate samples  based on
thin-layer,  gas, paper, high-pressure  liquid,  and column
chromatography and on electrophoresis {6-10). All of these
techniques begin with a solvent-solvent extraction utilizing
strong acid to partition the basic aza-arenes away from the
PAH. This procedure obviously discriminates against neutral
aza-arenes that might be present. To avoid this problem, we
have separated the aza-arenes from the bulk of the PAH by
alumina chromatography using gradient elution.
  Once  the compounds are separated from the PAH, the
identification of the exact molecular structures of aza-arenes
is still very difficult. Almost all assignments made  in the lit-
erature are, to some degree, ambiguous. Assignments  have
been based on gas chromatographic retention information and
on fluorescence, 1) V, or electron impact mass spectra (6-10).
In these analyses, all the possible isomers of a particular mo-
lecular structure have not been available. Hence,  by use ol
these techniques, there is no criterion by which the unavailable
isomers can be ruled out.
  To address the problem of determining the precise molec
ular structure of compounds produced by combustion, we hav(
                                                      32

-------
developed a method based  on charge exchange-chemical
ionization mass spectrometry (CE/CIMS). We have estab-
lished that isomeric PAH and aza-arenes have characteristic
mass spectra when 5 or 10% methane in argon is employed as
the reagent gas for CE/CIMS (11, 12). The two most promi-
nent ions in the reacting plasma are C^Hr/ and Ar+. A mole-
cule reacts either with a strongly acidic C2HS+ ion to give the
protonated molecular ion (M + 1), or it reacts with an Ar+ ion
to give the molecular ion (M). The relative rales of these two
reactions determine the intensity ratio of the two ions. A high,
positive correlation was found between ionization potential
and M + 1/M ratio (//). We have also established that this
relationship holds for some aza-arenes and for some methyl-
ated PAH (12). Aza-arenes that  have ionization potentials
differing by more than 0.1 eV can be distinguished using this
method.
  The great potential of this technique lies in its predictive
ability. The relationship between ionization potentials and
M + 1/M intensity ratios can be established for all available
isomers. Calculation of ionization potentials from molecular
orbital theory will then allow prediction of the M + 1/M ratio
for unavailable isomers.

Experimental
  Combustion Conditions. Soot was obtained  in  two
ways:
  A solution of 32% pyridine in «-xylene (6% fuel nitrogen)
was burned in a wick-fed, alcohol-lamp burner. Soot was
collected on the exterior of a precleaned, water-cooled, filter
flask. The soot was removed by scrubbing the flask with
CH2CI2-soaked glass wool. The soot, glass wool, and associated
CH2C12 were put in a preextracted Soxhlet thimble and ex-
tracted with 200 mL CH2CI2 overnight. The extract volume
was reduced to 1 mL on a rotary evaporator operating at 30
°C and 15 torr. Samples were stored in the dark at 6 °C.
  A Meker burner was modified by replacing the air inlets
with oxygen feeds. The fuel inlet was attached to a stainless
steel tube through which a benzene-methane mixture was
passed. The fuel mixture was preheated to vaporize the ben-
zene, and the plumbing was heated (300 °C) to prevent con-
densation. Methane and 02 flows were maintained by critical
orifices, and the benzene flow  was measured before vapor-
ization by passing the liquid through a  capillary tube in which
the upstream and  downstream pressures were carefully
measured. The liquid flows corresponded to Reynold's num-
bers of 800-1000 through the capillary tube. Thus, the flow
was always laminar and proportional to the pressure drop
through the tube. The benzene was  doped with sufficient
pyridine (5.3%) to give a fuel containing 1% nitrogen. Fuel
equivalence ratios of 4.0 and 4.5 were used, the CH4/CeH6 ratio
was 1.5, and the cold gas velocity was 31.8 cm/s. Soot was
collected by a water-cooled spray probe (13) and was trapped
in glass wool packed filters. The organic compounds were
extracted first with acetone and then  with CH2C12.
   Column Chromatography. To eliminate interferences
caused by PAH and to obtain an enrichment of the nitrogen
compounds, the soot extracts were fractionated as follows: The
sample (1 mL) was added to 1 g of neutral alumina (activity
grade 1, ICN Pharmaceuticals) in a 25-mL beaker, and the
CH.jCI'j was allowed to evaporate at  room temperature (15
min). This precoated alumina was added to the top of a col-
umn prepared from 5 g of alumina with hexane as the solvent.
Six fractions were then eluted (Table I). These fractions were
then concentrated on a rotary evaporator for analysis.
   Instrumentation. A Hewlett-Packard  5730A gas chro-
matograph equipped with dual nitrogen-phosphorus flame
ionization detectors (FID)  was used for GC analyses. A
Hewlett-Packard 5982A mass  spectrometer interfaced to a
5933A data system was utilized for gas chromatographic mass
spectrometry. Charge  exchange-chemical ionization mass
spectrometry  was performed  with 5%  methane  in argon
(supplied by Matheson Gas Products). The flow of carrier gas
was 10 mL/min. and the ion source temperature was-195 ± 2
°C. The ratios of the protonated molecular ion to the molec-
ular ion were established by adding all spectra scanned within
a given GC peak.


Results and Discussion
  The two different  combustion systems gave  virtually
identical results; therefore, they will not be distinguished in
the following discussion. As indicated in Table 1, all of the
aza-arenes were collected in fractions 3 and 4. The gas chro-
matogram of fraction 4 was extremely complex. Because this
fraction contained less than a third of the aza-arenes and be-
cause the identities of most of the components in this fraction
corresponded to those reported elsewhere (10), further anal-
yses of this fraction have not been pursued.
  Fraction 3 contained most of the aza-arenes; gas chro-
matograms of this fraction are shown in Figure 1. The upper
trace was obtained with a normal FID, and the lower with a
nitrogen specific FID. Comparison of these two traces show
that the nitrogen-FID and the normal  FID respond equally
to the early eluting peaks, indicating that nitrogen is present
in all of ihese constituents. The later eluting peaks were de-
tected only by the normal FID,  indicating that  they  are
probably PAH.
  The electron impactand CE/CI  mass spectra of peaks 1-7
are given in Table II. The spectra of peak 1 were interpreted
as those of either 1- or 2-cyanonaphthalene, and authentic
samples of these two compounds  were obtained (K and K
Laboratories). The exact retention time (by coinjection) and
the  electron impact and CE/CI mass spectra of 1-cyanona-
phthalene were identical with those of peak I.  The identifi-
cation of peak 2 as 2-cyanonaphthalene was proved in a similar
fashion.
  Peaks 3-6 all show molecular weights of 177 (Ci:iH7N) (see
Table II). Based on analogy and on the abundant presence of
the parent hydrocarbon, these peaks have been tentatively
identified as the four isomers of cyanoacenaphthylene.
Table I. Fractionatlon of Soot Extracts
    acllon        Solvent
     1     Hexane
     2     30% Benzene, 70%
            hexane
     3     70% Benzene, 30%
            hexane
     4     Benzene
     5     Benzene
     6     CH2CI2
 Vol
(mL)

 30
 30

 30

 30
 50
 50
Total
                                     % PAH" % Az«-ar«n.a
 0
64
12
          15
 2        7
 0        0
 0        0
78       22
  • Percent of total material in the various fractions as measured from the normal
 FID or nitrogen-specific FID responses, respectively.
                                                      83

-------
           _l
Figure 1. Gas chrornatograms of fraction 3, isolated from organic
compounds produced by combustion of solution of 32% pyridine in
o-xylene
GC conditions: 1.8 m X 6.3 mm o.d. glass column packed with 3% OV-17 on
80/100 mesh Supelcoport. programmed from 100 to 310 °C at 8 °C/min. Upper
trace obtained with normal FID, bottom trace with nitrogen specific FID
These are not available compounds, and proof of their struc-
ture must await their synthesis. Peak 7 has a molecular weight
of 203 (C]SH9N) and probably represents several unresolved
cyanophenanthrene isomers.
  Based on the GC data, we estimate that 80% of the aza-
art'nes in fraction 'I are ryanonaphthalencs, cyanoacena-
ptuhylenes, and cyanu|jhenanthrenes. These compounds are
the most abundant class of organic nitrogen compounds in the
effluent from these (lames.  Multiring, nitrogen helerocyclic
compounds (such as acridine or  phenanthridine) that have
been commonly observed in airborne paniculate matter (10)
are a minor component (<7%) in these effluents.
  Cyanonaphthalenes have not  been found in comhustion
effluents before. They have, however, been found at trace
levels in cigarette smoke tar (1-1, 15), in petroleum (16), in tar
from the low-temperature pyrolysis of coal (17), and in an-
thracene oil (18). In all of these rases, it is presumed that cy-
anonaphthalenes are formed from the pyrolysis of other ni-
trogen compounds  (19). The biological activity  of  these
compounds is not clear. There have been reports thatcyano-
naphthalenes cause irregular mitoses and chromosomal ab-
errations (14, 20), that they have insecticidal properties (21.
22), and that they are toxic to the eggs of body lice (2.'0. On
the other hand, testing by a quantitative forward mutation
assay using 8-azaguanine resistance in  Salmnnclla typ/ii-
murium (24) indicates that these compounds have less than
1% of the activity of henzo[a]pyrcne on a molar basis (2.5).
  The environmental significance of our results is twofold.
(a) The relative amount of cyano-arenes being produced by
the combustion of a nitrogen containing fuel is significant (see
Table I), and some of these compounds may be biologically
active. Emission of large amounts of such  compounds into the
environment would seem to be  undesirable, (b)  Although
several cyano-arenes are being produced in flames, it  is not
known  if they are  environmentally  persistent.  Nitrogen
functional group analyses by KSCA have indicated the pres-
ence of the cyano functionality in certain urban air paniculate
samples (26), but specific cyano-arenes have not been found
in the atmospheric environment. Cyano-arenes may well he
present in air particulates, but because of the analytical lim-
itations outlined above, they have not yet been detected. We
suggest that the quantity (if any) of cyano-arenes in the am-
bient air environment should be  measured using techniques
designed for these compounds. Cyano-arenes may be more
prevalent than the  multiring,  nitrogen  heterocyclic  com-
pounds studied in the past, and their environmental  chemistry
and toxicity may warrant at least equal attention.
Table II. Mass Spectra of


M + 1
M
M- H
M-CN
M-HCN
Others


M
M + H
M + 2
M + CH3
M + C2H5
M + 30
Others

1
m/e
154
153
152
127
126
125


153
154
155
168
182
183
156

i
Int
13
100
7
5
17
4


38
100
15
9
35
6
6

Peaks 1-7 in El (Top)
;
m/e
154
153
152
127
126
125


153
154
155
168
182
183
156

i
Int
13
100
8
5
15
3


40
100
14
9
37
5
3


m/e
178
177
176
151
150
179
175
149
177
178
179
192
206
207
180
208
and CE/CI (Bottom) Modes
Peak
1
Int
15
100
15
9
18
5
6
4
41
100
25
6
34
14
14
6

m/e
178
177
176
151
150
179
175
149
177
178
179
192
206
207
180
208
•
Int
16
100
8
6
15
7
4
6
45
100
18
8
40
8
9
6

m/e
178
177
176
151
150
179
175
149
177
178
179
192
206
207
180
208
5
Inl
18
100
11
4
13
4
7
7
38
100
17
13
45
11
10
3

m/e
178
177
176
151
150
179
175
149
177
178
179
192
206
207
180
208
6
Int
19
100
11
4
18
4
6
3
42
100
16
16
44
11
8
5
7
m/e
204
203
202
177
176
201
175
151
203
204
205
218
232
233
206
234

Inl
22
100
13
6
10
13
7
6
46
100
17
11
41
9
10
5
M + 1/M«         2.63
  • Corrected lor 13C.
                               2.37
                                              2.33
                                                             2.10
                                                                            2.49
                                                                                          2.27
                                                                                                         2.01
                                                      84

-------
Acknowledgment
  The authors thank Jack B. Howard and William J. Kausch,
Jr., for the design and operation of the second burner sys-
tem.
Literature Cited

(1)  Pott, P., "Chirugical Observations", p 63. Hawkes. Clarke, and
  Collins, London, England, 1775.
(2)  National Academy of Sciences, "Particulale Polycyclic Organic
  Mailer", Washington, D.C.. 1972.
(3)  Hoffmann, D., Wynder, E. L., in "Chemical Carcinogens". C. E.
  Searle, Ed., pp 324-65, Amer. Chem. Soc. Monograph 173, Wash-
  ington, D.C., 1976.
(4)  Sawicki, E.. McPherson, S. P., Stanley, T.  W., Meeker, J.. Elbert.
  W.C.,lnt.J. Air Water Pnltut , 9,515(1965).
(5)  Dipple, A., in "Chemical  Carcinogens",  C. E. Searle. Ed., pp
  245-314, Amer. Chem. Soc. Monograph 173, Washington.  D.C.,
  1976.
(6)  Sawicki, E., Stanley, T. W., McPherson, S., Morgan, M., Taianta,
  13,619(1966).
(7)  Sawicki, E., Guyer, M., Engel, C. R.. J  ChrnmatoRr.. 30. 522
  (1967).
(8)  Engel, C. R., Sawicki, E., ibid . 31, 109 (1967).
(9)  Cautreells, W., van Cauwenberghe, K., Almas. Environ., 10, 447
  (1976).
(10) Dong, M. W., Locke, D. C., Hoffmann, D.,  Environ. Sri. Technal.,
  11,612(1977).
(11) Lee, M. L., Hites, R. A., J. Am. Chem. Sot:., 99,2008(1977).
(12) Hites, R. A., Dubay, G. R., in "Carcinogenesis", P. W. Jones, Ed.,
  Vol III. pp 85-7, Raven Press, New York, N.Y., 1978.
(13)  Prado. C,. P.. Lee. M. I,., Hiles. K. A.. Hmill, D. P., Howard, J.
  B.. Sixteenth (Int.) Symp. on Combustion, pp 649-61, Combustion
  Institute. Pittsburgh, Pa. 1977.
(141  Izard, C., Moree-Testa, P.. C« .-laid. Sri.. SIT. I). 272, 2581
  (1971); (Vii'm. Abstr.. 75,61573j.
(15) Benner. .1.. Keene, C. K., Holt.T. W.. in 4th Tobacco and Health
  Workshop Conf. Proc., pp 408-20, Univ. of Kentucky, Lexinglon,
  Ky.. 1973. Oicm. Abslr.. 79, 896691).
(16)  Hartung. G. K.. Jewell.  I).  M.,  Anal. Chem. Ada. 27, 219
  (1962).
(17) Andre, J., Duth. P.. Mahieu, J., Grand'Ry, E. H.. Hrennst  Chem..
  18,369 (19671, Chem. Abslr . B8, 318(i9y.
(18)  Prokseh,E..Ocs(crr. Chfm.Zlg..  67, 105 (1966); Chem. Abslr  .
  64,19251g.
(191  Patterson..!. M., Haidar, N.  K., Smith. Jr., W. T., Chem. Ind.
  (London), 1975, p 128.
(201 Bhalla, P. 11. Arnold, K. C.. Sahharwal, 1'. S., J. Hered., 65, 311
  (1974).
(21) Swingle. M. C.. Mayer, K. L., (Jahan. J. R.,<7. Econ.'Entomol.,
  37,672 (1944); Chem.'Abslr.. .19. 7H4lt>).
(22) Maytr. E. L., Robertson, (I.. Nelson. R. H.. WcxxIward.C. F,. Hur.
  r'.ntomol.  Plant  Quarantine. E-836 (1952);  Chem. Abstr.,  4f>,
  525 la.
(23)  Eddy. G. W . Carson, N.  B..  J- Kenn. Entnmui. 41, 31 (1948);
  Chem.'Abstr.. 42. 5156g.
(24) Skopek.T. R., Liber, H. L., Krolewski,.!. J./Fhilly. W. ('.., /'roc.
  :Vo( Acad. .SVi , 75,410(1978).
(25)  Kaden. I). A., Thilly, W. (',..  Massachusetts Institute of Tech-
  nology. Cambridge, Mass.. private communication. 1977.
(26) Chang. S. C,., Novakov, T., Atrnus. Environ.. 9, 495 (1975).

Reeeil'ed for ret'ieu' November '23.  1977. Accepted February 2-1, /y78.
Wor/i supported 6v the rinuirtmmcntnl Protection Agency (Grant
KHIM2-I2} and the Department  of Energy ((.Irani EE-77-S-02-
•1267)
                                                           85

-------
                                TECHNICAL REPORT DATA
                         (Please read liiur.it'nons on the reverse br/ore completing}
1. REPORT NO.
 EPA-800/7-76-167
                           2.
4. TITLE AND SUBTITLE
Combustion Research on Characterization of Participate
 Organic Matter from Flames
                                6. PERFORMING ORGANIZATION CODE
                                 3. RECIPIENT'S ACCESSION NO.
                                 5. REPORT DATE
                                 August 157 b
7. AUTHOR(S)

R.A. Kites and J. B. Howard
                                8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Massachusetts Institute of Technology
Department of Chemical Engineering
Cambridge, Massachusetts  02139
                                 10. PROGRAM ELEMENT NO.
                                 EHE624A
                                 11. CONTRACT/GRANT NO.

                                 Grant R803242, Task 1
 12. SPONSORING AGENCY NAMF AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle  Park, NC  27711
                                 13. TYPE OF REPpRT AND PERIOD COVERED
                                 Task Final; 8/74-8/77	
                                 14. SPONSORING AGENCY CODE
                                  EPA/600/13
15. SUPPLEMENTARY NOTES IERL_RTP project OfflCCF IS John H.
541-2476.
                                 Wasser, Mail Drop 65, 919/
16. ABSTRACT
          The report gives results of a study of the formation and emission of soot and
polycyclic aromatic hydrocarbons (PAH) from both laminar flames and a turbulent
continuous-flow combustor. Decreasing the atomizing pressure decreased the amount
of soot and PAH produced in the latter system. Benzene fuel gave more soot and PAH
than did  kerosene, and the PAH from benzene were more substituted. Nitrogen- and
sulfur-doped fuels did not alfect total soot production. The distribution of soot and
PAH within the flame was consistent with the  concept that certain PAH may serve as
intermediates in soot formation.  The PAH formed during combustion of coal, wood.
and kerosene were separated and identified by capillary column gas chromatography
and mass spectrometry. The PAH from coal combustion were found to  be similar
to airborne PAH from Indianapolis,  a high coal consumption area; those from kero-
sene combustion were similar  to airborne PAH from Boston, an area of low coal con-
sumption and high consumption of petroleun derived fuels. Nitrogen containing fuels
primarily produce polycyclic aromatic compounds in which the nitrogen is in a cyano
substituent (approximately 70(7o).  This contrasts with the observation that nitrogen
heterocyclic  compounds  are the major nitrogen containing compounds in airborne
particulate matter.
17.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
Pollution
Combustion
Research
Flames
Properties
Organic  Compounds
Soot
Aromatic Polycyclic
  Hydrocarbons
Benzene
Kerosene
Coal
Wood
Nitrogen
                                          b. IDENTIFIERS/OPEN ENDED TERMS
Pollution Control
Stationary Sources
Particulate
                                             c. COSATi 1-icld/Group
13B
2 IB
14B
                        07C
          2 ID

          11L
          07B
18. DISTRIBUTION STATEMENT
 Unlimited
                                          19. SECURITY CLASS (This Report)
                                          Unclassified
                                             21. NO. OF PAGES
                                                  CO
                     20. SECURITY CLASS i Tliis pane)
                     Unclassified
                                             22. PRICE
EPA Form 2220-1 (9-?3)
                                        86

-------