EPA-R3-72-020
July 1972 ECOLOGICAL RESEARCH
Studies on Polycyclic Aromatic
Hydrocarbons in Flames
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington, D. C. 20460
-------
EPA-R3-72-020
Studies on Polycyclic
Aromatic Hydrocarocarbons
in Flames
by
Ronald Long
Department of Chemical Engineering
University of Birmingham
Birmingham, England
Grant No. 5R01-AP00323
Program Element No. 1A1008
Project Officer: Dr. Charles Walters
Division of Chemistry and Physics
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND MONIROTING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
July 1972
-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the
content necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
ABSTRACT
The analytical method developed in the early stages of the work involving
Soxhlet extraction of the particulate matter followed by column chromatography
and then programmed-temperature gas chromatography (with the use of
u. v. spectrophotometry to identify individual polycyclic aromatics) has
been simplified and made more rapid.
«•
The improved procedure has been used to analyse soot samples in later
work and, inter alia, the p. c. a. h. in soot samples from three
fire fighting schools in the U.S. A. The presence of appreciable amounts
of known carcinogens in these suggests a possible health hazard to
personnel exposed.
Exploratory work has been carried out on the use of the integrated
ion-current technique in high resolution mass spectrometry to determine
picogram quantities of p. c. a. h. introduced into the ion source of a
GEC-AEI, M.S. 9. instrument, a low electron voltage (13 ev) ensuring
that only the molecular ions are formed. Such a method has been shown
to be potentially valuable in air pollution studies in conjunction with thin
layer chromatography, or in certain cases oa its Own.
Samples have been withdrawn from rich oxy-acetylene and oxy-ethylene
low pressure flat flames using quartz micro-probes. These have been
analysed by mass spectrometry and gas chromatography. There is a
marked similarity in the flames and the concentration profiles of, inter alia,
diacetylene and other polyacetylenes, phenyl acetylene and styrene
and certain polycyclic aromatics have been determined. The results suggest
111
-------
that acetylene plays an important intermediate role in the formation of
p. c.a.h. The results relating to the formation of polyacetylenes confirm those
of previous workersbut those relating to p. c. a.h. differ in important respects.
The mode of formation of the polycyclic aromatics is considered.
Although it has not proved possible to undertake that part of the programme
involving the use of organo-metallic additives, contemporary research'
in Germany and in England has been reviewed briefly.
IV
-------
LIST OF PUBLICATIONS RESULTING FROM
RESEARCH SUPPORTED BY GRANT AP OO323
1. Gas chromacographic analysis of polycyclic aromatic hydrocarbons
in soot samples. B. B. Chakraborty and R. Long.
Environmental Science & Technology I, 828, (1967).
II, 217, (1968).
2. The formation of soot and polycyclic aromatic hydrocarbons in
diffusion flames. B. B. Chakraborty and R. Long.
Part I. Combustion and Flame, J£,226 (1968).
Part II. " " 12, 237, (1968).
Part UI " " ^2, 469, (1968).
3. The formation of soot and polycyclic aromatic hydrocarbons
in ethylene diffusion flames with methanol as additive.
B. B. Chakraborty and R. Long. Combustion and Flame. 12, 168, (1968).
4. Soot formation in ethylene and propane diffusion flames.
P. Dearden and R. Long. J. Applied Chem. 1_8, 243, (1968)
5. The flux of polycyclic aromatic hydrccarbons and insoluble matter in
pre-mixed acetylene - oxygen flames.
E. E. Tompkins and R. Long. 12th Symposium (International) on
Combustion. The Combustion Institute, Pittsburgh. (1969) page 625.
6. Mass spectrometic and chromatographic techniques for the determination
of polycyclic compounds.
J. R. Majer and R. Perry. Plenary Lecture I. U. P . A. C. Conference,
Cortina, Italy, on Chemical Aspects of Air Pollution. Pure and Applied
Chemistry 24, 685 (1970). Also published in 'Air Pollution', Edited
by V. Cantuti, I. U. P. A. C. Butterworth (1971)
-------
,7. The use of thin-layer chromatography and mass spectrometry
for the rapid estimation of trace quantities of air pollutants.
J. R. Majer, R. Ferry and Miss M. J. Reade. J. Chromatog.
48, 328, (1970).
8. Detection of geometrical isomers by fractional sublimation in a
mass spectrometer. J. R. Majer and R. Perry. J. Chem. Soc.
(Section A) 822, (1970).
9. The use of mass spectrometry in the analysis of air pollutants.
R. Perry, R. Long and J. R. Majer. Paper CP-37E. Second
International Clean Air Congress, The International Union of
Air Pollution Prevention Associations, Washington D. C.
December 1970.
Ph D. theses (University of Birmingham)
1. P. Dearden, May 1967.
'The formation of soot during the incomplete combustion of
hydrocarbons in a diffusion flame1.
2. B. B. Chakraborty, February 1967.
'Formation of polycyclic aromatic hydrocarbons during incomplete
combustion of hydrocarbons in a diffusion flame.
3. E. E. Tompkins, May 1968.
"The formation of polymeric materials and polycyclic aromatic
hydrocarbons in pre-mixed acetylene/oxygen flat flames.
4. B. D. Crittenden, February 1972.
'The formation of polycyclic aromatic hydrocarbons in rich
pre-mixed oxy-acetylene and oxy-ethylene flames. '
vi
-------
ACKNOWLEDGMENTS
This report includes work carried out by the following, apart from the principal
investigator.
R. Perry,B.Sc., Ph. D.:-
E.E. Tompkins.B. Sc. , Ph.D.:-
R. J. Martin, M. Sc. , Ph.D. :-
B. B. Chakraborty B.Sc., Ph.D. :-
P. Dearden, B. Sc. , Ph.D.:-
B. D. Crittenden,B. Sc. , Ph. D.
J. R. Majer, Ph.D. , D.Sc. — — — — Consultant on mass spectromctry.-.
Mrs. M. Hill (formerly Miss M. J. Reade)- assistant to Dr. J. R. Majer on
mass spectrometric work.
In the preparation of Part IV of this report, the writer has drawn heavily
from the Ph.D. thesis of his former research student B. D. Crittenden.
VI i
-------
CONTENTS
INTRODUCTION 1
PART 1. IMPROVEMENTS IN THE GAS CHROMATOGRAPHIC ANALYSIS OF
POLYCYCLIC AROMATIC HYDROCARBONS (P.C.A.H.) IN SOOT
SAMPLES 7
PART II. GAS CHROMATOGRAPHIC ANALYSIS OF POLYCYCLIC AROMATIC
HYDROCARBONS (P.C.A.H.) IN SOOT SAMPLES FROM FIRE-
FIGHTING SCHOOLS IN THE U.S.A 9
PART 111. THE INTEGRATED ION-CURRENT TECHNIQUE IN MASS SPEC-
TROMETRY (APPLIED TO P.C.A.H.) H
PART IV. THE MODE OF FORMATION OF POLYCYCLIC AROMATIC HYDRO-
CARBONS IN RICH PRE-MIXED FLAT FLAMES 15
PART V. BRIEF REVIEW OF USE OF ORGANO-METALLIC AND METAL
CONTAINING ADDITIVES IN SUPPRESSING SOOT AND
POLYCYCLIC AROMATIC HYDROCARBONS IN FLAMES 173
APPENDIX I. NOMENCLATURE OF POLYCYCLIC AROMATIC HYDRO-
CARBONS 1-1
APPENDIX II. THE FLUX AND CONCENTRATION OF A SPECIES IN A FLAME H-l
APPENDIX III. HITHERTO UNIDENTIFIED P.C.A.H. FOUND IN FLAME SOOTS IH-1
Vlll
-------
LIST OF TABLES
Table Page
1.1 Range of Polycyclic Aromatic Hydrocarbons Identified by
Tompkins and Long2 36
4.1 Flames Employed in this Study 55
4.2 Mass Measurement and Assignment of Species Present in the
Batch Samples 61
4.3 Identification of G.L.C. Peaks in the 'Cold1 Trap Chroma-
tograms 62
4.4 'Mass Measurement' of Mass Spectrum Peaks Resulting from the
'Cold' Trap Samples 68
4.5 Identification of G.L.C. Peaks in the Extraction Filter
Chromatograms 69
4.6 'Mass Measurement' of Mass Spectrum Peaks Resulting from the
Extraction Filter Samples 70
IX
-------
LIST OF FIGURES
Figure Page
1.1 Flow Pattern at the Edge of a Flat Flame Burner ...... 19
1.2 (a) Concentration Profiles of Diacetylene in Flat Acetylene-
Oxygen Flames of Different C 2^:62 Ratios ......... 23
1.2 (b) Concentration of Acetylene and Polyacetylenes in the
Burned Gas of Rich Acetylene-Oxygen Flames ........ 23
1.3 (a)-(c) Concentration Profiles in a Flat Acetylene-Oxygen
Flame ........................... 24
1.3 (d) Emission of C2, CH, and OH, and Continuous Emission in
the Same Flame as Function of the Height Above the Burner . 24
1.4 fa") Absolute Concentration of Carbon, the Mean Number of
Particles and Their Mean Diameter in a C-H.-Oo Flame. ... 25
1.4 (b) Concentration Profiles of Polyacetylenes in the Same
Flame ........................... 25
1.4 (c) Emission of C2, CH, OH and Continuous Emission Profiles
in the Same Flame ..................... 25
1.5 Reaction Scheme Proposed by Homann and Wagner^ for the
Formation of Polyacetylenes and 'Carbon' from Acetylene
and Free-Radicals ..................... 31
1.6 Flux of Total Polymeric Material VS. Height Above Burner
Surface .......................... 33
1.7 Flux of Soluble Material VS. Height Above Burner Surface . 33
1.8 Flux of Insoluble Material VS. Height Above Burner Surface. 34
1.9 Flux of Polycyclic Aromatic Hydrocarbons VS. Height Above
Burner Surface ...................... 34
1.10 Composite Plot of Flux of Insoluble Material, Flux of
p.c.a.h. and of Temperature Respectively VS. Height Above
Burner Surface ...................... 35
1.11 Reaction Scheme Presented by Chakraborty and Long^ to
Account for the Formation of Pcah and ' Carbon * in Diffusion
Flames ........................... 43
2.1 Sample Collection Systems ................. 53
/d>, \3
4.1.1 I— — VS. Height Above Burner Surface ......... 57
4.2.1 Corrected Temperature VS. Height Above Burner Surface ... 58
(Acetylene Flames)
-------
Figures Page
4.2.2 Corrected Temperatures VS. Height Above Burner Surface
(Ethylene Flames) 59
4.3.1 Part of a Mass Spectrum Which Results from the Introduction
of a Typical Batch Sample into the Source of the GEC-AEI
MS9 Mass Spectrometer 63
4.3.II Chromatogram of 'Cold1 Trap Sample Collected from Flame
1 at a Sampling Height of 18.7 on 64
4.3.Ill Chromatogram of Extraction Filter Sample Collected
from Flame 2 at a Sampling Height of 16.5 cm 65
4.3.IV Chromatogram of Extraction Filter Sample Collected from
Flame 2 at a Sampling Height of 16.5 cm 66
4.3.3 (a)-(b) Acetylene Concentration Profiles 71
4.3.3 (c)-(d) Acetylene Concentration Profiles 72
4.3.4 (a)-(b) Ethylene Concentration Profiles 73
4.3.5 (a)-(b)(Water + Carbon Monoxide) Concentration Profiles . . 74
4.3.5 (c)-(d)(Water + Carbon Monoxide) Concentration Profiles . . 75
4.3.6 (a)-(b)Oxygen Concentration Profiles 76
4.3.6 (c)-(d) Oxygen Concentration Profiles 77
4.3.7 (a)-(b) Methylacetylene Concentration Profiles 78
4.3.7 (c)-(d) Methylacetylene Concentration Profiles 79
4.3.8 (a)-(b) Propylene Concentration Profiles 80
4.3.8 (c)-(d) Propylene Concentration Profiles 81
4.3.9 (a)-(b) Carbon Dioxide Concentration Profiles 82
4.3.9 (c)-(d) Carbon Dioxide Concentration Profiles 83
4.3.10 (a)-(b) Diacetylene Concentration Profiles 84
4.3.10 (c)-(d) Diacetylene Concentration Profiles 85
4.3.11 (a)-(b) Vinylacetylene Concentration Profiles 86
4.3.11 (c)-(d) Vinylacetylene Concentration Profiles 87
4.3.12 (a)-(b) Triacetylene Concentration Profiles 88
4.3.12 (c)-(d) Triacetylene Concentration Profiles 89
XI
-------
Figures Page
4.3.13 (a)-(b) Benzene Concentration Profiles 90
4.3.13 (c)-(d) Benzene Concentration Profiles 91
4.4.1.3.1 (a)-(b) Benzene Concentration Profiles 92
4.4.1.3.1 (c)-(d) Benzene Concentration Profiles 93
4.4.1.3.2 (a)-(b) Tetra-acetylene (?) Concentration Profiles .... 94
4.4.1.3.2 (c)-(d) Tetra-acetylene (?) Concentration Profiles .... 95
4.4.1.3.3 (a)-(b) Toluene Concentration Profiles 96
4.4.1.3.3 (c)-(d) Toluene Concentration Profiles 97
4.4.1.3.4 (a)-(b) Phenylacetylene Concentration Profiles 98
4.4.1.3.4 (c)-(d) Phenylacetylene Concentration Profiles 99
4.4.1.3.5 (a)-(b) Styrene Concentration Profiles 100
4.4.1.3.5 (c)-(d) Styrene Concentration Profiles 101
4.4.1.3.6 (a)-(b) Methyl Styrenes (?) Concentration Profiles .... 102
4.4.1.3.6 (c)-(d) Methyl Styrenes (?) Concentration Profiles .... 103
4.4.1.3.7 (a)-(b) Trimethylbenzenes (?) Concentration Profiles . . . 104
4.4.1.3.7 (c)-(d) Trimethylbenzenes (?) Concentration Profiles . . . 105
4.4.1.3.8 (a)-(b) Indene Concentration Profiles 106
4.4.1.3.8 (c)-(d) Indene Concentration Profiles 107
4.4.1.3.9 (a)-(b) Dihydronaphthalenes (?) Concentration Profiles . . 108
4.4.1.3.9 (c)-(d) Dihydronaphthalenes(?) Concentration Profiles . . 109
4.4.1.3.10 (a)-(b) Naphthalene Concentration Profiles 110
4.4.1.3.10 (c)-(d) Naphthalene Concentration Profiles Ill
4.4.1.3.11 (a)-(b) 1-Methyl Naphthalene Concentration Profiles. ... 112
4.4.1.3.11 (c)-(d) 1-Methyl Naphthalene Concentration Profiles ... 113
4.4.1.3.12 (a)-(b) Biphenyl Concentration Profiles 114
4.4.1.3.12 (c)-(d) Biphenyl Concentration Profiles 115
xn
-------
Figures Page
4.4.1.3.13 (a)-(b) Acenaphthylene Concentration Profiles 116
4.4.1.3.13 (c)-(d) Acenaphthylene Concentration Profiles 117
4.4.1.3.14 (a)-(b) Fluorene Concentration Profiles 118
4.4.1.3.14 (c)-(d) Fluorene Concentration Profiles 119
4.4.2.3.1 (a)-(c) Indene Concentration Profiles 123
4.4.2.3.2 (a)-(c) Naphthalene Concentration Profiles 124
4.4.2.3.3 (a)-(c) 1-Methyl Naphthalene Concentration Profiles .... 125
4.4.2.3.4 (a)-(c) Acenaphthylene Concentration Profiles 126
4.4.2.3.5 (a)-(c) Fluorene Concentration Profiles 127
4.4.2.3.6 (a)-(c) 1.8,4.5-bi-(etheno-) napthalene (?) Concentration
Profiles 128
4.4.2.3.7 (a)-(c) Phenanthrene + Anthracene Concentration Profiles . 129
4.4.2.3.8 (a)-(c) 4.5-Methylene Phenanthrene Concentration Profiles . 130
4.4.2.3.9 (a)-(c) Fluoranthene Concentration Profiles 131
4.4.2.3.10 (a)-(c) Pyrene Concentration Profiles 132
4.4.2.3.11 (a)-(c) Benzofluorenes Concentration Profiles 133
4.4.2.3.12 (a)-(c) Methyl Pyrenes Concentration Profiles 134
IV.4.3.1 Composite Plot of Mole Fractions of All Species in Flame 1. 136
5.1 Proposed Flow Diagram for the Formation of Aromatic Species
and Simple Pcah from Acetylene 157
Xlll
-------
STUDIES ON POLYCYCLIC AROMATIC
HYDROCARBONS IN FLAMES
INTRODUCTION
It has been known for many years that benzo (a) pyrene is a potent
carcinogen and that some other polycyclic aromatic hydrocarbons
e.g. dibenz (a,h) anthracene are also carcinogenic.
benzo (a) pyrene dibenz (a,h) anthracene
The livers of many animals are able to metabolise a wide range of
molecules and it has recently been shown that some polycyclic aromatic
hydrocarbons can be converted initially to epoxides, which can be
detected as products of metabolism.although further metabolism
gives either the diol or derivatives.
Thus, in the case of benz (a) anthracene, the 5. 6- epoxide can be
formed which is extremely carcinogenic and mutagenic, although
benz (a) anthracene itself is only regarded as being weakly carcinogenic.
non-
f carcinogenic
compounds
benzo(a) anthracene 5. 6 r epoxide
(carcinogenic)
Polycyclic aromatic hydrocarbons are well-known to be associated with soot
or particulate carbonaceous matter resulting from the incomplete
combustion of fossil fuels. Whilst cigarette smoking is a most important
causal factor in luig cancer there might well be an air pollution factor
too, since the frequency of the disease is higher in urban that in rural
areas.
1
-------
pollutants must also be recognized.
A considerable amount of work has been carried out by G. M. Badger
and his associates on the mode of formation of benzo(a) pyrene and
other polycyclic aromatic hydrocarbons during hydrocarbon pyrolysis
but comparatively little has been published on the mode of formation
of polycyclic aromatics during incomplete combustion in flames..
The objectives of the present research were as follows: -
1. To develop appropriate analytical methods for poly cyclic aromatic
hydrocarbons in soots and particulates and in particular to study
methods employing gas chromatography and mass spectrometry.
2. To study the formation of these compounds in both pre-mixed
and diffusion flames of simple hydrocarbons and the factors
influencing this. In particular, stable compounds likely to suggest
the mode of formation of polycyclic aromatics were to be sought.
3. To seek organo-metallic additives which might lead to a redaction
in the formation of p. c. a.h. since pyrolysis and oxidation processes
can often be influenced by catalysts and additives.
This report comprises five parts; Part IV is the more detailed one
as our publications and progress reports have already dealt with the
earlier work in some detail, and consequently it is not considered
worthwhile reporting this again here. Part V is a brief review
of the work of others since, unfortunately, the loss of kpy personnel
precluded work in this area.
-------
Part I:- deals with improvements made in the gas chromatograpliic
method of analysis of polycyclic aromatic hydrocarbons
in soots since the description of the earlier work by
B. B. Chakraborty and R. Long. Th« use of a solvent
*
of low boiling point (and toxicity), rotary evaporation
under vacuum at a low temperature, the discarding of
the initial separation by column chromatography, the
use of an internal standard in the gas chromatography
and of high resolution mass spectrometry in the
identification of individual p. c. a. h. , have all been
improvements.
Part"II:- applies the above method to the analysis of soots
collected from fire-fighting schools in the U.S.A.
These soots are shown to contain appreciable amounts
of known carcinogens and consequently exposure to
the smoke may well constitute a health hazard to
personnel.
Part III:- describes an offshoot of the present work viz. the
application of the integrated ion-current technique
in mast spectrometry to the resolution of isomeric
compounds not readily separable by gas chromatography.
Thus, benzo (a) pyrene might be estimated in the presence
of benzo (e) pyrene and perylene. In conjunction
with other techniques of separation such as gas
chromatography or thin layer chromatography this
* Compared with benzene or chloroform
3
-------
method seems to offer potential. The combination of
T. L.C. with high resolution mass spectrometry results in
a highly sensitive and rapid method of analysis for
polycyclic aromatic hydrocarbons.
Part IV:- the main section, describes work carried out with the
aim of finding out more about the mode of formation
of polycyclic aromatic hydrocarbons during incomplete
combustion in flames. The earlier work already
published by P. Dearden and R. Long and by B. B. Chakraborty
and R. Long respectively, is not considered here;
attention being paid to as yet unpublished work.
A relatively quick and reliable method has been devised
for withdrawing samples, at different heights above
the burner, from rich oxy-acetylene and oxy-ethylene
low pressure flat pre-mixed flames. High resolution
mass spectrometry has been used for the analysis
of more volatile species and programmed - temperature
gas chromatography has been used for the analysis of
less volatile species including polycylic aromatic hydrocarbons.
Both u. v. absorption spectroscopy and mass spectrometry
have been used to identify species in the latter case.
The results show that, in general, the same
products are formed in rich oxy-acetylene and
oxy-ethylene pre-mixed flames, suggesting that the
-------
polycyclic aromatic are formed by the same route
in each case. The results confirm the important
role of acetylene in hydrocarbon flames and in
particular the results are in general agreement with
the pioneering work of K. H. Homann and H. Gg- Wagner
who found that polyacetylenes reach a maximum
concentration at the end of the oxidation zone and
that soot formation begins at the zone where the
polyacetylene concentration falls off. The present work
supports a scheme of G. M. Badger for polycyclic
aromatic formation during pyrolysis in that certain
compounds with a two carbon atom side-chain attached
to a benzene ring appear to represent "key" aromatic
species formed from acetylene, although the reactions
involved are likely to involve free radical, rather than
molecular'species. Such substances believed to be
important as precursors of polycyclic aromatics include
phenylacetylene and styrene. The experimental evidence
does nck favour ethylene or 1.3 butadiene as important
intermediates. Although only stable species can be
sampled, these are considered to give useful information
as to likely steps involved in the formation of the
polycyclic aromatic hydrocarbons in flames.
Part V:- While the effect of introducing oxygen into the combustion
air, or to the hydrocarbon itself, in a diffusion flame
-------
has been reported earlier, no progress has been
made in examining the effects of organo-metallic
additives, owing to loss of key personnel at certain
stages of the work. However, recent work in Germany
by G. Spengler and G. Haupt has indicated that
reduction in soot and poly eye lie aromatic hydrocarbons
by this means is feasible and work on the suppression of
soot and p. c. a. h. in flames by the use of metal
additives has recently been published. D. H. Cotton,
N. 'J. Friswell and D. R. Jenkins claim that the
mechanism of action of the alkaline earth metals
is one of gas-phase catalysis of the decomposition
of hydrogen or water vapour giving rise to the formation
of hydroxyl radicals which remove soot or precursors
of soot.
-------
Part I Improvements in the Gas Chromatographic Analysis of Polycyclic
Aromatic Hydrocarbons in Soot Samples.
The gas chromatographic method described by B. B. Chakraborty and
Ronald Long (Environmental. Science and Technology, I, 828-834, (1967)
has been improved and made more rapid in the following way
i) The use of methylene chloride (in place of chloroform) as solvent
*
in the extraction process enables this to be carried out at a
temperature below 40 C and so helps to prevent the possibility of
any further reaction in the Soxhlet apparatus.
ii) This solvent is readily removed from the extracted polycyclic aroma-
tics by rotary evaporation under vacuo at 0°C thus avoiding loss of
more volatile hydrocarbons. (Such as benzene, phenylacetylene,
naphthalene).
iii) The initial separation by column chromatography docs not seem to
be necessary and the analysis is thus speeded up considerably by
direct injection of an extract of soot into the G. C. column.
Thus, a Soxhlet extraction of '0. 5 g. of soot is carried out fcr
six hours using 250 ml of methylene chloride as solvent. The
resultant solution is evaporated down under vacuum at 0 C and the
residue is dissolved in 2 ml. of methylene chloride. 25^1. of this
solution are used for the programmed temperature gas chromatography
as described previously.
iv) The use of 3 methyl phenanthrene as internal standard increases
accuracy and also aids in the preliminary identification of components
by retention time ratios.
-------
v) High resolution mass spectrometry is used in identifying the component(s)
of certain gas chromatographic peaks, (ef. Part IV)
The above method has been used in Part II
-------
PART II Gas Chromatographic Analysis of Polycychc Aromatic
Hydrocarbons in Soot Samples from Firefighting Schools
in the U. S. A.
Peak No. in Gas
Chromatogram
Compound
Quantities in mg per gram soot.
City A City B City C
(Norfolk) (San Francisco) (Philadelphia)
1
2
3
4
5*
6*
7
8
9
10
11
12
13
14
15
16
17
Naphthalene
Acenaphthylene
Fluor ene
Phananthrene/
Anthracene
Methyl phenanthrenes
4, 5 - Methylene
phenanthrene
Fluor anthenes
Pyrene
Benzofluorenes
Methyl pyrenes
Benzo (m, n, o)
fluoranthene
Benzoanthracene/
Chrysene
Benzo fluor anthenes
Benzo(a)pyrene
Benzo(e)pyrene
and Perylene**
Indeno (1,2,3-c.d)
pyrene
Benzo (g, h, i )
perylene
Anthanthrene
0. 15
0.05
0.05
0.4
0.4
0. 15
0.3
0.6
0.1
0.05
0.2
0.4
0.5
0.9
0.4
0.4
0.05
0. 15
0.05
0.05
0.4
0.4
0. 15
0.3
0.6
0.1
0.05
0.2
0.4
0.5
0.9
0.4
0.4
0.05
0.6
0.4
0.4
2.4
3.0
1.8
4.6
7.8
1.2
1.0
2.0
4.6
3.5
4.5
3.3 -
2.5
0.9
0. 15
0. 15
0.8
2.2
4.3
1. 7
6. 1
8.6
1.5
0.15
2.0
2.0
1.3
1.2
none
detected
it
it
* identified by reference to earlier work.
** mass spectrometric measurements indicate about 50% benzo (a) pyrene.
(We are indebted to Dr. A. R. Siedle (U.S. Navy Preventive Medicine Unit,
2, Norfolk, Va for the samples, provided at his initiative).
-------
Part III. The Integrated Ion-Current Technique in Mass Spectrometry.
In conventional quantitative mass spectrometry, calibration
of the instrument is carried out for a specific compound by maintaining
a reservoir of the vapour of this compound at constant pressure whilst
allowing the compound to leak from the reservoir into the ion source
of the mass spectrometer at a very slow rate. The sensitivity of the
instrument may then be expressed in terms of the ion current measured
at any selected m/e value for a given partial pressure of the compound
in the inlet reservoir. Such a method is quite unsuitable for the
examination of samples of relatively in volatile solid materials. It is
then more appropriate to evaporate completely a small weighed sample
of the solid material and to record the ion-current at a selected m/e
value during the course of this evaporation. The sensitivity of the
instrument may then be expressed in terms of the integrated ion-current
obtained at the selected m/e value for a given weight of sample evaporated.
When a voltage scanning orquadrupole mass spectrometer is
being used, it may be possible to tune into the appropriate mass before
the ion-current at that m/e value is established. This is not so
simple when a magnetic scanning instrument'is used and it may be
necessary to use a calibrating substance, usually the vapour of hepta-
cosa fluoro tri-n-butyl amine to calibrate the mass scale. The peak
switching facilities of the instrument can then be used to tune into the
required mass before evaporation begins and the effects of any slight
instrumental drift are obviated. The record obtained consists of a
series of peaks drawn at one second intervals, the height of which
correspond to the instantaneous ion-current. It is the envelope of this
11
-------
record which is proportional to the integrated ion-current and hence to
the amount of sample evaporated.
When the sample consists of two substances the mass spectra of which
both contain a peak at a specific m/e value the integrated ionncurrent may
show fine structure because the rates of evaporation of the two compounds may
i
be different.)This effect is most marked when1 recording the molecule ion-
l
current of a substance which can exist in two isomeric forms.
Using a GEC -AEI, MS9, mass spectrometer J. R. Majer and
R. Perry have shown that by the above technique anthracene and
phenanthrene can be clearly separated. Benzo (a) pyrene is found
in products of incomplete combustion along with two comparatively non-
carcinogenic isomers, , benzo (e) pyrene and perylene. By the above
technique benzo (a) pyrene can be separated either from perylene or from
benzo (e) pyrene, although benzo (e) pyrene cannot be separated from
perylene.
Synthetic mixtures of polycylic aromatic hydrocarbons have
- • •
been separated by thin layer chromatography for subsequent determination
in the above manner by J. R. Majer, R. Perry and M. J. Reade.
Calibration curves were constructed using standard solutions of each
compound and these were linear in all ranges between 10~" and 10-12 g€
Provided the position of the eluted spot sample is known. , these
t
authors claim that it is possible to transfer samples in the picogram
range from the chromatogram to the mass spectrometer. Extraction of
the poly cyclic aromatic from the cellulose adsorbent can be carried out
in a micro centifuge tube with as little as ZOjml of solvent, enabling
a large part of the eluted sample to be used in the mass spectral assay.
12
-------
R. Perry, R. Long and J. R. Majer have published a paper
indicating the sensitivity and versatility of the integrated ion-cur rent
technique with polycyclic aromatics by comparing the analysis of a soot
extract obtained by mass spectromety with that obtained by gas chromatog-
raphy. For gas chromatography the concentration of the solution used
was a hundred times that used for mass spectrometry and a 25 A 1 aliquot
was used compared to a 5 *i 1 aliquot used in the integrated ion-current
method.
The usefulness of the mass spectrometer as a qualitative instrument
for the identification of polycyclic aromatic hydrocarbons in soot
samples has been demonstrated by M. Chaigneau, L. Giry and
i v
L. -P. Ricard who have shown that 75 compounds can be identified
in soots, including 34 hydrocarbons many of which are polycyclic
aromatics. The above mentioned quantitative technique, however, seems
to offer some potential in the field of air pollution studies.
References - Part III
i) J. R. Majer and R. Perry. J. Chem. Soc. (Sect. A) 822(1970)
ii) J. R Majer, R. Perry and Miss M. J. Reade.
J. Chromatog, 48 328 (1970)
iii) R. Perry, R. Long and J. R. Majer. Paper CP-37E. Second
Internat. Clean Air Congress. Washington D. C. Dec. 1970
iv) M. Chaigneau, L. Giry and L-P. Ricard
Chimie Analytique 5^, 187, (1969)
13
-------
Part IV The Mode of Formation of Polycyclic Aromatic Hydrocarbons
in Rich Pre-Mixed Flat Flames
(This section of the Report is based on the Ph. D. thesis of
B. D. Crittenden, without appreciable alteration. It is proposed to
publish a paper on these results at a later date - the conclusions
drawn in the paper might well be slightly different from those
reported here when the results have been re-examined critically).
15
-------
1.1 Rich Produced Flat Flames
1.1.1 Premised and Diffusion Flames
21
Gaydon and Wolfhard have stated that no flane nay be characterised
as being either purely premized or purely diffusions! in nature.
For example, rich premised Bunsen-type flames depend on the outer diffusion
flane cone for their stability and diffusion flames depend on a premizing
22
region near the burner rim for their stability . Reducing the pressure of
a diffusion flame increases the premizing region at its base and at
sufficiently low pressures, such a flame may be Indistinguishable from
one of a premized nature. The similarity between premized and diffusion
flames is more apparent in the multiple diffusion burner of the type used
23
by fieri and Wilson '\ as the tube spacing is reduced the array cf tiny
diffusion flames becomes Indistinguishable from a premized flame.
1.1.2 The Flat Prefixed Flaae
The requirements for such a flat flaae aret
(a) that the gases entering the burner should be perfectly
mixed and should leave the burner port with a perfectly
uniform Telocity distribution, and
(b) that the flow of gases leaving the burner port should
remain undisturbed by the behaviour of the hot products.
If these two requirements are satisfied and if the velocity of the
gas mixture leaving the burner port is equal to the burning velocity of
the gas mixture, the flame is perfectly flat and does not require
stabilisation.
16
-------
24
Fowling la his studies of the accurate measurement of low
burning velocities firat designed a burner which would stabllioe auoh •
flames* Flat flames of faster burning nixtares may be stabilised over
25
cooled porous plate burnero of the type used by Botha and Sp&lding ,
26 27 28 29 P 2
Kydd f Homann and co-workers " t Yumlu , and Tomkins and Long •
In auoh burner oysteaa. heat ia transferred to the cooled burner plate aa
the flaoe approaches it. The temperature of the burner is little affected
if it is efficiently cooled although the temperature of the gases in
contact with it is considerably reduced. Thus the flaae speed is
reduced and as a result, the flame rapidly and automatically takes up a
position of equilibrium a short distance away from the burner plate where
it loses Just enough heat to reduce the flaae speed to that of the gas
stream.
The thickness of the reaction zone in prefixed flat flames depends
on both the pressure and the burning velocity. Fristroa and Westenbarg'
present the following empirical relationship for an estimation of the
flame front thickness (L) for fuels burning in oxygen or air.
where L • the flame front thickness, cm.
P • the pressure, atmospheres
V • the flame speed, cm/sec
If it is assumed that the flame speed is a constant then the flams
front thickness is inversely proportional to the pressure. Conooqu*>...ly,
reducing the pressure of a rich prefixed flat flame extends the reaction
zone to allow u
-------
1,1.3 The I-seudo-one-diaensionnl Flame
The results of particle track studies^ demonstrate that in
practice flat flames are not truly one-dimensional and that the flame
is stabilised on the burner by a toroidal vortez of gases (see Figure 1.1).
Levy and Veinberg*1 have shown that the stability of the flame is
diminished when (a) a column of inert gas flows upwards around the burner
and (b) too narrow a chimney IB used.
52
Howeveri Fristrom, Grunfelder and Favin' have concluded that a
low pressure, lean premized ozy-aethane flame is sufficiently one-
dimensional to allow quantitative detcrulnation of flow velocities, mole
fractions and temperatures. Fenimore, Jcces and Moore'' have shown aleo
that samples withdrawn through critical flow quarts probes give the same/
OiCiH ratios at various radial positions close to the burner surface in
54
rich premized flat hydrocarbon flames. Singer and Grumer^ have
demonotrated that rich propane/air flat flameg are one-dimensional close
above the blue reaction cone but become progressively leas so higher in
the burned gaees. This is because as the burned gases rise, they become
cooler and the flame cross-sectional area decreases! consequently, e£ge
effects become more pronounced and radial diffusion may account for poor
28
flatness. Bonne and Wagner have also shc-m that rich premlxed flat
flames are poorly one-dimensional with respect to carbon formation high
in the burned gases of the flame.
Owing to the above-mentioned change in flame geometry and also to
the expansion of the gases leaving the burner port (Figure 1.1) there
is some difficulty in defining a flame diaaeter and consequently a flam*
cross-sectional area. The effective diameter of a flat flame burning on a
18
-------
IV. Figure 1.1 Flow Pattern at the Edge of a Flat Flame Burner
(reproduced from Reference 3)
19
-------
circular burner may be defined in one of three vayst
(i) a* the diameter of the flat region alone
(11) M the dlaaeter of the flat region + (2)' z the baae
projection of the upturned edge
(lii) as the diameter of the whole flame projected onto a
horicontal plane.
Za this studyf the eroaa-eectional area of the flame, A. ( at a
height h oae above the burner ourface ia that area derived from the
equationi
"K 2
*h • Tx <
where d. ia the optically measured diameter of the flame at a height
h COM above the burner surface.
1*1.4 Reaction Products in Rich Premised oxy-aeetyleneand
oxy-ethylene flames
we «g
Bonne and co-workers'" and later Toapklns*^ have found that there are
three diatinct conee in rich proaixed ozy«aoetylene and oxy-ethylene
flame«» naaely,
(a) the non-luminous zone directly above the burner Buxface(
which la some time a oal}.ed the preheat eone
(b) the blue-groen reaction or oxidation* cone
* In prefixed oxy-acetylene and oxy-ethylene flanes, Bonne, Uoaann and
Wagner'*7 have found that the region between the blue and yellow sonee
coincides well with the cone where the oxygen is nearly totally consumed.
20
-------
(a) the yellow-orange eonc of the hot burned gases*
Thesv worktrs have found also that there is no separation of the
57 58
blue and yellow tones as is reported by Killikan", Street and Thomas^,
and Fenlmore, Jones and Koore who have worked with premixed flames of
fuel/air mixtures.
By means of a combined mass spectrometer and moltcular beam
59
sampling system, Homann and co-workers" hare been able to detect many
hydrocarbons present in the flaae gases* Since the wall influence on the
particles reaching the ion souroe of the mase spectrometer via suoh a
sampling ays lea is very low Bonne, iiomann and Wagner were able also to
detect some free-radicals. In more recent work Homann and Wagner hare
classified three groups of hydrocarbons which are found to be products
in these flamest
Group 1 t acetylene and polyacetylenes with mass numbers ranging
from 26 to 146
Grcup 2 i polycyolio aromatic hydrocarbons (peah) with mass
numbers ranging from 78 to approximately 500
Group 5 i reactive pcah with side chains and containing more
hydrogen than purely polycyclio aromatic hydrocarbons
(mass nuitbere ranging from 150 to greater than 350).
In sufficiently rich oxy-aoetylene and oxy-ethyleno flames,
unsaturated hydrocarbons with the generic foroular C4 H. have been
zn 2
identified , with values of n ranging from uni
-------
in suob flames JJonne et al'' have snown that the concentrations of these
compounds increase with increasing fuel/oxygen ratios (Figures 1*2 (a)
and 1.2 (b)). Whilot the polyacetylenes are formed late in the blue
(or oxidation) zone it appears that in an oxy-ethylene flame, their
59
fonaafcion is preceded by that of acetylene . The concentrations of
acetylene and the individual polyaoetylenes reach maximum values at the
end of the blue reaction zone And then decrease to a final, roughly
constant value in the burned gases (Figures 1,} (c) and 1.4 (b)). The
maximum concentrations of the polyacetylenes as veil as their concentrations
in the burned g&sea increase with increasing fuel/oxygen ratios55 but the
ratio of the maximum concentrations to final concentrations decreases.
(It should be noted at this stage that increasing the fuel/oxygen ratio
9T Xfi
decreases the temperature at corresponding points in such flames .)
The ratio of maximum concentration to final concentration of individual
polyaoetylenes becomes greater with an increasing carbon content of the
polyaoetylene (Figures 1.3 (o) and 1.4 (b)).
From the measured acetylene, poly&cetylsne and hydrogen concentrations
in the flames considered, Bonne et al" were able to show that relation-
ships exist between these molecules in the form of equilibria.
"Equilibrium constants" were calculated for reactions such as
MA ^ C4B2 * H2
where the "equilibrium constant", K, is given by
22
-------
0.1
L
CJ
01
I!
nit
10 K 30 <0 50
Nttffnt aoovt thf o&fitf
JV«Figure 1.2 (a) Concentration profiles of diacetylene in
flat acetylene-oxygen flames of different
02^:02 ratios, (p = 20 mm.Hg; flov velocity,
50 cm. sec"1)
Pttctnt by Bright
(0 mm abort thr burnrr
v lw
Mok % at C3«j in Ihf unburnrd gat
'IV.Pigure 1.2 (b) Concentration of acetylene and polyacetylenes
in the burned gas of rich acetylene-oxygen
flames as function of the initial 02^:02 ratios,
The hatched line indicates the range in which
the yellow luminosity becomes visible
(reproduced from Reference 35)
23
-------
OS tor7
o;
X to SO
fcj
0 10 10 30 X 30 tO 50
Height obov* th« burner [mm]
IY. Pigxrre 1 .3 (a)_(c) Concentration profiles in a flat
acetylene-oxygen flame (CjII>:Ot = 0.95) at a
pressure of 20 nun UK and a flow velocity of the un-
bunicU gas of 50 cm sec"1. The figure doca not
include all of the components which could be
measured, (d) Emission of Ci, CII, and OH, and
continuous omission in the same flume as function
of the height above the burner.
(reproduced from Reference 35)
24
-------
(a)
Soot massf/action by
absorption
electronmicrogr.
in
400
200
0
0.020
0.015
0.005
0 20 40
C2(SKSA)
60 (mm)
f
.S
fci
I
CH(3672A)
> OH (3064 A)
20 40 60 (rrm)
Height abort the burner
IV, Figure 1.4 (a) Absolute concentration of carbon, the mean
number of particles and their mean diameter in a
C2H2~°2 flame (C2H2S°2 = 1*4; P = 20 mn>.Hg.; flov
velocity, 50 cm. sec ). (b) Concentration profiles
of polyacotylenes (C^, C^, C H ) in the tame flame
(c) Emission of C , CH, OH and continuous emission
profiles in the same flame.
(reproduced from Reference 35)
25
-------
Using these "equilibrium constants", these workers have calculated the
•nthalpioa of formation of C.B. (diaoetylene) and CgH. (tri-acetylens)
froai thoir elements to be 109 - 0*7 kcal/Bole aod 163 - 1«2 kcal/aols,
respectively.
In the discussion of their results Bonne, Hoaann and Va/jner'5
emphasise the possible importance of polyaoetylenea in the formation of
•carbon1 in rich premized flanes. They have claimed that in the zone
where the 'carbon' particles grow, there are no other hydrocarbons present
in concentrations large enough to account for the 'carbon' which appears
in the soot. (See Figures 1.4 (a)-(o)). This view ia supported by the
fact that increasing the fuel/oxygen ratio tends to increase the
individual polyacetylene concentrations, tho number of 'carbon* particles
and the anount of 'carbon* produced .
In premized flat oxy-acetylene flamea the acetylene concentration
doss not fall to eoro but beyond the blue cone maintains a steady value
(Figure 1.3 (a)) whilst in premized flat oxy-ethylene flames the ethyiene
59
concentration reaones eero at the ond of the blue sons .
Other low molecular weight hydrocarbons found in these flames
include pro^ylene and oethyl^acctylcno which reach thoir •aaxiaum
concentrations within the oxidation zone and are destroyed before all the
oxy&cn is consumed. The auiue is true for dimethylacotylene and vinyl-
acotylene, the uaxiauo concentrations of which precede that of diacetylene
in the ethyiene flauc; the regions) oi" aoxirium ooncentroLtiun of vinyl-
acotyleue and diacetylene coincide in the acetylene flase.
26
-------
The results of iionno, hoinann and ,-agner ohow that there is a
smooth transition between non-soot-fonainj and BOOt-forains flamea as la
indicated by the concentration profiles of the polyacetylenes and free-
radiOcils aod by the emission and absorption profiles of the 'carbon1
particles.
In the range of fuel/oxygen ratios 0*6 to 1*2, the maximum concentration
of Oil radicals la reduced from 2 x 10** mole fractions by one order of
35
magnitude . It seems likely that free radicals play an important part in
the oxidation reactions since radicals such as C H, CD, 0, IT CO and ItCO
disappear rapidly at the end of the blue eone (figure 1.3 (b)). Their
disappearance is explained by the formation of heavier radicals which
either cannot be detected in the burned gases or are adsorbed onto the
surface of 'carbon1 particles. In fact, electron spin resonance studios
have shown that soot particles have many unpaired electrons, presumably
because they contain free-radicals^ .
In Killikan'a experiments with prefixed ethylene/air flames'* infra-
red emission profiles show that acetylene is produced in the blue sons
well before soot luminosity occurs and that its concentration in the
burned gas remains constant, in agreement with the results of liomann et al.
Killikan found also that Just downstream of the oxidation zone, the
•
OH concentration is about eight tines its equilibrium value. Since its
concentration was also found to be 10 times that of the 0 radical it
•
secies likely that the CH radic&l is resjonsible for the oxidation reaction!
and that it may account for the dark space which he observed between the
blue and the yellow nones (see page 6).
27
-------
Fenicioro and Joncn h?vc studied fuel-rich ncetylenc/oxyjen finaes
to find that the rate of decay of acetylene in an oxygen containing
atnosjihere at tcnj.eratures between H27°C and 1727°C nay bo represented
by i
J ^^ ^^ ^ O ^i» * ^^ OB ^^ X
dt •• < 2 "^ *.-y*.y2~'
These workers have also noted that the terra /"*CH_7 m°y not tc replaced by
/"«17. f«7«« /"°? J»
42
Recently Cotton, Friswell and Jenkins have confirmed th?t alkaline
earth motels reduce soot fomation in a propane diffusion flame end have
BU&CFBted that the oode of action of barium, calcium or strontium is the
gas-phase catalysis of the decomposition of hydrogen or water vapour to
produce CH radicals* In prefixed flanrn where radical concentrations ar«
greater than equilibrium, catalysis will tend to reduce concentrations
towards equilibrium values.
In order to establish the concentration profiles of hydrocarbons in
Group 2 and Croup 3, Bonne, liomann and V
-------
The maao spectra which rcaul't from heating the 'curbon1 collected
froa heists well into tiie yellow zone of the flame ehow major peak* at
Bass numbers which correspond to thoue of the polycyclic aromatic
hydrocarbons of Group 2{ the nn.so spectra which result from heating the
•carbon' collected Just beyond the blue eone show the preornce of a larger
nusber of hydrocarbons with mass numbers ran^ln^ froa 150 to $001
corresponding the the hytiroc.irbcrio of Group J« It was Eu~jeated that
these latter hydrocarbons are polycyelic aromatic hydrocarbons with side
chains and which are core hydro^enated than the parent poah.
The concentration profiles of hydrocarbons of Group 2 are very
•ieilar and Hoaann and Wagner otated that their respective concentrations
steadily increase in the burned J&BOS. They are forced later than the
polyaoetylenea and their individual concentrations lie on average between
—o
that of CjJ!- and C-J*2 with a total concentration of about 10 mole per
cent of the burned &RACB. Since the concentration of each pcah of Group 2
«
increases without going throu/ii a onxinum value* hooann and Vainer have
concluded that these pcah are not important interxodiatco or 'nualei* for
•carbon1 formation in an acetylene flame.
Homann and Wagner could not accurately establish the concentration
profiles of the hydrocarbons in Group 3 but they were able to ohow that
tx,ey do not survive in the hot £as behind the oxidation zone. Unfortunately,
they oould not also identify individual compounds in this r-roup which are
present in individual concentrations of 10 ' oole fractions. However,
since they are deotroyed just beyond the blue zone they considered that
the iiydrocarbous of Group 3 are intermediates or 'nuclei1 for 'carbon*
particles.
29
-------
HoMinn and '. a.-picr have pro.csrd the rpnctlon cchme outlined In
Figure 1.5 for the formation of r.olyr.cetylenes ?>_nd 'carbon' from acetylene
and free-ra«iicals. During the formation of thr large poly.icotyleno
molecules, the average else of the radicals involved increases. Thry hnv«
•u^rested that ouch lar^c radicals night react with each other or with
higher polyacetylenes to form ' afc;gre&ates ' t possibly involving ring
• •
closures! to produce the hydroctrlons of Group J, oinco the oaxinun
concentration of hydrocarbons in tr.is tiroup lies in a region where the
concentration of the polyacetylenes is decreasing and the formation of
poah of Croup 2 and 'carbon* is just beginning. The reactive hydrocarbons
of Group 3 Are probably still free -radical in nature and Hoa&nn and Wa-ner
have au^ges ted that they mi^ht tjrow to srir.ll 'cnrbon' particles by further
addition of polyacetylenes. Ihis view is supported by the fact that the
hydro(T«n content of the soot decreases whilst the particles axe still
growing.
2
Tompkina and Long have collected 'carbon' at various heights in
rich preaixed oxy-acetylene flames similar to those employed by iiomann
27 2B
and co-workers " by .withdrawing samples isokinetically through a
F -
relatively larje water-cooled funnel. The total collected natcrial was
extracted witn chloroform for subsequent separation and identification
by gas-liquid chroma tography and ultra-violet spoctroacopy.
The flux* profiles of the following are presented ao functions of
the sampling height in the flame
* In flancs where very Mrh concentration cjra'lientK exist, flux profiles
Bay bo very different from concentration profiles, as is explained in
Appendix II.
30
-------
Badlcal reactions
with CB and
Acetylene - * Polyaoetylenea
polyacetylene radicals
Addition of Addition of C2&2 and
a radical polyaoetylenes, cyclisation
» branched radical »•
polyoyolio arooatio
hydrocarbons
I further addition
-»> reactive, partly cyclic of polyacetylenes
hydrocarbons, hydroeen-rloh
(group 3)
Addition of small soot
particles and polyacetylenica
temper process, inactivation
aall aoot particles
polyoyolic aromates
by surface reactions
t»
.large soot particles (inactive, 2^0 A)-
Agglomeration of large soot particles to chain-like
aggregates, slow growth of carbon amount by
heterogeneous decomposition of C2U2 and polyacetylenes
(activation energy JO-40 kcal/mole).
IV. Figure 1.5 Reaction Scbomo l'ropos«d by lioraunn OIK! r'a^ner1 for
the f on jut ion or j'oly.\cctylonea and 'Carbo
-------
1. total collected material (/iuure 1*6)
2. chlorofora soluble caterial (Figure 1*7)
3» chlorofora insoluble naterial (Figure 1.8)
4. total pcah (Figure 1.9)
All these flux profiles reach naximum values just beyond the blue
oxidation zone after the naxinua temperature has been reached (Figure 1.10).
Noneof the fluxes decreases to xero but each reaches a roughly constant
value until a height of about 16.0 en. in the flnme when the fluxes of
soluble material and pooh increase again. The fluxes of ell the above
materials increase with increasing fuel/oxygen ratios (Figures 1.6 to 1.9)*
2
Table 1.1 shows the range of poan identified by Toropkins and Long in
aoots sampled, from rich prefixed oxy-acetyler.e flanes. from tneir results
it is apparent that there are two Uiotinot regions of pcah fortnation in the
fluae. The lirat occurs near the end of the blue zone where the temperature
is rapidly rising and where the acetylene and oxygen concentrations are
35
falling . Just beyond this region the pcah are destroyed, in
disagreement with Hoaann and ..agner who have stated that the concentrations
of these species increase rapidly behind the oxidation zone without going
through any maxima. The second region of pcah formation occuro ouch
2
higher in the flame where it ia su^catcd that they rai^ht be formed by
pyrolysia of residual acetylene. It io worth noting that the temperatures
within tnia region (900° - 650°C) are those favourable for pcah
*f
foruation '.
The decline in the flu* of ?cun after the initinl concentrfttion
maxiauia appears to agree with the decline in the concentration of Croup 3
\
i
32
-------
X
t
Flarre 1 1.2
Flame 2 1 c
Flame 3 , Q
Q *• 8 12 16 20 24 28 32 36
Height above burner surface-cm.
(IV. Figure 1.6 Flux of total jwlymcric material vs. hc'ight above burner surface.
26
24
22
20
18
* Flame 1
X Flame 2
t Flame 3
IV. Figure 1.7
8 12 :S 20 21. 28 32 36
Height above burrer-cm.
f
Flux of soluble material vs. height above burner siirfaw.
(Reproduced from Reference 2)
33
-------
F'.an-.e 1.
X Flame 2.
•*• Flame 3.
0 4 8 12 16 20 u 28 32
Height above burner - cm. «•
IV . Figure 1.8 plux of insoluble material vs. height above burner surface.
4400
3600
^2000
I
* 1600
800
400
C2H2/02
• .Flame 1 1-2
X Flame 2 1.5
t Flame 3 1»8
0
8 \'i 16 20 24 28
Heigh! above burner surface -cm.
32
Figure 1.9 Flux of polycyclic aromatic hyilrocarlions vs. lu-i^hl above burner
surface.
(Reproduced from Reference 2)
34
-------
x FLUX OF INSOLUBLE
• FLUX OF PC.A.H.
4- TEMPERATURE.
12 16 20 24
Height above burner- cm.
;IV. Figure 1.10
Composite plot of flux of insoluble material, flux of p.c.a.h. and of temperature respectively vs.
height above burner surface, (for Flame 2).
(Reproduced from Reference 2)
35
-------
IV. TABLE 1.1 Range of polyoyolio aromatic hydrocarbon* identified
by Tcmpkina and Long?
1
2
5
4
5
6
7
8
p
10
11
12
15
14
15
16
17
16
19
20
p.o.a.ht Identified
naphthalene derivatives (?)
aoenaphthylene
fluorene
aoenaphthene
aoenaphthylene derivatives (?)
phenanthrene (oa. 80>) +• antiiraoene (ca. 20;*)
4,5-aethylene phenanturene
fluoranthene
fluoranthene ieomera (?)
pyrend
benxofluorenea (?)
•ethyl pyrenes
benBO-(n-n-o)-fluoranthene
benz(a)antnracene (ca.OO^)) + ehrysene (
-------
hydrocarbons as reported by Hoaann and Vainer • However, Tompklns and
2
Long hare shown that the concentrations of ths posh never actually
decline to sero. These results apparently contradict those reported
by Horaann and Wagner since the majority of pcah identified by Toapkins
and Long are devoid of Bide chains and are thus probably rather unreactive.
The chloroform insoluble material which is foraed early in the flaae
has an E/C ratio approximately equal to 1*0$ which compares with that
obtained by Hoaann and Vagner for the *carbon1 particles formed just
beyond the blue reaction zone. On account of the fact that the ohlorofom
insoluble oatnrial low in the flane has an K/C ratio similar to that of
2
the fuel itself, ?oapkins and Long thought it unlikely that this material
could be formed from polyacetylenee; lioaann , however, pointed out that
higher polyaoetylenes can easily add on smaller hydrocarbon radicals to
font a larger radical which in turn can add on further acetylene with loss
of some hydrogen to account for the H/C ratio of unity.
Feniaorev Jones and Kooro*5 have worked with quenched flat flames
of various fuel ond oxygen mixtures at various pressures to find that
in an oxy-*thylene flaoe, the ratio of methane to aoetylene in the burned
gas is about 0*3* whilst in an oxy-aoetylone flame no methane can be
detected. In both flames they found that ti*s ratio of benzene to
aoetylene in the burned gas is 0.006 - O.C02. It is interesting to note
that these workers did not find a burned gas (from the fuels they
employed) which is free of acetylene, thus emphasising the possible
importance of this compound in flaae checietry.
37
-------
1*2
of Foloyclio ~ro:aatic rlyJrooar'oon Forxation
Various hypotheses hare been proposed for the fornation of poah
froa siople fuel molecules, a certain amount of e&phasls having been
laid on oonpounds considered to be preoureora of the poah. The roles
of coapounda suoh as acetylene* ethylcne and 1.3-butadiene and their
corresponding radicals are discussed in sections 1.2.1 to 1.2.4.
1.2.1 The Role of Acetylene
She possible iaportanoe of acetylene in the ohcaiatry of rich
premised flat flames haa been pointed out earlier (Section 1.1.4).
Groll* considered the divalent radicals of acetylene (UC • CH)
and aoetylenio compounds (HC • CH) to be lorortant precursors of poah
in the Tapour phase pyrolysis of hydrocarbons. Be considered that a
species suoh as UC • CH would more readily polyreerise to beniene rather
than form acetylene
i.e. J HC • CH
Groll proposed the following soheties for the formation of etyrene
(a significant prouuct in the pyroiysis of propylene) and naphtn&lenei
UC • CH
CH
styrene
2HC
CH
CH=CH.
CH=CH.
.c:
101 CH
C* "
H
* et°J
naphthalene
, 38
-------
Be also proposed that pcah such aa anthracene and phenanthrene
night be foraed via similar laeohaniBBB.
Stehling, Frazee and Anderson ' have considered it unlikely that
acetylene would polyaeriee directly to benzene which would then react
further in the manner proposed by Groll , since benzene is not very
47
reactive. Instead, Stehling et al^' have suggested the following free
radical processes for the formation of the phenyl radical and other
aronatio speciesi
H H C9H9 H H 1!
(l) C-B «• CJ,-*- 'C-C-CsCH * » HC-C-C-C-C-CH
phenyl radical
Sufficient energy night then be released fort
."or *
and
B B C-H. H H B fl
(ii) -CgH + CgHg—* -C-C-CSCH —=-^- -C-C-C-C-CsC
v. ~w "". . ***
Hydrogen stripping could then occur together with condensation to
aromatic molecules.
39
-------
46
Cullla, Klnkoff and flettleton have proposed the following
reaction mechanism to account for the formation of polymeric material in
the pyrolysie of acetylene
CH.CH
49
In more reoent work Collie and Franklin hare proposed a similar
meohaniam for the polymerisation process which involves the
electronically excited triplet state of acetylene (denoted C-H-*) rather
than the free -radical CH=CH
surface
32H2
surface
C.B.*
C.H. (vinylacetylene)
44
The existence of other species may be explained thuss
C4V * CA
C6H6*
etc.
40
-------
50
Badger haa proposed a step-wise mechanism for the formation of
various peah from acetylene during pyrolysias
CH5CH ^ CH2«CH-CH«CH2
113-butadlen»
unit
pyrene
bensso(a)pyrene
Other poah might be formed
^
benco(b)fluoranthene
benBo(e)pyrene
41
-------
Badger5 has pointed out that ouch mechanisms may not be the only
ones Important in the formation of any one pfiah and that other species •
might also bo formed by decomposition of any of the above mentioned
speoieo to a variety of free-radicals which might reaot in different ways.
Although the importance of species such as acetylene and the acetylenyl
radical in the flame chemistry have been pointed out by Fenicorc et al5^
52 35
and Anderson , the results of Bonne Honann and Wagner' show aloo that
the fuel is initially broken down at least partially into C.. fragments to •
fora C- species such as aethylacetylene and propylene.
Chakraborty and Long55 have presented the reaction soheae shown in
Figure 1*11 to account for the formation of poah and •carbon' in their
experiaents on diffusion flames'*"" and the work of others in this
2 56
laboratory ' . Although it is apparent that no specific mechanion can
56
account for their resultsf Dearden and Long' were able to show, however,
that acetylene IB a significant reaction product in rich ethylene and
propane diffusion flames.
Since fcarbon' doos not arise preferentially from a two-carbon
fragment in explosion flames of propane-2-<3C5' and since the presence of
the OH radical has been established in flames' t Lindsey has suggested
,•
that poah might be formed either direotly from the CH radical or
indirectly via acetylene*
1.2.2 The Role of Ethylene
In their research on the pyrolysls of paraffins, Hague and Wheeler''
found that the production of ethyleme reaches a maximum at temperatures
42
-------
Aliphatic fuel
pyrolysis
acetylene butadiene
PCAH ^
(relatively stable)
\
\
\
X
\
I
vinylacetylena
diacetylene
I
favoured by
of
fcorOujatod frre radicals
polyacetylcnea
and relatively
low tcuperatures
( 1000°C)
polymerisation
polymers
N
\
\
oyclisation and
dehydrogeuation
favoured by removal
of U2 and by high
temperatures
( 1000°C)
polybenzcnoid
structures (radicals?)
perl-oondenscd aromatic
structurea, 'carbon*
IV. Figure 1.11 heaction Scheme Frrbtiited by ChakreL-orty and Ion,?3'
to account lor tl.t. /oiu:ition of I c
-------
between 7UJ°C an>l 7>0°C. l:.oy C'l-o fo^id Uut the yic-1'.ii of el'iy
uro pro^orlJUrul to tho jjs.-rcc.-nt. ._e jiclcio of ].cnh foir-od Inter in the
reactiono lc«di>i<; them to believe that the following chain-lengthening
and cyolisation oech&iiiso nicht account for their resultsj
2 C
1-butouc
1.3-butf.dicne
CH
C!I2aCH-CMoCH-C!TsCH2 •*• H.
2U,
This mechanism is very similar to that proposed by Kinney and Crovley ,
which hao been reproduced, in part, bclov.
CH,,
CH2
II 2 '
012
ethylene
11C
I
HG
1,2-butadicne
bonsene
etyrenc
-totrene
44
-------
The relatively hijh yield of toluene in the pyrolysis of C. and C.
hydrocarbons Is explained by the mothylation of LJ-butadienej Klnney
and Crowley have suggested that methyl radicals might be formed directly
from ethylene.
1.2.3 The Importance of a C. Species
The importance of a C» apeeles in the formation of pcah from simpler
molecules Is significant since Oro and Han report that about 97£ of the
total amount of aromatic hydrocarbons syntheslsed from methane at 1000°C
la composed of hydrocarbons with even nucbers of carbon atoms. This may
be explained by the fact that at such high temperatures, acetylene and other
Cy species are tbenaodynamioally more stable than methane and other C~
species*
(& d ^
Badger and co-workers ' ' have studied the formation of aromatic
hydrocarbons at high temperatures* and as a working hypothesis suggested
the following reaction scheme for the formation of a compound such as
benco(a)pyrene.
^ ^
(01 J, — [O
C
C
c
i
c
a
tetralin,
benzo(a)pyrecs
These workers have suggested that intermediates oi^ht be formed from
fragments larger than a Cy species and that complex hydrocarbons need not
neceanarily break down to C- species before re-synthesis to pcah.
45
-------
Lowcvcr, t*o rrr-,]<~ of tteJr ryrilyris rr, crlrynts44*51'64"85 with
hy
-------
pa
nn ct nl aloo cug^ftst th::t (a) f]uorr.ntlinnr ni-ht bs formed
from accnaphthylene» (b) pyrsne nisht be foracd from styrenfi and
(c) ohrysene ni^ht be forr.cd from indene by the following aechanisna.
GE
I
CH
dehydrog.
fluoranthene
(b)
dehydrog.
pyreno
(c)
dehydro^.
chryaene
Heobanlba (a) la supported by the reoults of Dor^mann ° and Kloeteel and
SO
hertel' vtio have found that acooaphthylene reacts readily at temperatures
between 140°C and 200°C wltL dienes to yield hydrogen substituted
fluoranthene derivatives.
An objection to the dicne synthesis reactions for the formation of
72
poah has been raised by Badger and 3potswood' vno have found that there
la no significant increase in the yields of benco(a)pyrene and
47
-------
benzo(c)j.yrcna :,hen 1.3-butaJicna la t:yrolyccd In the presence of pyrsne
v.iiour nt 700°C.
Proas the results of their vork on the thermal decomposition of
- -j Q«
ethylene at temperatures between 977 and 1577 c Kozlov and Knorre7
have concluded that butadiene whioh ie first formed frox ethylene 10
docoopooed to acetylene;
C2H4 * ¥4 — C4B8 — °fy * H2
1-butone butadiene
C4B6 — C2H2
48
-------
1.5 3uii--"jy oT Lh.TtfT 1
The resultD of previous rcucarch loth on polycyolic aroualic
hydrocarbon and 'carbon* formation iu rich prefixed flat flakes suggest
that the fuel is initially broken down, in part, to both C.. und C_ unitet
Higher molecular weight species mi^ht then be formed by chain lengthening
59
processea similar to those which have been proposed by Hague and Wheeler ,
Kinney and Growley and Stehlinj et al . These long-chain molecules,
by a process of dehydro^enation and cyclisatlon might account for tha
formation of pcah and other aromatic molecules in rich preaixed flat
flan:es. The chain lengthening processes are evident in the results of
*c 128
Hooann and co-workers and Kistiakovsky and co-workers ; it is well-
known, also, that polyacetylenes are unstable compounds and readily
polymerise . Homann and co-workers think it likely that the
polyacetylenes whicl. are found in rich premised hydrocarbon flt^neo ni(.iit be
precursors of 'carbon* although it is possible that such molecules are
by-products of other reactions.
Botn acotylcne and ethylene (and their correoponding radicals) have
been proposed as intenuodi^tes in the formation of pcah although Roaann's
results susjeat that acetylene is the more important species. Pyrolyaia
of residual acetylene in rich prefixed flat flames might account for the
2
secondary region of formation of pcah t especially since temperatures
AA
within this region are favourable for pooh formation generally .
1.J-Butadiene is present in relatively small concentrations in the
•xn
reaction Bone of rich prefixed oxy-acetylene and oxy-ethylene flanes ,
but under the conditions encountered in the blu? cone of such flames it is
likely that butadiene would decompose to acetylene rather thin polyaerise
(section 1.4«7)« The absence of other oleflnio compounds, notably the
49
-------
cyclic uoT.pounl8, eu;"TOto that the pyrolysis of ethylenw do«*« not occur
within the flnne.
Tho results from the pyrolysis of acetylene show that at high
temperatures and at ohort contact times diacetylene formation la favoured
and the presence of relatively lar^e amounts of thie compound in the blue
reaction cone of rich prcnixed flames su-jgests that the following overall
reaction mi:-;ht occur.
C4H2 * H2
Vinylacetylene, which is foraed prior to diacetylene in the reaction cone
of such flanes is forced generally at lower temperatures under pyrolysis
conditions*
The fact that biprienyl and other poly-phenyla have not been found in
measurable quantities botJi in ricu preoixed and in rich diffusion '
yj-Jsi floaos suggesta thai benzene pyrolyuis does not occur and that
tliio compound is a relatively stable by-product. The relative unrenctivity
of benzene and pcah ouch as naphthalene, anthracene, pyrenc, etc.* suggests
that the pcali which are present in flame soots are likely to be eido-
produots rather than important intermediates in the formation of 'carbon1*
However, once forced* polycycllc aromatic uydrooarbous nay well be pyrolysed
to some extent in the turned ^aaes in the flame to produce species of higher
molecular weight and possibly carbonaceous material also.
17 18
She works of Arthur and co-workors ' and Lindsey show that similar
pcah, althougn in differing quuntitalive distributions, are formed in
diffusion flanes of a wide variety of fuels. This suggests that these
compounds are stable by-products of the flame (cf. Croup 2 hydrocarbons in
50
-------
'j \.orl:) and that the actual laccliinisa of fornatioa cuy be
different-for each fuel.
There is coco uncertainty us to the nature cf the reactive poeh
(Croup 3) which Hozann uud co-workerc have claimed exist in the blue
reaction zono of prefixed flat flooes although it is accepted that the
more stttblo pcuh which nay be extracted frou flame ooota are devoid,
162
generally, of eidp-chains. Accent work carried out by Kern and Spenjler
has ehown that opccies such £S pr.f?r,yL\cetylenBf styrcne and toluene exist
in the cases of ricli hexane diffusion flakes.
51
-------
IV - 2
Experimental
The burner employed is of the type used originally by Botha and
25 2728
Spalding and later by Homann and co-workers ,' to stabilize flat
flames of various rich fuel/oxygen mixtures at reduced pressure and
36
also by Tompkins for rich G£ H2/02 flames.
Samples are withdrawn from the flame via fused silica microprobes
in two distinct ways as shown in Figure 2. 1.
The batch collection system enables the major products of the
flame (i.e. hydrogen, water, carbon monoxide, carbon dioxide;etc.)
to be sampled quickly and efficiently. The continuous flow collection
system allows samples to be withdrawn from the flame over relatively
long periods, thus enabling sufficient quantities of species such as
poly eye lie aromatic hydrocarbons to be collected for subsequent analysis.
IV - 3
Analytical Methods
Full details of the mass spectrometric and gas chromatographic
analyses o' the samples are given in the Ph.D. thesis of B. D. Crittenden.
IV - 4
Results and Discussion
IV - 4 - 1 Introduction.
IV - 4 - 2 Temperature measurements.
IV - 4 - 3 Batch and Continuous Sampling - Results and Discussion
52
-------
to pump S..
s/s and P.T.F.E.
connecting lines
Probe
(a) Batch
Premixed Gases
'10
to pump
batch sample
collection vessel
liquid nitrogen
P.T.P.E. connection line
extraction
filter
to pump S«
^ liquid
nitrogen
(b) Continuous
Premixed Gases
'cold1 trap vessels
IV. Figure 2.1 Sample Collection Systems
53
-------
IV - 4
Results and Discussion
4-1 Table 4. 1 shows the types of flame examined and the types
of sample withdrawn from them. It was not possible to obtain quantitatively
reproducible results from the batch samples unless argon was introduced
into the flame as an "internal standard".
In the flames there were geometrical effects caused by the
column of hot gases contracting as they became cooler. Since the
change in density may be assumed to have the same effect on the
concentrations of all species, including argon, in the flames employed,
the concentration (in mole fractions) of species obtained from the
batch sampling results are the actual mole fractions in the flame gases.
However, in the continuous sampling method, species are withdrawn
from the flame with no reference to the flame "internal standard".
Consequently, high in the flame gases, concentrations of species are
apparently too large and corrections must be applied to take the change
in density into account.
In this study the height above the burner at which the flame in
question has its maximum cross-sectional area is taken to be the
reference height, h max. If the diameter of the flame at height h max is
d max and the measured concentration at a sampling height h, (where the
flame diameter is dh) is X, then the concentration corrected for the
geometrical effects : X1 is given by
X' = X. $• ) 3
dmax
54
-------
Table 4-1
Planes Employed in this Study
Flaae
C2H2 flow
1/fflin NTP
C^ flow
1/nln NTP
02 flow
1/mln RTF
Ar flow
I/Din HTP
Total flow
1/mln NTP
Mixture strength
Fuel/oxygen
ratio
Free sure
DUD Hg
1
3.58
•
2.76
0.5
6.84
0.308
1.3
40
2
-
3-58
2.76
0.5
6.84
0.254
1.3
40
3
3.58
•»
2.76
-
6.34
0.308
1.3
40
4
3.32
-
3.02
0.5
6.84
0.364
1.1
40
5 6
3.32
3-32
3.02 3.02
0.5
6.84 6.34
0.303 0.364
1.1 1.1
40 40
Typea of Saaple Collected
Plane
Batch
Continuous
•cold1 trap
1 2
V J
J J
3 4
'
y
5 6
y
*
Continuous
filter extract
55
-------
The term ( h ) is cubed since the geometrical effects are
d
caused by a change in density of the hot column of gases.
Figure 4. 1. 3 shows a plot of ( h ) versus the height above the
d
burner surface for Flames 1,2, • 3,4, 5, and 6 together with the
values of max.
In all the flames studied there is no indication of a 'dark space1
as has been observed by several workers. The oxidation or blue zone
of oxy-acetylene flames is a blue-green colour although it is
referred to in the text as being blue. The oxidation zone of an
oxy-ethylene flameis, in fact, blue.
The flames listed in Table 4. 1 have been chosen so that direct
comparisons may be made between similar oxy-acetylene and oxy-ethylene
flames which have been run both at different fuel/oxygen ratios and also
at different mixture strengths.
The mixture strength, A , is defined as:
\ _ stoichiometric fuel/oxygen ratio
actual fuel/ oxygen ratio
IV - 4 - 2
Temperature Measurements
Temperatures were measured by fine wire, platinum: platinum 13%
rhodium thermocouples and the results were corrected to zero hot junc-
tion diameter. [(Waggener(235)] Measurements were made at discrete
positions in the oxy-acetylene and oxy-ethylene flat flames and the re
sults are shown in Figures 4.2.1 and 4.2.2.
It is apparent .that :-
i) the maximum temperature occurs just beyond the blue oxidation K
ii) the addition of argon reduces the temperature
56
-------
V 3
Figure 4.1-1 ("5 ) vs. height above burner surface
dnax (cn)
Flarae 1
Flame 5
Plane 4
Flame 6
•: 10 12
height above burner surface - cm.
20
-------
~T i.-": ••--- •
.-: i •.•r-.-.:- •
:-:!-;..;- I..-:... L.:,:. i:j::
...
~...i:.__.
400
300
2 CO
100
'•
Figure 4«2.1 Corrected Temperatures vs
•Height above Burner
2.0 4.0
6.0
i
• - '• :• : i : I
i i 1 i
Surface (acetylene flames)
8.0
10.0
12.0
.14.0. ..!._
Height Atove the Burner Surface - cm._i
.0
20.0
-------
above Burner surface
Height Above the Burner Surface
-------
iii) increasing the fuel/oxygen ratio decreases the temperature.
IV - 4-3 Batch and Continuous Sampling Results & Discussion
Since quantitative determinations using the MS9 mass spectrometer
were achieved only by use of an argon 'standard1 in the flame. The
mole fractions of the species listed in Table 4.2 have been determined
for Flames 1,2,4, and 5. Table 4.2 lists the nominal mass number,
measured mass, and assignment for all species whose concentrations in
the flame allow a peak intensity greater than that of the background
signal to be recorded.
Figures 4. 3. 3 et seq show the mole fractions of various species in the
flame as a function of sampling height above the burner.
Continuous Sampling Results
Identification of peaks in the gas chromatograms of the 'cold trap1
products is listed in Table 4. 3. The symbol (,? ) has been used to
indicate that a peak has not been positively identified.
Preliminary identification of compounds present in the "cold trap1
samples wa. by mass spectrometry at both 70 eV and««10 e V.
On the mass spectra which have been recorded at an ionising
220)
electron voltage of/wlO (used so that there is negligible fragmentation
ions at the following mass numbers have significant intensities.
m/c = 78, 92, 102, 104, 106, 116, 118, 128, 130, 136, 142, 146,
152, 154, and 166.
60
-------
Table 4»2
Kasa r.'-aouroinent and Anil -nnent of opfcloa Prespnt In the Batch Samples
Bominal Mass
2
16
16
26
28 I
26 II
28 III
32
40 I
40 II
42
44
50
52
74
78
(HB i)
(83 1)
(Ml)
(S3 1)
(HB1.2)
(;i91,2,5)
(HB1.2)
(as i)
(K31.3)
(-a 3)
(HB 1)
(NB 4)
Measured Mao* Actual Mass Assignment
(oV) (reference 242)
not measured
N M
n n
H H
N M
H H
II II
H II
tt N
40.031 (70) 40.031
42.045 (70) 42.047
not measured
50.015 (70) !>0.016
52.032 (70) 52.031
74.013 (70) 74.014
78.046 (70) 70.047
Hydrogen
Methane
Water
Acetylene
Carbon Monoxide
Nitrogen
Ethylene
Oxygen
Argon
CJI * Methyl -icetylone/
propadiene
C.H,* Propylene
Carbon Dioxide
C.H2 Di acetylene
C.H Vinylacetylene
CxH* Tri acetylene
CgiL. Benzene
NB 1 Peaks at these rc/e values have not been 'mass measured' since
there is no ambiguity in their identification.
NB 2 The three ppnks at m/e ° 28 nre usually separated on the
photographic trace.
liB 3 The two peaks at m/e » 40 arc ucually separated on the
photographic trace.
HB 4 The peak at m/e = 70 has been assigned to benzene rather than
aliphatic moleculen since benzene has beon found to be present
in the continuous collection cample?.
KB 5 Nitrogen is preoont in the raaos spectrometer us part of the
•residual air1 (see Section 3.1.^.2.2.2).
61
-------
Table 4-3
Identification of G.L.C» i-caka In the 'Cold' Trap Chroaatograma
C»L.C« Peak Compound Identified
1 unidentified solvent impurity
2 benzene
3 tetra-acetylene (?)
4 toluene
5 phenylacotylene
6 etyrene
7 methyl styrones (?)
8 trimethylbenzenea (?)
9 indene
10 unidentified
11 dihydronaphtlmleneB (?)
12 naphthalene
13 1-methyl naphthalene
14 biphenyl
15 aeenaphthyleno
16 fluorene
(?) denotes that this compound has not been positively
identified.
62
-------
1
1
s
71
I
B
I
L_
,
2
0
I
.
25v
26
X
JO
28
*"
1
,
35
1
40
i
N
V
>
III
1 |
Jill
74 78
1 iiiii.iiiiiiriilniii .,,
50
49
1...
ii i i
Top Trace
(most sensitive)
Middle Trace
Bottom Trace
.(least sensitive)
Figure 4.3.1. Part of a mass spectrum which results from the introduction of a typical batch sample
into the source of the GEC-AEI MS 9 mass spectrometer
N.B. The spectrum has been recorded (at 70eV) at an increased speed (for reproduction
purposes) so that the resolution of the three peaks at m/e = 28 is not observable.
-------
d>
3
o
O
o
•H
o
n
1 '2
V,
01
01
12
14
15
00
•s
16
VJ
X/V.
Figure 4.3.II
Chroraatogram of 'Cold1 Trap
Sample Collected from Flame 1
at a sampling height of 18.7 cm.
(25 pi injection)
The Identification of numbered
peaks is given in Chapter 4.
10 20
.Retention Time (mins.)
40
-------
°fl
17
18
19
20
31
33
34 35
30
Retention Time (mins.)
40
-8-
50
Figure ^3.111.
Chronatograrc of Extraction
Filter Sanple Collected
from Plane 2 at a
sampling height of
16.5 cm.
(25 pi. injection)
N.B. No 3-rethyl
phenanthrene has been
added to the sample
The Identification of
numbered peaks is giv;r.
in Chapter 4.
-------
'17
Ul 0)
CO1
18
19
VAJ
20
23
21
22
o
3-methyl phenanthrene
' (internal standard)
CO
30 40
Retension Time (mins.)
50
Figure 43.IV.
Chromatograra of Extraction
Filter Sample Collected
f roir. Plane 2 at a sen-.pl Lr_g
height of 16.5 c^i.
(25 ul. injection)
N.B. 3-methyl
phenanthrene has been sdicd
to the sample (cf.Figure ';.}
The Identification of
numbered peaks is given
in Chapter 4.
37
-------
Collection of Gat, Chro.r.atographic Fractions
For a positive identification of peaks in the gas chromatograms, fractions
have been collected as they eluted from the chromatographic column for
subsequent identification either by mass spectrometry or by ultra-violet
absorption spectroscopy.
Concentration Profiles of 'Cold Trap' Products
Concentration profiles of several species are shown in Figures
4.4. 1. 3. 4 et seq. The broken lines represent the concentrations of each
species which would result if there were no geometrical effects due to the
contraction.-, of gases in the flame.
( Cf. Section IV - 4-1).
Identification of Gas Chromatographic Peaks in the Extraction Filter
Samples
Preliminary identification of the species in the extraction filter samples
has been carried out by introducing such samples in solution form
into the source of the MS9 mass spectrometer via the direct insertion
probe. On the mass spectra which have been recorded at 10 ev ions > at the '
following m/e values have significant intensities:
m/e = 116, 1Z8, 130, 136, 142, 146, 152, 154, 166, 168, 176, 178, 190,
202, 216 and 226.
Each of the above ions has been 'mass measured1 by comparision
of its mass with that of an ion from the standard (heptacosafluoro - tri - n -
butylamine) which has been introduced via the cold inlet system. The results
are presented in Table 4. 6.
Identification of gas chromatographic fractions Irom the extraction
filter samples is summarized in Table 4.5.
67
-------
Tabl* 4*4
'Kane i-'easureudnt' of Mass Spectrum Peaks resulting from the
'Cold' i'rap Samples
Nominal Haaa
73
92
102
104
106
116
118
120
128
130
136
142
146
152
154
166
Measured Haas
78.046
92.060
102.044
104.058
106.074
116.063
118.074
120.009
128.060
130.070
136.126
142.082
146.108
152.063
154-C-79
166.075
Ionising
(eV)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
Actual Mass
(ref. 242)
78.047
92.063
102.047
104.062
106.078
116.063
118.078
120.093
128.062
130.078
136.125
142.078
146.110
152.063
154.078
166.078
Aealgncient
C10H10
C11H10
C12H10
68
-------
Tablo 4.!?
Identification of C.L«C» Peaks in the Detraction Filter Chromatograms
C.L.C* Peak Cottpound Identified
17 unidentified solvent impurity
18 indene
19 naphthalene
20 1-DBthyl naphthalene
21 acenaphthylone
22 fluorene
23 1.8t4«5-bi-(etheno-)naphthalene (?)
24 anthracene •*• phenanthrono
2$ unidentified
26 unidentified
27 4«5-ttothylone phenanthrone
'28 unidentified
29 fluoranthone
pO a fluorantheno isomer
31 pyrene
32 benzofluorcnos
33 nothy 1 pyronns
34 bonzo(ono)fluoranthene
35 ehryseno t bor/s(a)anthracene
36 a pyreno doriv.itive (?)
37 uni(]pntif5nd
(?) denotes that this compound has not been positively identified
69
-------
Table 4*6
'Mass Measurement * of t-'.ass Spectrum Peaks Recultin^ fron the Kxtraotion
Filter wamulea
Noainal Mass
116
128
130
136
142
146
152
154
166
168
176
178
190
202
216
226
Ko&Burod Haas
116.063
128.060
130.078
136.126
142.052
146.108
152.063
154.079
166.075
163.055
176.059
178.077
190.078
202.079
216.096
226.074
ionising
(eV)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(70)
(13)
(13)
(13)
(13)
(13)
Actual Haas
(ref. 242)
116.063
128.062
130.078
136.125
142.078
146.110
152.063
154.073
166.078
168.057
176.063
178.078
190.078
202.078
216.093
226.078
ABBignment
CA*
^10^8
C10H10*
C10H16*
C1lV
C11H14*
Ciaaa*
C12H10*
Vio*
c12ueo*
°14H84
C14H10
C H *
C15H10
C16H10+
C17H12*
C1flH, *
70
-------
Figure 4»3«3 (a)-(b) Acetylene Concentration
Profiles
0.8
•p
s
0.6
-------
Figure 4.3.3 (c)-(d) Acetylene Concentration
Profiles
12 ..:. ... 16
Sampling Height Above Burner - cm.
0.4
i-'g
•S
o
0.2 •
0.1 -
(d)
j Flame 5
CH/0 =1.1 (+Ar)
"!::::.I:::.;•.* *+ ^
; 4 6 1? 16 20
Sampling Height Above Burner - cm.
72
-------
Fieure 4.%4 Ethylene Concentration Profiles
(a)-(b)
4>
O
)
0.8
0.6
(a)
g
« 0.4
' ;• ' . . _li. !*>
-;. •.:••.:•...: ; -c
... ..
-..:_...._^.j;_j_.-._-j 0.2
Flame 2
o
o
e—oo-o
12
16
20
Sampling Height Above Burner - cm.
0.8
... . ..... -...,._., -r .
: '. -j : :.:;j:- :. _. 0
• :.. . ~ ;"::'i" :i'-~ '""^|
-P
.-: o
-.—-.; n /:
H U.O
•;•.-• fc
;Flame 5
--!-C H /00 = 1.1 (+Ar)
4 e 1? 16
Sampling Height Above Bi:rner - cm.
73
20
-------
Figure 4.%5 (a)-(b) (Water + Carbon Monoxide)
Concentration Profiles
I
oj 0.8
-------
Figurr 4.3.!' (c)-(d) (Water + Carbon Monoxide)
Concentration Profiles
o
•^
-p
o
0
^
fe
(c)
o> 0.8
rH
O
K
Flame 4
B2/02 =1.1 (+Ar)
•H
g 0.6 •
. : £
£
d
o
• - :• °
H A A .
."-; \~r 5 °*4
+
" ' "--:-/. •-:-;:- . .. .-:. -•- ^
•;— .•:.-— • . .:•-.-:••— i H 4»
> 0 2.
:: ; : . *r;- ~ •• ». o«* •
: • ; •••••••• •;;...• • ;
. ' .1. r
,.;;;•::!:.;: i .• T i ': ::[";.;• • • :- :^ -•
!•! "" .1 . .; - .. .. :•"• •• :~\~ i •-..- :: .i '. :..
I t l ' ',--•-
" ."77^ 7"*r"!™7"-~"l~^r' ~'~~ ] • ~,"~T™- !T'"~"' '"T"!"
.-,•• : :[:•;: . • -|:.. :• -;;; •;..-•. : :: ;. v: •::: •.:::.
: ••:.-.:! '•':' : \ -.:•'••": :f:-;i;'" : • • --- -
•;!•'.; i: ":" -\': a
_„ ,' :: ;:^... :{_.•.. :'.... ..'.., ,.T. o
: •.:•.::!.••:., . ,v , :;..) ;:;.! .- ( :. ^| ,
: '" 1
::•:•: .. ;:..— •-,•- M
.• I :.:.. •:::: . . fr,
••••-. Aft-
-.'...•• fl) VtO
rH
:-.-.:• -7-...:-. ; rv o
• . .
..;. •• •;•'. . . .: . ;: •;- :f ;; c>
^
.^j . x
. •' ; ':> ' :'.!.!•:'•. '.-('.: '•
o-r-rrrrr^ — °-^ — n- -, fl>
p.T0i--TSr*^O ^ o -*fc— o
o^£^
i
t ^ :"::: !
• • ! ': ' i I.1
i , . . . .
• . ' '...',.' . •
"• ' i .•..:••(.;;••• ' :'.
..:•'..: '•'. '.. :.\ •
; . '. "'.'.'',.'-'.'..['
)
4 8 12 16 -2
Sampling Height Above Burner - cm.
:-' :. • ' .. . : : ••;•;••
.: ..' :..-..: .•;-.•.;
(d)
r»ia_d c
C H /O =1.1 (+Ar)
• ' : -~ ", ' " "I . •" • :"-' • • '
.. : ; :;
. .: ' • . .i ._: j . . •;:'.. .L._.
;
:
" .
0
• ;
1 i
•
•
•a 0.6-
-------
Fifiure 4.3.6 (a)-(b) Oxygen Concentration Profiles
C H2/02 = 1.3 (+Ar)
8 ...J. 12 .16 20
ing Height Above Burner - cm.
4 8 12 16
Sampling Height Above Burner - cm.
-------
Figure 4.3.6 (c)-(d) Oxygen Concentration Profiles
.. . Flame 4
.--:-:-. : C H /O =1,1 (+Ar)
, ... ...... . tt <-
Sampling Height Above Burner - cm.
-------
Kethylacetylene Concentration
Profiler,
;_ 4 --;;; .......8.......;..„-, 12 16
Sampling Height Above Burner - cm.
20
Sampling Height Above Burner - cm.
78
-------
Figure 4.3^7 (c)-(d) Methylacetylene Concentration
Profile
„_
: - 4: 8 : 12 16
20
;j Sampling Height Above Burner - cm.
1
&4
0.004
0.003
"£ 0.002 •
0.001
4 8 12 16
Sampling Height Above Burner - cm.
79
20
-------
•; Figure 4.J.O (a)-(b) Propylene Concentration Profiles
0.004 '
d
o
o
B
0.003
0}
c
H 0.002
^>
••: 0.001
(a)
Flame 1
C2H2/02 = 1
4 .... 8 --.. 12 16
_ Sampling Height Above Burner - cm.
20
0.004
§
J
0.003
(b)
Flame 2
/0 n 1.3 (+Ar)
s
0.002 •
0.001
o
O
. . _J. ...L-ii L . ,. j...
4 8 12 16 20
: Sampling Height Above Burner _ cm.
80
-------
Figure 4«5«6 (c)-(d) Propylene Concentration Profiles
: 4 ;- : 8 :. 12 16
20
: Sampling Height Above Burner - cm.
g 0.005
0
rH
§ 0.002
4)
I
0.001
(a)
Flame 5
C2R4/02
4 8 12 16
Sampling Height Above Burner - cm.
81
20
-------
Figure 4.3.9 (a)-(b) Carbon Dioxide Concentration
Profiles
4...-:.. . . -8 . _ 12 16
Sampling Height Above Burner - cm
4 8 12 16
Sampling Height Above Burner - cm.
82
-------
Figure 4.3.9 (c)-(d) C.irbon Dioxide Concentration
Profile's
Sampling Height Above Burner - cm.
4 6 12 16
Sampling Height Above Burner - cm.
Plarae 5
'"• • W°2-
20
-------
Figure 4.3.10 (a)-(b) Diacctylene Concentration
Profiles
Sampling Height Above Burner - cm
16
20
Flame 2
C2H /02 =1.3 (+Ar)
4 8 12 16 20
' Sampling Height Above Burner - cm.
84
-------
Figure 4.5.10 (c)-(d) Diacctyl-no Concentration
Profiles
4 9 1?
Sampling Height Above Burner - cm.
" " 85
•>r\
-------
Figure 4.3.11 (a)-(b) Vinylacetylone Concentration
Profiles
:'."::'. i::::. r
T-r .r:-.g-
-:i 4 •;•.:-;. • 8 : 7 :i-12 -• •;:• 16
Sampling Height Above Burner - cm.
o
2
3 0.003 •
•g 0.002 -
•'•::• S
0.001 •
(b)
Flame 2
C2H4/02 =
4 8 1? 16
Sampling Height Above Burner - cm.
86
20
-------
Figure 4.J.11 (c)(d) Vinylacrtylone Concentration
Profiles
Sampling Height Above Burner - cm.
A 8 19
Sampling Height Above Burner - cm,.
i*7
-------
Figure 4.3-12 (a)-(b) Tricxcetylone Concentration
Profilej
a
o
•H
. tl
••g
•s
g
5
0)
§
0.003 -
.002
.001 •
4J
§
a)
':
O. - -
1
l\
\o
\
\
\
D
1
- . -
•° "
;
T?1 om
-------
Ficure 4.3-12 (c)-(d) TrLv-lyl-nc Conccntrntln
Prorilns
: •' C
j ••••; o
$ 0.0016
B[::±;E:J
4)
:.! • ..: ...:.'.... .:.'-.; . r-«
"a, 0.0012
"fV-i-W:- g
•J?
; - ••:—:• «)
•-T-- -S 0.0006
•-•• .-• • ! M
_::..:._i'..•: •-'—_:__;;_ii—i_ 0.0004
(c)
o\
el od
Flame
C2H2/02 =1.1 (+Ar)
O O
o o - e
G>
o
i ''..:. I'
s
0.004 •
(d)
Flame 5
C2H4/02 =
.:• 4 8 12 16 20
; .; : Sampling Height Above Burner - cm.
O.OOJ
•S
4)
O
a
-H
0.002
0.001
4 8 12 16
Sampling Hcirht Above Bnrnrr - era.
20
ii'j
-------
Figure 4.5.13 (a)-(b) Benzene Concentration Profiles
C H2/0 = 1.3 (+Ar)
: L ... & 0.0006
g 0.0004
S
&..
- 0.0002
8 . 12 16
Sampling Height Above Burner - cm.
C2H4/02 "
4 8 12 16
Sampling Height Above Burner - cm.
-------
Figure 4.5.15 (c)-(d) Benzene Concentration Profiles
(c)
0.0008-
Flame 4
H/0 = 1.1 (+Ar)
4 8 12 16
it..'1 : . • • : .
Sampling Height Above Burner - cm.
20
--n
0.0008
a
o
g 0.0006
0)
r-l
I
g 0.0004'
8
0.000?
Flame 5
=1.1 (+Ar)
4 3 12 16
Sampling Height Above Burner - cm.
91
-20
-------
Figure 4.4.1.3.1 (a)-(b) Benzene Concentration Profiles
Sampling Height Above Burner - cm.
.. Flame 2
0.004
4 8 12 16
Sampling Height Above Burner -'cm.
-------
Figure 4.4.1.J.1 (c)-(d) Benzene Concentration Profiles
0.0004
Sampling Height Above Burner - cm.
4 8 12 16
Sampling Height Above Burner - cm.
93
-------
Figure 4.4.1.3.2 (a)-(b) Tetra-acetylcne(?) Concentration
Profiles
c
o
••g 0.0016
£
(a)
Flame 1
*o/0o =1-3
0.0012
0
0)
C
u
0.0008
.0
O ©
0.0004
O
o
o
.: 4 ...•... 8 12 16
Sampling Height Above Burner - cm.
20
Flame 2
=1.3 (+Ar)
20
Sampling Height Above Burner - cm.
94
-------
Figure 4.4.1.3.2 (c)-(d) Tetra-acetylene (?) Conccntrati'on
Profiles i':
a
o
0.004
flj
Flame 3
>„ = 1.3
a>
r-t
O
0.003
£ 0.002
;_jii-i f i.
rt
Q)
0
O
o
©o
o
•
e>
12
16
20
'Sampling Height Above Burner
c
o
•g 0.0008
n)
£
0.0006
(d)
Flame 6
C2E2/02 =1.1
o>
c
0)
o
CO
0
4 8 12 16
Sampling Height Above Burner - cm.
95
?0
-------
Figure 4.4.1.3.5 (a)-(b) Toluene Concentration Profiles
Flame 1
C2V°2 = 1.3 (-i-Ar)
0.0002
4.... _•: j8 : 12 ..... 16 20
-.-Sampling Height Above Burner - cm
._:.„-..—_. . 0.00016
8 0.00008
4 . 8 12 16
Sampling Height Above Burner - cm
-------
Figure 4.4.1.3.5 (c)-(d) Toluene Concentration Profiles
0.0004
T —
c
o
t)
r-1
O
S
0.0003
0.0002
0.0001
(c)
Flame 3
1t3
O
^-s 4 8 .12 16 20
-___-._ij-Sampling Height Above Burner - cm.
.::.'.:!.-! : , ': - • •
~~-~~~"~ ~.7}:.. " ' ••••'•- ;• -• ' ••• :':: :,- .'. . L.'-;.1. i' :
I ._.. '.j;__...^;J.;i!.-_ .•":: ^i_!l-i- •-•--.-• :.... -'—'.:-•'-'••' •--"-'••
„.__!_.... 0.00008
|
4>
s
•£ 0.00006
4)
So. 00004
0.00002
4 8 12 16
Sampling Height Above Burner r cm.
97
-------
Figure 4.4.1.3.4 (a)-(b) Phenylacetylene Concentration
Profiles
C2H2/02 = 1'
_._.::_——:..»*< 0.0001
.4 8 12 16
•"T" Sampling Height Atove Burner - cm,
0.0008
« 0.0006
0.0004
: i
!
i i
0.0002
.(_„.._.' ______ - _-_! - ; - — J ..... Sampling Height Above Burner - cm.
98
-------
Figure 4.4.1.5.4 (c)-(d) Phenylacetylene Concentration
Profiles
—•-[—T~-—trr " Sampling Height Above Burner - cm.
B 0.00016
.;_...;_..._j.__ .__: •-. o .....
-•'• -•.- - S 0.00004
4 8 12 16
Sampling Height Above Burner - cm.
99
-------
Figure 4.4.1.J.5 Styrene Concentration Profiles
(a)-(b) '
CLH0/00 = 1.3 (+Ar)
...: Sampling Height Above Burner - cm
T—: . [-•.••• *
, i 4 8 12
Sampling Height Above Burner - cm.
''—•-• -Lioo; ;
20
-------
Figure 4.4.1.3.5 (c)-(d) Styrene. Concentration Profiles
c2H2/o2 = 1.3
„:._•..... 0.00004
4 8 .12 16 20
"Sampling Height Above Burner - cm.
0.00004
0.00003
o>
rH
O
Q)
S 0.00002
CO
0.00001
Flame 6
=1.1
12 16
Sampling Height Above Burner - cm.
101
20
-------
Figure 4.4.1.5.6 (a)-(b) Methyl Styrcncs (?)
Concentration Profiles
Flame 1
= 1.3 (+Ar)
4 8 12 16
Sampling Height Above Burner - cm.
g
•rt
o 0.00008
n
£
Flame 2
=1.3 (+Ar)
i-l
O
K
0.00006
ro
-------
Figure 4.4.1.3.6 (c)-(d) Methyl Styrenea (?)
Concentration Profiles
4 8 .12 16
-Sampling Height Above Burner - cm.
-p 0.00012
4 8 12 16
Sampling Height Above Burner --cm.
103
20
-------
Figure 4.4.1.3.? (a)-(b) Trimethylbenzenes (?)
Concentration ^Profiles
c
o
•H
g 0.000016
Q)
,0.000012
n
8
V
8
0.000008
•p
Q)
B
E-l 0.000004
Flame 1
=1.3 (+Ar)
.. _. 4 8 12 16 20
"Sampling Height Above Burner - cm.
Flame 2
=1.3 (+Ar)
8 12 16
Sampling Height Above Burner - en.
104
-------
Fit,.re 4.4.1.3.7 (c)-(d) Trimothylbonzenea (?)
Concentration Profiles
D2V°2 = 1'
Sampling Height Above Burner - cm.
o
•H
{* 0.000016
PM
O
a
• 0.000012
w
fl)
<1)
C
,0 0.000008
£
o
G
•r)
M
0.000004
4 8 12 16
Sampling Height Above Burner - cm.
20
105
-------
Figure 4.4.1.3.8 Indene Concentration Profiles
(a)-(b)
4 ......: 8 .... 12 16 20
Sampling Height Above Burner - cm.
4 8 12 16
Sampling Height Above Burner - cm.
106
-------
Figure 4.4.1.3.8 (c)-(d) Ir.dcne Concentration Profiles
0.00020
-p
«> 0.00015
(ki
T~ g 0.00010
0
• "S
._:-.;„.:...2.0.00005
4 -'8- .-.-.': 12 16
- Sampling Height Above Burner - cm.
20
4 8 12 16 20
: Sampling Height Above Burner - cm.
107
-------
Figure 4.4.1.3.9 (a)-(b) Dihydronaphthalenes (?)
Concentration Profiles
o
•H
0.00004
0)
rH
O
CO
0)
C
0)
l-l
rt
x
•£
C
o
0.00003
0.00002
0.00001
(a)
Flame 1
>V°2 =
.-._... 4 -8. :-::; 12 16
-r; Sampling Height Above Burner - cm.
20
0)
0.00004
w 0.00003
a
g
p, 0.00002
id
a
0.00001
(b)
Flame 2
c2H4/o2 = 1.3 (+AP)
0-
4 8 12 16
Sampling Height Above Burner - cm.
20
108
-------
Fogure 4.4.1.3.9 (c)-(d) Dihydronaphthalenes(?)
Concentration Profiles
C2V°2 = 1*
o 0.00004
4 •-• 6 : 12 1.6 20
Sampling Height Above Burner - cm.
P, 0.000004
4 8 . 12 16
Sampling Height Above Burner - cm.
109
-------
Figure 4.4,1.3.10 (a)-(b) Naphthalene Concentration
Profiles
Flame 1
=1.3 (+Ar)
o
. 4 8 12 16
--Sampling Height Above Burner - cm.
4 ., .-. ,
- Sampling Height Above Burner
110
-------
Figure 4.4.1.3.10 (c)-(d) Naphthalene Concentration
Profiles
g 0.00016
•H
1
(c)
Flame 3
C2H2/02 = 1.3 .
o
K
. s
rt
*>
I
:- «S
! . .
0.00012
0.00008
-0.00004
4 ...i. 8 ...:. 12 16
Sampling Height Above Burner - cm.
20
4 8 . 12 16
. Sampling Height Above Burner - cm.
Ill
20
-------
Figure 4.4.1.3.11 (a)-(b) 1-Methyl Naphthalene
Concentration Profiles
::;- 4 8 12 16
Sampling Height Above Burner - cm
•4 .8 12 , 16 20
Sampling Height Above Burner - cm.
112
-------
Figure 4.4.1.J.11 (c)-(d) 1-Methyl Naphthalene
Concentration Profiles
4 •:. 8 _ .12 1
-Sampling Height Above Burner
I."
5 0.000016
4»
0)
o 0.000012
0)
c
Flame 6
C2H2/02 = 1*1""
•3
I
0.000008
"S 0.000004
481?
! Sampling Height Above Burner
i •
113
16
• cm.
20
-------
Figure 4.4.1.5.12 (a)-(b) Biphenyl Concentration Profiles
0.000016
(a)
.!:.' Flame 1
C2H2/02 = 1,
4 ; 8 12 16 20
'Sampling Height Above Burner - cm.
0.000016
fi 0.000012
d>
t-i
o
i;
o
-a
•H
n
0.000004
(b)
Flame 2
•j C2H4/°2 = 1'
4 8 12 16 20
Sampling Height Above Burner - cm.
114
-------
Figure 4.4.1.3.12 (c)-(d) Biphcnyl Concentration Profiles
C2H2/02 = 1'
••;• ••' a 0.000016
I O
•H
: +>
a, 0.000012
~; 0.000008
- — 0.000004
( •aaeasi'e o*e> &e>
-^Sampling Height Above Burner - cm.
-v 0.000016
8
..-;.—: .;:-:-. : 0.000004
4 .8 : 12 16
Sampling Height Above Burner - cm.
-------
Figure 4.4.1.3.15 (a)-(b) Acenaphthylene Concentration
Profiles
d
0.00016
Flame 1 :.
T;- ;C2H2/02 = 1.3 (+Ar)
...... ...... ,g 0.00012
... v
•S 0.00008
0)
- o
0.00004
....I.....:- ! ....
:Sampling Height Above Burner - cm.
•t
-~-'TJ- g 0.00016
•H
•p
!. *>
.,i... ".' M.
0)
0.00012
s
!
-------
Figure 4.4.1.3.13 (c)-(d) Acenaphthylcno Concentration
• Profiles
4 ' 8 ; .:,- •: 12 16
Sampling Height Above Burner - cm.
4 8 12 16 20
Sampling Height Above Burner - era.
117
-------
Figure 4.4.1.3.14 (a)-(b) Fluorene Concentration Profiles
rassrrra
•— -- a, 0.000012
o 0.000008
fa
0.000004
CO
o
&
4 8 12 16 20
''-•p Sampling Height Above Burner - cm.
118
-------
r :•::.-;
{-..:.• -|
fri'Tr."."'.
[•::•; ;
Figure 4.4.1.5.14 (c)-(d) Pluorene Concentration Profiles
0.000016
•P
s
- S 0.000012
0)
go.
L-i, .L...;. : ..^.. -.-.:—^ .-0.000004
4 ..:.•.. 8 - 12 16 ; 20
Sampling Height Above Burner - cm.
-:,..J_._v:..l ;.•:::;
Flame 6
C2H2/°2=
1.1
Sampling Height Above Burner - cm.
i
119
-------
Although other higher molecular weight poah (e.g. the benzop^renca)
have been found In Boot, insufficient quantities of these compounds are
sampled in this study to allow detection by gas chromatography.
4.4*2.3 Concentration Profiles of fjctraction Filter Products
The concentration profiles of the larger peaks in the extraction
filter products are presented in Figures 4.4.2.J.I to 4*4*2.% 12. Since
G.L.C. Peaks 30, 54. 35, 36 and 37 are often too email to be measured with
any reasonable degree of accuracy* the concentration profiles of the
species which produce these peaks are, therefore, net presented.
The concentration profiles of the following peaks are presented!
Indene
naphthalene
1-Wethyl Naphthalene
Aoenaphthylene
Fluorene
1.8,4*5-bi(etheno-)naphthalene(?)
(n/e » 176)
Anthracene 4 Phenanthrene
4.5-nethylene phenanthrene
Fluoranthene
Pyrene
Benzofluorenes
Methyl Pyrenes
in Figures 4«4.2.3.l(a)-(o)
in Figures 4*4.2*3.2(a)-(c)
in Figures 4«4.2.3.3(a)-(c)
in Figures 4.4.2.3*4(a)-(c)
.In Figures 4.4*2*5.5(a)-(c)
in Figures 4.4*2.}.6(a)-(o)
in Figures 4*4*2.3.7(a)-(c)
in Figures 4.4.2.3*8(a)-(o)
in Figures 4.4.2.5.9(a)-(o)
in Figures 4*4*2.3*10(a)-(o)
in Figures 4.4.2.3.1l(a)-(c)
in Figures 4.4.2*3«12
-------
involved in measuring euoh lov concentrations*
The oonoentration profiles of all species ore very similar for
Fionas 1 and 3 (the oxy-aoetylone floaes). For Indene, naphthalene,
1-aethyl naphthalene* fluoreno, 1.8,4.5-bi«(etheno-)naphthalene (?)»
fluoranthene and pyrene the results show that the addition of argon makes
little difference to the concentration although for species such as
aconaphthylene, anthracene + phenanthrone and the benzofluorene fraction,
the addition of arjon tends to increase the oonoentration. Cne might
expect, as in the case of 4»5-niethylene phenanthrene, that the addition of
argon to a flooto would tend to reduce concentrations.
For the lower molecular weight species collected in the extraction
filter (i.e., indene, naphthalene, 1-mothyl naphthalene, aoenaphthylene
and anthracene + phenanthrene) the concentrations of speoieo in the yellow
cone of Flames 1 and 3 (the acetylene flames) renain. roughly constant
although for the higher molecular weight species (i.e., 1.6,4.5-bi-(etheno-)
naphthalene, 4«5~nethylcne phenanthrene, fluoranthene, pyrene and the
benzofluorono fraction), there is some increase in concentration. There is
a slight increase in the concentration of fluorene in the yellow zones of
Flames 1 and J. This increase in concentration may be attributed to the
pyrolysis of lower molecular weight upocies in the flace although the
results of this study do not indicate clearly what these ppecies might be*
Ac may be seen in Figures 4.4.2.3.1 to 4.4.2.3.12 there ie an initial
rapid Increase in concentration of all species in the blue zone of the
oxy-acetylene flanos and a maximua concentration is reached before the end
of this zone* These species are then destroyed to a certain extent (but
not completely), possibly by OH-radicals or other oxygenated radicals.
(N.fl. tfcie is the region in tho flame whore tho oxygen concentration rapidly
falls to sero) The initial peak in concentration io also ehown in Flaiae 2
121
-------
(the oxy-ethylane flame) for several species including indene,
naphthalene, 1-methyl naphthalene, aoenaphthylene, fluorene, 1.8,4.^-bi-
(etheno-)naphthalene (?), 4.5-oethylene phenanthrena and fluoranthene.
Th« concentration profiles of anthracene •*• phenanthrene and pyrene
apparently show a fairly steady rise throughout this flane, although when
these profiles are corrected for the change in density of the flame gases,
a ffiaxlmum value ia observable beyond the blue region.
122
-------
Figure 4.4.2.3.1 (a)-(c) Indene Concentration Profiles
g 0.0010
•H
•P
O
Sampling Height Above Burner - cm.
. __ .
4 : 8 12 16
"/""'Sampling Height Above Burner - cm.
:". 0.0010
•H 0.0008
.......... , .......... -p.,.,:,-
4,8 12 16
Sampling Height Above Burner - cm.
123
-------
% Figure 4»4«2.3.2 (a)-(c) Naphthalene Concentration
Profiles
+>
g 0.0006
fe
V o» .
, -3
S 0.0004
':: +>
i •£
Flame 1
0.0002
;;..:.:,- 4-:,:-:ri 8 K : 12 M6
Sampling Height Above Burner - cm.
.20
4 8 12 16
Sampling Height above Burner - cm
20
124
-------
R
O
Figure 4.4.2.3.3 (a)-(c) 1-Methyl Naphthalene Concentration
Profiles
0.00012
4.8 12 '16
Sampling Height Above Burner - cm.
125
-------
Figure 4.4.2.3.4 (a)-(c) Acenaphthylene Concentration
Profiles
o 0.00003
•g 0.00002
S .
S
a>
•^ 0.00001
I.
6
lame 1
X.
0
O o
.4 8 12 16 20
. -.Sampling Height Above Burner - cm.
-.---- o 0.00003
..!..;...;„_:.., L.!... S
4 8 12 16 20
Sampling Height Above Burner - cm.
126
-------
§
; Figure 4.4.2.5.5 (a)-(c) Fluorene Concentration Profiles
..i
0.000012
°s
0)
0.000008
o 0.000004
:. 4 .: 8;.. .::.r. 12 _...16
H Sampling Height Above Burner - cm.
20
Vvl
'•I
~0.00006
i. a .
ir-::
r-'-Tr
o
:fi
-^ 0.
«>
o>
C
h 0.00002
3 .
PH.
:&
Sampling Height Above Burner - cm.
4 8 12 16
--Sampling Height Above Burner - cm.
127
20
-------
f:-:rf;
L. . - : . _„
IIS
i . ..i : • .: i!'
r
Figure 4.4.2.5.6 (a)-(c) 1.8,4.5-bi-(ethono-)naphthalene('
Concentration Profiles
0.000015
0.000010
§
0)
r-l
(tf
f,
•P
jc a
p. o
d -H
c +>
x-x o
I (4
O h
£^
J3 4)
•f> rH
«> O
YK
•H
|? 0.000005
•
•*
a?
.4 _; 8 .. 12 16
Sampling Height Above Burner - cm.
20
:! C-
"rs
'•' 0)
20
- Sampling Height Above Burner - cm.
0.000015
. . •:•*»
a a
S o
^^•rt
-P
§ S 0.000010
Q) M
-"So
2*
|t 0.000005
't i;;:
00 :
4.8 12 16-
Sampling Height Above Burner - cm.
128
20
-------
. 0)
. o
• ; fli
•:S:.I
*>
oj Ci
o
+ -H
o
£
Figure 4.4.2.3.7 (a)-(c) Phenanthrene + Anthracene
Concentration Profiles
0.00003
0.00002
0.00001
- —0.00012
§ ,
TFrr
. ... 4»-
C (4
b £
. . 4 . 8...._ , .:;. 12 , .. . 16
,- -Sampling Height Above Burner - cm.
20
0. oooos
- «
. ::;i :•••.[
T-i-rrra"
_r_ .. i
0.00004
_L_:.:4L....:::.. .. 8..-.- ... 12 '. . 16
--;-;-Sanipling Height Above Bxirner - cm.
20
0.00003
. .. . . .
: ; . •• :.:.! .-:::.-. O
- •' + £
88
I
„, 0.00002
0.00001
(o)
Flame 3 T~
4 8 . 12 16
Sampling Height Above Burner - cm.
129
20
-------
~! Figure 4.4.2.3.8 (a)-(c) 4.5-Methylene Phenanthrene
Concentration Profiles
0.(
4 8 . 12 16
Sampling Height Above Burner - cm.
130
-------
4.4.2.3.9 (a)-(c) Fluoranthene Concentration
Profiles
o 0.000002
-Sampling Height Above Burner - cm.
12 ... ._ 16
-----™ i~ 0.000003
—--•——-•—--£ 0.000002
4 8 12 16
Sampling Height Above Burner - era.
131
-------
Figure 4.4.2.3.10 (a)-(c) Pyreno Concentration Profiles
g 0.000012
-:- — J?o.
% 0.000008
o
K
8
0)
000004
4 8 :..., 12 .16.
Sampling Height Above Burner - cm.
20
'—- -0.00006
•""': •' 0.00004
-^---•--rr-^ 0.00002
— Sampling Height Above Burner - cm.
4 8 12 16
Sampling Height Above Burner - cm.
20
132
-------
Figure 4.4.2.3.11 (a)-(c) Benzofluorenes Concentration
Profiles
4 , : 8 -12 16
- Sampling Height Above Burner - cm.
133
-------
Figure 4.4.2.3.12 (a)-(c) Methyl Pyrenes Concentration
^0.000012 Pr°file3
CD
0 C
C O
. f*
fi
0.000008
,C 0)
-P i-l
£.§•0.000004
(a)
Flame 1
O
4 ...... . 8 . .,.:;... 12 .;. 16 20
—--Sampling Height Above Burner - cm.
J.:..L_
sg
0.000004
:£fc
* 8) :
+> •-)
• SS 0.000002
(b)
Flame 2
o;_.
;::;!:;;:::: o
:;; :f;rr::'.:
o
Q
(•-
t::_
T-;©
8
Height
12
Btirner
16...
- cm.
20
t:_.:
-- -- ------ 0.000006
n
0) C
c o
0) -H
^1 '
£*
•P rH
0) O
r
(o)
Flame J
G-
e°
O G)Q?>
O
O
O
4 8 12 16
Sampling Height Above Burner - cm.
20
134
-------
IV.4.3 Discussion of Results
Figure IV.4.3.1 gives an indication of the amount of each species
formed in a typical acetylene flame (Flame 1). The mole fractions of
species vhlch are found to be present in both the •cold' trap products and
the extraction filter samples (i.e. indene, naphtnalene, 1-methyl naphthalene
and fluoreoe) have been combined. The concentration profiles which are
presented in Figure IV.4.3.1 are uncorrected for the change in flame gas
density high above the burner (see Section 4.1). Nevertheless, it has been
shown (earlier in Chapter 4) that several hydrocarbon species(notably
methane, methylaoetylene/propadiene, vinylacetylenet benzene, toluene*
phenylaoetylene, styrene, indene, dihydronaphthalenes (?), naphthalene*
biphenyl, aoenaphthylene, iluorene, 1.6,4.5-bi-(othenoO-naphthalca«, 4.5-
methylene phenanthrene, fluoranthene and pyrene) .do show a genuine
secondary increase in formation high in the flame. A further olight
increase in the concentration (mole fraction) of pcah or othr aromatic
species in the yellow zone is not unexpected since this is the region
where
(1) the oiygen concentration (see Figures 4.5.6(a)-(d)) and the OH
radical conoetration (Bonne et el*') are zero.
(2) the acetylene concentration is still relatively high.
(mole fraction -0.1, see Figures 4.3»3(a)-(d)).
;.
(3) the temperatures are favourable to poah formation
(see Figures 4*2.1 and 4.2.2 and Reference 44).
«
Although the secondary Increase in concentration of a species uuch
as phenylacotylene mijht appear lar^e (/- O.OOC01-0.001 mole fractions)
135
-------
Figure IV.4.3.1 Composite Plot of Mole Fractions of All Species in
Flame 1 (uncorrected for flame geometry)
"(z = sampling height" aboveburner;_surfa.ceiLa'iCtnr) ',' ".
i- CO)
=E€===r==3i~::~ ^SsSSiE^^EE^S Diacetylene
~Propyl
>•- Naphthalene
lethyl
hthalene
ethylene
Fluorene KiiFhenfinthrer.e
TFluoranthene
16
?0
136
-------
euch an increase would have a negligible effect on the actual acetylene
concentration profile ulnco the mole fraction of this letter species is
relatively so much higher (fO.1). It nay be postulated, therefore, that
althcujh the acetylene concentration profiles generally do not show an
observable downward trond in the yellow zones of the flames considered,
acetylene pyrolyais probably accounts for the secondary increase in the
formation of aromatic species ana pcah. (it is worth noting at this
eta£0 that the apparent secondary increase in flux of chloroform-soluble
material and pcah (see Figures 1.7 and 1»9) in rich prcoixed oxy-acctylene
o
flames as reported by Tocpkins and long Bay well be due mainly to the
contraction of flame gasee upon cooling^ In general, the concentrations
of species found in equivalent oxy-ethylenc flames are similar in value to
those of species found in oxy-acetylene flcceaf however the concentration
profiles are somewhat different owin^ to the different geometry of the
oxy-ethylene i'looes.
A comparison between the flux fractions of species presented by
2
Tompkina and Long and the mole fractions of species presented in this
study can be made since at the points of maximum and minimum concentrations
G, a f (see Appendix II)
where G. is the flux fraction of si.-eoles i
and f. is the weight fraction of species i
(the weight fraction cf species i is derived froa tLe cole
fraction of species i).
Although the flames studied are not identical (either in flovrate or
in operating pressure) a comparison hes been made- between /"laze 5 of this
137
-------
study and a hypothetical flame (between 1 and 2 In loapkins and Long18
•tudy) which has been corrected to a total flow of premixed gases of
6.04 litres/Bin, (from 9.53 lltrea/min.).
2
From the results of Tonpkins and Long the naxinum flux of the
following typical species have been calculatedi
Species Actual Measured flux Corrected Flux
(Total Flow 9.53 l./min. (Total Flow 6.64 l./min.
t* Mi1?) W KIP)
2 2
jag/cm hr. jig/cm hr.
aoenaphthylene 120 66
anthracene 100 ' 72
+ phenanthrene
pyrene 130 94
The total naes flux of species through this hypothetical flame
6 2
(assuming a uniform cross-sectional area) is 4 x 10 ug/ca hr.
Therefore the maximum flux fractions of acenaphthylene* anthracene +
phsnantLrene and pyrone are 22 x 10~ , 16 x 10" and 24 x 10" ,
respectively.
In this otudy the maximua oole fractions of aoenaphthylene v anthracene
phenanthrene and pyrene are 34 x 10** , 6 x 10~ and 10 x 10~ t
respectively. Assuming an Lverage molecular weight of species in the
flaae to be 25 (reasonable since the flame gases are conposed mainly of the
permanent gases) » then the luaiiuuu weight fractions of acenaphthylene t
anthracene •»• phenanthrene and pyrene are approximately 204 x 10 ,
42 x 10" and 60 x 10~ t respectively. A comparison between the naxlmuB
flux and wcijit fractions is set out belowi
138
-------
«?n.«i.« Flux Fraction Max0 Weight Fraction
species (Tonpklna and Long2) (thls Btudy)
acenaphthylone 22 x 10 204 x 10
anthracene 18 1Q-6 2 1Q-6
+ phenantnrene
pyrene 24 x 10~6 60 x 10"6
Although the maxiaum flux and weight fractions are comparable for
anthracene + pcenanthrene and pyrcna, considering the errors in measuring
•uon low quantities! the relatively high value for the maximum weight
fraction of acenaphthylene indicates the necessity to collect the vapours
as veil aa the coot from flames| nuch of the acenaphthylene would be lost
in th« collection method employed by Tompkins and Long. One interesting
significance of the similarity between the maxlaur* flux and weight fractions
of anthracene + phenanthrene and pyrene is that the former was obtained
from the weight of species extracted from soot and the latter was obtained
froa the weight of vapours condensed from a rich preaixed flat oxy-
acetylene flame. This su&jests that pcah species nay be adsorbed onto the
surface of soot or 'carbon1 particles upon sampling by relatively
2
inefficient devices f this is in fact supported by the results of
Hofflann et al since these workers have found that when 'carbon1
(collected from rich premixed flames) is heated in a vacuum the residue is
found to be practically pure carbon, indicating tftat all the hydrogen is
bonded to relatively small hydrocarbon compounds.
Although only relatively stable species are analysed from thn samples
withdrawn froa flanea using the techniques descrioed in Chapter J,
undoubtedly the blue oxidation zone represents a recion where many free-
radical species are present. "With tlie present techniques one can only
139
-------
speculate about the sequence of events from the nature of the stable
compounds found, but this does not imply that the actual reactions are not
free-radical in nature.
It is most probable that no single mechanism can account for either pcah
formation or 'carbon* formation in rich premised fleaes althougn there
are certain significant observations that oua be made from both the results
of this study and those of otiier vorkers.
It cannot be said with certainty that all unoxidised fuel molecules
'
are Initially broken down to acetylene although several
have found relatively large amounts of this compound in flame gases t (In
fact acetylene is produced commercially by the partial combustion of
met cane with subsequent quenching (BASF, formerly called oachsse, process)).
The fact that in the oxy-ethylene flames employed in this study the othylene
concentration falls very rapidly to a very low value In the blue zone whilst
the acetylene concentration rises (see Figures 4»3»3(b)+(d) and
Figures 4.3«4(n)+(b)) shows that a large percentage of the original ethylene
fuel is converted to acetylene (most probably via free-radical reactions)}
in high tempers ture pyrolysis experiments carried out on ethylone (e.g. in
shook tubes), acetylene is generally a major product. The low temperature
(•v 60G-10GO°C) pyrolyais of ethylene generally yields species such as
1-butene, 1 . 3- butadiene and cyclo-olefins. Hone of these compounds has
been identified in the present work, although it aay be argued that the
inability to detect a species may imply that this species is too reactive
to be sampled by the techniques employed. This argument may be refuted in
the present case, however, since by usinT such sampling techniques it has
been shown possible to detect and estimate (with a certain uegree of
experimental scatter) such species aa diaoetylene, triacatylone and
140
-------
tetra-acetylene (v) which themselves are known to be extremely reactive.
Thus| it way be concluded that ethylene and other olefinlo species such ao
1.5-butadiene are not important species in the formation of pcah in rich
premised flames*
fcthylene nay be consumed initially both by pyrolysis-type reactions
as the temperature riaeo and clso by conbuetion since this is the region
in the flame where oxygen and Oli radicals are present.
The presence of methane, particularly in the blue oxidation zones
of oxy-aoetylene and oxy-ethylene flameo, shows that sooe of the fuel is
also broken down into C. units. These speciea, again, rcay be the result
either of the combustion reactions or of pyrolysis-type reactions. One
significant result of this study and that carried out by Fenicore et al"
ia that there is a ouch greater formation of methane in oxy-ethylene
48
flames than in equivalent oxy-aoetylene flanes. Cullis et alH have
presented a reaction scheme which suggests that methane may be formed by
toe deaydrogenation of polymeric groups during the pyrolysia of acetylene.
This may well explain the presence of methane in an acetylene flane but it
would not account for increased methane foruation in an ethyleno flame even
if all the etbylene were converted initially to acetylene, one explanation
may be that aetnane is formed to a greater extent in the combustion
reactions in etl^ylene flames stjcji in the combustion reactions in acetylene
flames.
Since it seena very likely that some (" 25/5) of the etnylcne iucl is
converted to acetylene in the blue gone of rich premixed flames the fate
of this latter compound must be studied. Acetylene can polymerise to
benzene but the relatively low reactivity of benzene suggests that it is.
141
-------
a stable by-product rather than a precursor of condensed-ring aromatic
hydrocarbons. & very acall tuaount of biphenyl in the flanes studied suggests,
however, that some benzene pyrolysis does occur probably via phenyl radicals*
(Verphenyle have not been detected in this study).
Chain-lentftheDing processes are evident in this and in previous
39
studies . The identification of any individual polyacetylene is
unambiguous by mass erectrometry since identification by 'mass measurement1
39
is 100/f positive. Although iioaonn and co-vorsers report the presence of
polyacetylenea up to the twelve carbon compound C12H_, in the present study
it is likely that polyacetylenea of a higher order than CQH2 cannot be
detected sincei-
(l) they are too reactive
(2) they are present in too low concentrations
The polymeric nature of the chloroforai-insoluble material collected
low (1.5-1.9 co.) in rich prfir.ixcd oxy-acctylene flames may well be
accounted for by the polymerisation of one or more polyacetylenesi
45
Hooann' has pointed out how such reactions together with the addition of
acetylene can account fox the H/C ratio of approximately unity for this
»carbon1 (see Section 1.1.4).
The oechanlsju of the formation of polyacotylenes has not been studied
in detail except in the special case of dlacetylene. This compound is
rery often formed in preference to vlnylaoetylene as a product during
the high temperature pyrolyuia of acetylene) the results of the present and
TO
other studies" have snown that in rich prentixea hydrocarbon flames
vlnylaoetylene is, in fact, formed prior to diaoetylene and that the
142
-------
concentration of the latter Increases vhilot the concentration of the
former is decreasing. This suggests that at the high temperatures
prevalent in the blue eone of oxy-acetylene and oxy-ethylene flames
dehydrogenation of vlnylacetylene is occurring. 3y analogy, one might
pro;
-------
The next stage (after tee Cg-C. species) In the step-wise synthesis
is postulated to be a C,-C. species (IV) wuich could well be a phenyl-
butadiene or related radical.
H
CH
I
.CH
or
r,/
'H.
C
I
CH
or
'CH
II
,C
phenylbutadienes
A species whoas molecular weight is equal to 1JO (that of phenylbutadiene)
has been detected in the present work, although its identification is not
certain (however, its fonaula is C1(^1Q). It is possible, however, that
phenylbutadiene (or its related radical) might stabilise as a dihydro-
naphthalene upon sampling since naphthalene has been prepared from
247
phenylbutadiene by passing this compound through red-hot tubes .In
Badger*s reaction scheme compound V nay therefore be a dihydronaphthalene
rather than tetralin (a tetrohydronaphthalene), especially in flames where
dehydrogonation reactions are favoured. No tetralin has been detected in
46
the preoent studies of flames. Groll^ has proposed that naphthalene may
be formed Indirectly from acetylene via a dihydronaphthalene (see Section
1.2.1).
112
Stehling and co-workers pyrolysed acetylene with atyrene as an
additive to find that there is a slight increase in naphthalene formation
144
-------
at 600°C« thus suggesting that the following overall reaction (cf. Badger
£2 65
et al ' ) nay be occurringi
B B
CB,
HC^CH
B
'-c
H
VIII
B
252
Kany years ago, Berthelot synthesised anthracene from benzene and
atyreno, thus suggesting that the following overall reaction cay occurs
•CH,
+ 2H.
The presence of phenylacetyleue is not unknown in products of
incomplete ooobostion or in products of pyrolysis-type experiments!
particularly in the case of benzene*
(1) In rich premised benzene/oxygen flames Hooann and co-workers"
have found tnat the concentrations of species such as phenyl-
ao«tylenet indene, methyl naphthalene and biphenyl pass through
mazioa and decrease in ihe burned gases of the flames.
Acetylene and polyaoetylenes are formed also in such flames.
(2) 7ery recently. Kern and Spengler have reported tha presence
of both phenylacrtylene and styrene (in roughly equal concentrations)
in the products formed in hexane diffusion flames.
145
-------
(3) Using an electric arc in benzene, Kuller and £anninger hare
found evidence for the formation of i/henyl ana acetylenyl
radicals since the products include (apart from 95A1 unreaoted
benzene) phenylacctylene, biphenyl, diacetylene (and some
higher acetylenes) together with hydrogen and acetylene.
The initial breakdown of fuel (both acetylene and ethylene) into
C~ species, which has been mentioned previously, nay account for the
formation of species with odd numbers of carbon atoms, e.g. toluene, indene,
and 1-methyl naphthalene. The possible importance of indene in the
formation of higher molecular weight aronatio species in flaaes has been
2b4
Bulgestea by tavies and Scully . These workers have found that when indene
is injected into a rich towns gas/air ^remixed flame, soot formation is
strikingly high and they have suggested that the following reactions might
occurs
-CH a CE*
•a,
cleavage along
dotted line
This radical may also be forced by the dchydrogenation of
o-oethyl styrene, a species which has been suggested to be present in the
flane gases of rich ozy-aoetyiene and oxy-et/jylene flames in the present
work, (see Section 4.4.1.2). Javies and Scully ^ have also proposed that
benz(a)anthracene and chxysene eight be fenced from the above-mentioned
diradioal in the following manners
146
-------
ben»(a)anthracene
ohryecne
Theee vorkera have also r«port«d that when etyrene la added to the
tovna gas flense instead of indeno, euoh less aoot ia forced Indicating
that the CH- group in the radical obtained from indene plays an
Important r&le. Thle ia supported by the fact tnat the 8OOt yield with
254
'
et>rece io alao lens than that vith toluene
Jiariee and Scully have concluded that in euch flcioea,
(1) benzene ringa favour aoot foraatioa
(2) afUc'jcd
^rou;.a procoto aoot fonution even further
(?) polyoondeneed aroo&tic hydrocarbons favour eoot formation.
112
Point (2) la au. ported by the results of Stealing and co-workers |
these workers have found that in the pyrolynie of acetylene the rate of
diaapuearanoe of thla coapound la accelerated by the addition of 2-cethyl
naphthalene. However, toluene doee not appear to effect the rate of
acetylene disappearance*
147
-------
Daniels '* has concluded from the results of Street and Thomas5 that
In premixed flames alkyl groups attached to benzene rings increase the
amount of oxygen required to suppress •carbon* formation. This nay be due
to the participation of side-chains in ring-closures yielding pcah in the
flanes, thus giving rise to compounds from which it is difficult to suppress
'carbon' formation. Lang, Buffleb and Zander report that the pyrolysia
of alkyl substituted aromatic hydrocarbons occurs via direct nuclear
condensation, the primary bonding occurrino acrofis the alkyl group.
Kinney has suggested, for instance* that toluene can condense to
blbeneyl whioh in turn may produce anthracene or phenanthrene.
CH
phenant'irene
Mo bibenzyl (mol. wt. = 162) has been found In the present work
QC£ 5^T
although Ingold and Lossing and Blades et al have shown by mass
speotrometry that beneyl raolcala are fonaed during the pyrolyais of
toluene (wnioh naa been found in both oxy-aoetylene and oxy-ethylene
flames).
148
-------
Hoiaann and Wagner have eugjested that polycycllo aromatic
hydrocarbons auch as anthracene, pheuanthrene, pyrene, eto. (i.e. the
ao-oolled Group 2) cannot be important intermediates or 'nuclei1 in the
formation of 'carbon1 in acetylene flames since the rate of 'carbon'
formation decreases to »ero whilst the concentration of these species
increases. (From their results on the pyrolysis of benzene Sakai et al
have concluded also that 'bare aromatic' molecules cannot be intermediate
compounds in the formation of coke or tar in the pyrolyois of petroleum
hydrocarbons at temperatures around 600 C.) The results of the present
study show, however, that the concentrations (mole fractions) of most poah
do not rise steadily in the burned gas; instead the mole fractions of
several poah increase to maximum values in the blue zone, then fall to
lov but definite values in the burned gas. Similar results, usins a
different and leas satisfactory sampling technique, have been obtained
2
by previous workers .
There ia no evidence in the present study for the presence of the
eo-called "Croup 3" hydrocarbons (reaotive poah with side chains and
containing more hydrogen than the parent poah) in the blue zones of oxy-
acetylene and oxy-ethylene flames, although it should be pointed out that
the concentration of an individual species within this group is of the
•7 2S8
order of 10 of a mole fraction* Although Honann and Wagner consider
Group 3 species "to be important intermediatea for the formation of solid
particles" in rich premized flat flar.es, there is some uncertainty as to
the lover mass number limit of this Group, since this has been reported as
1 259 258
being "150, "90 t and 250 . Grouu 3 hydrocarbons have been clairced to
be detected by mass spectronetry both by evaporating soot samples in vacuo
and also directly from the flaae, thus suggesting that such molecules are
relatively stable species.
149
-------
It is difficult to understand how Hoaann and Wagner could streos the
possible importance of t&ia troup of compounds la 'carbon* formation and
not to comment on the presence of species such as phenylacetylene, etyrene,
etc., which have been shown (in the present work) to be present in
an
relatively appreciable quantities, liomann and co-workers" have shown,
however, that speoies such as fhenylacetylene are preaent in rich
premised benzene/oxygen flanes and that the concentration proflleo of such
species are similar to those found (in the present vork) in similar rich
preoiied ojy-aeetylene flames.
150
-------
CONCLUSIONS
IV.5.1. Sampling Techniques and Analysis
The results of this Investigation demonstrate that by the use of high
resolution mass apeotrometry and progranuned-teaperature gas-liquid
chromatograpby, it is possible to detect a stable species whose mole
fraction in the flame gases is approximately 10 , thus obviating the
necessity for a complex molecular beam sampling/mass spectrometer system
TO
as has been used previously by Hooann et al on work of a similar nature.
These workers have shown, however, that such a system does enable the
concentrations of a few free-radicals to be determined.
A technique has been devised for using the GEC-AEI M59 mass
spectrometer as a quantitative instrument by admitting into the combustion
chamber (along with the premized gases) a known flowrate of an inert
reference gae (argon).
Fro^raoaed-teaperature QIC techniques have been shown to provide a
rapid and efficient means of separation and determination of polyoyclio
aromatic hydrocarbons and other compounds which have been identified.
A technique has been employed to provide an efficient means of collecting
gas chromatographio fractions for the Identification of species by both
mass speotroac*ry and U7 absorption speotroscopy. It has been found
necessary to use both of these latter techniques for the Identification
of certain species present ir. the flame gases.
The similarity between the present batch results and those results
reported by the only previous workers in the field" confirms that
sampling via quart* mioroprobes in the manner described is efficient in
terms of reaction quenching. The results of this study indicate, also,
that such microprobea sample species which are present in the gaseous phase
151
-------
rather than associated with the solid phase ('carbon') thus suggesting that
a large proportion of the pcah are adsorbed onto the surface of 'carbon*
2
particles when this material is withdrawn from flames .
IV.5.II. Formation of Pol/acetylenes
The presence of polyacetylenes in rich premised oxy-aoetylene and
oxy-etiiylene flames has been confirmed but* because of their relatively
high reactivity, concentration profiles cannot be determined very
accurately. It cannot be established whether equilibria exist between
these compounds and hydrogen as Bonne, Eonann and Wagner'5 have suggested.
It is difficult to propose a mechanism which accounts for th* formation
of high molecular weight polyacetylenes although as Eomann and Wagner have
pointed out, such a mechanism is probably free-radical in nature* It is
possible taat polyacetylenes are formed by the dehydrogenatlon of polymers
with the generic formula (C2H_) ; these species are most probably free-
radical in nature. When x • 2 the polymer (radical) may stabilise as
vlnylaoetylene or be dehydrogenated to dlacetylene; when x » J the polymer
(radical) may either stabilise as benzene (cyollsation) or be dehydrogenated
to triacetylene, etc. It is not proposed that such polymers are formed by
the direct polymerisation of acetylene but rather by the combination of
acetylenyl and polyacetylenyl radicals.
IV.5.III. Polycyclic Aromatic Hydrocarbon Formation
The concentration (mole fraction) profiles of pcah throughout the
2
flame support, in general, the results obtained by Tompklns and Long
152
-------
whose sampling technique was nover claimed to be other than relatively
crude. The resulta of the present study and those reported by Tompkina '
and Long show that, in general, the concentrations of individual poah do
not rise steadily in the burned gas (yellow 2one) of either oxy-aoetylene
or oxy-othylene flanes* Thus some doubt oust be oast upon the validity of
those results reported by Homann and Vaguer which suggest that they do.
With the exception of species such as toluene, phenylacetylene, styrene,
methyl styrenes (?), 1-methyl naphthalene, methyl pyrenes, etc., there is
little evidence for the presence of many alkyl substituted pcah in either
the blue or yellow zones of rich premised oxy-acetylena and oxy-ethylene
flameo* There is no evidence of the so-called 'Group 3' hydrocarbons that
Botaann and Wagner have claimed are formed in and are destroyed by the end
of the blue cones of such flames although these species nay be present in
concentrations of less than 10~ mole fractions*
The presence of several significant compounds such as phenylacetylene,
etyrene and the ccapound whose molecular weight is 130 is particularly
interesting since this euggests that the reaction mechanisms proposed by
62 6^
Badger and co-workers ' may well account, at least in part, for the
formation of pcah from the two-carbon species, acetylene. It is unlikely,
however, that such reaction schemes solely account for the formation of
higher molecular weight pcah in rich prefixed flames since onoe formed,
pcah may be pyrolysed to .five higher aromatic species as is indicated by
the results of Lang, Buffleb and Zander '.
IV.5.IV Concentration Profiles
With a few exceptions, the concentration (mole fraction) profiles of
153
-------
many species In rich prcmixed oxy-acetylene and oxy-etnylene flames are
•imilar* The typical profile nay be divided into three parts as follovsi
(a) The initial rapid formation of the speoiea in the pre-heat tone
and in the less-hot part of the blue oxidation zone; this probably
involves the pj-rolysis of lov molecular weight species in the
presence cf oxygon and Oh radicals. Some combustion of hydro-
carbon species will also be taking place, simultaneously, of
course.
(b) The destruction of some compounds to low (sometimes cero)
concentrations due to combustion and oxidstion reactions
•
(presumably by attack by OH radicals, chiefly).
(c) The secondary increase in concentration; this most probably
being due to the pyrolyais of residual lover molecular weight
species, in particular acetylene whose concentration in the
burned gases of all the flumes studied is relatively high.
The temperatures within this region are favourable to poah
formation (Figures 4.2.1, 4.2.2, Ref. 44).
.IV.5.V. The Importance of Acetylene and Lthylene
Apart from a few compounds (e.g. methane) the concentrations of
products are similar in both oxy-acetylene and oxy-etnylene flames* This,
together witn the fact that relatively large amounts of acetylene are
formed in oxy-ethylene flames, suggestu that etnylene is first converted
(in part, since ito&e ethylene is oxidised in combustion reactions) to
acetylene which is then responsible for the formation of the many products
of higher molecular weight found in both types of flame.
I 54
-------
Ho (or at least, relatively very little) ethylene, 1.3-butacUene,
1-butenef etc* are present in the rich preoixed oxy-acetylene and oxy-
ethylene flanes, thus sug^estix)*; that the following mechanism does not
ooour in flamest*
ethylene —* 1. J-butadiene —+• products
(However, this does not rule out the possibility that such a mechanism
Bight possibly account for the formation of poab in pyrolysis experiments
where the degree of dehydrogenation is leso than in flanes.)
IV. 5.VI 'Carbon' Formation
It has not been the intention in the present study to investigate
the mechaniso of 'carbon* formation although several observations are
worthy of mention*
(1) The different types of 'carbon1 collected at various heights in
both oxy-acetyleno and oxy-ethylene flames are similar in
2
appearance to those reported by Tompklne and Long .
(2) The results of the present work suggest that pcah are adsorbed
onto the surface of 'carbon* particles when these are collected
from the flame through filters, etc.
(j) The similarity of products in both oxy-acetylone and oxy-
ethylene flames 6U£gestB that 'carbon* is formed via the sane
mechanism in both cases.
(4) from the results obtained in this study it is not possible to
say whether polyacetylenes, Dare poah or other aromatio species
155
-------
(with side chains) arc important intermediates in the fonsation
of 'carbon1 since different mechanisms and compounds may be
important at different heights in the flasea. The fact that
aromatic molecule* vith side chains tend to promote soot
formation in flames (see Section 4*4*9 indicates that this type
of species may play an important role in 'carbon1 formation.
However, it is most likely that, as other workers have concluded,
tar* aromatic specieu are tost probably relatively stable by-
products of the reactions rather than 'nuclei* for 'carbon1
formation.
IV.5.VII. Mechanisms for Polyacetylene and Pcah Formation
The reaction scheme outlined in Figure 5*1 summarises the results
of this study and those of others deemed to be relevant| it is apparent
that a number of reaction schemes any account for the formation of poah
although some indication is given of how lower molecular weight species
such aa phenylaoet/lone, dlhydronaphthalene (?) etc., are formed.
It has' not been possible to establish the nature of the actual
species taking part in the complex reactions involved but it seems nost
likely that these are free-radical in nature; thus the stable compounds
identified can at this stage only give an indication of the free-radicals
involved.
156
-------
equilibrium with each other and H2 (refs 35 and 128)
Hydrocarbons in general
polyacetylenes
I dehydrog.
°2E2
\
C1 species
Methane
Cullis et al43
-. Toluene
- phenyl
radical
tetra-acetylene
t
diacetylene
triacetylene
Free-radical
Free-radical
Free-radical
Free-radical
- <°2H2>2
QI Pcah
s>x
phenylacetylene
44
vinylacetylene
dihydronaphthalc-nas
chrysene <
benz(a)anthracene
IV. Figure 5.1 Proposed Flow Diagram for the
Formation of Aromatic Species and
Simple Pcah from Acetylene
1-methyl naphthalene
benzo(a)pyrene
-------
IV. LIST OF REFERENCES
1. Ilomann K.H. and Wanner H.Gg. , 11th Symp.(lnt) on Combustion,
The Combustion Institute, Pittsburgh, 371|1967«
2. Tompklna E.E. and Long li., 12th Syap.(lnt) on Coabustion,
The Combustion ID Ue, Pittsburgh, 625,1969*
3. Chakraborty fi.B. and Long R. , Combustion and Plame, 12.226,1968.
4. Searl T.2. , Casoidy F.J., King W.R. and Brown Q.A.,
Anal.Chem., ^2, 9 54, 1970.
5. Sawioki E. , Stanley T.W., Elbert V.C., Meeker J. and HoPheraon 3.,
Atm. Environ. , .£,131,1967.
6. Schoental H., in "Polycyclio Hydrocarbons", Vol. 1, (Ed. Clar £•),
Aoadeoio Press, London, 1964.
7. Bailey E.J. and fungal K. , Brit. J. Cancer, J2, 548, 1550.
6. Comains B.T., Waller R.E. and Lawther P.J., Brit.J.Industr.Msd.,
9. Hoffman D* and Wyndcr E.L. • Mail. Cancer Institute Monograph Ro*9i
(Ed. Savioki E. and Caesel k'.)» 91,1962.
10. Dikun P.P., ProbleoB Cnkol. (U.5.S.R.) ,6,64,1960.
11. Vyndcr E.L. and Hoffman D., Cancer, J^,1079,1959.
12. Cook, J.V., Bevett C.L. and EiGger I*, J.Chem.Soc., 595,1935.
13* Kacaiond E.G. and Horn 0., J.Aracr.Ked.ABsn., Ll66«1294i1950.
14. Cook J.W., "Tobacoo iimoke end Lun^ Cancer", Royal lost, of Chen.,
Lecture Series Ho. 5, 1961.
15. Lindsey A.J., Brit.Hed.J., ^06,1959(vol. 1).
16. Kern J. and Spongier 0., Erdol u Kohle, Ercjas Potrocheaie, 23.813.1970*
17. Arthur J.R., Commino B.T., Gilbert J.A.S., Lindsey A.J. and Napier D.H.,
Combustion and Flaae, ,2,267, 1958.
18. Lindsey A.J., Combustion and Plane, ^,261,1960.
19. Ray S.K. , and Long R. , Combustion and Flame, ^,159,1964.
20. "fionenclature of Organic Cheaistry", 2nd Ed., Internatlon Union of
Pure and Applied Chemistry, Euttervortha,
London, 1^66.
159
-------
21. Gaydon A.G. and Uolfhard E.G., "Floors, their Structure* Radiation
and Temperature" f 3rd Ed. t Ciiapman and Hall,
London, 1970.
22. Kinbara T. and Hakanura J., 5th Symp. (int) on Combustion,
ileinhold, New York, 285,1955-
23. Berl U.G. and Wilson W.E., Nature, 191.580.1961.
24. Fowling J., Fuel, 16,25,1949.
2^. Botha J.P. and Spaldlng 3.3. , Proa. Roy. Soe., A22§,71t1954.
26. Kydd P.H., Coabuation and Plane, JJ, 133, 1959.
27* Hoffiana K.H. and Wagner 3.Gg. , Ber.Bunaenges.Phys.Chea. 169 ,20,1965.
28, Bonne 0. and Wagner H.Gg., Ber.Bunsenges.fhys.Chee., 69«35tl965.
29* Tualu 7*3., Combustion and Flame, 11,190t1967.
JO. Fristrom H.H. and Westenberg A.A. , "Plane Structure", KcGraw-flill,
Btw York, 1965.
J1* Levy A. aad Weinberg F.J., 7th Syop. (Int) on Combustion,
Buttervorths, London, 296,1959.
32. Friatroo fl.M., Grunfelder C. and Favin S., J.Phys.Chem. , 64.1566.1960.
33* Fenimore C.P., Jones G.W., and Moore G.E. , 6th Symp. (lot) OB Combustion,
Keinnold, Hew York, 242, 1957.
34. Singer J.K. and Cruiser J., U.S. Bureau of 1'ines, Report of
Investigations No* 6007, 1962.
39* Bonne U., Homann K.E. and Vainer H.Cg. , 10th Symp. (int) on
Combustion, The Combustion Institute,
Pittsburgh, 503t1965-
36* Tonpkina E.B., Ph.D. Thesis, University of Birmingham, 19<>7.
37* Killikon B.C., J.Phys.Chem., £6,794,1962.
5fl. Street J.C. and Thomas A., Fuel, 54. 4.1955.
39, Hooann K.H*, Hoohizukl H. and Wagner H.Gg. , Z.Phys.Chen.,B.F.,
40. Ingram D.J.E. , Tapley J.G., Jaokson R., Bond R.L. and Murnaehan A.R. ,
Kature, J[H» 797,1954.
41* Fenlmoro C.P. and Jones G.W. , J.Cheo.Phys., 41,1887,1964
160
-------
42. Cotton D.L., Frinw?! N.J. <-n(; j.-nkins D.'!., Combustion and Flame,
.11, B7.19
43* Bonne I', and ;:onf>nn K.H., Kurzn'ichrichton, Gottlnjon, No.4,p.1»1965.
44* Badger G.N.. Kinber /..W.L. and Novotny J., Auatr.J.Chcc. ,17. 776, 1964*
45« Hoaann K..H. , Consent on paper presented by To&xins and Long.
(iiefersnce 2)
46. Croll U.P.A., lad. and Lne.Cheia. , 25,704»1933.
47* Stehliivj P.C,, Frazoe J.D. and Anderson It.C., 6th Symp. (int) on
Coa'ouotion, Ileinhold, Rev York, 247,
48* Cullia C.F., Minkoff G.J. and Hettleton K.A. , Trans. Far. Soo.,
49* Cull is C.F. and Franklin K.H. , Proc.Hoy.Soo. , A260, 1)9, 1964.
50. Badger O.K., "The Chemical Basis of Carcinogenic Activity",
Charles C. Thooaa, oprin Ferguson S.E., Combustion and flame, J.,431,1957.
58. Gay don A. C. and Wolf hard H.G., Proo.noy.Soo. ^199,89,1949.
59* Hajue E.ll. and Wheeler H.V. , Fuel, .8,^60,1929.
60. Rinney R.E. and Crowley D.J., Ind. and L".ns.Cho«. , 46.230«1'J54«
61. Oro J. and Ban J. , Soience, 153.1393.1966.
62. Badger G.K., Buttory R.G., Kimber H.W.L., Lewis C.1'., Koritz A.G.
and iiapier I.M., J.Chea.Soo. ,2449, 1958.
6j. Badger CJJ., Kimbcr R.V.L. and Spotawood r.I!., Nature, 187.663.1960.
64. Badger G.K. and Kimber R.W.L., J.Chem.Soc., 2455,1958.
65. Badger G.h. and iuttery a.G., ibid., 2458,1958.
161
-------
66. B&dtc-r G.N. and Spotst-cofi T.M., ibid., I63r,,1959«
6?. Badger C.r. and Kir. tor R.W.L. , ibid, 266,1960.
68. Badger G.H. and Kimber K.V.L. , ibid., 2746,1960.
69. Badger G.K., Levio G.E. and Hauler I.tt., ibid., 2825,1960.
70. Badger G.K. and opotuvood T.M., ibid., 4420,1960.
71. Spotevood T.K., ibid., 4427, 1960.
72. Badger G.h. and apotBuood T.K., ibid., 4431 t I960.
73. Badger C.K. and Hovotny J., ibid., 3400,1961.
74. Badger G.h. and Kovotny J., ibid. , 3403,1961.
75. Badger G.K. and timber R.W.L., ibid., 3407,1961.
76. Badger G.K. , Kimber R.W.L. and Eovotny J., Auatr.J.Chea. .1^.616.1962.
77. Badger G.H. and Sovotny J.v ibid., .16,613,1963.
78. Badger G.H. and Novotny J., ibid., ^,623,1963.
79* Badger G.M., Jolad S.D., and Spotawood T.M., ibid., .12,771,1964.
60. Badger G.l-i., Donnelly J.E. and Spotsvood T.K., ibid., JI, 11 38, 1964.
81. Badger G.M., Donnelly J.K. and Spotswood T.K., ibid., .££,1147,1964.
82. Badger G.H*, Jolad 3. 1), and cipotewood T.H., ibid., 19,63.1966.
63. Badger G.h., Jolad S.D. and Spots wood T.M., ibid., J^, 93. 1^66.
84. Badger G.M., Jolad S.D. and opotawood T.tt., ibid., .20,1^29,1967.
8$. Badcer G.K., Jolad S.L. and Spotswood T.M., ibid., 20,1439,1967.
86. Meier au Kocker E., Brennatoff-Chenie, 47t(Ho.2).55.1966«
67* Kunugi T., aakai T., Soma K. and oasalci Y., Ind. and Eng.Chem.
Fundonpntals, JB, 374, 1969.
86. Weizmann C., Bergraann £. , Hu^^ett V.F. , Steiner K. and Sulcbacner A.,
Ind. and Lng.Uh«a.,
89. Bergaann, L. Hature, 161.609.1948.
90. Kloetsel M.C. and Martcl H.E., J.Amer.Chem.Soo., 22,4786,1950.
91. Koalov G.I. and Knorre V.G., Combustion and Flcjte, ^,253,1962.
162
-------
92. Street J.C. and Thomas A., Fuel, J&,4.1955.
93. Gill D.V., B.C.I-.H.A. Konthly Bull., .22,487^958.
94. Vhelon P.P., Paper Presented to the Inst. of Hech.Fngre, 1960.
95* Palmer H.B. and Cull is C.F., "Chemistry and Physica of Carbon",
Vol. 1, (Ed. V,nlker P.L.), Edward Arnold,
London, 1965.
96. Homann K. E., Combustion and Flame, 11,265.1967.
97. Uwes V.B., Proc.Roy.Soo.,
98. Porter G., 4th Symp.(lnt) on Combustion, Williams and Vilkina,
Baltimore, 243,
99. Cole fi.J. and Wlnkof f C.J., Proo.Roy.Soo., A239.2B0.1957.
100. leaner P.A., Robinovitch H.J. and ilafalkes 1.5. , 8th Sjrmp. (Int) on
Combustion, Willians and Wilkins,
Baltimore, 801,1962.
101. Veatbrook E.A., Hellwig If. and Anderoon R.C., 5th Symp. (Int) on
Combustion, Reinhold, New York, 631,1955*
102. Runnel K. and Veh P.O., Arch. Eisenhuttenwesen, 1^,409.1941.
103* Thoraoe A., Combustion and Flame, j3, 46,1 962.
104. Milberg M.L\ , J.Wiys.ChoB. , 6$, 578, 1959.
105. Sveitzer C.W. and Heller G.L., Rubber World. 154.055.1956.
106. Gordon A.S., timith S.K. and KcNesby J.H. , 7th Uyap. (Int) on
Combustion, Butterworths, London, 317, 1^59*
107. Arthur J.R. and Napier D.I!., 5th Symp. ( Int) on Combustion,
Reinhold, New York, 303, 1955*
108* Egloff G. ,. "The Reactions of Pure HyJrocarbona", Am.Cheo.3oo.
Monograph No. 73, Reinhold, Now York, 1937.
109. Steaoie E.W.R. , "Atomio and Free Radical Reactions" TO!. 1, 2nd Ed.,
Reinhold, Rev York, 1954.
110, Kinkoff G.J., Hewitt D.M. and Rutledge P., J.Appl.Chen., 2,406,1957.
111. Silcocks C.O., Proo.iloy.Soo., A242. 41 1.1957.
112. Stehllng P.O., Fracee J.i). and Anderson R.C., 8th Symp. (Int) on
Combustion, Williams and Vilkina, Baltimore,
774.1962.
113. Munoon M.3.B. and Anderson R.C.', Carbon, J., 51, 1963.
163
-------
114. no« K.C. and Anderson R.C., J.Phys.Chmu, 6
115. Cull la C.F. end Head I. A. Trana.Far.Soo., 66,920,1970.
116. KinkoffG.J., Canad. J. Cham. , j£,131,1958.
11.7* Kinney C.H. and Slyoh K.S., Proo. 4th Conf. on Carbon, Perganon,
Oxford, 301.19CO.
118. Slysh R.S. and Kinney C.R., J.Phys.Cheau, 65. 1044 • 1961.
119. Palmer H.B. and Donnish F.L., J.Fhys.Cheau, 60,1533,1964*
120. Valeh A.D., Disc. Far. Soc., .2,16,1947.
121. Burrows I.K. and Linda ey A.J., Chen, and Ind., 1395»1961.
122. Cooiaias B.T., Atn. Environ., jj, 565, 1969.
123. Johnaon G.L. and Anderson R.C., Froc. 5th Conf. on Carbon, Vol. 1,
n, Oxford, 395|1962.
124* Coats F.H. and Anderson R.C., J.FhyB.Chen. , 63.1340.1959.
125. Crenne E.F., Taylor H.L. and Patterson V.L., J.Fhye.Chem., 62,236,1958.
126. Aten C.F. and Greene E.F., Combustion and Flame, J^,55|1961.
127. Skinner G.3. and Gokoloaki E.K., J.Fhys.Chen., 6^,1952,1960.
128. Gay I.D., Klatiakowsky G.B., Kichael J.7. And Niki H.,
J.Chcns.rhys., 4^,1720,1965.
129* Bradley J.N. and KistiakowsScy G.B. , J.Chen.PhyB. , 35i264t196l.
130. Takita S., Mori T. and Tanaka I., J.Phys.Chen., 12,2929,1969.
American Petroleum Institute Research Project Ho. 44, Carnegie Preae,
Pittsburgh, 1953.
132. Howard V.B., Wood B.ll. and Kaltcnbacher E.J., Che&.irlng.Progr.,
50,1961.
133* Miller S.A., "Ethylene &nd its Industrial Derivatives H,
(Ed.Hiller S.A.), Ernest £enn, London, 1969.
134. Kolera M.J. and Stubbs F.J., J.Chem.Soc., 381,1952.
135. Towell G.D. and Martin J.J., A.I.Ch.r.J., 2,693,1961.
136. Dchlgren G. and Douglas J.E. , J.Amer.Cben.Soc., 80,5108.1956.
137. DeKaleioeye J.T., Kellottee H. and Delbourgo R., Bull.Soc.Chlm.Krance,
2266,1965.
164
-------
138* Benson S.V. and Haugen O.K., J.PhyB.Chesu, H,1735»1967«
139* Halstead H.P. and *ulnn C.P.« Trana.Par.Soo. , 6^,103,1966.
140. Skinner G.fi. and Sokoloski E.N., J.PhyB.Cheau, .64, 1028, 1960.
141* Skinner G.B., Sweet E.G. and Davis S.K., J.Fhys.Chen. , 75,1,1971.
142. Gay I.D. , Kern H.D.V Klstiakowsky G.£. and Niki E.v J.Chem.Phye.,
4S.2371.1966.
143. Killer I.?, and Churchill S.V., A.I.Ch.E.J., .8,201 ,1962.
144. Collie C.F., Read I.A. and Triaa D.L., 11th Symp. (int) on
Combustion, The Combustion Institute Pittsburgh,
391,1967.
143* Heffernan M.L. and Jones A.J., Chem.Coan., Ho. 4,120,1966.
146. Boa E.G. and Palmer H.B., J.Phye.CheB. , 69,658,1965.
147* Ollison V.M. and Anderson B.C., Carbon, JJt 459, 1966.
146* Kurphy M.T. and Euggan A.C., J.Aoer.CheB.Soc., 7.1*5547.1949.
149* Gil-ay £., Shabtai J. and Steokel F., J.of Chen, and Eng.Data,^,98,1960.
and Ind. and En^.Chea., 52.31.1960.
150. Kurd C.D. , "The Pyro lysis of Carbon Compounds", Am.Chea.Soc.,
Monograph Bo. 50, The Chemical Catalog Co.,
Hew York, 1929.
131. Klnney C.B. "The Cheeistry of Petroleum Hydrocarbons'*, Vol. 2.,
(Edited by i Brooks B.T. , Boord C.E. , Kurtz S.S.,
and Schnerling L.), Reinhold, Hew York, 1955.
152. Gordon A.S., "Combustion and Propulsion", 5th AGARD Colloquium,
(Edited by: Eagerty H.P., Lutz 0.,
Jaumotte A.L. and Penner S.S.),Pergaoon,
London, 1963.
153. CulliB C.F., Fuel.Soo.J., Sheffield University, Ji,7,196%
154* Sakai T., Vada 3. and Kunugi T. , Ind. and Eng.Chem. , Proc.Des. and
JO, 305, 1971.
155* Bauer S.H. and Aten C.F., J.Chem.Phye., J22,1253,1963.
156. Hou £.C. and Palmer B.B., J.Phys.Chem., 62,863,1965.
157* Lang E.F., Buffleb E. and Zander M., Erdbl u Kohle, Erdgas,
Petrooheole, 16,944.1965.
158. Klnney C.H. and DelBcl E., Ind. and Lng.Chem., ^6,548,1954.
165
-------
159. Mullery.G., Helv.Chim.Acta, 8,626,1925.
160. Straus F. and Xollek L., Ber.Deut.Chem.Gesell, Jj£, 166411926.
161* "Faraday's £ncyolopaedia of rlydrocarbon Compounds", Cheaindex,
Butterworths, London, 1957*
162* Kern J. and Spongier G., Erdol u Kohle, Erdgae, Potrochenie,
21,615,1970.
163. Hillikan R.C., Combustion and Flame, jj,349tW.
164. Cole D.J. and Kinkoff G.J., Combustion and Flame, J.,241,1957.
16$. Spokee G.H., 7th Symp. (int) on Combustion, Butterworths, London,
229,1959-
166. PristroB R.H. and Westenberg A. A., Combustion and Flame, .1,217,1957*
167* Westonberg A.A., Raezer S.D. and Friatrom ii.H., Coabustion and
Flame, J.467,1957-
168. Friedman R. and Cyphers J.A., J.Chen.Fhys., 2^,1875,1955.
169. Singer J.M. and Gruraer J., 7th Synp. (int) on Combustion,
ButtervorthB, London, 559,1959.
170. Prey F.E. and Yant V.P. Ind. and Eng.Chem., 1^,492,1927.
171. Smith 3.R. and Gordon A.S., J.HjyB.Chea., 60,759,1956.
172. Hoffman D» and Vynder E.L., Cancer, J^.93,1962.
173. Berenblnm I., fiature, 156.601.1945.
174. Savioki £., Chemiet-Anayst, .5J.24,56,88,1964.
175. Petrowitm U.J., Chenikcr-Zeituog, 6S,235»1964.
176. Gunther F.A. and Buzzetti F. Residue iteviewe, J^,90,1965.
177* van Langenaeerech A., Chioie Analytique, jO,3«196u.
176. Chakraborty B.B. and Long a., iinviron.Sci. and Teoh., J.,828,1967.
179. Moore G.E., Thocias R.S. and Konkman J.L., J.Chromatog., 26.456.1957.
160. Koore G.E., Katz M. and Drowley W.B., J.Air Poll.Control Ass.,
JI6,492,1966.
181. ZdrojewBki A., Iiubois L., hoore G.ci., Thonas R.3. and Kouknan J.L.,
J.Uhroaatog,, ^6,317,1967.
182. Lindstedt G«, Ato.Environ., .2,1,1968.
166
-------
183. Lijinaky W.f Anal. Chem. , J2t 684 ,1960.
184. Dubois L., Corkery A. and Honknan J.L. , Int.J.Alr Poll., ,2,236,1960.
. l-'ukai J"5.| Tebbens B.D. and Thomas J.F., Anal. Chen. , 36,1126,1964*
166. Sawicki E., Stanley T.V..t Elbcrt W.C. and Pfaff J.D., Anal. Chen.,
j6,497,1964.
16?. Savicki £., Stanley T.V. and Libert W.C., J. Chroma tog., 20,548,1965.
168. Sawioki L. , Stanley T.*., McPherson S. and Morgan K., Talanta,
.13,619,1966.
169* Stromberg L.L1. and Wldmark G., J.Chrooatog.. 47.27.1970.
190. T6th L. , J.Cbromatog., 50.72, 1970.
191. Major J.3. , Perry R. and Reade H.J., J.Chronatog. , ,4,8,328,1970.
192. Janak J. and Kubecova 7., J.Chronotoc., ,53,132,19^8.
193* Lao J. and Berg A., J.Chrorantog. , 20,168,1965.
194* Lljineky W. , Dorasky I.I. and Vard J., J.Gae.Cbroo., ^,152,1965.
195* Spengler G. and ilaupt C., Ilrdol u Lohle, Erdgaa Fetroeheale,
196. Cantuti V. , Carton! G.F., Libert! A. and Torri A.C., J. Chroma tog.
197* Carugno N. and Roaii S. , J.Caa Chrooa, .2,103,1967.
198* Perry H. , Long R. and Kajer J.il., Paper presentnd to the 2nd
International Clean /.ir Congress of the
International Union of Air Pollution Prevention,
Aeen. , Weshington, 1970.
199* Beeuon J.U. and Pecsar R.K., Anal.Chca. , 4,1,1678,1969*
200. Chakraborty 6.B. , Kilburn K.L. and Thornton H.E., Chen* and Ind. ,
672,1971.
201, Oro J., Han J., and Zlatkia A«, J.Gaa Chron., ^,480,1967.
202. DeMaio L. and Corn M. , Anal.Chea., J|8, 1J1, 1966.
203. BhatlaK., Anal. Chen., ^,609,1971.
204. J)avia H.J., Anal.Chea., ^0,158J, .968.
203. GuTernator G.C. Ill, Cagcr F.L., Robb E.V. end Barley A.,
J.Gaa Chron., jj, 36J, 1965.
167
-------
206. Lijineky W., fonsky I., h&son (J., Ra^ahi H.T. and Safari T.,
Anal. Chen., J£,952,1963.
207. Vllmshurst J.H., J.Chromatog., 17«50t1965«
206. Lyona L.E., Morris &.C. and Warren L.J., J.Phye. tinea., 72,3677.1968.
209. Lovelock J.E., Zlatkie A. and Becker R.S., Nature, 195.540.1962.
210. Liberti A.f Carton! C.P., and Cantuti V., J.Chroaatog. , .!£, 141,1964.
211. Searl T.D., Caaeidy F.J., King W.II. and Brown H.A., Anal.Cbea. t
212. Terry R»t Private C
213* Chortyk O.T., Sohlotzhauer V.S. and Stcdraan.R.L. , J.Gas Chrom. ,
1,394,1965.
214. Soloman P.W. , Anal.CLea. . 3,6. 476.1964*
215* Gump B.H., J.Chroaatog.Sci., 2,755,1969.
216. Lindsay A.J., Anal. Chin. Aota., .20,175,1959 and ^1,101,1959.
217. Cooper S.L., Analyst, j£, 573, 1954.
218. Sawieki £.v Fiber t W.t Stanley T.W., Mauser T.B. and Pox P.T.,
Int.J.Air Poll., ^,2t>3,l96o.
219« Cbaigneau M., dry L. and fiicard L.-P., Chimie Analytique,
il, 187, 1969.
220. taunan M.f Chaigneau M. , Valdener C., Herts C. and Giry L.,
CoDpt.itend.Acad.aci. .Paris, 260,5795.1965.
221. Oro J. and ban J., J.Caa Chrom., ^,480,1967.
222. Majer J.H. and Perry E., J.Chea.Soo.(A), 622,1970.
223. Majer J.H. and Perry R., Pure and Appl.Cheo. ,
224. Lewis B. and von Elbe G. in "Temperature, its Measurement and Control
in Science and Industry", Am.lnst. of Physics,
Reinhold, Hew York, 1941.
225. Millikan R.C., J.Opt.Soo.Aaerlca, ^,535,1961.
226. Fery Ch., Conpt.Rend., 157.909.1952.
227. Kurlbaunf., Phyz.Z.,
226. Hottel P.O. and Brou.jhton P.P., Ind. and rng.Chen.(anal.Fd.),
1,166,1932.
168
-------
229. Ornstein L.8. and Brinkn&n I!., K.Akad. Amsterdam, ^4 1 33, 409, 1931.
230. Olsen H.L. Jrd Jyiap. on Coabustion and Flana an.i Explosion
Phenoiaena, Williams and Wilkins,
Baltimore, 663,1749.
231. ShirodkarA., Fhil.Ha&., 1^,426,1933-
232. Ueltnann R.S. and Xuhns P.W., jJ.A.C.A. Tech. Rote, 25G0.1951.
233. Marlov L.G., Nisevanger C.R. and Cady W.li., J.Appl.Phys., .20,771,1949.
234* Flock E.F., Clean L.O. and Freeze P.J). , 3rd Sytop. on Combustion and
Flame and explosion Phenomena, Wllliaua and
Wilkina, Baltimore, 65$, 1949.
235* Vaggener W.J., Verh.Phys. Gea zu Berlin, ,1^76,1895.
236. Taybulevekii A.M., Int.Cheo.Ens.t 1.346,1967.
237* Friedman R., 4th Syap. (int) on Combustion, Williams and Wilkins,
Baltimore, 2^9 » 1953.
238. Leah A.S* and Carpenter 21., 4ta Gymp. (int) on Combustion,
V/illiaaa and Wilkins, Baltimore, 274,1953.
239* Kaskan W.E., 6th Synp. (Int) on Combustion, ileinhold, New York,
134,1957.
240. Griffiths E. and Awbory J.H., Froc.Boy.Soc., A125.401.1929.
241. KaJerJ.R., Talanta JJ, 537 1 1970.
242. Beynon J.B. and Williams A.'.:. "I'.aeo and Abundance Tableo for Use
in Haas Jpectrorsotry", Ulsevior, London, 19'J
243* Lavrov M.V. , Korobov V.V. and Filipi-ova V.I., "Thermodynamics of
Gasification and Gao-liynthcuia iieactions",
, Oxford, 1963.
244* Edwards a. A. and Fageraon I.b., Anal.Chem., J7.1630.1965.
245* "Eiaht Peak Index of Kass Spoctnl Data", Vol. 1, Hass Spectroraetry
Data Centre, A.'«..;.L., AldenaBton, 1970.
246. Stull D.R. , "Styrene, ito Polymers, Conolyiacrs ami Deriv -.livea",
(Ed. by: Uoundy ^.H. and Boyer R.f.),
Reinhold, }
-------
230. Wellcave,L., rnviron.Sci. and Tech., .2,948,1969.
. Uundhkaa V., U.13. Patent, 2,317,806, (1945).
232. Berthelot M.P.3., "Lee Carbures d« Hydrogens", Gauthier-Villars,
Paris, 1901.
233. Kuller,?.C. and Bannincer A., Helv.Chim./ctft. , 10, 765, 1927*
234* Bavies R.A. and Scully H.B., Coobuatlon and Flame, 10, 165*1966.
233* Daniels P.H. Combutticn and Flame, 4, 43,1960.
236. Ingold K.U. and Losaing F.r., Canad.J.Cheffi., JH* 30, 1933*
237* Blades a., Blades A.T. ana Steaole B.W.H., Canad.J.Chem.,
.52,290,1954.
238. Uooiann, K.I], and Watiner H.Gg. , Proe.tioy.5oc., A 507. 141 ,1968.
239* Uomann K.H. , Angev.Chem. Internal. Ed., 2*414,1963.
260. American Petroleum Institute, Research Project No. 44*
261* AD.Soo. for Testing and Materials, "Index of Kass apeotral Data", 1965.
170
-------
Errati
The following references b&ve been duplicatedi-
EOB. 3 end 54
16 and 162
J8 and 92
171
-------
Part V A Brief Review of the Use of Organo-Metallic and Metal
Containing Additives in Suppressing Soot and Polycyclic
Aromatics in Flames
The effect of introducing oxygen into the combustion air, or into the
hydrocarbon fuel itself, in a diffusion flame was studied as an early part
of the present programme of work. The results emphasized the
undesirability of either a general or local depletion of oxygen in the
diffusion-flame combustion of hydrocarbons since this leads to an increase
in the formation of poly cyclic aromatic hydrocarbons, including the
carcinogen benzo (a) pyrene. Oxygen enrichment of combustion air
can greatly reduce the concentration of p. c. a. h. in the soot. If
sufficient oxygen is added to the fuel itself, p. c. a. h. can be eliminated
from the soot.
It was hoped to continue this work by examining the effects of
organo-metallic additives on the formation of soot and p. c. a.h. in
flames. Pyrolysis and oxidation processes are well-known to be
influenced by catalysts and additives and it was hoped to attempt to
reduce the formation of p. c. a. h. during flame combustion by the use of
these.
Unfortunately the loss -of key personnel with experience in the
synthesis of metal chelates precluded the continuation of this aspect of
the work.
a
However, recent work in Germany by G. Spengler and G. Haupt
has indicated that reduction in both soot and polycyclic aromatic
hydrocarbons by the addition of compounds containing metals is feasible.
Compound such as methylcyclopenta dienyl manganesstricarbonyl
iron pentacarbonyl and ferrocene when introduced into diffusion flames of,
173
-------
atomized fuels, reduced soot and p. c. a. h. This interesting piece of
woork also claims that the formation of acetylene in the flame was in
no way influenced by the additives used.
In later work, these authors studied 29 organo-metallic a.nd
certain organic compounds in a single cyclinder 4'stroke diesel engine.
Methylcyclo penta dienyl manganese tricarbonyl and the iron pentacarbonyl
seemed to be the most effective additives in reducing soot and p.c.a.h.
whilst of the organic additives, cyclohexanol nitrate was the most
effective, although less so than the metal-containing additives.
b
M. W. Shayeson found that organo-metallic compounds of
barium, maganese and iron were the most effective smoke reducing
fuel additives but that the effectiveness was a function of engine design
and power level. (J. P. 5. fuel was used for the tests and a jet engine
was operated in a test cell).
c
A review of burner fuel additives by K. C. Salooja has
recently appeared and the author points out that despite much interest
in smoke suppressants over many years, the mechanism by which tnese
additives act, in any of the applications has not been explored.
However, also very recently, a very interesting paper
d
by D. H. Cotton, N. J. Friswell and D. R. Jenkins has appeared .
They report measurements on the effects of forty metals on the amount of
soot emitted by a laboratory scale propane diffusion flame. The alkaline
earth metals were amongst the most effective and it will be remembered
that over the last few years a number of proprietary additives containing
barium compounds have been produced, and claimed to be effective as diesel
174
-------
fuel additives for example.
A semi-quantitative mechanism is proposed to account for the
action of the alkaline earth metals: its basis is that these metals undergo
a homogeneous gas-phase reaction with hydrogen or water vapour in flame
gases. Hydrogen atoms so produced will react rapidly with water vapour
to give hydroxyl radicals, so that the net effect of either decompostion
will be to produce .OH radicals. These will then be effective in rapidly
removing soot or soot precursors.
There seems to be no evidence on the effect of these additives
on polycylie aromatic hydrocarbons associated with soot, but one would
expect these or their precursors to be removed by .OH radicals too.
Whilst interest is being shown in soot suppressing additives, their
possible toxicity as exhaust products and the effects of solid products
on engine operation must always be borne in mind.
References - Part V
a) G.Spengler, L. G. Haupt, Erdol, Kohle, Erdgas, Petrochemie
22, 679, (1969)
b) M. W. Shayeson, S.A. E. Trans. 76,2687, (1968)
c) K. C. Salooja, J. Inst. Fuel XLV, 37, (1972)
d) D. H. Cotton, N. J. Friswell, D. R. Jenkins, Combustion & Flame,
17, 87, (1971)
175
-------
APPENDIX I KocenclatuTC of I'olycvslic AToaitic Hydrocarbons
The nomenclature of pcah relevant to this study is given below.
9O
The names used are according to I.U.P.A.C. (1957) Rules and the
compounds are listed in order of increasing molecular weights (in
parentneeea)* The symbol [Oj *•" U8e(* to denote a benzene-ring
configuration.
Zndene (116)
Haphthalene (I2d)
r ^^ ^
roTo
1-Kethyl Naphthalene (142)
2-Mothyl naphthalene (142)
Aoenaphthylene
Aoenaphthena
1-1
-------
fliphonyl (154)
0>-<0
Fluorene (166)
^^ ^r
oTTo
i-Cyclopenta (£,g) acenaphthylene (176)
[1.8, 4.5-bi-(etheno-) naphthalene]
Phenanthrene (178)
Anthracene (17«)
x^ ^^^ ^v^
loToTo
4>5~aetbylene phenantbrene
H
H
Fluoranthaae (202)
1-2
-------
Acephenanthrylene (202)
Aceantbrylene (202)
Pyrcne (202)
Benzo(a)fluorene (216)
Benzo(b)fluorene (216)
oYTojA
Benso(o)fluoreiM (216)
£eoeo(aoo)fluoranthent (226)
1-3
-------
Triphenylenc (228)
Chryaene (228)
Benz(a)anthracene (226)
Perylene (232)
Beozo(k)fluoranlhene (252)
Jteneo(a)pyr«no (252)
Bonso(e)pyrene
Coronane (300)
1-4
-------
APPLUJIX II The Flux and Concentration of a Species In a Flame
167
Experimental results have shown ' that microprobes constructed
30
according to the method deocrlbed by frletron and Westenberg sample tbe
concentration and not tbe flux of a species in a simple aybtarn. Although
the concentration gradients which exist in flames are much greater than
167
those which were act up *n the simple experiment by Westenberg et al 'f
it may reasonably be assumed that such oicroprobes sample concentrations
and not fluxes in flames.
Concentration
The concentration, N., of a species 1 at a point in a flame is defined
as the number of moles of the species present per unit volume at the point
in question. If V. is the mass per unit volume of species i with molecular
weight K. at tns point, then
tfi " *iMi
If p is the total mass density at the point, then
P - r w1
and the mass faction 'f.' of species 1 is given by
If X. is the mole fraction of species 1 at the point, then
where fl is the mean molecular weight of all species at the point.
II-l
-------
Since the various forms of concentration are defined solely aa
quantities! concentration is u scalar variable.
The basic flux variable of a species i is the vector which defines
the number of moles or grams of the species passing through unit area of
the flame per unit tine as viewed by a stationary observer.
50
The species mass flux is defined' by
for a one-dimensional flame
where v • the mass average velocity
?. m the diffusion velocity of species i created by a
concentration gradient.
Th» species mass flux fraction G. is defined byt
IA (T * V
1 ' f,
X.H
In the absence of a concentration gradient V. » 0 and G. • f..
The diffusion velocity V. in one-dimensional form for a species i
50
present in an excess of carrier (.;-« J is ^iven by
II-2
-------
. H - -
M d» ^ M'
vhere B • the total number of moles present per unit volume at the
plan*.
D.. • the binary diffusion coeffioient for the species i in the
*3
excess carrier gas j. (D.. is dependent on temperature*)
8 • the height above the burner surface.
Thus „ XiKi
or X.H. D.. dX
01 • (1
This equation shows clearly the effect of a changing concentration
gradient on the flux fraction of species i. In a region of increasing
*1
concentration -r- is positive. Therefore G.< F., an effect caused by the
species flowing against the concentration f'«dient.
dX
In a region of zero concentration gradient -T— e 0 md G. • f.
In a rejion of decreasing concentration gradient -r"- is negative and
thus
II-3
-------
APPENDIX III,
Hitherto Unidentified Polycyclic Aromatic
Hydrocarbons Found in Flame Soots
a
I B. B. Chakraborty and R. Long reported an unknown " derivative
of pyrene" isolated from soot. This led to an exchange of letters between
R. S. Thomas and J. L. Monkman and the above authors;
Thomas and Monkman suggested it was a methyl pyrene, but as methyl
pyrenes had already been identified as correspond- tig to an earlier
b.
peak in the gas chromatogram, Chakraborty and Long suggested that the
unknown pyrene derivative they had found was of higher molecular
weight than the methyl pyrenes.
c.
L. Wallcave then reported an apparently identical compound
which he had isolated from a coal tar pitch. The u. v. absorption
spectrum of this substance had the seemingly characteristic peaks at
378 and 358 rr\M reported by Chakraborty and Long.
Earlier M. J. Lyons ' had separated, interalia, by adsorption
chromatography, a compound in a gasoline soot sample, in a diesel soot
sample and in a general atmospheric soot sample, which gave spectral
maxima, as follows, 376. (368), 355, 338, 324, 310, 291, 278 and which
he desmated "orange compound (pyrene derivative?)"
The u. v. spectrum of the "derivative of pyrene" reported by Chakraborty
and Long shows several of these absorption maxima and is presumably the
same compound. Based on his u. v. spectrum, on the chromatographic
behaviour, and especially on the molecular weight as determined by mass
c.
spectrometry (228), Wallcave proposed that the compound in questiop
was Cyclopcntfi (c, d) pyrcr.c (cr accpyrcne).
228
III-l
-------
II During the present work, in extraction filter products, B. D. Crittenden
(Ph.D. thesis 1972) found a gas chromatographic peak, the mass spectrum
of which showed two principal peaks at m/e values of 168 and 176
respectively (at 70 e. v. )
Since neither the methylbiphenyls no- biphenyl methane (molecular weight
= 168) have a peak at m/e = 176 in their fragmentation mass spectra
it was assumed that the gas chror-atogr^nhic peak was due, at least
in part, to a polycyclic aromatic hydrocarbon of molecular weight 176 and
it was suggested that this might well be.-
C14 H8
176
e.
Rather strangely, K. H. Homann and H. Gg. Wagner in their study of
rich premixed flat flames report the concentration change of only one.
polycyclic aromatic hydrocarbon throughout their flame. This, a
species with molecular mass 176 and having the formula C, . H0, was
14 o
said to be of medium concentration relatively to the others. They did
not comment on its structure. A species, molecular weisht (by mass
f.
spectrometry) of 176, was also found by E. E. Tompkins and R. Long
in rich pre-mixed acetylene - oxygen flames.
The compound above appears n.iver to have been isolated or
g
synthesised. In 1952 A. G- AnrWson Jr. and R. H- Wade reported
the synthesis of 'pyracene1 or 1. 2 - .dihydrocyclopenta (f, g) acenaphthene
and mentioned that an attempt was in progress tc introduce a double bond
in-:
-------
into each of the peri-rings in pyracene to form 'pyracylene1
Acenaphthene I O I O I is a main constituent of coal tar and
•can easily be dehydrogenated to acenaphthylene
However, the analogous compounds 'pyracene1 and'pyracylene'
have nct> been isolated or identified in coal tar on any fraction of
coal tar.
III-3
-------
APPENDIX REFERENCES
a. B. B. Chakraborty and R. Long
Environ. Sci. Technol.
I. 828, (1967)
b. R. S. Thomas, and J. L. Monkman ibid
2, 217 (1968)
b. B. B. Chakraborty and R. Long
c. L. Wallcave
d. M. J. Lyons
ibid
2, 217 (1968)
ibid
3, 948 (1969)
Symposium, 'Analysis of Carcinogenic
Air Pollutants' Aug. 1962. Nat.
Cancer Inst. Monograph No. 9.
U. S. Dept. of Health, Education
& Welfare (1962).
e. R. H. Homann and H.Gg. Wagner
f. E. E. Tompkins and R. Long
jt. A. G. Anderson Jjr. and R. H.
Wade
Eleventh Symposium (International)
on Combustion. The Combustion
Institute, 1967 p. 371.
Twelfth Symposium on Combustion.
1969 p. 625
J. Amer. Chem. Soc.
74, 2274, (1952)
III-4
------- |