CHEMICAL IDENTIFICATION
OF THE ODOR COMPONENTS IN
DIESEL ENGINE EXHAUST
Final Report (Year 3) to
COORDINATING RESEARCH COUNCIL
and
ENVIRONMENTAL PROTECTION AGENCY
Arthur D Little, Inc.
-------
CHEMICAL IDENTIFICATION OF THE ODOR COMPONENTS IN
DIESEL ENGINE EXHAUST
Final Report (Year 3) to
Coordinating Research Council
and
Environmental Protection Agency
CRC Project; Cape-7-68
EPA~Contract EHSD 71-18
June 1971
Report No. ADL 62561-5
Arthur D Little, Inc
-------
ACKNOWLEDGEMENT
Identification of the compounds responsible for diesel exhaust
odor has been an exciting challenge. Verification and application of the
basic data becomes especially meaningful as fuel and engine variables are
studied. We would like to thank the members of the CAPE-7 Committee for
their guidance in the direction of the program and for their encouragement:
Dr. Roderick S. Spindt, Gulf Research and Development
Company; Roger C. Bascom, Cummins Engine Company;
Gerald J. Barnes, General Motors Research Laboratories;
Fred J. Hills, Mobil Research and Development Corporation;
Dr. Ernest W. Landen, Caterpillar Tractor Company;
John E. Sigsby, Jr., Environmental Protection Agency;
and Otto Zeck, International Harvester Company.
We also wish to thank Drs. Issenberg and Eissenbraun for their
assistance in providing valuable reference samples of oxygenated compounds.
The assistance of Drs. Joseph Bendoraitis and Alan Peters of the
Mobil Research and Development Corporation in supplying the aromatics
computer analysis program is greatly appreciated.
iii
Arthur D Little, Inc
-------
TABLE OF CONTENTS
ACKNOWLEDGMENT iii
LIST OF FIGURES vii
LIST OF TABLES ix
I. SUMMARY 1
A. Introduction 1
B. Results 1
C. Recommendations 3
II. BACKGROUND 5
III. CHEMICAL IDENTIFICATION OF"THE SMOKY-BURNT ODOR COMPLEX ... 11
A. Sample Preparation 11
B. Gas Chromatographic Separation and Resolution 13
C. Chemical Identification of Odor Significant Species ... 13
D. Studies on Reference Oxygenated Compounds 14
E. Discussion of Results 14
IV. QUANTITATIVE ANALYSIS METHODS 19
A. Exhaust Sample Collection 19
B. Liquid Chromatography 20
C. Total Organic Gas Chromatographic Analysis 20
D. Analysis of Chemical Classes 22
1. Aromatics Oily-Kerosene Odor Complex 22
2. Oxygenates Smoky-Burnt Odor Complex 23
Arthur D Little, Inc
-------
TABLE OF CONTENTS (con't)
V. VARIABLES STUDY 27
A. Engine Load .... 27
1. Exhaust Odor 27
2. Chemical Analysis of Exhaust Samples 30
3. Odor of Exhaust Samples 32
B. Fuels 35
1. Analytical and Odor Evaluation 35
2. Engine Study of Soltrol 200 40
VI. ODOR/CHEMICAL ANALYSIS CORRELATION 43
VII. REFERENCES 45
VIII. GLOSSARY 47
APPENDIX A - ODOR PROFILE TECHNIQUE 49
APPENDIX B - DIESEL EXHAUST SAMPLING SYSTEMS 55
APPENDIX C - LIQUID COLUMN CHROMATOGRAPHY (LCC)
SEPARATION PROCEDURES 61
APPENDIX D - SILICONS AND CARBOWAX CHROMATOGRAMS
OF SMOKY-BURNT FRACTIONS 69
APPENDIX E - STRUCTURE ASSIGNMENT METHODS 94
APPENDIX F - BASIC HIGH RESOLUTION MASS SPECTROMETRY-
ODOR DATA ON ODOR SIGNIFICANT SMOKY-BURNT SPECIES ... 110
APPENDIX G - SMOKY-BURNT STRUCTURE DATA ORGANIZED BY R+DB VALUES .. 127
APPENDIX H - OXYGENATED REFERENCE COMPOUNDS 139
vi
Arthur D Little, Inc
-------
LIST OF FIGURES
Figure Title
1 Experimental Arrangement of Diesel Engine,
Odor Test Room, Exhaust Sampling and
Collection System ,
2 Standard Diesel Exhaust Workup Procedure
3 Modified Smoky-Burnt Isolation Sequence
4 Procedure for Isolation and Characterization
of the Smoky-Burnt Fraction 12
vii
Arthur D Little, Inc
-------
LIST OF TABLES
Table Title Page
1 Smoky-Burnt Summary Observations 16
2 Reproducibility of Total Organic GC Method 21
3 Examples of an R+DB Analysis of the Smoky-Burnt
High Resolution Mass Spectral Data 25
4 Engine Operating Parameters Under Various Load
Conditions 28
5 Examination of Diesel Exhaust Odor at Various
Loads 29
6 Sample Distribution from Various Load Conditions
Silica Gel Trapped Exhaust 31
7 Aromatic Composition Analysis of Fractions from
Various Engine Loads 33
8 Test Room Odor Evaluations of Samples from
Variable Load 34
9 Physical/Chemical Characteristics of
Alternative Fuels 36
10 Aromatic Composition Analysis of Alternative
High Aromatic Fuels 37
11 Odor Test Room Evaluation of Fuels and
Aromatic Fractions 37
12 Summary of Soltrol 200 Engine Test Results 41
13 Comparison of Oily-Kerosene Composition and
Odor Intensity 44
IX
Arthur D Little, Inc
-------
I. SUMMARY
A. INTRODUCTION
This report represents the results of a continuing effort to
chemically identify the odor components of diesel exhaust. The study is
sponsored jointly by the Coordinating Research Council and the Environmen-
tal Protection Agency as part of their support of programs designed to
obtain a detailed understanding of diesel exhaust and its odor.
The basic methodology for characterizing and analyzing the diesel
exhaust odor was developed during the previous effort on this program. (1,2)
The approach developed involved sensory characterization as an integral
part of the study. The original exhaust odor was characterized as consist-
ing about equally of oily-kerosene and smoky-burnt odors. The experimental
approach for analysis of the odor species was to collect large volumes of
exhaust by concensation at 0°C, extract the organic portion of the exhaust
from the non-odorous portion by liquid column chromatography (LCC). The
odorous species separated into two major odor fractions with roughly equal
odor intensities oily-kerosene and smoky-burnt which represented the
original exhaust odor.
During the preceding year the principal components responsible for
the characteristic oily-kerosene portion of the exhaust odor were identi-
fied. Identification of the oily-kerosene odor complex was achieved by
resolving the LCC odor fraction using a two-stage gas chromatographic (GC)
method and final analysis of the resolved odor species by high resolution
mass spectrometry. Using the basic chemical and odor data achieved by
analysis of the sample as well as odor studies on selected reference
compounds, the chemical classes associated with the oily-kerosene odor
complex were found to be: alkyl benzenes, idans/tetralins, and indenes.
Additional sample resolution was required for application of the
same methodology used for the oily-kerosene LCC odor fraction to the
smoky-burnt fraction. The balance of the preceding year was devoted to
developing an improved gradient LCC method for the smoky-burnt odor
fraction.
B. RESULTS
During this past year, the details of the methodology required for
identification of the smoky-burnt odor fraction were completed and applied
to the analysis of this fraction. All of the odor-significant species in
this fraction have been identified. While several paraffinic oxidation
products were recognized as important odor contributors, the most important
smoky-burnt odor species are those associated with the partial oxidation
products of compounds found in the aromatic fraction of the diesel fuel.
-1-
ArthurD Little, Inc
-------
Of the species identified, the greatest contribution to the smoky-burnt
odor character appears to be from the higher molecular weight components
and those with multi-functional substitution. Feel factors (irritation,
pungency) are frequently associated with the lower molecular weight
members of a particular chemical class. Our analysis of this odor fraction
was aided by the study of a large number of oxygenated reference compounds,
many of which had appropriate odor character and intensity.
In summarizing our odor and chemical identification results for
the smoky-burnt odor complex we find that:
The smoky odor character is most consistently associated
with hydroxy and methoxy indanones with some contributions
from methyl and methoxy phenols.
Burnt odors are associated with furans and alkyl
benzaldehydes.
The oxidized oily character is usually ascribed to
alkenones, dienones, hydroxy cyclocarbonyls, and indanones.
4 Irritation factors seem most frequently to be associated
with the lower molecular weight phenols. Some benzeldehydes
and methoxy benzenes may also contribute to this sensation.
While some unsaturated aldehydes contribute to a portion
of the exhaust odor complex, the most abundant exhaust
aldehydes do not appear to contribute significantly.
Neither sulfur nor nitrogen containing species contribute
to the smoky-burnt odor complex. Although such species
were observed during portions of the analyses, none were
associated with exhaust odors.
The primary emphasis during the initial phase of the program was
on qualitative methods for the identification of specific resolved
odorous exhaust species. Emphasis during the balance of the year was
placed on development of quantitative methods for the measurement of the
identified odorous species. Research was also initiated on the influence
of fuel and engine variables on the exhaust odor chemistry. One purpose
of these studies is to verify the previous identification data by deter-
mining whether any new species are observed under conditions other than
the fixed operating conditions, for which the chemical odor data were
obtained. A second purpose is to develop a correlation by which one will
be able to express diesel exhaust odor by the measurement of selected
groups of chemical species.
Various steps of the sample work-up and analysis procedure have
been modified so that good reproducibility has been obtained in the
measurement of odorous exhaust fractions. A new sample collection
-2-
ArthurD Little, Inc.
-------
procedure which will allow us to work more efficiently with smaller samples
of exhaust is being evaluated. Quantitative analysis of each of the major
classes of odorous compounds in the oily-kerosene odor fraction has been
achieved by application of a computerized matrix analyses of the low re-
solution mass spectral data for that fraction. An analagous method for
representing the relative amount of each of the odorous chemical classes in
the smoky-burnt fraction is under study.
Engine and fuel variable studies have been initiated. Preliminary
studies show that the analytical fractions accurately reflect the observed
change in odor character and intensity with load. Further, some initial
data in dicate that measurement of the indan/tetralin group as a total
appropriately reflects the level of kerosene odor in the exhaust. Several
tentative fuels have been studied in the odor test room for selection of
the most appropriate ones for detailed study. An initial examination of
the exhaust from a very low aromatics content fuel is encouraging in terms
of the correlation between odor and chemical analysis.
C. RECOMMENDATIONS
1. Work on a complete quantitative analytical method for the
collection and analysis of diesel exhaust should be completed.
2. Analytical methods should be developed for the analysis of
of the aromatic (oily-kerosene) and oxygenated (smoky-burnt)
odor fractions in a manner which will reflect the correlation
between exhaust chemistry and odor.
3. Study the influence of fuel and engine variables on the nature
of the diesel exhaust in detail in order to verify the chemical
assignments, provide the variables for establishing a correla-
tion method, and aid in determining the origin of the odor
species.
-3-
ArthurD Little, Inc
-------
II. BACKGROUND
This section briefly reviews the approach developed during the
previous years of our research program.
The success of our program is attributed to an approach which
combines the sensory talents of odor chemists with the capabilities of
analytical chemists to produce a scheme for identifying the trace odor
components in the complex diesel exhaust.
The test facility used for these studies is shown schematically
in Figure 1. The engine used in these studies is a 71-series 4-cylinder
diesel.* For all of the chemical identification work, it has been
operated under constant conditions of 1800 rpm and 33% load from a large
supply of No. 1 diesel fuel. Fuel and load have been varied for the
specific study of these variables. The exhaust from the engine can be
passed into an adjacent aluminum-lined test room for sensory evaluation
or through a sampling system for 'collection analysis. A typical odor
description using the ADL odor profile method (See Appendix A) is:
TIA** 2
Oily 2
Burnt 2
Kerosene 1^
Eye irritation /
Nose irritation /
(600/1 dilution)
Thus, in terms of the overall perception of diesel exhaust odor,
the odor quality and intensity is comprised about equally of an oily-
kerosene and a smoky-burnt odor complex.
For that portion of the program concerned with chemical identifi-
cation of the odor components, our preferred sample collection approach is:
Exhaust
Aliquot
t
9
Hot
Particulate
Filter
k
9
Condenser
0°C
k
f
Silica
Gel
k
9
Pump
*Detroit Diesel Allison Division of General Motors Corporation, Model 4154N
**Total Intensity of Aroma
-5-
Arthur D Little Inc
-------
75 KW
GENERATOR
FUEL
MEASUREMENT
4-71 N
DIESEL
ENGINE
AIR SUPPLY AND
MEASUREMENT
WATER
ACOUSTIC WALL
FLEXIBLE
SEGMENT I
I
ENGINE EXHAUST
ANTECHAMBERJ
DOORS'
ODOR TEST ROOM
CIRCULATING
^ FANS
INJECTION PQRTS
4-
POST MUFFLER SAMPLING POINTS
CONDENSERS
AND SAMPLE
COLLECTION
SILICA
GEL
EXHAUST
TO
ROOF
Figure 1 EXPERIMENTAL ARRANGEMENT OF DIESEL ENGINE, ODOR TEST ROOM,
EXHAUST SAMPLING AND COLLECTION SYSTEM
D
c:
-------
Large volumes of the exhaust sample are passed through unheated
^-inch stainless steel pipe, a heated fiber glass filter, and a condenser
kept at 0°C. All condensate is collected in a round bottom flask and
kept at 0°C. The remaining gases are swept through a silica gel column.
Initially, the oirginal system utilizing two Friedrich condensers in series
and a diaphragm pump was used to give a sampling rate of about 1000 liters/
hour. Later, a high volume sampling system using a single condenser and
carbon vane pump was assembled to provide a 10,000 liter/hour sampling rate
(See Appendix B.I). Using either method, condensate from 20,000 - 60,000
liters of exhaust was usually collected for the generation of analytical
samples.
The standard basic sample workup procedure used for the identifica-
tion of the oily-kerosene odor fraction is shown in Figure 2 and represents
the "normal" distribution of the entire sample. Detailed modifications of
this procedure have been made to fit the particular needs of the program,
but the procedure still properly reflects the entire process for identifi-
cation of individual exhaust odor species.
The oil phase which separates from large volumes of exhaust conden-
sate is combined with the pentane and chloroform extracts to provide the
concentrated organic portion of the diesel exhaust. This organic concen-
trate is first separated into major fractions by silica-gel/liquid-column
chromatography (LCC) yielding a major non-odorous fraction (LCC-1) and two
major odorous fractions - oily-kerosene (fraction LCC-4) and smoky-burnt
(fraction LCC-10) (see Appendix C for details of the LCC step). Other
non-odorous species in these fractions are separated, and the remaining
odorous species are resolved into individual chemical species by two stages
of gas chromatography (GC). The first stage uses a silicone column and the
second a Carbowax column. The chemical structure of odorous compounds is
determined by high resolution mass spectrometry (HRMS). This procedure was
used as described for the identification of the oily-kerosene odor species
reported in the last final report (2).
The standard LCC procedure as described did not provide sufficient
sample resolution for analysis of the smoky-burnt odor complex by the two
stage GC-HRMS system. Therefore, a modified sample workup procedure was
developed for identification of the odor species in this fraction. The
essential changes have been to work up only the chloroform and pentane
extracts of the aqueous condensate independently and separate the odor
fractions utilizing a more detailed LCC gradient elution. These modifica-
tions are shown schematically in Figure 3. Using the modified procedure
a gain of a factor of 5-10 in the odor/mass ratio was obtained in the
eluted smoky-burnt odor fractions compared with the standard procedure are
given in Appendix C.2.
Throughout the program, we have depended on the odor profile
technique to determine whether samples of diesel exhaust maintained their
odor character after each analytical step.
-7-
ArthurD Little, Inc
-------
Exhaust
Condensate
ODOR FRACTIONS
Oily-Kerosene = LCC-4
Smoky-Burnt = LCC -10
OV-1 Silicone
Gas Chromatography
i Odor Regions
Carbowax
Gas Chromatography
High Resolution
Mass Spectrometry
FIGURE 2 STANDARD DIESEL EXHAUST WORKUP PROCEDURE
-8-
ArthurD Little Inc
-------
Exhaust
Aqueous Condensate
Pentane
Pentane Extract
I
LCC
Micro Column
Gradient Elution
Smoky-Burnt
Fractions
Aqueous Phase
Chloroform
Chloroform
Extract
Water
Discard
FIGURE 3 MODIFIED SMOKY-BURNT ISOLATION SEQUENCE
-9-
ArthurD Little Inc.
-------
III. CHEMICAL IDENTIFICATION OF THE SMOKY-BURNT ODOR COMPLEX
The basic scheme described in "Background" using the modified
separation procedures has been applied to an identification of the major
species contributing to the odor of the smoky-burnt diesel exhaust fraction.
A. SAMPLE PREPARATION
Four separate collections of large volumes of diesel exhaust were
made to complete this portion of the project. Samples 51 and 52 (45,000
liters each) were collected with the original 1000 liter/hour trapping
system. Samples 56 and 60 (50,000 liters each) were collected with the
new high volume sampler designed to increase the experimental efficiency
(see Appendix B.I for details). The modified sequence used for identifica-
tion of the smoky-burnt odor species is represented completely in Figure 4.
Earlier work had shown that virtually all of the oily-kerosene
odor species in the exhaust condensate were extracted by the pentane and
chloroform extracts. Because of its more favorable odor-to-mass ratio,
identification of the smoky-burnt odor species was completed by working
primarily on the chloroform extract of the condensate. Sufficient research
was done on the pentane extract portion to obtain the contribution of the
odorous compounds present in that extract to the total smoky-burnt exhaust
odor.
The chloroform extracts from each of the sample collections and the
pentane extract from sample 56 were subjected to the gradient elution LCC
procedure described in Appendix C.2. The identification results are thus
based on the analysis of four separate exhaust sample, namely:
Chloroform Extracts
Sample 51C fractions 34/35
Sample 52C fractions 34/35/36 + sample 56C fractions 39/40
Sample 60C fractions 39/40/41
Pentane Extract
Sample 56P fraction 40
These samples appeared to be consistent with all of our earlier
observations, and we feel that they accurately reflected the important
smoky-burnt odor fractions of the original diesel exhaust.
-11-
ArthurDLittldnc
-------
Diesel Exhaust
Condensate
Pentane Extract
Water Phase
For Sample 56
Only Follow
Same Route as
Chloroform Extract
Chloroform
Extract
Gradient Liquid
Chromatography
on Silica
Smoky-Burnt
Fractions
GC on Silicone
(OV-1)
_0dor
"Profile
Trapped Silicone
Peaks
GC on Carbowax
(SP-1000)
High Resolution
MS
Odor Profile
FIGURE 4 PROCEDURE FOR ISOLATION AND CHARACTERIZATION
OF THE SMOKY-BURNT FRACTION
-12-
Arthur D Little Inc
-------
B. GAS CHROMATOGRAPHIC SEPARATION AND RESOLUTION
Our previously developed techniques were used for the final isola-
tion, odor characterization and chemical identification of individual
smoky-burnt odor species (see Ref. 2). Thus, odor profiles were determined
on the samples described above after gas chromatographic resolution on an
OV-1 silicone column. The most significant odor areas were defined and
then designated for trapping and further resolution. Trapping of the
selected areas was done using short lengths (1/8" x 5") of stainless steel
tubing containing OV-1 column packing at room temperature. Then, the trapped
areas were rechromatographed on an SP-1000 Carbowax column (Supelco Company).
The SP-1000 column is a significant improvement over previous Carbowax
columns allowing components to elute at about a 50°C lower temperature and
showing four to five times less column bleed at 200 - 250°C.
Initial studies on Sample 51C were done using one aliquot for GC-
odor and a second for GC-MS. The experimental procedure was then modified
to improve reliability and sample efficiency and the remainder of the
studies were carried out with the SP-1000 GC effluent split three ways to
allow simultanelus GC-Odor-HRMS studies. All of the pertinent chromatograms
that we obtained during this work have been included in Appendix D. The
silicone chromatograms for each sample appear first, followed by the Carbowax
chromatograms of each trapped peak.
C. CHEMICAL IDENTIFICATION OF ODOR SIGNIFICANT SPECIES
High resolution mass spectra (HRMS) were obtained on each of the
species eluting from the Carbowax column which were considered to be signi-
ficant odor contributors. These data provided the basic information for
assignment of chemical structures to specific exhaust odor compounds.
However, wherever possible the structural assignments were also made by
correlations developed by the study of reference compounds and full appli-
cation of all of the available odor and gas chromatographic retention data.
A detailed discussion of the methodology used is presented in Appendix E.
The basic data derived from the studies are summarized in Appendix F. These
data include reference to the original silicone peak which was trapped and
its nominal OV-1 elution temperature, the Carbowax temperature program
conditions used, and elution temperature, the photoplate exposure number
(peak number), the odor descriptors, and the HRMS data for the parent mole-
cular ion.
In addition, the approximate percentage of the individual components
in the total samples have been estimated qualitatively along with possible
structure type assignments for most of the observed species. In a few cases
chemical structural assignments are definite (where indicated), but for most
of the species the structures assigned represent our best present estimate
of the probable chemical class.
As described in Appendix E, one of the most useful ways we have
found for organizing these data is by means of the rings plus double bonds
(R+DB) classification system (3), a means of representing the structure of
-13-
ArthurD Little, Inc
-------
a compound by means of its hydrogen unsaturation. Appendix G presents the
data for the smoky-burnt fraction according to an organization based on
the R+DB values observed for each identified odor species.
D. STUDIES ON REFERENCE OXYGENATED COMPOUNDS
The high degree of similarity in the basic fragmentation behavior
of some of the phenols, benzaldehydes, indanones, and indanols in particular
has made it difficult to clearly distinguish between them on the basis of
the mass spectrometry data alone. Therefore, the attempts to assign
structure types have been considerably aided by a study of the GC behavior
of reference compounds on the silicone and Carbowax columns. The retention
data on the 144 oxygenated reference materials studied to date are listed
in Table H-l of Appendix H. The GC data have been extremely useful in
placing limits on structural considerations and in establishing correlations
between structure types.
Several of the reference compounds were also studied in the odor
test room (Table H-2, Appendix H) for further confirmation of the consistency
of the chemical structure/odor assignments. The phenol-based materials
tend to smell medicinal, but as alkyl substituents are added they appear to
be more reminiscent of odors observed in diesel exhaust. Several of the
benzaldehydes and indanones (tetralones) have odors consistent with portions
of the smoky-burnt odor complex. The concentrations at which the model
compounds are detectable in the test room are consistent with the concen-
trations of compounds observed in the diesel exhaust fraction.
E. DISCUSSION OF RESULTS
Using the program described above, we have accumulated a wealth of
data odor evaluation, gas chromatographic retention behavior, and high
resolution mass spectrometric composition on the smoky-burnt odor complex.
This section will present the important highlights of the work, while the
specific odor composition details can be found primarily in Appendices F
and G.
In attempting to describe the state of our knowledge, it is impor-
tant to stress that there are several levels of confidence that can be
assigned to various segments of the data. We feel very strongly that we
have developed a basic understanding of the types of compounds in diesel
exhaust which cause the smoky-burnt character of the total odor. In some
cases, we also have identified specific compounds within a compound class.
However, most of the specific compound identification assignments are
tentative. We intend to continue firming up these individual assignments,
but we do not feel that changes in identification of specific chemicals
will have any significant impact on the course of the total program beyond
adjusting our thinking in quantitative terms. Therefore, we are proceeding
on to other phases of the study and continuing the identification studies
at a low level of effort.
-14-
ArthurD Little; Inc
-------
Based on the original diesel exhaust odor characteristics (oily,
kerosene, smoky-burnt, and irritation factors) our primary concern has
been to identify components which are recognizable in the exhaust as smoky
character notes and secondarily as burnt character notes. There is some
interest in oily or oxidized oily notes as well as sour, leather, linseed,
and naphthenate descriptors. Each of these are detectable at varying times
in some of the liquid column chromatography fractions. In addition, the
added resolution obtained by the first and second stages of gas chromato-
graphy permits one to recognize a great variety of odor characteristics.
The latter problem taxes the observers' descriptive language and confuses
his ability to recognize characteristics because of the variety of rapidly
appearing and changing odors. The goal of the odor analyst during this
phase of the work is to recognize and indicate only those odor components
which are most importantly associated with the original odor profile.
We have found in this program that many of the same types of chemical
structures appear to relate to different aspects of the odor. Some of this
redundancy undoubtedly is real but considering the difficulties facing
the odor chemist some may represent interference during the odor examination..
The chemical and odor observations on the smoky-burnt fraction are
summarized in Table 1. This table lists the approximate percentage for
which' each structure class was observed in the chloroform (C) and pentane (P)
extracts, as well as showing that the total amount of the measured species
account for approximately 20% of each of the extracts. We believe that much
of the remainder of the extract is represented by similar materials which we
did not happen to measure. In addition, there undoubtedly are also several
types of species, such as some residual alkyl benzenes and naphthalenes
which account for some of the mass but do not contribute to the odor. This
list provides the major structure classes, which are primarily alkyl-
substituted species. The generic name also includes the hydroxy and methoxy
derivatives of these classes. The "carbon range" column indicates the range
of carbon numbers which we observed for each of the classes.
The odor intensity and character notes of the chloroform and pentane
extracts were found to be different. In the chloroform extract, a complex
of smoky-burnt is the major odor. This is supported by character notes des-
cribed as oxidized oily, sour and naphthenate, and detectable levels of a
sensory impression of irritation or pain. In the comparable pentane extract,
the odor mixture is more complex and consists of kerosene-related odors,
which are dominant, with supporting notes of oily and burnt smoky.
In conjunction with these odor differences, a comparison of the
chemical data obtained from the chloroform and pentane extracts revealed that
the initial exhaust condensate extraction process was somewhat selective.
There is a greater abundance of the more polar species in the chloroform
sample, such as the phenols and hydroxy/methoxy indanones, while the less
polar materials such as the dienones are found preferentially in the pentane
extract. Even in cases where phenols, for instance, were isolated in both
fractions, the pentane extract tends to contain the less polar homologs as
indicated by their earlier elution from the Carbowax GC column (see
appendix F for details).
-15-
ArthurD Little Inc
-------
TABLE 1
yb
/o
C P
1.0 1.1
1.0
1.4
0.9
0.7 0.1
4.9 1.0
2.7 6.8
1.6
6.1 2.2
Smoky-Burnt Summary
Structure Class3 R + DB
Alkenone 2
Furan 3
Dienone 3
Furfural 4
Methoxy benzene 4
!
Phenol 4
Benzaldehyde 5
Phenyl ketone
Benzofuran 6
Indanone 6
Observations
C Range Principal Odor Contribution
C5-CU Oxidized oily
CG~CIO Irritation, burnt
Cg-Ci3 Sour, oxidized oily
Cg-C7 Burnt, oily
Ce-Cg Smoky, pungency
C?-Cl2 Burnt, irritation, tarry,
particle size
C7-C13 Burnt, pungency
^8~^9 Particle size
Cg-C13 Metallic, smoky, sour
(plus indenols)
1.0 1.6 Indenone
Leathery, tarry, burnt
0.4 4.9 Naphthaldehyde
Cjj-Cjj Particle size
a. includes hydroxy and methoxy derivatives, most with alkyl
substitution.
b. percent of each class in the chloroform (C) and pentane (P)
extract
c. see Appendix E f or discussion of R + DB.
-16-
ArthurDLittleJnc
-------
Summarizing this another way:
Hydroxy and methoxy indanones are most consistently
described as smoky, while aethyl and methoxy phenols
may also contribute to this character note.
Burnt odors are associated with components described
as furans and possibly alkyl benzaldehydes and aceto-
phenones.
The oxidized oily character note is usually ascribed
to alkenones, dienones, hydroxy cyclocarbonyls and
indanones.
Irritation seems to be most frequently associated with
the lower molecular weight phenols. Some of the alkyl
benzaldehydes and methoxy benzenes may also contribute
to this sensation.
Odor observations consistently seem to indicate that
the feel factors (pungency, irritation) are associated
more with the lower molecular weight members of a
class, and the odor character is associated with the
more highly substituted or more poly functional
derivatives.
The HEMS data acquired for all of the odor significant peaks has
also been examined in an effort to determine whether sulfur or nitrogen-
containing compounds are present in any of the areas we have examined.
We were unable to find any such species in the data. We consider this
observation to be significant in view of the fact that we were able to
detect species present in the exhaust at about 1 ppb and based on odor
threshold data would have observed the species if they had contributed
to the odor.
Finally, several quantitative studies were carried out with refer-
ence compounds to determine whether any major odor components of the
smoky-burnt complex could have been lost in the liquid column and gas
chromatographic steps. The studies show good material recoveries and
indicate that it is unlikely that any important odor species were selectively
lost in the procedures.
-17-
ArthurD Little, Inc
-------
IV. QUANTITATIVE ANALYSIS METHODS
The next phase of our research program involves a study of fuel and
engine variables. The purpose of these studies is to determine the complete-
ness of our chemical structure/odor assignments to date and to provide the
variation in exhaust data by which correlation schemes between composition
data and odor may be evaluated. The analytical approach used for the iden-
tification phase is usable for initiating these studies, but the approach is
complex and time-consuming and not designed primarily for repetitive use.
We have, therefore, initiated a study to develop a simpler ( but not
a "Black Box") quantitative method which will be appropriate for use first
in our own laboratories for expediting the variables study and ultimately
in other research laboratories. Each step in the procedure is being studied
in detail. This portion of the study is still in progress and results in
some areas are still in the tentative stage.
The overall scheme under study involves
Exhaust Sample ^
Collection and Isolation
Liquid
Chromatography
Aromatic
(Oily-Kerosene)
and
Oxygenated
(Smoky-Burnt)
Fractions
and basically eliminates the gas chromatography steps for the final analysis.
The scheme continues to rely upon liquid chromatography as the most effective
means of separating the odorous and non-odorous compounds and providing odor
fractions which are directly amenable to analysis by mass spectrometry.
Individual studies to date on each of these steps are discussed in the
following sections.
A. EXHAUST SAMPLE COLLECTION
Although the condenser systems used for the study to date are effec-
tive in obtaining a representative portion of the diesel exhaust, we have
known for some time that the exhaust odor is not collected quantitatively by
that method. We had observed earlier that a bed of silica gel placed after
the condenser was effective in removing all of the diesel exhaust odor.
We have begun to explore means of collecting samples by direct
adsorption on substrates such a silica gel. Our studies to date on silica
gel itself are quite encouraging and suggest that this approach will
ultimately provide the preferred method. The method is described in detail
in Appendix B.2.
-19-
ArthurD Little, Inc
-------
Approximately 720 liter samples of diesel exhaust have been collected
directly on 25 grams of silica gel over a one-hour collection period. We
have found that the odorous exhaust species can be effectively removed
from the silica gel by a pentane/acidic methanol elution.
For the 33% load engine condition, the total organic extract (TOE)
obtained by this method was approximately 250-300 mg/Kfc of exhaust. These
data contrast with an average 15-25 mg/K& obtained with the condenser
system. Thus, an order of magnitude gain has been made in sample collection
efficiency. Significantly, the total extract from the gel trapping method,
when examined in the odor test room at an equivalent of 21 liters of exhaust,
has an odor intensity and quality which is nearly the same as a direct
21 liter of exhaust sample.
This new collection procedure is much more efficient than the con-
denser systems as well as meeting the quantitative requirements of the over-
all program. Further, since the odor fractions will initially be analyzed
as a total group only, 500-1000 liter samples of exhaust are sufficient.
B. LIQUID CHROMATOGRAPHY
A micro version of the standard liquid chromatography procedure has
been developed consistent with the needs for analyzing the 1000 liter silica
gel trapped samples. The elution sequence has been simplified and established
so that the odor-significant fractions still elute in the LCC-4 (oily-
kerosene aromatics) and LCC-10 (smoky-burnt oxygenates) fractions, as they
did under the original procedure. The procedure is detailed in Appendix C.3.
C. TOTAL ORGANIC GAS CHROMATOGRAPHIC ANALYSIS
The basic preliminary analytical data we wished to obtain was the
total organic mass found in the solvent extracts and the LCC fractions. We
had previously obtained these data from the temperature-programmed GC
analysis on the ten-foot silicone column. This method had the problem that
it was time-consuming for the data desired and difficult to measure the
eluting area since the base line was sometimes difficult to define. There-
fore, a new assay procedure was developed using a short column and rapid
temperature programming.
The column is a IV x 1/8" stainless steel column packed with the
same 10% OV-1 used previously and used in our P-E 900 with FID detection.
The column is maintained at ambient temperature for three minutes to allow
the solvent to elute and then heated ballistically to 250°C over a three-
minute period. During this heating period, the sample elutes as a relatively
unresolved peak which is well defined and easy to measure quantitatively.
All total mass data have since been obtained by this method. The reproduci-
bility of the assay is acceptable, as can be seen in the data shown in
Table 2, which was obtained by repeated analysis of the same samples on the
dates shown.
-20-
ArthurD Little, Inc
-------
Table 2
Date
12/18/70
12/23/70
12/29/70
1/5/71
1/14/71
Average
Reproducibility
Fuel Oil
Response
(sq. in/pg)
27.1
30.7
26.6
27.1
24.5
27.2±3.5
of Total
Pentane
Extract
(62-P)
50.4
45.7
52.8
49.4
_
49.6±4
Organic GC Method
a
LCC-1
(64-LCC-l)
9.1
8.0
11.1
8.6
11.5
.0 9.711.8
-i <->____
LCC-10
(64-LCC-10)
1.40
1.40
1.34
1.05
1.24
1.29±0.24
a. mg/1000 £ as FOE values
-21-
ArthurD Little, Inc
-------
At the present time we are still using diesel fuel to calibrate
the flame response to obtain weights (as FOE, Fuel Oil Equivalent) of all
the silica fractions. If an appropriate mixture can be found, we may use
separate calibration mixtures for the paraffin, aromatic, and oxygenate
fractions. The error, however, introduced by using fuel for all fractions
is minor.
D. ANALYSIS OF CHEMICAL CLASSES
1. Aromatics Oily-Kerosene Odor Complex
According to the chemical identification/odor data described in
the second final report (2), a measurement of the indans and tetralins
present in the exhaust samples should provide a measure of the kerosene
odor note while the alkyl benzene concentration should reflect the oily
note.
The most efficient means of obtaining those data at the present
time is by means of a low resolution mass spectral analysis of the LCC
aromatic fraction. The petroleum industry has developed computer programs
for the matrix analysis of such aromatic fractions and the version developed
by the Mobil Research and Development Corporation was kindly supplied to
us for our studies. The Fortran program has been converted to process data
on our laboratory Hewlett-Packard 2116B computer. A typical analysis of a
condensate LCC-4 oily-kerosene aromatics fraction derived from the No. 1
diesel fuel looks like:
Sample: Experiment 66 LCC-4
Class
Alkyl benzenes
Tetralins, indanes, indenes
Naphthalenes
Acenaphthenes, fluorenes, etc.
Phenanthrenes, anthracenes, etc.
The exhaust and fuel samples from the No. 1 diesel fuel do not
normally have a significant amount of the last two classes. These classes
are important, however, in higher boiling fuels and may reflect additional
odor character notes.
This form of data analysis basically provides a fairly rapid means
of estimating the idan/tetralin concentration in odor samples to determine
the degree of correlation between composition and odor. We have just begun
to examine the correlation data and will need to study more samples derived
from various load and fuel conditions to fully assess the appropriateness
of this approach.
-22-
ArthurD Little; Inc
-------
2. Oxygenates Smoky-Burnt Odor Complex
It is much more difficult to begin to establish an assay method
for the smoky-burnt odor than the oily-kerosene odor species for several
reasons. First, there appear to be five to seven general compound classes
we wish to measure out of a total of about twelve versus two out of
three for the aromatics, and many of these have overlapping group function-
ality. Secondly, we have not yet progressed to the point that we know with
the certainty that was established for the oily-kerosene species just what
classes we primarily wish to measure. Finally, the detailed isomeric and
functional group choices of many species have not yet been verified, and
the compounds are only known with certainty by their mass spectral and
chromatographic data. For these reasons and the reason of efficiency in
using the same technique for analysis of both odor fractions, we have begun
to explore in a preliminary way the suitability of mass spectrometry for
analysis of the smoky-burnt odor classes.
Examination of the odor-chemical structure data indicates a prelim-
inary first choice of wishing to measure alkenones and furans (R+DB 2,3)
for oxidized oily, and phenols (R+DB 4), benzaldehydes (R+DB 5), and
indanones and tetralones (R+DB 6) for the burnt and/or smoky odor components
of the LCC-10 exhaust fraction. We have examined both the high and low
resolution mass spectra of LCC-10 fractions to determine whether data
reflecting the concentration of these classes could be obtained in a manner
similar to the aromatics program.
Upon examination of the data we find that the intensity of many of
the significant mass values for these species in the low resolution spectra
is contributed to substantially by the spectrum of ions with an aromatic
composition, and there is a great deal of fragmentation leading to similar-
ities from different chemical classes. A typical example is the distribu-
tion of compositions found by high resolution at mass 132, the parent ion
mass of indanone:
Relative
Intensity Precise Mass R+DB Composition
20 132.0452 2
60 132.0560 6 C9H80
20 132.0808 1 C6H1203
50 132.0939 5 C10H12
We have also examined the spectra at low ionizing voltage (15 ev,
compared to the normal 70 ev) in the hopes of enhancing the more easily
ionized oxygenated species. Although the overall degree of framentation
has decreased somewhat under these conditions, and the spectrum somewhat
simplified, the results still did not improve sufficiently to enable us
to use the low resolution approach.
-23-
ArthurD Little Inc
-------
In the course of this analysis, however, it did become apparent
that the complete high resolution mass spectra contained all the infor-
mation one needs for this interim analysis period. The data provides the
potential for highly refined compound class analysis to test various
chemical class-odor correlations through application of proper computer-
ized manipulation of the data. A small effort has been put into some
preliminary computations, and the results and future potential appear very
attractive.
For instance, since the chemical composition data are completely
resolved by virtue of the exact mass differences between various composi-
tions, it is a simple matter to instruct a compilation of all desired
oxygenated species. Similarly, one can organize particular data in various
formats to explore the potential of conducting the analysis in that manner.
In some preliminary attempts, the smoky-burnt oxygenated fraction
has been organized by R+DB value to list all appropriate molecular ions
based on the identification data. The compilation routine was instructed
to exclude complicating data such as hydrocarbons, 13C isotope peaks,
fragment ions (odd mass), and all ions less than mass 94, the lowest mole-
cular weight observed in any of the identified exhaust species. Table 3
shows examples of the output obtained for three of the odor significant
R+DB classes from an exhaust condensate LCC-10 smoky-burnt sample.
The tables list the intensity (HGT) of each ion relative to the
most abundant ion in the entire spectrum, the precise mass (DET. MASS),
calculation error in millimass units (0.3 = 0.0003 mass units), the R+DB
value, a value X to be ignored, and the elemental composition.
The R+DB 2 (DB = 2) class represents the alkenones and hydroxy
alkenones associated with the oxidized oily odor note, while the R+DB 4
class represents primarily phenols. The R+DB 6 class represents indanones
associated with the smoky and burnt odors.
After examining several sets of data in this manner, the potential
looks very good for obtaining detailed analysis of the smoky-burnt mixture
in this manner with a minimum investment in effort.
As the odor correlation develops and we learn more precisely what
chemical classes we wish to measure, it may be appropriate to develop
alternative means of more directly measuring those species. It certainly
is a requirement of a simpler analytical method to be used in other
laboratories. However, at the present time this approach appears to offer
the maximum potential for obtaining detailed quantitative data from a very
complex sample based on the sample chemistry. Further refinements will be
made in the analysis scheme as the rationale develops.
-24-
ArthurD Little Inc
-------
Table 3
Examples of an R+DB Analysis of the Smoky-Burnt
High Resolution Mass Spectral Data
ANALYSIS FOR
HOT
OUT. MASS ERROR R+D3 X
C12CU H
5
2
3
2
2
1
1
1
1
1
1
1
93
00
12
14
126.
1
23
140.
1
1
42
56
07135
05058
08746
06662
10253
08229
11868
09842
11438
-1
-1
-1
-1
-1
-1
-1
-0
n
*
.
.
.
.
.
82
S'*
36
45
93
44
44
96
65
2
2
2
2
2
2
2
2
2
»
00
00
00
00
00
00
00
00
00
2
2
2
2
2
2
2
2
2
fc
r;
7
6
8
7
9
8
9
0
U
0
0
0
0
0
0
0
10
8
12
10
14
12
16
14
16
1
2
I
L.
1
2
1
7
9
SUM HGT =
18
ANALYSIS FOR OR=
HGT
DET. MASS ERROR R+DB X
C12C13 H
SUM :1GT =
6
3
7
5
6
4
1
3
3
1
2
2
1.
1
1
1
1
1
1
1
1
94.
96.
OB.
10.
122.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
24.
26.
36.
3d.
40.
50.
52.
54.
64.
66.
7B.
80.
92.
94.
04069
01967
05635
03606
07203
G5181
03078
08742
06667
04692
10326
08291
06278
11939
09858
13464
11450
151C8
13053
-1.
-1.
-1.
-0.
-1.
-0.
-G.
-1.
-1.
-0.
-1.
-0.
-0.
-0.
-c.
-1.
-c.
-0.
-0.
18
46
16
70
14
o2
91
40
41
42
20
81
21
7?
80
12
52
34
15
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4.00
4 . 00
4.00
4. 0'.?
4.00
4.00
4.0'.-)
6
6
6
6
6
6
6
6
6
6
6
6
6
6
o
0
6
6
6
6
5
7
o
8
7
6
Q
8
7
*, ~\
}
:j
>. 1
i -'
I. <-
II
1 3
1 ">
^ C
0
0
0
0
Q
0
0
0
0
.-\
'.'
0
o
!\
o
{':
-\
0
0
;*
6
4
6
6
10
8
6
12
10
8
i1-*
12
10
in
;4
IS
lo
50
-25-
ArthurD Little, Inc
-------
ANALYSIS FOR 03=
Table 3 (cont.)
HGT
DET. MASS ERROR R4-DB X
C12C13 H
3
6
1
8
5
1
q
8
3
3
4
2
2
2
1
1
1
1
104.
L
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
18.
20.
32.
34.
36.
46.
43.
50.
60.
62.
64.
74.
76.
73.
88.
90.
202.
02553
04173
02079
05617
03526
01536
07152
05127
03085
0-83C3
06723
04679
10380
C8301
06112
11954
09378
13531
-0.
-0.
-0.
-1.
-1.
-0.
-1.
-1.
-0.
-0.
-0.
-0.
-o.
-0.
-I.
-0.
-0.
-0.
68
13
34
34
52
68
65
16
84
79
85
55
67
71
88
58
60
45
6.00
6.00
6.00
6.00
6.00
6.00
6.00
6.00
6.00
6.00
6.00
6.00
6.00
6. 00
6.00
6.00
6.00
6.00
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
7
8
7
9
6
7
1C
q
3
1 1
1C
9
12
li
1C
*. 3
i 2
1.4
0
0
0
0
0
0
0
0
0
T
0
0
0
0
Q
o
0
0
4
fc
4
8
6
4
10
8
6
12
10
e
14
12
10
16
14
18
1
i^
?.
1
t_
3
1
2
3
1
2
3
1
2
3
^
X
(r_
i
SUV HGT =
67
-26-
ArthurD Little, Inc
-------
V. VARIABLES STUDY
The study of engine and fuel variables became a major portion of
our research effort during the last part of the program. The primary
objectives of these studies are to develop the chemical class-odor correla-
tion necessary for instrumental measurement of diesel exhaust odor and to
verify and determine the completeness of the identification data obtained
to date under our fixed operating conditions.
These studies are still in the preliminary stages. Out initial
objectives have been to examine the exhaust and analytical fractions derived
from the exhaust under the variable conditions to determine how the exhaust
odor is reflected in the separated odor fractions, both qualitatively and
quantitatively. Then each primary odor fraction will be analyzed according
to the procedures described in Section IV D. Our first effort will be to
determine the quantitative correlation between chemical composition and
odor in the resolved fractions. Once this correlation is achieved, an
effort will be made to establish the same type of correlation in the total
organic extract of the diesel exhaust.
A. ENGINE LOAD
1. Exhaust Odor
The effect of engine load on exhaust odor has been studied in
detail by examining the exhaust produced with the engine at 10, 33, and 90%
load operating at a constant 1800 rpm, using the No. 1 diesel fuel. Some
relevant experimental engine operating parameters for these load conditions
are given in Table 4.
Odor profile composites for the three load conditions (10%, 33%
and 90%) appear in Table 5. These profile descriptions represent a summary
of the results of numerous studies at each of the load conditions. The
terminology is selected to best represent the consensus of the panel des-
cription as well as to differentiate the character of the three load
conditions.
The 33% load used for all of the former chemical identification
studies has a moderate total intensity of aroma, which is easily recognized
in the test room and described in general as diesel exhaust. The primary
characteristic as previously noted is the smoky-burnt character note at a
moderate intensity. This is supported by an oily odor described as oxidized.
The kerosene odor note is the third descriptor in the profile which, from
our previous work, was shown to be associated with aromatic hydrocarbons.
This note appears to vary somewhat from sample to sample in intensity, either
because of concentration or as a result of odor blending. The feeling sen-
sations are primarily nose irritation (a stinging or pain sensation in the
nose) and some slight eye irritation.
-27-
ArthurD Little, Inc
-------
Table 4
Engine Operating Parameters Under Various Load Conditions'
Load
(%)
Exhaust
Temperature
(OF)
Fuel
Consumption
(Kg/hr)
Collected
water
ml/lOOOa
10
33
90
290
380
615
5.5
8.2
.4.6
4.7
15
29
a. No. 1 Diesel Fuel, 1800 rpm.
b. Average amount of water collected/1000£ of exhaust
by high volume condenser sampling systems.
-28-
ArthurD Little; Inc
-------
TABLE 5
EXAMINATION OF DIESEL EXHAUST ODOR AT VARIOUS LOADS
10% Load
33% Load
90% Load
TLA 2
Burnt-smoky 2
Sour oxidized oil 1*2-2
Kerosene
Sooty particle
feel
Eye irritation
Nose irritation
TLA
Smoky-burnt
Oily (oxidized)
Kerosene
Nose irritation
SI. eye irritation
TIA
Smoky-tarry
Hot oily
Metallic (acrid)
Nose irritation
Headache
2
2
a. 20 liter samples
b. // = more intense than /
-29-
ArthurD Little Inc
-------
In comparison at 10% load, although the total intensity of aroma
is still moderate, a sour oxidized oily character note appears to be of
primary importance. The smoky-related character note which remains as the
most intense odor is somewhat sweeter and described as burnt-smoky to
contrast with smoky-burnt. The kerosene odor notes are less apparent than
at 33% load, varying between a just perceptible and a slight intensity.
In addition to the sweetness of the burnt-smoky descriptor, there is a
sooty, almost particular, feel which appears to be characteristic of the
10% load. This is evident both in filtered and nonfiltered exhaust samples.
Irritation appears to affect the eyes as much or more than the nose under
this load condition.
With 90% load, although again the total intensity of aroma is
moderate, it seems to be fuller and heavier, possibly associated with
higher molecular weight oxygenated compounds. The smoky character note is
further described as tarry to indicate this increased heaviness and fullness.
Under these conditions, the oily aroma is qualitatively quite different from
the 10% load and is described as hot lube oil rather than as sour or oxidized.
This appears to be consistent with the change observed by the increasing load
situation noted from 10% to 33% load. At 90% load, the panel does not
describe kerosene or fuel-related odors, but the term "hot metal", "stove
pipe" or "metallic-sharpness" is noted which may relate to an acrid or
pungent character of the exhaust.
The nose irritation at 90% load appears to be the highest of the
three conditions examined. There are in addition occasional references to
headache produced with 90% load condition which is not generally recognized
at the lower load conditions. This is often perceived as pain just behind
the eyes.
2. Chemical Analysis of Exhaust Samples
Several initial sample collections and fraction odor evaluations
were made at the three load conditions using the condenser sampling system
in order to initiate our studies in this area. These studies were described
in the third quarterly progress report. Since the quantitative aspects of
the silica gel trapping system (Appendix B.2.) were so much improved over
the condenser system, our emphasis has since been on evaluation of data
collected using this exhaust collection system. The collection results of
several samples from the three load conditions are given in Table 6.
The reproducibility in the total organics (TOE) collected at the
10 and 33% load conditions is good. However, the 90% TOE values vary
considerably. The greatest amount of water is generated at the high load
condition and we have noticed an inverse dependence on the TOE values and
the amount of water collected on the silica gel. Efforts are currently in
progress to minimize this effect. There is still a fair spread in the
LCC-4 and LCC-10 values for any given load condition and studies are under
way to determine the cause of this variation. Except for the obvious
variation in TOE values, the effect of load on the fraction distribution
-30-
ArthurD Little Inc
-------
Table 6
Sample Code
Gel 3
Gel 4
Gel 9
Gel 5
Gel 6
Gel 10
Gel 7
Gel 8
Gel 11
Sample
T naH
10%
10%
10%
33%
33%
33%
90%
90%
90%
Distribution
Silica Gel
TOF
230
230
300
250
240
290
50
100
150
from Various Load
Trapped Exhaust
mg/K£ present
____ T r*p T
LCC-1
160
220
126
207
219
169
32.3
65.9
82.1
Conditions
in
b
LCC-4
4.2
6.9
10.8
4.8
15.6
19.3
3.7
5.7
11.1
LCC-10
9.2
7.7
10.3
7.4
7.3
8.8
11.5
6.0
8.6
a. Total Organic Extract before fractionation
b. LCC-1 = Paraffins, LCC-4 = Aromatics, LCC-10 = Oxygenates
-31-
ArthurD Little Inc
-------
is not obvious from these data. The sources of variation in the data will
first have to be understood before any clear correlations can be achieved.
The quantitative chemical class composition of each of the aromatic
fractions was determined by the mass spectrometric computer analysis routine
and the results are reported in Table 7. These data at the present time
show sufficient variation within a load condition that any correlation of
load and composition is precluded. However, the samples and data do provide
a valuable interim means to evaluate the correlation between odor and
composition of specific samples.
3. Odor of Exhaust Samples
The odor evaluations reported in Table 8 represent collections on
silica gel at the 10, 33, and 90% load conditions obtained for comparison
with early studies of condensate samples. The most significant finding is
the obvious improvement in recovery of odorants, so that the total organic
extracts are recognizable at slight to moderate intensities when examined
in the test room at the equivalent of 20 liters.
When examining fractions from aqueous extracts, test levels equi-
valent to 200 liters of exhaust were normally used. In some instances
these were increased to 400-liter equivalents to obtain definitive descrip-
tions. It is also apparent from the relative intensities shown in Table 8
that the 33% load yields a relatively higher odorant recovery, possibly
because this has been the standard process.
From the odor profile examination of the TOE from the gel trap at
33% load, there is still some loss evident in the TIA, in comparison with
the total exhaust. Oxidized oil odor notes are detected in the same relative
intensities as found in the total exhaust. Kerosene is just perceptible and
is significantly lower, while some general irritation is evident.
LCC-4 exhibits very slight intensities of both kerosene and oily
aromatics, which might be expected when compared with the TOE. LCC-10
indicates a slight loss of smoky burnt compounds. Both oxidized oil and
smoky odor notes are detected at slight intensities, and eye and nose
irritation are again noted.
At the 10% load, the characteristic sour oxidized oily aromatic is
apparent in the TOE with a burnt, smoky secondary odor. No kerosene aroma-
tics are detectable at 20-liter equivalent, which is consistent with the
33% load findings. In the examination of the LCC-4 fraction, kerosene
per se is not identified, but a very slight level of a sweety solventy aroma
and some pungency are noted. The oxygenated fraction (LCC-10) is dominated
by a sour oxidized oil character and a smoky sooty odor both observed at
intensities at least as high as found in the total extract. Eye and nose
irritation are apparent as well.
-32-
ArthurDLittldnc
-------
Table 7
Aromatic Composition Analysis of Fractions from
Various Engine Loads
Composition
Engine Load, %
Gel Sample
Chemical Class
Alkyl Benzenes
Indans/Tetralins/Indenes
Naphthalenes
FOE
10
349
46 43 47
8 10 16
44 49 39
7
33
5 6 10
49 42 44
6 18 26
42 44 33
16
90
7 8 11
52 44 42
3 9 24
41 46 37
6
-33-
ArthurD Little, Inc
-------
Table 8
Total
Exhaust
Total Organic
Extract
(TOE)
u>
Aromatic
LCC-4 Fraction
Oxygenate
LCC-10 Fraction
-i
a
TEST ROOM ODOR EVALUATIONS OF SAMPLES FROM VARIABLE
Gel A - 10% Load
TIA
Sour oxidized oil
Burnt smoky (part.)
Kerosene
Nose irritation
Eye irritation
TIA
Sour oxidized oil
Burnt sooty
Irritation
TIA
Solventy sweet
Oily
Pungent
TIA
Sour oxidized oil
Smoky (sooty)
2
1*5-2
2
1
1-1*5
1
1
*5
*5
1*5
1
Gel 6 - 33% Load
TIA
Smoky tarry
Oxidized oil
Kerosene
Nose irritation
Eye irritation
TIA
Smoky burnt
Oxidized oil
Kerosene
Irritation
TIA
Solventy kerosene
Oily
TIA
Oxidized oil
Leathery
2
2
1*5
1-1*5
1*5
1*5
1-1*5
*5-l
*5~1
*5-l
1-1%
1
1
LOAD3
Gel 8 - 90% Load
TIA
Burnt tarry
Kerosene
Oily (metallic)
TIA
Smoky (sooty)
Burnt
Oily metallic
Irritation
TIA
Oily
Kerosene
Caramel
TIA
Smoky (sooty)
Oily
1*5-2
1*5
1-1*5
1-1*5
%
1
%
1
1
%
Eye irritation
Nose irritation
Smoky
Nose irritation
Eye irritation
a. 20£ samples of exhaust or the equivalent of the analytical fractions.
n
-------
At 90% load, the dominant odor note both in the exhaust and in
the TOE is a heavy smoky character although at a noticeably lower
intensity in the latter. The sharp oily metallic characteristic is also
detectable at a very slight level in the total organic extract, but it was
just perceptible in the total exhaust.
Kerosene was not detected in the TOE, and its presence was ques-
tionable in the LCC-4 fraction which exhibited little relation to diesel
exhaust. The oxygenate fraction produced the lowest TIA of the set and
exhibited the smoky, sooty odor observed in the total organic extract at
a slight intensity. Some irritation was noted but less than the two
comparable samples from 10 to 33% loads. This is consistent with the
sensations observed in the original exhaust.
B. FUELS
1. Analytical and Odor Evaluation
We have compared the odor characteristics of three new diesel fuels
of varying aromatic content to the No. 1 diesel fuel presently used as well
as several aromatic-free paraffin mixtures. The alternative diesel fuels
studied were:
East Coast Diesel No. 2
Heating Oil
Midwest Diesel No. 2
The Phillips Petroleum Company has available several aromatic free
paraffin mixtures of narrow (Soltrol's) and wide (Base Oil's) boiling
ranges. The Soltrol 170,200 and Base Oil No. 1 have properties similar
to those of the diesel fuels and were also examined for odor characteristics.
The Phillips paraffin mixtures have a considerably cost advantage for our
testing program compared to similar quantities of heptane or cetane.
Some of the physical and chemical properties of these materials are
given in Table 9. The aromatic portion of the diesel fuels was isolated
in the LCC-4 fraction by the liquid chromatography method for odor study.
The observed contribution of this fraction to the total sample mass deter-
mined by the GC method is also included in the table.
The detailed composition of the aromatic portion of these fuels
was determined to provide a test of the odor character-composition correla-
tion developed for the oily-kerosene odor complex and is reported in
Table 10.
The fuels and LCC-4 fractions were examined in detail in the odor
test room. None of the Phillips paraffins had any recognizable odor, and
each would be acceptable for study as a paraffinic fuel. The boiling range
data indicate the Soltrol 200 or Base Oil No. 1 would provide the best
general diesel match. We had a preliminary preference for the Soltrol 200,
-35-
ArthurDLittklnc
-------
Table 9
Physical/Chemical Characteristics of Alternative Fuels
Fuels
No. 1 Diesel (Year 2)
East Coast Diesel No. 2
Midwest Diesel No. 2
East Coast Heating Oil
Soltrol 170
Soltrol 200
Base Oil No. 1
Sp. Gr.
Distillation
Range (°F)
10-90%
50%
Aromatics
- Composition (%)-
Olefins3 S
Silica 4
NMR
0.832
0.826
0.852
0.854
396-497
424-566
418-586
430-585
424-460
460-495
357-549
436
482
506
498
442
478
453
20.8
24.1
34.7
35.4
2.8
1.8
2.2
2.0
0.19
0.20
0.29
0.24
11
10
18
16
13
14
23
25
CO
a. volume %, FIA
b. weight %, FOE
c. mole %
o
-------
Table 10
Aromatic Composition Analysis of Alternative High Aromatic Fuels
Compound Class
Alkyl benzenes
% Composition in
No. 1 East Coast Mid West East Coast
Diesel No. 2 Diesel No. 2 Diesel Heating Oil
32 12 3 2
Indans/Tetralins/Indenes
38
13
Naphthalenes
Acenaphthenes, etc.
Phenanthrenes, etc.
32
52
16
61
21
10
70
17
-37-
ArthurD Little, Inc
-------
since it is chemically simpler and would, therefore, somewhat simplify the
analyses. One engine test was made with this fuel. In the course of the
analyses, however, we found that the Soltrol 200 did have a small aromatic
content. Future studies will be done with the base Oil No. 1 which was
found to be free of aromatics.
The test room data for the high aromatics fuel samples is given in
Table 11. Of the four samples, the Midwest diesel has the highest total
intensity of aroma and the most complex odor at this concentration. It
has both kerosene-type and tarry odors with an oxidized oily note at a
slight to moderate intensity. It produces both eye and nose irritation and
has some naphthenate-related odor. The heating oil is similar to this with
a slightly lower total intensity of aroma and somewhat less complexity. At
close to moderate intensity, the kerosene is the dominant identifhing odor
and is somewhat lighter than that observed with the Midwest diesel. This
character note is supported by an oxidized oily note which does not have
linseed overtones. Naphthenate is present at a slight level and tarry-
related odor notes are present, but just detectable as in the Midwest diesel
fuel sample.
The existing No. 1 diesel fuel also has a moderate total intensity
of aroma. Kerosene is the primary character note and this is supported by
an oily linseed-related character at less than moderate intensity. There
is a very low level of naphthenate but no tarry aroma in this sample. Both
eye and nose irritation are evident.
The lowest total intensity of aroma is observed with East Coast
Disel No. 2, which has an intensity of slight to moderate (1^) this is
cominated by the kerosene odor and the associated oily linseed-type aroma.
Musty naphthanate is also apparent in this sample at a slight intensity.
Interestingly, although eye irritation is evident in the sample, nose
irritation is not observed. Of the four fuel samples, then, this has the
lowest intensity of odor and feeling sensation.
The LCC-4 fractions derived from each of the fuels correspond
fiarly well with the odor character and description of the original samples.
There is in general some slight reduction in total intensity of aroma in
the samples as well as in the recognizable intensities of the character
notes. This is particularly true in the Heating Oil fraction. Kerosene
continues to be the dominant character note in the Midwest diesel fuel
fraction. This is supported by a tarry, almost phenolic odor note. The
tarry odor character may be associated with the acenaphthenes, etc. present
in this sample and not experiences before in the studies with the No. 1 fuel.
The oily note does not appear to be oxidized as in the total sample, which
may indicate some presence of oxygenated components in the fuel. Musty
naphthenate is present as are eye and nose irritation.
With the Heating Oil, the total intensity of aroma is reduced by a
slight amount in the LCC-A fraction as are the kerosene and oily odors. The
naphthenate appears to be slightly tarry in character and is the character
note of highest intensity in this sample, which correlates with the charac-
ter observed in the Midwest diesel fuel. Both eye and nose irritation are
apparent.
-38-
ArthurD Little, Inc
-------
TABLE 11
Odor Test Room Evaluation of Fuels and Aromatics Fractions
Fuel
No. 1 Diesel (Year 2)
Fuel
2
Kerosene
Oily painty
Naphthanate
Eye irritation
Nose irritation
Odor
LCC-4 Fraction
2
2 Oxidized oily sour
1*5-2 Kerosene
*5-l Tarry naphthanate
/ Eye irritation
/ Nose irritation
Headache
1*5
rl
East Coast Diesel No. 2
East Coast Heating Oil
1*5
Kerosene
Oily painty
Musty naphthanate
Eye irritation
1*5-2
Kerosene
Oxidized oily
Naphthanate
Tarry
Eye irritation
Nose irritation
1*5
1*5
1
1*5-2
1-1*5
1*5
Kerosene 1*5
Rubbery 1-lJg
Tarry ^-1
Oxidized oily 1
Eye irritation /
Kerosene (solv) 1
Oily 1
Tarry naphthanate l-l*s
Eye irritation . /
Nose irritation /
Midwest Diesel No. 2
Kerosene
Tarry
Oxidized oily
Naphthanate
Solventy
Eye irritation
Nose irritation
2
1*5
1*5
1
h
Kerosene 1*5-2
Tarry phenolic 1*5-2
Oily 1
Musty naphthante *5-l
Eye irritation /
Nose irritation /
a. 150 yl injected into 12,600 S, test room.
-39-
ArthurD Little, Inc
-------
The existing No. 1 fuel has a moderate total intensity of aroma in
the LCC-4 fraction, but the oily character, which is slightly oxidized,
appears as the first character note in the profile description. Kerosene
is at a similar intensity but appears as the second character note suggest-
ing a slightly lower importance of this characteristic. There is a slight
level of naphthenate or tarry-naphthenate and both eye and nose irritation
are observed. The headache effect was also observed in this sample.
The East Coast Diesel No. 2 fuel is the lowest in total intensity
of aroma and is the only sample in which we observe a new character note in
the LCC-4 fraction. Kerosene is the dominant aroma, and there is a charac-
ter note described as rubbery which was not observed in the total sample.
It is recognized at a slight intensity in the LCC-4 fraction. Tarry or
naphthenate is detected at a very low level and oxidized oily is present at
a slight intensity. Again, eye irritation is the only irritation factor
noted in this sample.
Of primary importance in this examination is the apparent demonstra-
tion of reasonable recovery of the identifhing odor characteristics in the
LCC-4 fraction and the correlation in odor character as well as the general
correlation in odor intensity of these fractions with the total fuel sample.
It is interesting to note that, in some preliminary experiments, the sol-
venty kerosene odor can be recognized with 1.5 y£ of sample in the test
room representing a 1/100 dilution of the normal test room concentration.
2. Engine Study of Soltrol 200
An initial study of the exhaust odor and analytical characteristics
of Soltrol 200 was carried out to study the characteristics of an aromatic
free fuel. In the course of the study we found that the fuel did indeed
have a small aromatic content and future studies will be carried out with
the aromatic free Base Oil No. 1.
The results obtained from this study are summarized in Table 12.
The fuel consumption rate is close to that observed with the No. 1 reference
diesel fuel. The amount of organic material collected from the exhaust is
slightly higher than that observed for the diesel fuel (390 mg/K£ vs. 260).
The liquid chromatographic fractionation shows that the aromatic content of
the exhaust sample is the same as the original fuel. The composition
analysis of the aromatic fractions is also approximately the same suggesting
that all of the aromatics came from the fuel. Further studies of a completely
aromatic free fuel will resolve any remaining questions about the synthesis
of aromatics in the combustion process. It is interesting that the oxygenate
content of the exhaust sample is the same percentage as the aromatic content
and may just be a coincidence which does not have any direct bearing on the
aromatic content.
The odor of the Soltrol exhaust is quite different than that of the
diesel fuel exhaust being lower in total odor intensity (TIA 1% vs. 2-2^)
and totally different in character. The kerosene odor note is completely
absent, as expected, based on the sample analysis. The exhaust odor is
-40-
ArthurD Little, Inc
-------
Table 12
Summary of Soltrol 200 Engine8 Test Results
Fuel Consumption: 8.8 Kg/hr (vs. 8.2 average for No. 1 Diesel)
Silica Gel (Gel 16) Total Organic Extract: 390 mg/KX,
Liquid Chromatographic Analysis:
LCC Fraction Soltrol Fuel Gel 16 Exhaust
Si02-l
-4
-10
98%
1.3%
0.0%
96%
1.5%
1.6%
Aromatic Composition Analysis
Chemical Class Soltrol Fuel
Alkyl benzenes 48%
Indans/tetralins/indenes 30%
Naphthalenes 16%
Odor Characterization (20
Aromatic Exhaust Fraction
56%
21%
21%
Exhaust
TIA 1%
Smoky candle 1-1%
Sour oxidized 1
Irritation //
Total Organic Extract
TIA 1-1%
Smoky candle 1%
Sour oxidized %-l
Irritation /
Oxygenates
Gel 16 LCC-10
TIA 1%
Smoky candle 1%
Sour oxidized %-
irritation //
No odor observed in LCC-4 aromatic fractions
a. 33% load
-41-
ArthurD Little. Inc.
-------
characterized by a moderate intensity of smoky candle and sour oxidized and
a strong irritation factor. This same odor character is also found in the
total organic extract and seen to be due entirely to the oxygenate fraction
which has essentially the same odor as the original exhaust.
We expect that the observed odor character can be accounted for
primarily in terms of alkenone and furan (R+DB 2 and 3) type species. The
sample will be analyzed in detail once the oxygenate analysis scheme
(Section IV D.2.) is completed.
-42-
ArthurD Little Inc
-------
VI. ODOR/CHEMICAL ANALYSIS CORRELATION
Our attempts to establish a quantitative odor/composition correla-
tion are still in the very preliminary stages. However, several observa-
tions have been quite encouraging in this regard. Sufficient odor and
quantitative composition data have been obtained from the engine load studies
using the No. 1 diesel fuel to explore the correlation between the oily and
kerosene odors in the LCC-4 aromatic exhaust fractions and the amount of
alkyl benzenes and indans/tetralins present in the samples.
From the composition analysis data we were able to calculate the
amount of each of these chemical classes present in the test room when the
odor of the LCC-4 fraction was determined. These data can then be compared
with the sample odor intensity as shown in Table 13. The data have been
obtained from our studies on exhaust collected by both the condenser and
silica gel systems and represent a wide range of concentration of samples
studied.
Although there is not yet a great deal of resolution or accuracy on
either the composition or odor scales, it is apparent that the samples with
the highest concentration of indans/tetralins have the highest kerosene odor
intensity while those with the lowest concentration correspondingly have
the lowest odor intensity. The two samples with intermediate concentrations
of these species are not differentiated on the limited odor intensity scale.
It is important to remember at this point that there is a logarithmic rela-
tionship between sample concentration and odor intensity so that minor
differences in sample concentration will tend to be unresolved on the odor
intensity scale.
The first and last two sets of results for the alkyl benzenes show
the same relationship to the oily odor intensity, but the one intermediate
observation does not fit the trend.
Overall, these first attempts at establishing a correlation between
composition and odor have been quite encouraging and suggest that we are
proceeding in the proper direction. The most appropriate point is that our
data analysis schemes provide maximum flexibility in shifting the direction
of our correlation emphasis without requiring any fundamental change in the
data acquisition steps.
-43-
ArthurD Little, he
-------
Table 13
Comparison of Oily-Kerosene Composition and Odor Intensity5
Sample
Gel 4b
Gel 8b
Gel 6b
Cond 66
Cond 67
£
Cond 65
Kerosene
yg of Odor
Indanes/Tetralins Intensity
14 %
12 %
54 jf-1
68 h
160 %-l
300 h-l
Oily
yg of
Alkylbenzenes
60
50
120
70
170
520
Odor
Intensity
)(
%-l
-
h
h
a. LCC-4 fractions studied in odor test room
b. 20JI sample
c. 200S, sample
d. 400£ sample
-44-
ArthurD Little, Inc
-------
VII. REFERENCES
1. Chemical Identification of the Odor Components in Diesel Engine
Exhaust, final report July 1969, CRC Project CAPE-7-68, HEW
Contract PH 22-68-20.
2. Chemical Identification of the Odor Components in Diesel Engine
Exhaust, final report June 1970, CRC Project CAPE-7-68, HEW
Contract No. CPA 22-69-63.
3. F. W. McLafferty, "Interpretation of Mass Spectra, Benjamin,
New York, 1966.
A. H. Budzikiewicz, C. Djerassi and D. H. Williams, "Mass Spectrometry
of Organic Compounds," Holden-Day, San Francisco, 1967.
5. A. Cornu and R. Massot, "Compilation of Mass Spectral Data",
Heydon and Son, London, U. K., 1966.
6. "M.S.D.C. Series Mass Spectral Data", A.W.R.E., Aldermaston,
U.K. 1966 - cont.
7. A. 0. Lustre and P. Issenberg, J. Agr. Food Chem. 17, 1387 (1969),
8. W. Fiddler, R. C. Doerr and A. E. Wasserman, ibid, 18, 310 (1970)
-45-
ArthurD Little, Inc
-------
VIII. GLOSSARY
The following terms have been used frequently in the text and
are summarized here with their definitions, for the convenience of
the reader.
Cond. - Abbreviation for condensate representing the sample collection
used resulting in an aqueous condensate from the diesel exhaust.
GC - Gas Chromatography, used for sample comparison and quantitative
measurement.
Gel - Sample code name for exhaust samples collected using the silica
gel method.
FOE - Fuel Oil Equivalent, the quantity of exhaust species present in
a sample as measured by the flame ionization detector response
when compared to the response calibration with fuel oil.
HRMS - High Resolution Mass Spectrometry, used for chemical identifi-
cation, and quantitative mixture analysis of the oxygenate
fraction.
LCC - Liquid Column Chromatography, used as the means of separating
the paraffin, aromatic, and oxygenate fractions of the organic
extract from the exhaust condensate. The procedure results in
a series of fractions LCC-1, LCC-2, etc. Fractions LCC-4 and
LCC-10 contain the aromatic and oxygenate exhaust odor complexes,
TIA - Total Intensity of Aroma, see Appendix A for details.
TOE - Total Organic Extract, the total organic exhaust species isolated
from the sample collection by solvent extraction.
R+DB - Rings plus Double Bonds, a representation of chemical structure
type by expressing the degree of hydrogen unsaturation (see
Appendix E).
-47-
ArthurD Little, Inc
-------
APPENDIX A
ODOR PROFILE TECHNIQUE
-49-
ArthurD Little, Inc
-------
APPENDIX A
ODOR PROFILE TECHNIQUE*
1. SUMMARY
The standard diesel exhaust sample has been defined as a 2Jg-min.
post-muffler exhaust aliquot of 212. taken during normal engine operation
and a 25 kw (33%) load after warm-up. The profile analysis of the stand-
ard sample diesel exhaust was consistent from day to day, but the odor of
diesel exhaust did show some differences with variations in engine oper-
ation. Preliminary studies indicate that within the normal procedural
time interval between sampling and examination, there are no detectable
losses. Indeed, the odor appears to persist with only slight change
for over one hour. Total profile characterization is consistent with
the odor observed when traveling behind a bus, which confirms our belief
that the mode of engine operation provided a representative sample for
analytical studies.
The description of the diesel exhaust odor in the test room with a
dilution ratio of 600:1 can be described by three character notes: oily,
represented by technical grade hexadecane among other standards; burnt,
which, although similar to a low dilution of propionaldehyde, phenol,
and cresol, is produced in fuels with partial oxidation at elevated
temperatures, and kerosene, which is the top odorous component of the
fuel and may be described as having sweet, sharp, sour, tarry, and sol-
vent components. In addition to these odor characteristics which
appeared in the slight-to-moderate intensity range at this dilution, two
feeling sensations - nose irritation and eye irritation - were apparent.
As implied by the descriptive terminology used, some of the odor
characteristics are present in the fuel itself. The odor characterization
of a 150y£. aliquot of fuel, which (by computation) is equivalent to the
amount of fuel burnt to produce the 21-2. sample of exhaust, produces an
odor in the test room at least as strong as the exhaust odor. The domi-
nant odor characteristic is kerosene, with the oily note being less
intense and the burnt aroma barely detectable. With the diesel exhaost,
the oily and burnt aromas are primary character notes and kerosene a
supplementary factor.
2. ODOR PROFILE METHODOLOGY
The odor Profile Method of analysis has proven useful in flavor and
odor studies in a wide range of food and nonfood products. The Profile
Method, which originated at Arthur D. Little, Inc., 25 years ago, is a
* Taken from Final report of first year's work, July 1969, "Chemical Iden-
tification of the Odor Components in Diesel Engine Exhaust."
-51-
ArthurD Little, Inc
-------
semiquantitative and qualitative description of the odor sensation. The
total odor sensation can be described by six character notes. The method
is qualitative in that there is verbal description as to the odor quality(s)
perceived. The order of appearance of odor character notes indicates the
other odor qualities present as a function of time on a microsecond basis.
The intensity of each character note (as well as the Total Intensity
of Aroma, TIA) is rated on a four-point scale ranging from threshold-)(,
slight-1, moderate-2, to strong-3 intensities. It has been our experi-
ence that for the odor intensity to increase by one unit (i.e., from
slight to moderate), a ten-fold increase in concentration is required.
The threshold intensity indicates that the character note detected is
just recognizable. The basic four point scale of threshold- ) ( to strong-3
intensity can be expanded into a seven point scale with experienced
panelists by the use of one-half ratings. Thus, the full scale of
intensity rating of the odor strength is summarized below.
Numerical Rating Intensity
)( Threshold (recognition)
h Very Slight
1 Slight
1*5 Slight to Moderate
2 Moderate
2Jg Moderate to Strong
3 Strong
The presence of feeling sensations is indicated by a check mark (/)
without any effort to describe their intensity. Four trained analysts
form the odor profile panel. The sample to be analyzed is presented to
the panelists in a standard manner. In this study, each of the four
panelists entered the odorized test chamber independently of one another
and sniffed the air three times. Each then recorded his observations
on the odor character notes perceived, their order of appearance, and
their intensity. After the observations in the test room, the panalists
gathered to discuss their results. Reference was made to odor standards
to relate the various verbal descriptions used and to develop common
language in describing the odor quality. Reference odor standards may
be single chemical species or may refer to a mixture of chemicals.
The panel's results were then composited into an odor profile that
summarized the odor observations of the four panelists and indicated the
odor quality, the order of appearance of the characteristic notes and
their intensities.
3. ODOR TEST ROOM
The Odor Test Room consisted of an antechamber, an odor chamber,
and supporting equipment such as fans, ducts, activated carbon, air
intake, and air exhaust motors. The air is treated with activated car-
bon (C-42 cannister from Dorex) and provided a low-odor background
-52-
ArthurD Little, Inc
-------
diluting medium and was also used to flush odorized air from the chamber
and acclimate the four panel members to a low odor background. The
odor chamber where the odor studies were carried out was an aluminum-
clad room with a volume of 12,600£. Previous studies have shown polished
aluminum to be satisfactory for odor studies because it has a low odor
background. Fans in both the odor chamber and antechamber ensured
adequate mixing and assisted in flushing the test room with odor-free air.
The sequence of events occurring prior to an odor observation by the
panel in the test room is listed below.
a. Odor-free air is used to flush out the antechamber and odor
chamber.
b. The door connecting the odor chamber and antechamber is closed
thus sealing the odor chamber.
c. Diesel exhaust is injected into the odor chamber through a
sampling line by means of a swivel-jointed sampling system.
Three fans located in the odor chamber circulate the diesel
exhaust with the diluting air to ensure proper mixing. Five
minutes after injection the fans are shut down, and the odor-
ized air in the odor chamber is allowed to come to rest.
d. The four panel members then enter the antechamber where they
become acclimated to the low odor background air.
e. The panel then enters the odor chamber, one at a time, to make
observations.
f. The cycle is then repeated to prepare the odor test room for
the next observation. A 20-minute flushing period has been
found to be adequate for removing odor from the test room.
4. ODOR ANALYSIS - DIESEL TEST SAMPLES
Reference odor profiles on diesel exhaust itself have been developed
using a 21-£ injection of exhaust into the odor test room (600/1 dilution)
Because the condensation sample collection procedure collects only about
10% of the odor in the condensate, it is necessary to inject 210-S,
aliquots of the extracted fractions into the test room for their odor
evaluation.
-53-
ArthurD Little, Inc
-------
APPENDIX B
DIESEL EXHAUST SAMPLING SYSTEMS
-55-
ArthurD Little, Inc
-------
APPENDIX B
DIESEL EXHAUST SAMPLING SYSTEMS
1. HIGH VOLUME EXHAUST SAMPLING SYSTEM
A trapping system capable of accumulating large volumes of diesel
exhaust over a shorter period of time was constructed to aid our studies
in several areas. The small amount of organic material isolated in the
chloroform smoky-burnt fraction required 20,000 - 30,000 liters of ex-
haust aliquot for a single GC-MS experiment. Therefore, the trapping
system was redesigned so that we could more efficiently trap the amounts
of sample required for the identification phase of the program. The high
sampling rate also enabled us to more efficiently study the engine
and fuel variables and simplified the evaluation of the quantitative analy-
tical methods.
With our original trapping system, we were only able to collect
about 1000 liters per hour, and the accumulation of 50,000 liters of ex-
haust condensate entailed the inconvenience of maintaining the system in
continuous operation for at least 48 hours. Thus, we have modified our
trapping system to effect a ten-fold increase in collection capacity, i.e.,
volume throughputs of about 10,000 liters per hour. This increase was
accomplished by a general scale-up of our original system without incor-
porating any basically new, and therefore unknown, approaches. The new
system is shown schematically in Figure B-l.
The exhaust gas is channeled into two streams which can be filtered
simultaneously through the large, heated, double filter. The dual filter
holder is a section of 12" diameter x 3" stainless steel tubing which has
been fitted with two sets of stainless screens slightly inset to support
the 11-inch glass fiber sheets. The sheets also act as gaskets between
the main body and the headers. Each header is fitted with three flat
heaters (600 watts per side) to maintain the filter at the exhaust gas
temperature.
The flow of filtered exhaust gases is directed down through a 1-inch
opening into a QVF Model HE-4 glass condenser heat exchanger containing
5 square feet of heat exchanger surface. The exhaust condensate is col-
lected in a cooled 5-liter, one-necked flask situated immediately below
the condenser and connected to it by a tee. The residual exhaust gases
are drawn through the dry gas meter by the carbon vane pump and vented
to the atmosphere. Pressure drop through the system is measured by a
manometer just downstream of the meter to allow for pressure corrections
if necessary. The temperature of the condenser and flask is maintained
at 0°C by circulating coolant from a reservoir which is chilled by a
1.5 hp compressor.
-57-
ArthurD Little, Inc
-------
(1st Quart.)
Post Muffler Manifold
Coolant
Reservoir
1.5 hp
Compressor
Dry
Gas
Meter
Carbon
Vane
Pump
1
Vent
Figure B-1 - High Volume Diesel Exhaust Sampling System
-58-
Arthur D Little Inc
-------
2. SILICA GEL COLLECTION METHOD
Exhaust is sampled from the exhaust manifold by means of a heated
particulate filter and heated lines to and from a teflon diaphragm
Dynapump which provides a 0.5 cfm sampling rate. The exhaust sample
is passed from the pump through a tube (approximately 2 cm dia. x
10 cm long) containing 25 g of silica gel (Fisher Scientific, 14-20 mesh)
which has been acid washed (pH 1-2) and activated at 110°C. A slight
odor and hydrocarbon breakthrough is noted after the collection of 720 SL
in one hour. A distinct yellow-brown color is observed at the top of
the gel trap where the silica is first contacted with the exhaust.
The absorbed sample is extracted from the gel by treating the silica
in the collection tubes successively with 50 ml of pentane and 50 ml of
10% MeOH/H 0 solution which is 0.01N in l^SO^. The aqueous acid methanol
extract is re-extracted twice with 5 ml aliquots of CHC1-. Analysis
of the pentane and chloroform extracts suggests that 90% of the organic
sample collected is extracted by the pentane. This extract consists
primarily of the hydrocarbons. The aqueous acid methanol, on the other
hand, is required to complete the extraction of the oxygenated species.
It should also be noted that the greater bulk of the sample collected
on the gel trap consists of condensed water vapor and that the amount
of organic sample collected bears an inverse relation to the amount of
water trapped. No attempt has been made in the course of the present
experimentation to control the water accumulation.
-59-
ArthurD Little, Inc
-------
APPENDIX C
LIQUID COLUMN CHROMATOGRAPHY (LCC) SEPARATION PROCEDURES
-61-
ArthurD Little, Inc
-------
APPENDIX C
LIQUID COLUMN CHROMATOGRAPHY (LCC) PROCEDURES
1. STANDARD PROCEDURE*
In the standard first-stage separation technique, the sample is
subjected to silica liquid-column chromatography. The detailed experi-
mental conditions are as follows using the 76,500 liters of exhaust col-
lected in Experiment 25 as an example:
Column Conditions. 1.8 cm ID packed for a length of 20 cm with
silica, Grade 950 (Fisher Scientific, 60 to 200 mesh) activated at
110°C for two hours.
Sample Preparation. A small volume (15 ml) or organic condensate
extract was mixed with 15 ml silica and transferred to the top
of the column.
Elution Scheme. Refer to Table C-l.
Handling of Various Fractions. All fractions were allowed to
evaporate overnight at room temperature and the final volume was
adjusted to 7.65 ml in each case. Thus, for these examples repre-
sentative of 76,5002. of exhaust.
The fractionation of the sample, along with the elution scheme and
qualitative odor is given in Table B-l. Oily kerosene comes out in
fractions 4 or 5 and the smoky-burnt odor character comes out in
fraction 10. Since the total Sample 25 had a fuel oil equivalent
(FOE) of about 5,000 mg.** about 70% of the mass was in fraction 1,
14% in fraction 5, and only 3% In fraction 10. This procedure was used
in preparing the fraction for the identification phase of the smoky-burnt
odor studies.
*Taken from Final report of first year's study, ref. 1.
**Mass of sample as determined from flame ionization detector response
based on calibration with fuel oil.
-63-
ArthurD Little, Inc
-------
TABLE C-l
SILICA LCC ELUTION SCHEME AND ODOR OBSERVATIONS FOR SAMPLE 25
Fraction
1
2
3
4
Solvent
Comment
FOE (mg)
5
6
7
8
9
10
11
12
Pentane, 150 ml c Colorless effluent.
Pentane, 100 ml Colorless effluent
3,500
Odor
Odorless
Pentane, 100ml
Colorless effluent
700
Oily, kerosene
Benzene, 100 ml The yellow component starts
moving down upon addition of
benzene; collected effluent
was still colorless
Benzene, 100 ml Greenish yellow effluent .
Benzene, 100 ml Greenish yellow
CHCI3, 150 ml Light greenish yellow
5%MeOH/CHC 13, 100 ml Very light greenish yellow
10% MeOH/CHCI 3, 100 ml Very light greenish yellow
25%MeOH/CHCI3, 100ml Brown
50% MeOH/CHCI3 100ml Brownish yellow
MeOH, 125ml Yellow
a - Fuel oil equivalent; weight of sample based on GC response compared to fuel oil calibration
using the FID response from silicone column. Total FOE for Sample 25 =5000 mg.
b Qualitative odor screening observation.
c The eluted fractions were concentrated to 7.65 ml.
150
Smoky-burnt, oily
-64-
ArthurD Little, Inc
-------
2. GRADIENT ELUTION PROCEDURE
The gradient elution procedure was used for the isolation of the
smoky-burnt odor species from the separate pentane and chloroform extracts
of the exhaust condensate. This procedure was found to give a resolved
fraction having an improved odor/mass ratio, and resolution compatible
with the two-stage GC-HRMS analysis procedure.
Column Conditions: 0.7 cm I.D. packed for a length of 14 cm with
silica, Grade 950 activated at 110°C for two hours.
Sample Preparation; A small volume (5-10ml) of chloroform or
pentane condensate extract was mixed with 2ml of silica and trans-
ferred to the top of the column.
Elution Scheme: Refer to Table C-2.
Fraction Handling; All fractions were allowed to evaporate until
they had reached an equivalent concentration of approximately 10£
of exhaust/yJi.
The elution sequence and order of odor elution is given in Table C-2.
The effectiveness of the procedure for improving the smoky-burnt fraction
can be seen in a comparison of fraction 38 from this example with a
comparative smoky-burnt fraction obtained on the same initial sample
using the standard procedure. While both the standard and modified
odor fractions had a TIA of 1 and smoky-burnt odor intensity of 1, the
standard fraction contained 0.6yg/1000£ compared to 0.04yg/1000& for
the gradient fraction.
3. MICRO COLUMN PROCEDURE FOR QUANTITATIVE ANALYSIS
A new micro LCC procedure was developed to simplify the analysis
and meet the sample handling requirements of the 1000& exhaust samples
collected on the silica gel traps.
The concentrated pentane and chloroform silica gel extracts are
fractionated in a micro-column, which consists of a bottom tapered 16 cm
piece of 8 mm glass tubing fused to a 6 cm piece of 18 mm glass tubing
which serves as solvent reservoir. The column is packed with activated
silica Grade 950 (Fisher Scientific, 60-200 mesh) to a height of 12 cm.
The volumes of the solvents used for eluting were adjusted correspondingly
to establish chromatographic conditions similar to those of our standard
LCC fractionation. The elution pattern and odor characteristics are
shown in Table C-3.
-65-
ArthurD Little, Inc
-------
TABLE C-2
GRADIENT ELUTION FRACTIONATION OF SAMPLE 43-CHCL EXTRACT (3.000&)
(4)
Fraction No.
Solvent
Odor Notes3
1 to 6
7
8
9 to 12
13 and 14
15
16
17
18
19 to 28
29
30 to 31
32
33
34 and 35
36
37
38
39
40 to 42
43
44
45
46 to 48
49 and 50
51
52
53
Pentane, 5 ml
5%CHC\3,(l\ 5ml
5%CHCI3,5ml
10%CHCI3,5ml
20%CHCI3,5ml
20%CHCI3,7ml
30%CHCI3.5ml
30%CHCI3,5ml
30%CHCI3,5ml
30%CHCI3,5ml
30%CHCI3,5ml
35%CHCI3,5ml
35%CHCI3,5ml
40%CHCI3,7ml
40%CHCI3,5ml
40%CHCI3,5ml
40%CHCI3, 1 ml
40%CHCI3,2.5ml
50%CHCI3,2.0ml
50%CHCI3,5ml
50%CHCI3,5ml
50%CHCI3,5ml
75%CHCI3,5ml
75%CHCI3,5ml
CHCI3,5ml
5%MeOH(2),5ml
5% MeOH, 5 ml
10%MeOH,5ml
Kerosene
Oxidized oil
Smoky-Burnt
Smoky-Burnt
Sour
1. %CHCI3 in pentane
2. % MeOH in CHCI3
3. Blotter Strip
4. Taken from final report, second year,- ref. 2,
-66-
ArthurD Little Inc
-------
Table C-3
Micro-LCC Fractionation of Gel Trap Samples^ '
Fraction No.
1
2
3
4
5
6
7
8
9
10
Solvent
Pentane, 10.0 ml
Pentane, 2.5
Pentane, 2.5
Benzene, 11.0
Benzene, 2.5
Benzene, 2.5
CHC13, 5.0
10% MeOH/CHCl-, 2.5
10% MeOH/CHCl3, 2.5
10% MeOH/CHCl-, 2.0
Compound Type Eluted and Odor
Aliphatic hydrocarbons; odorless
Aromatic hydrocarbons; oily kerosene
Oxygenated compounds; smoky burnt
(a) The micro-column used consists of a short glass tubing of
0.6 cm I.D. and packed with activated silica Grade 950
(Fisher Scientific, 60-200 mesh) to a height of 12.0 cm.
The column volume is approximately 2.2 ml.
(b) The pentane and chloroform extracts of the gel traps were
concentrated to about 1.0 ml and then applied directly to
the top of the column.
-67-
ArthurD Little Inc
-------
APPENDIX D
SILICONE AND CARBOWAX CHROMATOGRAMS OF SMOKY-BURNT FRACTIONS
-69-
ArthurD Little, Inc
-------
APPENDIX D
SILICONS AND CARBOWAX CHROMATOGRAMS OF SMOKY-BURNT FRACTIONS
The silicone chromatograms and odor profiles of the working samples
are shown in Figures D-l to D-3. All chromatograms were run at a
temperature program of 4°C per minute from 70°C to 300°C. Samples
52C-34/5/6/ plus 56C-39/40 and Sample 56P-40 were analyzed on the
same 10' x 1/8", 10% OV-1 column (D2 and D3) while Sample 51C-34/5
was examined on a similar but older column. Use of the two columns
accounts for the difference in elution temperatures of the otherwise
similar peak patterns shown in Figure D-l as compared to the earlier
elution temperatures observed in Figures D-2 and D-3. The silicone
chromatogram of sample 60C was identical to that of samples 52C and 56C
(Figure D-2) and has not been duplicated here. We have assigned peak
numbers to similar peaks in each of the chromatograms as a constant
working reference.
From Sample 51C-34/5 and Sample 52C-34/5/6 and 56C-39/40, we have
trapped at the designated intervals over the region from peak 7 to peak
17. Peaks 18, 19, 20 and 21, 22, were trapped from Sample 60C. From
Sample 56P-40, we have trapped from peak 10 to peak 22. The regions
trapped are designated in the figures with broken lines. In our trap-
ping experiments, we have tried to cover the reasonably intense peaks
which are the significant odor contributors. Each trapped silicone
peak was re-chromatographed on a 10' x 1/8", 5% SP-1000 Carbowax column,
run under the temperature program designated in each chromatogram.
(Figures D-4 to D-19). Odor profiles and high resolution mass spectra
were obtained on each chromatogram. As described in the main text, the
odor profiles and HRMS were obtained on separate trapped fractions with
Sample 51, while the odor, GC response, and HRMS were all obtained
simultaneously with Samples 52, 56, and 60.
The species or areas selected for HRMS examination on the basis of
their odor are designated on the Carbowax curves by a code (2.0, 3.5,
9.0, etc.) which corresponds to the mass spectrometer photoplate exposure
number.
-71-
ArthurD Little, Inc
-------
S °0
MDkjr
mt rubbery
'(most coumarin
heavy burnt.
Mphthol-Hk«i burnt oil
pungent, nose 1rr1ttt1on
"3ieseT exhaust smokv
f __ smoky metallic
exhaust
"strong feeTing stnsation
oxidized oil
_ __ pungent and smoky
itronq smoky, diesel related
leathery sour, pungent
woody, burnt rubber, good smoky
jassy, particle size
smoTy, oxidized oil
smoky exhaust with particle size
HOky
linseed oil, smoky
smoky
slight smoky
fragrant, particle size
leathery, g00(j
sweet powdery
rubbery
fragrant, lavender, floral
related to leathery
I
S
5
1
_J
5!
.H
A
0)
bO
-72-
W
Si
Arthur D Little, Inc
-------
smoky
diesel exhaust
oxidized oil, oreen
oil cloth, pungent
almost burnt rubber
burnt, leathery
creosote, sour
sweet 3
'Oxidized oil
irri tation
phenoli c
burnt sweet, linseed
particle size
sliont oxidized oil
leathery
woody
sliqht burnt
rubbery
3
to
fe
CNI
n
D
bC
H
fe
-73-
ArthurD Little, Inc
-------
8 -
&
O
s
sulfide .
tallic
rubbery burnt
burnt cloth
smoky
burnt fuel. Dirt of
slightly burnt rubber
sour
oxidized oil
illght fragrance
coconut
tjllic, particle sizes
coumarin
chalky
green frtartncr
very sticky burnt'oil;
coconut, smoky MX
musky
peach aldehyde
~ C^
T
oxidized oi 1 ,
pungent, benzoin
green, sour, oxidized
smoky fat
?Tmoly fat
- (^ linseed oil
^ burnt fat
smoky burnt cloth;
^_5 aldehydlc
metallic, grease
oil
W\
X
\a
»
* 1
8
>
«r ""
Q
QJ
T M
grease ^« «l«Wy*
oxidized oil 7»^ pungent> etherea)
aldehydlc sour- Bung*"'
^, oxidised oil -
sTiqhtly smoky
oil cloth, burnt cloth
sour, linseed
pungent, sort of leathery
woody \
benzoin
pur,gent, slightly rubbery and burnt "
kind of smoky
iour ournt, burnt wood, burnt rubber
injsty, oxidized oi 1
spruce green
henlock
green, sour
musty
smoky charcoal, gassy
sour, rubbery
-74-
ArthurD Little Inc
-------
18 r.
51 SP-1000; 10' « 1/8"
100° tV-Sil.
~^l
tn
>
-i
Er
-7k
Figure D-4
CAMOMM CMMMTOMM OF SILICOK KAC 7 MB nm SMTI.E SIC-M/S
-------
-9L-
I
ft
in
i
s«et. fragrant, pungent
sour, slightly burnt, maybe benzofuran
ubbery
rv irritating
- something related to tarry
ounqent, burnt
nqent, spicy, sweet
fragrant
-) f
oil cloth
irriutlnq.
n, sour, burnt
Bllv linseedi pungent
sweet, almost buttery
^^ wood smoke
* f "3 phenolic
smoky, irritation
burnt, pungent
-------
-LL-
OQ
it
re
a
~
n
Oi
8-
a
*
0
oo 01
0) rt
3 O
-a oo
t* *"(
(D 01
Ln
t-0 O
O Hi
I
OJ 00
C- H-
~-^ O
ON O
3
w n>
a. >T3
(D
Ln 03
ON ?T
n
i t-1
u> o
vo
i-h
*- H
O O
like indanone, very qood odor, punaent. spicy
not mucn interest
fragrant, pungent
sweet, pungent,
kind of Kraft paper
particle size part of smokv with nrp»n kind of meet
ind of fattv . oiIv
-------
*i
(6
1
1
5
9
!
m
tn
s
i
V-_^^
\ t^^A»^t,c-t
.^f* particle size,
*~\-g-
C^:
4.oT~~> P'.rcjent, met
4*- Sj_
*-**-!
spicy
allic
-------
TJ
e
a
i
oo
c
-------
-------
-T8-
i-i
It
a
o
I
3
t_
«
f**\. /w/^->y
JJ f
asphyxiating, 5ws
y*/
-------
-38-
o
fi
i
burnt djpsel
-------
-C8-
C
n
ro
o
m
84
060
tit
fco
Ih'i I ^/^ r^. C.j
S.-.X '..^^ 7,Jl,
» I
X
u
{ f ^/^'^
jL^nt S^,tl. *>/ ,£^,.1
IK.
^^y
:?
^ f*.*.,*^ i#,i.\
rnt sweet, Du^qent
- An* /.. a^ it i &*-* N. ( o-u f
/
-------
*
8
i\
\
4
.cj
_<55£7~>I'_ pAt
g f r J
J-d
"22Jly aldehydk
"^ linseed, irrititioti
-------
-58-
i
re
a
I
g
§
%*-/ r /
-------
Dup|j)
-98-
H-
Og
5 -
-------
Tl
H-
OQ
c
n
[B
a
i
h^
CTN
*
U>
At
>_ r*+*-b . A*~£tU A.U ,
/' If 0
oily
1
-f~*
'
8 3
-------
-88-
4'*- '
O
s
s
*
s
-------
-68-
*d
H-
(70
a
CE
>
I'
.;
r.
-
;
/w »«*.,
r.
r
woody, irritation
particle size, i
-------
M
i
1
-06-
v-
\
f.Q
..//.- no '
i » L-*«-«'
""'^
^^>pungent, smoky
pungent, sour, oi1^
i S
-------
-T6-
p
ro
o
I
O
-n
CO
s
ro
ro
O
m
§
o
CO
NJ V>
s t
-------
-16-
a
I
r\j
n
01
-s
CT
O
I
X
01
o
-h
oo
n
o
re
0>
7T
on
|NJ
Co
no
oo
a>
6
ro
CTI
O
O
o
*
f-< >
1 _
-------
APPENDIX E
STRUCTURE ASSIGNMENT METHODS
-93-
ArthurD Little Inc
-------
APPENDIX E
STRUCTURE ASSIGNMENT METHODS
1. INTRODUCTION
The assignment of chemical structure type for the large variety
of odorous materials observed in the oily-kerosene and smoky-burnt odor
complex is a difficult task in view of the many choices possible. The
purpose of this discussion is to present clearly in one place the various
iterative processes used for assigning the selected structures for the
various odor compounds.
2. DISCUSSION
Our chemical structure assignments have been made using as many
of the following data sources as possible:
Odor
Gas Chromatography (GC)
Retention times on
Silicone OV-1
Carbowax SP-1000
High Resolution Mass Spectrometry (HRMS)
Molecular weight (MW)
Elemental composition (ELCOMP)
(rings plus double bonds, R + DB)
Fragmentation pattern
It is important to remember that we were restricted to these sources
of data due to the small amounts of sample available. Unfortunately,
we have not always had all three sources of information available to
use and more frequently have had to rely on HRMS and GC, HRMS and odor,
or HRMS data alone. However, in each case, the objective has been to
utilize all of the data available via an iterative process of interpre-
tation-confirmation and structure choice refinement.
-95-
ArthurD Little, Inc
-------
Structure assignment for the oily-kerosene odor species was a
considerably simpler task for the oxygenates in the smoky-burnt
fraction for several reasons. First, there were only three or four
principal structure types to identify, and they were tentatively rather
easily selected on the basis of HRMS. Verification of these assignments
was simplified in terms of the number of known reference chemicals which
had to be studied. Another factor which was very important was the simi-
larity of the identified species to the original fuel components - -
therefore, allowing confirmation based on previously well known and doc-
umented reference information.
The task for the smoky-burnt oxygenates is a much more complicated
one due to the greater number of structure classes possible. These data
reported in Appendix F were interpreted using the data input described
above to arrive at structure assignments. In several instances, it
was possible to make specific structure assignments based on comparison
with reference standards. In many instances, however, the structure
class assignments are still tentative and continually in the process
of being reevaluated as new data become available.
Generally, the first step was to restrict the possibilities as
much as possible from the HRMS data (MW and ELCOMP). This process
is discussed in detail in part 3 of this appendix. The rings plus
double bonds (R+DB) approach has been a considerable aid in organizing
this task. Detailed assignment of the geometric structure and function-
al groups then required extensive use of all possible available
reference data.
The next step was to find or generate data on the physical, proper-
ties or reference standards selected on the basis of the initial HRMS
interpretation. Gas chromatographic reference data were obtained on
the OV-1 silicone and SP-1000 Carbowax columns, and these data served
as reference points for the identified exhaust species. The reference
GC data were used first for establishing the presence of the specific
reference standard compound in the exhaust sample when a match was
achieved between the two sets of GC data and the HRMS data from the
exhaust sample and reference study.
However, but of equal importance, was the use of the GC data to
establish characteristic elution patterns for homologous series within
a structure class and to characterize the differences, when they existed,
between structure classes. This latter approach is probably most
important because of the exceptionally large effort that would be
required to confirm identifications based only on specific reference
compound matches, due to the many homologues and isomers possible.
In addition to information found in general references on the
combustion processes and products of paraffinic and aromatic hydro-
carbons, structure assignments, based on the HRMS data for each species,
have been aided by interpretation of the fragmentation patterns with
particular reference to the work published by McLafferty (3) and
-96-
ArthurD Little, !nc
-------
Budzikiewicz, Djerassi, and Williams (4). When available, the spectra
were compared with the reference spectra published in the compilation
by Cornu and Massot (5) and the AWRE set of spectra (6). In addition,
valuable assistance was found in comparing the analysis of wood smoke
by Lustre and Issenberg (7) and liquid smoke solution by Fiddler, Doerr,
and Wassennan (8).
Finally, confirmation of structure class assignments were obtained
by odor studies of the reference standards. These studies have confirmed
the odor significance of several structural types. The studies are
significant both in terms of odor differences between compound classes
and also in detailing the effects of geometric isomerism on odor
characteristics.
Most of the arguments presented above work also, of course, for
the exclusion of certain structure types either on the basis of their
GC or HRMS data and in that manner also help reduce the number of
possibilities one needs to consider.
Several examples are shown in Table 1 to demonstrate the nature
of the data available and the manner in which it was used to arrive
at chemical structure assignments.
-97-
ArthurD Little, Inc
-------
Table E-l
STRUCTURE ASSIGNMENT EXAMPLES
Observed/Assignment OVT SPT
Specific Reference Standard Match
1. 52C/56C Peak 9 116 160
m-Cresol 114 164.
2. 52C/56C 140 160
cinnamaldehyde 148 160
3. 52C/56C Peak 13 140 158
indanone 150 160
Assignment based on HRMS and extrapolated GC
60C Peak 18-20 166 194
allyl phenol
Assignment based on HRMS data
1. 52C/56C Peak 9
alkyl(butyl )furan
116
2. 52C/56C Peak 11
"benzaldehyde"
actually, methyltolyl-
ketone
126
124
137
MW
108
108
132
132
132
132
ELCOMP
134
134
124
134
C9H80
CgH80
CgHgO
CgH80
C9H10°
C9H100
C8H12°
C9H1Q0
R+DB
4
4
6
6
FRAGMENTS
107(-H)
matches
Comment
103(-CHO)
matches
104(-CO)
matches
133(-H). 119(-CH3)
matches
109,96,95
82(C5H60)
119(-CH3)
matches
reference
correction of
original assign-
ment as indenol
confirmation,
structure first
assigned on basis
of HRMS
consistent with
lower derivative of
dimethyallylphenol
which elutes in
region of peak 23,
SPT 191°
spectrum consistent
with butyl chain
fragmentation both
to eliminate C?H.
to give 96 and
straight cleavage
terminating in
stable furan ring
(82).
elutes too soon
from Carbowax to
be an indanol
Rejected possibilities
". Tentative assignment of hydroxybenzoic acid to MW 138(5) C7Hg03 species observed in 52C/56C Peak 10
rejected because reference standard acid exceded observed Carbowax elution temperature (standard
SPT > 220°, exhaust species SPT 152°). Structure still unassigned.
2. The following quinone structure was considered for MW = 136 (6) C^H^O^ species observed in 52C/56C
peak 13 but rejected because HRMS pattern did not possess required ions for loss of C2H2 and C3H20;
-98-
ArthurD Little Inc
-------
3. INTERPRETATION OF MASS SPECTRAL DATA
One of the most useful first interpretative aids in restricting
the possible structural assignments are the R + DB values listed in
the appendix tables. R + DB stands for "rings plus double bonds" and
is an interpretive aid taken from McLafferty's treatment of mass
spectral data (3). Values of R + DB are basically arrived at by a
simple analysis of the degree of unsaturation in a molecule with a
particular composition. For species containing only carbon, hydrogen,
and oxygen the values are arrived at numerically from the formula
R + DB = No. C atoms - %(No. H. atoms) + 1
Several examples will serve to demonstrate the utility of the values:
for an n-paraffin C,^,
R 4- DB = 6-7 + 1 = 0; i.e., the n-paraffin has no
rings nor double bonds
for a hexenone C.-H-.-.O
D 1U
R + DB =6-5 + 1=2; fitting a structure
CH3-CH2-CH2-CH = CH-CHO
Cyclohexanone also satisfies the R+DB
criteria having one ring and one double bond.
-99-
ArthurD Little, Inc
-------
a phenol C_H00 would have
/ o
R+DB =7-4+1=4 consistent with the structure
cm
These R + DB values serve to simplify the process of assigning
structures while examining a multitude of data because they
conveniently exclude certain possibilities in a manner easy to
remember once one becomes accustomed to the procedure. Thus, in a
trivial case, an R + DB = 7 species cannot be a paraffin, simple ketone,
etc. The values do not imply certain explicit structures but only serve
to restrict the possibilities. As the R + DB value increases, the struc-
tural possibilities expand considerably as one can find continuing new
ways to combine aromatic rings, carbonyls, etc.
We have compiled in Table E-2 some possible hypothetical structure
types which we feel are most likely, on the basis of our data so far,
to be considered in the smoky-burnt diesel odor complex. The table is
organized by R + DB for structures containing one, two, and three oxygen
atoms. The data are self-explanatory for the most part, but several
points to note are that alcohols are represented by -ols, ene refers to
an unsaturation, carbonyls (or .-ones) may be aldehydes and/or ketones,
and hydroxy and methoxy derivatives are occasionally referred to as
-oxy. L
The table is meant to indicate primarily what new species may be
considered as the R + DB value increases and does not exclude the com-
bination of lower R + DB value functionalities. For the 0» and 0_
cases, a dydroxy or methoxy derivative of an 0- case is always also
allowed. R + DB values are a routine output in the HRMS computer
routines for each ion whose composition is listed.
-100-
Arthur D Little. Inc
-------
TABLE E-2
POSSIBLE OXYGENATES - 0,
1-
R + DB Possible Structure Type
alcohols (O *, ethers (C»)
ketones (CL), aldehydes (C ), cyclic-ols (C-), epoxides (C~)
alkenones (C_), cyclic carbonyls (C,.)
dienones (C-), cyclic-ene-carbonyls (C-) , furans (C,)
phenols (C,), benzyl alcohols (C_)
phenyl carbonyls (C7), indanols (C_), allyl phenols (C.), dihydro-
benzofuran (C )
o
benzofurans (CQ) , indanones (Cn), indenols (C0) , phenylene-carbonyls (CQ)
o y y o
naphthols (Clf.), indenones (C )
naphthaldehydes (C,,)
dibenzofurans (C)
* Carbon number in parenthesis represents smallest number of
carbon atoms which first member in homolgous series may have.
-101-
ArthurD Little, Inc
-------
Table E-2 cont,
POSSIBLE OXYGENATES - 0^
R + DB Possible Structure Type
0 diols (C-), peroxides (C.), ethers (C.)
1 acids (aliphatic) (C~), esters (C.) , hydroxy carbonyls (C,)
2 alkene acids (C_), esters (C_), cyclic acids (C,) , dicarbonyls (,),
oxy cyclo carbonyl (C,.)
3 MCP's (O , cyclodiones (C_) , oxy furans (C.)
4 furfurals (C_), hydroquinones (C,), methoxy phenols (C7)
5 aromatic acids (C7) , hydroxy aromatic carbonyls (C ),
oxy indanols (Cq) , quinones (C,), allyl phenols (Cg)
6 phenylpropene acids (Cn) , dihydrocoumarins (CQ) , aromatic
y y
dialdehydes (C0) , oxy benzofurans (C0), oxy indanones (C0)
H o y
7 coumarins (iso) (Cq), benzofurfurals (Cg) , indandiones (C_),
oxy indenones (Cq)
8 naphthoquinones (C,n), hydroxy naphthaldehydes (C,,)
-102-
Arthur D Little Inc
-------
Table E-2 cont.
5
POSSIBLE OXYGENATES - 0^
R + DB Possible Structure Type
0 ethers (C.)
4 dimethoxyphenols (C0)
o
dioxyphenyl carbonyl (C7) , allyl phenols (CQ) , oxy benzoic
acids (C ), oxy quinones (C,)
oxy phenyl propene aldehydes (or oxy allyl phenyl aldehydes (C]0)
(plus other hydroxy and methoxy derivatives from
CL and 0- categories)
-103-
ArthurD Little Inc
-------
4. GAS CHROMATOGRAPHIC DATA
Several adjustments are necessary in comparing the GC data for the
exhaust and reference compounds to determine the degree of match of the
physical properties. While the adjusted retention temperatures of the
reference standards reported in Table 2 of this report are probably
accurate to about 2°C, the exhaust sample data are not known with the
same degree of accuracy. The peaks trapped from the silicone column
were collected over a 3° - 12°C range, and the average temperature
of the collection range was recorded. This spread, and the overlap
between peaks, requires, therefore, that any reference standard eluting
between + 3 - 6° of the trapped OV-1 temperature be considered.
The reference standards adjusted elution temperature (SPT) were
obtained on the SP-1000 Carbowax column using the 120°C initial
2°C/min 220°C hold program, while much of the exhaust identification
work was done on other program rates. The relationship between the
various Carbowax elution temperatures obtained by studying reference
standards under each of the conditions is, however, a smooth one and by
reference to Figure E-l, the exhaust data may be corredted to the same
base as the reference standards. The correlation between peak number
and elution temperature necessary for comparison of the exhaust and
reference standard data is shown in Figure E-2.
We have attempted to organize our GC retention data so that one
could differentiate between structure classes from these data in a
simple manner. Unfortunately, establising characteristic GC elution
patterns for the structure classes has not been as straightfoward as
one would wish. The data plotted in Figure E-3 serve to illustrate
this point. The figure represents the data available at this time on
one of the structure groups we have studied most extensively, the
phenols and methoxy benzenes. The elution temperature on the silicone
column (OVT) are plotted against the Carbowax (SPT) temperature for the
two related classes. While it is apparent that the free phenols as a
group elute at a higher temperature on the Carbowax than the methoxy
benzenes, there is a large spread in the data, and it is difficult to
establish a simple correlation. These results are not unexpected since
this type of correlation technique works best with increasing chain
length homologous series, and the exhaust species vary principally by
degree of substitution on an aromatic ring. The overlap of these data
with other structure classes can be seen from a few other reference
compounds also shown on the figure.
-104-
ArthurD Little, Inc.
-------
i
100°/2°
and
70°/4°
o
f temps.
c
D
S
(L
3
220
210
200
190
180
170
°C 160
150
140
130
120
110
100
1/1 correlation
x
/ i
100°/2° data
O 70°/4° data
I
I
110 120 130 140 150 160 170 180 190 200 210 220
120°/2° Program Rate
FIGURE E-I- TEMPERATURE RELATIONSHIP OF VARIOUS SP-IOOO ELUTION PROGRAMS
: ^ TO THAT FOR".THE 120°/2° PROGRAM RATE
-------
APE
°C
240
230
220
210
200
190
Temperature 180
Program
o 4%niiu o 170
160
150
140 -
130
120
110
100
90
Value Peak No.
_
-
2.9~
2.8-1
2.7-
2.6-
2.5-
2.4-
2.3-
2.2-
" Trip
1.9-
1.8-
1.7-
1.6 -
1.5 -
175-
__ 1.3-
11
~~ i.o
"~ 0.85
~~ 0.6 "
28
27
26
25
24
23
22
21
20
19
18
17
16
15
14
13
12
" 11
10
9
8
5*
3
2
1
\
\
\
°C APE Peak //
f
156
154
152
150
\148
\ ,
-1.7-
1.6-
_
-
13.0
12.8
12.6
12.4
12.2
12
\
\
Peak No.
\
FIGURE E-2: RELATIONSHIP BETWEEN OV-1 SILICONE TEMPERATURE, APE VALUE
AND PEAK NUMBER ELUTION SCALES
-106-
ArthurD Little, Inc
-------
190
180
170
SPT
(0°C) 160
i
M
o .
150
140
130
O
«»'
r*
r»_
3
80
a acetMfhenone
D indanol
X
D methyl naphthoqui none
Carbowax SP-1000 Program (SPT)
2 /m?n. 220°
Sili cone OV-1 Program (OVT)
4°/min
D cinnajnaldehyde
D Indanone
methoxy benzenes
© di methoxy benzenes
X phenols
9 methoxy phenols
10° 120 140 160 180 200 220
OVT (°C)
FIGURE E-3: ELUTION PATTERNS OF OXYGENATED REFERENCE STANDARDS ON SILICONE AND
240
CARBOWAX GC COLUMNS
-------
APPENDIX F
BASIC HIGH RESOLUTION MASS SPECTROMETRY-ODOR DATA
ON ODOR SIGNIFICANT SMOKY-BURNT SPECIES
-109-
ArthurD Little, Inc
-------
APPENDIX F
Basic High Resolution Mass Spectrometry-Odor Data
on Odor Significant Smoky-Burnt Species
The data included in Tables F-l to F-4 represent most
of the basic information obtained in the structure-odor study of
the three smoky-burnt samples 51C-34/5, 52C-34/5/6 plus 56C-39/40,
60C-39/40/41 and 56P-40.
The significance of each of the items in the tables is
as follows:
The first line after the title states that
Peak 9 - The peak area trapped from the silicone (OV-1)
column.
APE 0.8 to 0.9 - The silicone retention temperature rela-
tive to allyl £henyl ether.
0V 138° - 138 is the nominal eluting temperature (°C)
on the OV-1 column of peak 9.
SP 70 /4 - The temperature programmed conditions used for
the rechromatography of trapped peak 9 on the SP-
1000 Carbowax column an initial temperature of
70°C followed by a 4°C/min program rate. For the
sample run at the same program conditions used for
the reference standards, i.e. from 120°C at 2°C per
minute, a tolerance of + 2°C has been allowed for the
various peaks identified. For those traps which
were run at the different program conditions, a
larger tolerance value has been used; + 6°C for
4° 4° ~~
70 > 220°C and the 100 C » 220°C programs
2°
and + 4°C for the 100 » 220°C program conditions.
The column headings have the following significance:
% - This number represents the relative amount of the
particular species eluting on the Carbowax column
as compared to the total original smoky-burnt
sample.
Exp - The peak identification number of the species
studied, the number corresponds to the photo-
plate exposure number.
-ill-
Arthur D Little, Inc
-------
SPT - The elution temperature (°C) of the peak from the
SP-1000 column adjusted to a standard program con-
dition of an initial 120°C followed by a 2°C/min
temperature program rate.
Odor - Observed odor characteristic of the eluting peak.
The molecular weight (MW) of the species measured
by HRMS. R+DB is a structure interpretation air
meaning "rings plus double bonds." The interpre-
tive significance of the R+DB value is discussed
in Appendix E.
ELCOMP - Elemental composition of the indicated molecular
ion as obtained from the HRMS data.
Structure Type - Chemical structure of the measured species. In
most instances, the indicated structure types
represent our best present estimate of the most
probable structure. In some cases sufficient
supporting reference data are available to make
definite assignments. These cases are indicated
with an asterisk (*). The names represent only
the basic nucleus the degree of alkyl substi-
tution is determined by examining ELCOMP. Thus,
a CsHjoO methoxy benzene is actually a methyl
methoxy benzene. A CgHigO benzaldehyde is a
dimethyl benzaldehyde, etc.
An additional discussion of the significance of
the structure assignments is given in Appendix G.
-112-
ArthurD Little, Inc
-------
TABLE F-l
Mass Spectrometry - Odor Data - Sample 51C-34/5
%
0.1
0.2
0.1
0.3
0.6
0.4
0.4
0.4
0.8
0.2
1.3
0.6
1.0
%
0.2
0.3
0.6
0.6
0.1
0.1
1.0
EXP
1.5
2.0
2.5
4.0
5.0
5.5
6.0
6.5
7.5
8.0
8.5
9.0
9.5
EXP
7.0
7.5
8.0
8.5
10.0
11.0
11.5
Peak
SPT
126
128
132
135
137
139
140
141
144
147
149
159
165
Peak
SPT
138
140
142
143
150
155
158
7,8 APE 0.8 to 0.9 0V 128° -
Sweet, fragrant, particle size,
Odor MW(R+DB)
126(2)
134(4)
100(2)
142(2)
110(3)
124(3)
120(5)
138(4)
124(4)
124(4)
112(3)
122(4)
110(4)
108(4)
108(4)
- SP 70°/4
smoky
ELCOMP
C 8H i i^O
C i 0^1 't
C5H802
C8HlltO
C7H10°}
C8Hi20
C8H80
C8H1002
C7H802
C7H802
C6H802
C8H100
C6H602
C7H80
C7H80
0 (8%)
Structure Type
Cyclohexanone
A
t-butyl benzene
4-hydroxy-2-pentenone
Hydroxyalkenone
Furans
Acetophenone
A
1,2-dimethoxy benzene
Furan
Furan
MCP*
Furan
Methyl furfural
A
o-Cresol
A
m-Cresol
9 APE 0.95 to 1.05 0V 138° SP 70°/4° (6%)
Smoky, linseed
Odor MW
Oxidized oil, chalky 124(3)
Sour, leathery 96(3)
Burnt oil 100(2)
Linseed oil 114(1)
Leathery 126(3)
Irritation 96(3)
Woody 122(4)
ELCOMP
C8H120
C6H80
C5H802
C7H1itO
C7H10°2
C6H80
C6H10°
Structure Type
A
Dimethyl cyclohexenone
Dimethyl furan
Valerolactone
Heptanone
Methyl cyclohexanedione
Furan
A
2,6 dimethyl phenol
-113-
ArthurD Little, Inc
-------
Peak
10 APE 1.2 to 1.3
0V 144°
SP 70°/4° (5%)
Smoky
0.3
0.4
0.5
0.2
Exp
2.5
3.0
4.0
4.5
SPT
143
144
147
150
Odor
Smoky
Smoky
Pungent
Sour
Peak 11 APE 1.35 to 1.45
MW(R+DB)
136(4)
152(4)
134(5)
146(6)
132(6)
0V 152° ~
ELCOMP
C9Hi20
C0 HI Q 0
GI Q HI o ^
CgHsO
SP 70°/4° (3%)
Structure Type
Methoxy benzene
Dimethoxy benzene
*
Dimethyl benzaldehyde
Benzofuran
2-indanone
Gassy, particle size
0.4
0.2
5.5
8.5
155
170
Peak
Oxidized oil
Burnt phenolic
13 APE 1.5 to 1.6
220(4)
134(6)
0V 158°
CsS°
SP 70°/4° (6%)
Methoxy benzene
3-coumaranone
Leathery, sour, pungent
0.3
1.0
0.3
0.3
0.4
0.4
0.1
0.2
0.2
8.5
11.0
11.5
12.0
2.0
5.0
3.5
4.0
4.5
147
162
170
179
Peak
142
165
Peak
161
162
165
Oxidized oil
Burnt, smoky
Burnt , smoky
Pungent, burnt smoky
14 APE 1.6 to 1.75 -
Smoky ,
Chalky
Pungent, garlic
15 APE 1.75 to 1.8 -
Smoky ,
Metallic
Smoky , tarry
Pungent, garlic
152(3)
132(6)
136(5)
136(4)
- 0V 162° -
pungent
120(5)
132(6)
- 0V 165° -
pungent
148(5)
160(6)
146(6)
C10H16°
C9H80
C9H120
Dienone
*
1-indanone
*
Hydroxy acetophenone
*
Trimethyl phenol
- SP 100°/4° (6%)
C8H80
C9H80
Benzaldehyde
Indanone*
- SP 100°/4° (2%)
C10H120 .
C11H12°
C10H10°
Benzaldehyde
*
D ime thy 1-1- indanone
1-tetralone
-114-
ArthurD Little Inc
-------
TABLE F-2
Mass Spectrometry - Odor Data - Sample 52C-34/5/6 plus
%
0.3
0.3
0.4
0.5
0.4
0.2
0.3
0.5
1.3
Peak 9 APE 1.0 to 1.1
Oily, woody, burnt,
EXP SPT Odor
2.0 135 Smoky, irritation
2.5 137 Phenolic
3.0 139 Oily linseed, pungent
3.5 140 Green, sour, burnt
4.0 141 Fragrant
4.5 146 Pungent, burnt
6.0 151 Musty, burnt
6.5 153 Sooty
7.0 164 Burnt, pungent, garlic
0V 116° SP 100 °/ 2°
rubbery, particle size
MW(R+DB) ELCOMP
124(3) C8H120
122(4) C8H100
142(2) C8Hltt02
140(3) C8H1202
i oo / / \ r» u f\
Uo(A ) CQrljQ(J2
134(5) C9H100
124(4) C7H802
110(4) C6H602
110(4) C6H602
108(4) C7H80
Structure Type
4
Dimethyl cyclohexenone
*
Methylmethoxy benzene
Hydroxy alkenone
Oxy furan
*
Dimethoxy benzene
*
Dimethyl benzaldehyde
Furan aldehyde
Dihydroxy benzene
Dihydroxy benzene
*
m-Cresol
-115-
Arthur D Little Inc
-------
^CtL/JkW A fc WWLll
% EXP
0.1 2.0
0.1 2.5
0.1 3.0
0.2 3.5
0.2 4.0
0.1 4.5
0.2 5.0
0.1 5.5
0.1 6.0
0.3 6.5
Peak 10 APE 1.:
Phenolic,
SPT Odor
138 Fragrant
Pungent
140 Pungent
141 Fruity
142 Burnt was
142 Pungent
143 Pungent,
144 Oily, spj
148 Metallic
149 Linseed (
153 Smokv. ol
Phenolic, pungent, linseed, woody, smoky
0.1
7.0
155
Sour
MW(R+DB)
136(4)
138(3)
134(5)
96(3)
138(4)
134(5)
134(5)
138(4)
124(4)
136(5)
136(5)
122(4)
120(5)
138(5)
ELCOMP
C9H120
CgH^O
CgH100
C6H80
CsHioOz
C9H100
C9H100
C8H1002
C7H802
C8H802
C8H802
C8H100
C8H80
C7H603
Structure Type
Dimethyl anisole*
Furan
4
Dimethyl benzaldehyde
*
Dimethyl furan
Dimethoxy benzene*
*
Dimethyl benzaldehyde
*
Dimethyl benzaldehyde
Dimethoxy benzene*
Furan aldehyde
*
Hydroxy acetophenone
Isomer
Dimethyl phenol*
Acetophenone
-116-
ArthurDLittleJnc
-------
.fc MLS .klv
%
0.5
0.5
0.3
0.3
0.2
1.2
0.2
0.4
0.2
0.2
0.2
0.3
0.9
0.2
Peak 11
EXP SPI
7.5 145
8.0 141
9.0 156
9.5 160
10.0 167
10.5 172
Peak 12
1.5 149
2.0 150
3.0 160
4.0 164
4.5 166
5.0 168
5.5 169
6.0 175
Peak 11 APE 1.2 to 1.4 0V 126° SP 100°/2° (8%)
Oily, particle size, smoky
Odor
Oily, sour
Particle size
Pungent, metallic
Particle size, spicy
Smoky rubber
Particle size, pungent
MW(R&DB) ELCOMP
114(2) C6H1002
134(5) C9H100
124(4) C7H802
134(6) C8H602
122(5) C7H602
Structure Type
Hydroxy alkenone
t
Dimethyl benzaldehyde
Methoxy phenol*
Oxy benzofuran
122(4) C8H100 Xylenol*
SP 100°/2° (5%)
Burnt oily, burnt rubber, pungent
Musty
Burnt oil
Tarry
Particle size, metallic
Burnt sweet
Chalky, particle size
Linseed oil, pungent
Irritation
148(5)
156(2)
138(3)
152(4)
138(4)
134(5)
152(5)
136(4)
134(5)
134(6)
134(6)
122(4)
C10H120
C9H16°2
CgHj i,0
CgHj 202
C8H10°2
CgH100
C9H120
C8H602
C8H100
K
Dimethyl acetophenone
Hydroxy alkenone
Furan
Dimethoxy benzene
Methylmethoxy phenol*
*
Indanol
*
Dihydroxy acetophenone
Benzyl alcohol
*
Allyl phenol
*
2-coumaranone
Hydroxy benzofuran
Xylenol*
-117-
ArthurD Little, Inc
-------
*. I* LJ ^«_> *.
0.2
0.3
0.3
0.2
1.9
0.3
0.2
0.2
0.2
0.2
Peak 1
EXP SPT
1.5 148
2.0 149
3.5 155
4.0 158
4.5 160
5.0 161
5.5 167
6.0 169
6.5 174
7.0 177
L3 APE 1.5 to 1.7 0V U
Musty, green, metallic
Odor
Oily, naphthanate
Sour, particle size
Metallic, linseed oil
Burnt rubber
Irritation
Sweet , chalky
Sweet spicy
Green, fragrant
Linseed oil, naphthanate
Irritation
fO° SP 100°/2° (
, smoky
MW(R&DB) ELCOMP
148(5) C10H120
150(4) CjoH^O
160(6) CnH1?0
146(6) C10H100
138(4) C8H1002
132(6) C9H80
132(6) C9H80
134(5) C9H100
148(6) C9H802
146(7) CgH602
132(6) C9H80
136(4) C9H120
8%)
Structure Type
A
Ane thole
Methoxy benzene
Indanone
Indanone
Methylmethoxy phi
Indanone*
Cinnamaldehyde
Allyl phenol
Hydroxy benzofur;
Hydroxy Indenone
Indenol
Trimethyl phenol
136(6)
-118-
ArthurD Little, Inc
-------
Table F-2 cont.
Peak 14 APE 1.7 to 1.8 0V 145° SP 100°/2° (5%)
Sweet, oily, burnt
%
0.4
0.3
0.3
0.2
0.2
0.2
0.4
0.2
0.1
0.4
EXP
8.5
9.0
9.5
10.0
10.5
11.5
12.0
12.5
13.0
13.5
SPT
156
158
160
167
171
176
178
180
181
185
Odor
Smoky exhaus t
Sour, metallic
Sour, oily
Painty
Metallic, linseed oil
Oily, metallic
Sour, oily
Smoky
Smoky fat
Asphyxiating, sweet
MW(R+DB)
180(4)
168(3)
164(4)
148(5)
160(6)
168(3)
146(6)
166(4)
144(7)
146(7)
148(6)
150(5)
150(6)
152(5)
150(4)
134(6)
ELCOMP
C11H16C
C10H16C
CiiHie^
C10H12C
CUH12C
C10H16C
C10H10C
f U f
C10H14C
C10H80
CgHg02
C9H802
c9Hio°:
C8H603
C8H803
c!0Hll+(
C8H602
Structure Type
C11H16°2 Dimethoxy benzene
Methoxy phenol
Indenone
Hydroxy indenone
Hydroxy indanone
C9H1002 Methoxy benzaldehyde
*
Piperonal
Piperonyl alcohol
CIOHII+O Tetramethyl phenol*
Phthaldehyde
-119-
ArthurD Little, Inc
-------
Peak 15 APE 1.8 to 1.9 0V 150° SP 120°/2° (5%)
0.1
0.2
0.2
0.2
0.2
0.3
0.5
0.2
0.6
0.9
Smoky, oily, irritation
EXP SPT Odor MW(R+DB)
10.5 154 Burnt metal 178(4)
11.0 155 Chalky 180(4)
11.5 158 Smoky, metallic 174(6)
12.5 169 Smoky fat 146(6)
13.0 170 Irritation 164(5)
162(6)
14.0 182 Soft coal smoke 148(6)
14.5 183 Smoky 148(6)
Peak 16/17 APE 1.85 to 2.05 0V 158°
Smoky , sweet , waxy
EXP SPT Odor MW(R&DB)
2.5 144 Burnt sweet, pungent 134(5)
8.0 184 Smoky 162(6)
9.0 190 Leathery, sour 162(6)
ELCOMP
C12H180
C11H16°2
C10H100
C10H12°2
C10H10°2
C9H802
C9H802
SP 120°/2(
ELCOMP
CgH100
C10H10°2
C10H10°2
Structure Type
Methoxy benzene
Dimethoxy benzene
*
Indanone
Indanone
Hydroxy benzaldehyde
Hydroxy indanone
Hydroxy indanone
Hydroxy indanone
3 (8%)
Structure Type
Benzaldehyde
Hydroxy indanone
Hydroxy indanone
160(7)
Hydroxy indenone
0.4
9.5
192
Smoky
156(8) CUH80 Naphthaldehyde
-120-
ArthurD Little, Inc
-------
TABLE F-3
Mass Spectrometry - Odor Data Sample 60C-39/40/41
Peak 18/19/20 APE 1.95 to 2.15 0V 166
° - SP 120°
/2° (9%)
Smoky, burnt, metallic, oily
0.3
0.6
0.1
0.3
0.3
0.3
0.5
0.3
EXP SPT odor MW(R&DB)
3.5 188 Smoky 162(6)
156(8)
4.0 192 Oxidized oil 176(6)
4.5 194 Burnt tarry 148(6)
134(5)
5.0 200 Burnt rubber 162(6)
160(7)
Peak 21/22 APE 2.15 to 2.30 0V 182°
Smoky, burnt oil
9.0 190 smoky oil 148(6)
146(7)
9.5 191 smoky metallic 160(7)
10.0 193 pungent, oxidized 162(6)
148(6)
11.0 202 smoky, oily 160(7)
ELCOMP
C10H10°2
CUH80
C11H12°2
C9Hi00
C10H10°2
SP 120° /
C9H802
C9H602
C10H10°2
C9H802
C10H8°2
Structure Type
Hydroxy indanone
Napthaldehyde
Hydroxy indanone
Hydroxy indanone
Indanol
Hydroxy indanone
Hydroxy indenone
2° (6%)
Hydroxy indanone
Hydroxy indenone
Hydroxy indenone
Hydroxy indanone
Hydroxy indanone
Hydroxy indenone
-121-
Arthur D Little, Inc
-------
TABLE F-4
%
0.2
0.2
0.2
0.4
0.6
0.6
0.3
0.2
0.6
0.4
0.2
EXP
7.0
8.0
8.5
9.0
9.5
10.0
10.5
11.0
11.5
12.0
12.5
SPT
129
134
138
140
142
143
144
145
146
147
149
0.3
13.0
150
Spectrometry - Odor Data
APE 1.1 to 1.3 0V
- Sample
56P-40
124° SP 100°/2°
Oxidized Oil, linseed
Odor
Metallic, pungent
Oxidized oil, aldehydic
Burnt sweet
Linseed, irritation
Oily aldehydic
Fruity
Pungent
Burnt woody
Pungent, sour
Oxidized oil, fragrant
Irritation
Burnt sweet
MW(R+DB)
180(3)
166(3)
168(2)
154(2)
152(3)
150(4)
148(5)
138(3)
136(4)
134(5)
148(5)
134(5)
148(5)
134(5)
134(5)
148(5)
136(5)
132(6)
136(5)
150(5)
130(7)
ELCOMP
C12H20°
c 1 1H 1 8°
C11H20°
C10H180
ClQHlsO
C10HlltO
C10H12°
C gH i i^O
CgH120
C9H100
C10H12°
CgH100
C10H120
CgH1Q0
CgH1Q0
C10H120
CeH802
CgHgO
C8H802
CgH1Q02
C9H60
(6%)
Structure Type
Dienone
Dienone
Alkenone
Alkenone
Dienone
Methoxy benzene
Benzaldehyde
Furan
Methoxy benzene
*
Dimethyl benzaldehyde
Dimethyl acetophenone
*
Dimethyl benzaldehyde
Benzaldehyde
*
Dimethyl benzaldehyde
Benzaldehyde
*
Dimethyl acetophenone
*
Hydroxy acetophenone
Phenyl vinyl ketone
Hydroxy acetophenone
*
Methoxy benzaldehyde
Indenone
-122-
ArthurD Little, Inc
-------
4. fc» W -^ W t
%
0.3
0.3
0.4
0.3
0.2
0.2
0.1
0.1
Peak
EXP
3.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
lib,
SPT
133
145
147
149
153
159
160
167
lie & 12 APE 1.3
Pungent ,
Odor
Fragrant
Pungent, spicy
Sour, leathery
Sour dish rag
Solvent, linseed
Naphthenate
Chalky, spicy
Sour naphthenate
MW(R+DB)
166(3)
162(4)
148(5)
150(4)
148(5)
150(4)
148(5)
144(7)
150(5)
132(6)
132(6)
ELCOMP
CUH180
C12H18
C10H12°
C10HlltO
C10H12°
C10H11+0
C10H120
C10H80
C9H10°2
C9H80
C9H80
Structure Type
Dienone
Alkyl benzene
*
Dimethyl acetophenone
Methoxy benzene
*
Anethole
Methoxy benzene
Benzaldehyde
Indenone
Methoxy benzaldehyde
1-indanone
Indenol
2.0
Peak 13 APE 1.5 to 1.65 0V 140° SP 100°/2° (8%)
5.0 150
Pungent, oxidized oil, smoky fat
Burnt fat 162(5)
Benzaldehyde
148(5) C10H120 Benzaldehyde
Peak 14 APE 1.65 to 1.75 0V 145° SP 100°/2° (7%)
Oxidized oil, aldehyde
0.6
0.4
9.0
9.5
149
161
Aldehydic
Oily
-123-
194(3) C13H220 Dienone
180(3) C12H200 Dienone
148(5) C10H120 Benzaldehyde
162(5) CnH140 Benzaldehyde
Arthur D Little, Inc
-------
t*L?^^ A. T ^.WL&ht
Peak 15, 16, 17 APE 1.75 to 2.0 0V 152° SP 120°/2° (16%)
Smoky, linseed, burnt fat
%
1.1
1.1
0.8
0.5
EXP SPT
1.5 141
2.0 158
2.5 163
3.0 167
Odor
Sour
Oxidized oil, tarry,
pungent
Metallic, burnt
Pungent, leathery
Peaks 18/19 APE 2.0 to 2.25
0.3
1.2
0.1
6.5 185
7.0 187
7.5 192
Peaks
Burnt oil, pungent,
Particle size, smoky
Woody, irritation
Sour, leathery
20/21/22 APE 2.25 to 2.6
MW(R+DB)
194(3)
190(4)
180(2)
190(5)
178(4)
160(6)
158(7)
160(6)
0V 162°
ELCOMP
C13H220
C11+H22
C13H14
C13H180
C12H180
C.UH120
C11H10°
CnH120
SP 120°/
Structure Type
Dienone
Alkyl benzene
Diene
Benzaldehyde
Methoxy benzene
Indanone
Indenone
*
Methyl tetralone
'2° (13%)
green , smoky
172(8)
156(8)
160(7)
162(6)
0V 175°
C11H8°2
CUH80
C10H8°2
C10H10°2
SP 120° /
*
Methyl naphthoquinone
1-Naphthal
Hydroxy indenone
Hydroxy indanone
'2° (21%)
Burnt predominates
0.4
0.4
0.8
1.7
1.3
0.1
1.5 187
2.0 192
2.5 194
3.0 196
3.5 198
4.0 204
Smoky , burnt
Irritation, smoky
Smoky
Pungent, sour, oily
Smoky, particle size
Pungent , smoky
204(6)
200(7)
190(6)
186(8)
170(8)
170(8)
170(8)
174(7)
176(6)
C13H16°2
ClltH160
C12Hll+02
C12H10°2
C12H100
C12H100
C12H10°
C11H10°2
Cl 1H12°2
Hydroxy indanone
Indenone
Hydroxy indanone
Methoxy naphthaldehyde
Acetonaphthone
Naphthaldehyde
*
Ac e tonaph thone
Hydroxy indenone
*
Methoxy tetralone
-124-
ArthurD Little, Inc
-------
APPENDIX G
SMOKY-BURNT STRUCTURE DATA ORGANIZED BY R+DB VALUES
-125-
ArthurD Little, Inc
-------
APPENDIX G
Smoky-Burnt Structure Data Organized by R+DB Values
Attempts to find suitable means of assembling a summary
of the observed data have been frustrated by the great degree of over-
lap observed in structure-odor relationships. These attempts are
further confused by the presence of several functional groups on the
same nucleus. The one means which we have found convenient is to list
the observed data according to the compositionally defined R+DB values.
In this way, any subsequent revisions of specific structure assignments
would not require complete rearrangement of the data. All of the data
are summarized in this manner in Table G-l.
The data in the table are arranged in order of increasing
R+DB value. The possible hypothetical structure for given combina-
tions of oxygen from Appendix E are also included for referencing con-
venience. The data obtained from the chloroform and pentane extract
samples are separated by the double line ( ).
In examining the chemical data in Table G-l and observing
the repetition of certain species, it might at first seem that a single
species from the exhaust has been reported several times. The data in
the table has actually been assembled with care taken to eliminate dup-
lication of species from Appendix F which did have the same GC reten-
tion times in either the chloroform or pentane extracts. It is our
present interpretation that nearly each of the species listed repre-
sents a unique isomer of a given molecular formula. There is, however,
some duplication in reporting between the chloroform and pentane series.
-127-
ArthurD Little, Inc
-------
TABLE G-l
Individual Species Observed in the Smoky-Burnt Studies
R + DB CLASS 1
Q! - ketones, aldehydes, cyclic-ols
MW
a 114
ELCOMP
C7HlltO
Odor
Linseed Oil
Structure type
Heptanone
R + DB CLASS 2
0, - alkenones, cyclic carbonyls
02 - alkene or cyclic acids and esters, dicarbonyls
dicarbonyls, oxy cyclo carbonyl,
oxy alkenones
MW
ELCOMP
142
100
114
a 156
168
154
C8HlltO
C5H802
C6H10°2
C9H16°2
CUH200
C10H180
Odor
Linseed oil, pungent
Burnt oil
Oily sour
Musty
Metallic, Pungent
Oxidized oily, aldehydic
Structure Type
Hydroxy alkenone
Valerolactone
Hydroxy alkenone
Hydroxy alkenone
Alkenone
Alkenone
a. data above
data below
observed in chloroform extracts
observed in pentane extract
-123-
Arthur D Little, Inc
-------
Table G-l cont.
R+DB CLASS 3
- Dienones, furans, cyclic-ene carbonyls
- MPS,s, cyclodiones, oxy furans
MW
124
126
140
96
96
152
138
168
168
180
166
152
138
194
180
ELCOMP
C8H120
C7H1Q0
C8H12°2
C6H8°
C6H8°
C10H16°
C9Hl-*°
C10H16°2
C10H16°2
C12H200
C11H18°
C10H16°
C9H1ItO
C13H22°
C,,H,nO
Odor
Irritation, smoky
Leathery
Green, sour, burnt
Sour, leathery
Irritation
Oxidized oil
Pungent
Burnt oil
Sour, oily
Metallic, pungent
Metallic, pungent
Oxidized oily, aldehydic
Burnt sweet
Aldehydic
Aldehydic
Structure Type
Cyclohexenone
Cyclohexanedione
Oxy furan
Furan
Furan
Dienone
Furan
Hydroxy furan
Hydroxy furan
Dienone
Dienone
Dienone
Furan
Dienone
Dienone
-129-
ArthurD Little Inc
-------
Table G-l cont.
R+DB CLASS 4
Oj - Phenols, methoxy benzenes, benzyl alcohols
02 - Methoxy (hydroxy) phenols, furfurals
MW
122
152
220
136
122
136
138
124
108
110
110
136
138
183
124
122
124
122
136
152
138
122
150
136
180
164
166
150
178
180
150~
150
136
150
178
ELCOMP
C 8H 1 00
CgHi202
C15H240
C8Hi00
CgH^O
C8H10°2
C7He02
C7H80
C9Hi20
C8H1002
C8H10°2
C7H802
C8HioO
C9H120
C8H1002
C8H100
c10Hlit0
CgHi20
C11H16°
Ci0H1it02
C10H14°
Ci2Hi80
C11H1602
Ci0H11+0
C9H120
CigH1(+0
Cl2H18°
Odor
Woody
Smoky
Oxidized oil
Burnt smoky
Phenolic
Smoky
Burnt
Burnt
Burnt, pungent, garlic
Burnt, musty
Sooty
Fragrant
Fruity
Pungent, sour
Oily, spicy
Smoky, phenolic
Pungent, metallic
Particle size, pungent
Pungent, burnt
Burnt oil
Tarry
Irritation
Oily, Naphthenate
Irritation
Smoky, exhaust
Burnt oil
Painty
Smoky fat
Burnt metal
Chalky
Burnt sweet
Pungent, spicy
Sharp
Sour dish rag
Tarry , pungent
Structure Type
Dimethyl phenol
Dimethoxy benzene
Methoxy benzene
Trimethyl phenol
Methoxy benzene
Methoxy benzene
Dimethoxy benzene
Furan aldehyde
m-Cresol
Dihydroxy benzene
Dihydroxy benzene
Dimethyl anisole
Dimethoxy benzene
Dimethoxy benzene
Furan aldehyde
Phenol
Methoxy phenol
Xylenol
Phenol
Dimethoxy benzene
Methoxy phenol
Xylenol
Methoxy benzene
Phenol
Dimethoxy benzene
Methoxy benzene
Methoxy phenol
Phenol
Methoxy benzene
Dimethoxy benzene
Methoxy benzene
Methoxy benzene
Methoxy benzene
Methoxy benzene
Phenol
-130-
ArthurD Little Inc
-------
Table G-l cont,
R+DB CLASS 5
0, - Phenyl carbonyls, indanols, allyl phenols, dihydrobenzofuran
0 - Aromatic acids, hydroxy(methoxy) aromatic carbonyls, oxy indanols,
allyl phenols, quinones
0- - Oxy benzoic acids, dioxy phenyl carbonyls, dioxy allyl phenols,
oxy quinones
MW
ELCOMP
Odor
Structure Type
134
136
120
134
134
134
136
136
120
138
134
122
136
148
134
152
134
148
148
150
152
164
134
148
134
148
148
134
148
136
136
150
162
190
CgHi00
CeH8°2
C8H80
CgHiQO
CgHi00
C8H802
C8H802
C7H603
CgH100
C7H&02
C8H802
CgHioO
C8H803
C 10^120
CgHio02
CeH803
C 10^120
CgH100
C10H120
Ci0H120
CgHjgO
C8H802
C8H802
CgH1002
CiiHlttO
Ci3H180
Smoky
Burnt, smoky
Chalky
Fragrant
Pungent
Burnt wax
Metallic
Linseed Oil
Smoky
Sour
Particle size
Smoky rubber
Burnt
Musty
Particle size
Metallic
Burnt sweet
Oily, Naphthenate
Smoky exhaust
Oily, metallic
Smoky
Irritation
Burnt, sweet, pungent
Burnt sweet
Linseed, irritation
Oily aldehydic
Fruity
Burnt woody
Pungent, sour
Pungent, sour
Irritation
Naphthenate
Burnt fat
Oxidized oil
Benzaldehyde
Hydroxy acetophenone
Benzaldehyde
Benzaldehyde
Benzaldehyde
Benzaldehyde
Hydroxy acetophenone
Isomer
Acetophenone
?
Benzaldehyde
?
Hydroxy benzaldehyde
Acetophenone
Indanol
Dihydroxy acetophenone
Allyl phenol
Anethole
Benzaldehyde
Methoxy benzaldehyde
Piperonyl alcohol
Hydroxy benzaldehyde
Indanol
Benzaldehyde
Benzaldehyde
Acetophenone
Benzaldehyde
Benzaldehyde
Acetophenone
Hydroxy acetophenone
Hydroxy acetophenone
Methyoxy benzaldehyde
Benzaldehyde
Benzaldehye
-131-
Arthur D Little Inc
-------
Table G-l cont.
R+DB CLASS 6
DI - benzofurans, indanones, indenols, phenyl-ene-carbonyls
62 - oxy benzofurans, oxy indanones, phenyl propene acids,
dihydrocoumarins, aromatic dialdehydes
03 - dioxy allyl benzaldehydes
MW
146
132
146
134
134
134
160
146
132
132
148
132
136
160
146
148
150
134
174
146
162
148
148
162
162
176
132
132
132
160
160
162
204
190
176
ELCOMP
C9H80
C8H602
C8H602
C8H602
CUH120
C9H80
C9H80
C9H802
C9H80
CnH120
C9H802
C8H603
C8H602
ClOH10°2
C9H802
C9H802
C10H10°2
CnH1202
C9H80
CgH802
CgH802
CllH120
CHH120
C10H10°2
C12H11+02
Odor
Pungent
Sour
Pungent, garlic
Particle size, spicy
Chalky, particle size
Linseed oil, pungent
Sour, particle size
Metallic Linseed oil
Irritation
Sweet, chalky
Sweet, spicy
Linseed oil, naphthenate
Irritation
Sour, Metallic
Sour, oily
Oily, metallic
Sour, oily
Asphyxiating, sweet
Smoky, metallic
Smoky fat
Irritation
Soft coal smoke
Smoky
Smoky
Sour
Oxidized oil
Oxidized oil, fragrant
Chalky, spicy
Sour naphthenate
Metallic, burnt
Pungent, leathery
Sour, leathery
Smoky
Smoky, irritation
Pungent, smoky
Structure Type
Benzofuran
Indanone
Tetralone
Oxy benzofuran
Coumaranone
Hydroxy benzofuran
Indanons
Indanone
Indanone
Cinnamaldehyde
Hydroxy benzofuran
Indenol
?
Indanone
Indanone
Hydroxy indanone
Piperonal
Phthaldehyde
Indanone
Indanone
Hydroxy indanone
Hydroxy indanone
Hydroxy indanone
Hydroxy indanone
Hydroxy indanone
Hydroxy indanone
Phenyl vinyl ketone
Indanone
Indenol
Indanone
Tetralone
Hydroxy indanone
Hydroxy indanone
Hydroxy indanone
Methoxy tetralone
a. from CKJ tetralones may also be considered as possible
choices for indanones.
-132-
ArthurD Little, Inc
-------
Table G-l cont.
R+DB Class 7
0 - Naphthols, indenones
» - Oxy naphthols, coumarins, oxy indenones
MW
144
146
160
160
146
160
130
144
158
160
200
174
ELCOMP
C10H8°
C9H6°2
C10H8°2
C10H8°2
C9H6°2
C10H8°2
C9H6°
C10H8°
C11H10°
C10H8°2
C14H16°
C11H10°2
Odor
Painty
Metallic, linseed oil
Leathery, sour
Burnt rubber
Smoky oil
Smoky metallic
Burnt sweet
Naphthenate
Pungent, leathery
Sour, leathery
Smoky burnt
Smoky, particle size
Structure Type
Indenone
Hydroxy indenone
Hydroxy indenone
Hydroxy indenone
Hydroxy indenone
Hydroxy indenone
Indenone
Indenone
Indenone
Hydroxy indenone
Indenone
Hydroxy indenone
-133-
ArthurD Little, Inc
-------
Table G-l cont.
R+D Class 8
01 - Naphthaldehydes
0- - Naphthoquinones, oxy naphthaldehydes
MW
156
172
156
186
170
170
ELCOMP
CllV
C11H8°2
C11H8°
C12H10°2
C12H1()0
C12H10°
Odor
Smoky
Particle size, smoky
Woody, irritation
Irritation, smoky
Smoky
Smoky, particle size
Structure Type
Naphthaldehyde
Naphthoquinone
Naphthaldehyde
Methoxy naphthaldehyde
Acetonaphthone
Acetonaphthone
-134-
ArthurD Little; Inc
-------
APPENDIX H
OXYGENATED REFERENCE COMPOUNDS
-135-
ArthurD Little, Inc
-------
APPENDIX H
OXYGENATED REFERENCE COMPOUNDS
1. GAS CHRGMATOGRAPHIC STUDIES
In an effort to confirm the chemical structure assignments and odor
description associated with oxygenated compound classes related to diesel
exhaust smoky-burnt odor, we have sought to obtain as many oxygenated ref-
erence compounds as appropriate based on the exhaust identification data
and study their chromatographic behavior and odor characteristics. At the
present time, 144 compounds have been studied, 33 of these have been iden-
tified in diesel exhaust. We will continue to study new materials as they
come to our attention. Unfortunately, we have been unable to locate many
of the compounds we wish to examine.
The GC retention data from the silicone OV-1 and Carbowax SP-1000
columns and the species fundamental molecular weight and elemental compos-
ition provide a precise basis for comparing data observed in the exhaust
analysis. The data thus obtained for these materials is detailed in
Table H-l. The compounds are listed in order of increasing elution from
the silicone column. In obtaining the elution temperatures, primary elu-
tion standards were run with each material so that any changes in chromato-
graphic behavior could be referenced to a common base. The data in the
table are organized based on elution scales for each column. We have
found it convenient to reference the silicone elution temperature to that
of allyl phenyl ether, thus arriving at an APE value indicative of the
relative elution behavior of each of the materials. The relationship be-
tween the silicone elution temperature, APE value, and exhaust sample peak
numbers is shown in Figure E-2 of Appendix E. The data in Table H-l are
organized in order of increasing APE value and are listed to indicate that
the compounds listed between adjacent APE values have retention times in
that range. The Carbowax GC index is listed as an SPT (SP-1000 tempera-
ture) value and references all of the Carbowax retention temperatures to
a common program rate of an initial 120°C and a 2°C/min heating rate.
These data have been compared to the smoky-burnt identification data
reported in Appendix F. The requirement for a match used was that the two
sets of GC data and elemental composition fit, as well as the requirement
that the observed mass spectral fragmentation pattern be consistent with
the structure. Species identified as being present in the exhaust samples
-137-
ArthurD Little, Inc
-------
have been indicated with an asterisk (*). The data has been useful both
in confirming HRMS assignments or refining the isoraeric details and also
in eliminating such species as hydroxybenzoic acid and others which elute
outside the regions studied.
2. TEST ROOM ODOR STUDIES
Many of the reference materials were examined for odor characteristics
by study of the sample either directly in the bottle or on blotter strip.
Several of these had sufficient odor intensity and the appropriate odor
character to warrant quantitative study in the odor test room. The results
of these studies are listed in Table H-2. Most of the studies were carried
out with a solution (Soltrol 170 or methanol) injection of the material into
the odor test room to give a concentration ranging from 0.1 - 12 yg/m3.
Clearly many of the compounds studied have an odor character consistent
with the assignments made in the exhaust samples. Of further significance,
however, is that they also have an odor intensity comparable to the levels
calculated to be present in the exhaust samples studied in the test room.
-138-
Arthur D Little Inc
-------
Table H-l
Gas Chromatographic Data on Oxygenated Reference Standards
APE"
0.20
0.30
0.50
MW
0.60
0.70
0.8
Compound
2,5 dimethylfuran
2,4 pentanedione
4-benzoquinone
cyclohexanol
2-furylmethyl ketone
camphor
Y-valerolactone
* 4-hydroxy-2-pentenone
5-methyl-2-furaldehyde
benzaldehyde
phenol
3,4 & 3,5-cyclopentanediol
2-cyclohexene-1-one
2-methylanisole
* 1-methyl-cyclopentene-2-ol-3-one
2,3-benzofuran
1,4-cyclohexanedione
4-methylanisole
* phenylacetaldehyde
acetophenone
a. Retention on OV-1 relative to allyl phenyl ether.
-139-
ELCOMP R&DB
SPT
96
100
108
100
110
152
100
100
no
106
94
102
96
122
112
118
112
122
120
120
C6H80
C5H802
w c n 1 i v o
O H £,
C8H120
C6H602
C10H16°
C5H802
C5H802
C6H602
C7H60
C6H60
C5H1002
C6H80
C8H100
C6H802
C8H60
C6H802
C8H100
C8H80
C8H80
3
2
5
1
4
3
2
2
4
5
4
1
3
4
3
6
3
4
5
5
138
127
132
128
131
126
137
130.5
134.5
132
158
163.5
130
128
146
131
153
131
138
138
Arthur D Little, Inc
-------
APE
0.90
1.00
1.10
1.20
Compound MW
* o-cresol
salicylaldehyde
3-methylbenzaldehyde
benzyl alcohol
2,5-dimethyl-3-hexyne-2,5-diol
* m-cresol
p-cresol
allylphenyl ether (elution standard) 134
2,6-dimethylam'sole
* 3,5-dimethyl-2-cyclohexene-l-one
* 2,6-dimethylphenol
2-furanacrolein
* 1,2-dimethyoxybenzene
2-methylbenzyl alcohol
4-methylbenzyl alcohol
2,5-dimethyl phenol
1,3-dimethoxybenzene
* 4-methylsalicylaldehyde
5,5-dimethyl-l,3-cyclohexanedione
2,4-xylenol
* 3,5-dimethylphenol
* 2,5-dimethylbenzaldehyde
2-methyl-1,3-cyclohexanedione
4-methoxyphenol
1-phenyl-l,2-propanedione
2-methylbenzoic acid
* 2,4-dimethylbenzaldehyde
ELCOMP R&DB SPT
108
122
120
108
142
108
108
134
136
124
122
122
138
122
122
122
138
136
140
122
122
134
126
124
148
136
134
C7H80
C7H602
C8H80
C7H80
C8Hltt02
C7H80
C7H80
CgHjC)0
C9Hi20
C8H120
C8H100
C7H602
C8H1002
C8Hi00
C8H100
C8H100
C8H1002
C8H802
C8H1202
C8H100
C8H100
C9H100
C7H1002
C7H802
C9H802
C8H802
CgH100
4
5
5
4
2
4
4
5
4
3
4
4
4
4
4
4
4
5
3
4
4
5
3
4
6
5
5
157
140
137
148
146
164
162.5
133
130
137
156
148.5
141
159
157
164
143
145
168
163
172
143.5
179
186'
146
194
144
-140-
ArthurD Little, Inc
-------
APE
1.30
1.40
1.50
1.60
Compound
4(2-furyl)-3-butene-2-one
* 2-methoxy-4-methylphenol
* 2-hydroxyacetophenone
3,4-dimethylbenzaldehyde
3-methoxyphenol
* 3,4-xylenol
* 2-allylphenol
l-(2-furyl)-l,3-butanedione
3,4-dihydro-l-(2H)-naphthalenone
2-methoxybenza1dehyde
* 2,4-dimethylacetophenone
3-methoxybenzaldehyde
5-methylfurfuryl alcohol
* cinnamaldehyde
* 2-indanone
* 2-coumaranone
* 1-indanone
3,3,4,7-tetramethy1 -1-i ndanone
3-hydroxybenzaldehyde
* a-hydroxyacetophenone
2-hydroxy-3-methoxybenzaldehyde
MW
ELCOMP R&DB
1,3-indandione
5-indanol
2,6-dimethoxyphenol
* 2,3,5-trimethylphenol
cinnamic alcohol
1,4-cyclohexanedimethanol
-141-
136
138
136
134
124
122
134
152
144
136
148
136
112
132
132
134
132
188
122
136
152
146
134
154
136
134
144
C8H802
C8H1002
C8H802
C9H100
C7H802
C8H100
C9H100
C8H803
C10H80
C8H802
C10H120
C8H802
C6H802
C9H80
C9H80
C8H602
C9H80
C13H160
C7H602
C8H802
C8H803
C9H602
C9H100
C8H10°3
912^
C9H100
C8H1602
5
4
5
5
4
4
5
5
7
5
5
5
3
6
6
6
6
6
5
5
5
7
5
4
4
5
1
SPT
162
155
147
148
191.5
172.5
168
160
165.5
157.5
147
153.5
143
160.5
154
164
160
163.5
213
170
173
182
190
178
173.5
178
190
Arthur D Little, Inc
-------
APE
1.70
1.80
1.90
Compound
* Anethole
4-methoxyacetophenone
2-methyl-5-isopropylphenol
2-hydroxybenzoic acid
2-hydroxybenzyl alcohol
* piperonyl alcohol
2,3-dimethoxybenzaldehyde
3,3,7-trimethyl-l-indanone
2,3-dimethoxybenzyl alcohol
3-methoxybenzoic acid
2,3,5,6-tetramethylphenol
coumarin
4-hydroxybenzaldehyde
o-anisaldehyde
* 2-methyl-l-tetralone
5-methyl-2-hydroxybenzoic acid
* 3,3,5-trimethyl-l-indanone
3,4-dimethoxybenzaldehyde
1,2-cyclodecanedione
5,7-dimethyl-l-indanone
3-methoxy-4-hydroxyacetophenone
4,7-dimethyl-l-indanone
p-(l,1-dimethylpropyl)phenol
2,3,5,6-tetramethylacetophenone
* 1-tetralone
MW
ELCOMP R&DB
SPT
148
150
150
138
124
152
166
174
168
152
150
146
122
136
160
152
174
166
168
160
166
160
164
176
146
C]oHi20
CgHio02
CioHmO
C7H603
C7H802
C8H803
CgH1003
Ci2Hn»0
CgH1203
C8H803
CioHmO
C9H602
C7H602
C8H802
CUH120
C8H803
C12HlttO
CgH1003
C10H1602
CnHi20
CgHioOs
CnHi20
CnHi60
Ci2Hi60
Ci0Hi00
5
5
4
5
4
5
5
6
4
5
4
7
5
5
6
5
6
5
3
6
5
6
4
5
6
148
169
172
(b)
155
176
171
159
187
217
171
190
220 + 2.2 min
156
166
220 + 33 min.
159
189
143
174
204
175.5
187
165
166
-142-
ArthurD Little, Inc
-------
APE Compound MW ELCOMP R&DB SPT
2.00
coumarin 146 C9H602 7 190
* piperonal 150 C8H603 5 175
2-hydroxy-3-methylbenzoic acid 152 C8H803 5 (b)
3,5-dimethoxybenzyl alcohol 168 C9H1203 4 209
4-hydroxyacetophenone 136 C8H802 5 220+5 min.
2-hydroxy-4-methoxyacetophenone 166 C9H1003 5 179
5-acetylindan 160 CHH120 5 174 '
2,4-dimethoxybenzaldehyde 166 C9H1003 5 195
2,4-dihydroxybenzaldehyde 138 C7H603 5 220 + 12 min.
3,3,5,6-tetramethyl-l-indanone 188 C13H160 5 179
3,3,4,6-tetramethyl-l-indanone 188 C13H160 5 169
3,3,5,7-tetramethyl-l-indanone 188 C13H160 6 166
2.10
2,6-dihydroxyacetophenone 152 C8H803 5 220 + 11 min.
1-naphthol 144 C10H80 7 220+1 min.
2-methoxybenzoic acid 152 C8H803 5 212
2,3-dimethylbenzoic acid 150 C9H1002 5 206
5,6,7,8-tetrahydronaphthol 148 C10H120 5 202
2.20
2,4-dihydroxyacetophenone 152 C8H803 5 (a)
3,4-dimethoxyacetophenone 180 C10H1203 5 194
2,5,8-trimethyl-l-tetralone 188 C13H160 6 181
. 3,5,8-trimethyl-l-tetralone 188 C13H160 6 182
3,3,6,8-tetramethyl-l-tetralone 202 ClltH180 6 175
-143-
ArthurD Little, Inc
-------
APE
2.30
2.40
2.50
2.60
2.70
2.80
Compound
* 1-acetonaphthone
l,2,3,4-tetrahydro-2,5,8-trimethyl.
1-naphthol
4,5,7-trimethyl-l-indanone
4,6,7-trimethyl-l indanone
5-methoxy-l-tetralone
5,7-dimethyl-1-tetralone
* 2-methyl-l,4-naphthaquinone
6,7-dimethy!-1-tetral one
3,3,4,5,7-pentamethyl-l-indanone
3,3,5,6,7-pentamethyl-l-indanone
4,5,8-trimethyl-l-tetralone
* 2-acetonaphthone
2-hydroxy-l-naphthaldehyde
2-allyl-4,5-dimethylphenol
5,6-dimethyl-l-tetral one
* 6-methoxy-l-tetralone
1-naphthaldehyde
l,2,3,4-tetrahydro-8-isopropyl-
2,5-dimethyl-1-naphthol
2-allyl-4,5-dimethylphenol
6-methyl coumarin
4(e-methy1 ethylketo)phenol
MW ELCOMP R&DB SPT
170
190
174
174
176
174
172
174
202
202
188
170
172
162
174
176
156
218
Ci2HioO
C13H180
C12HlltO
C12HU0
CnH1202
C12HltfO
CnH802
C12H1IfO
ClttH180
ClltH180
C13H160
C12H100
CHH802
CnHlttO
C12HllfO
CHH1202
CnH80
C H 0
15 22
8
5
6
6
6
6
8
6
6
6
6
8
8
5
6
6
8
5
192
193
191
191
196.5
190
184
192
187
183
181
201
202.5
191
197.5
206.5
188
192.5
162 CnHlltO 5 191
160 C10H802 6 199.5
164 C10H1202 5
-144-
ArthurD Little, Inc
-------
3.00
Compound MW ELCOMP R&DB SPT
2,3-dihydroxynaphthalene 160 C10H802 7 (a)
2,7-dihydroxynaphthalene 160 C10H802 7 (a)
3.20
2-hydroxy-2-methyl- 188 CnH803 8 213
1 ,4-naphthoquinone
(a) The compound does not elute off the Carbowax column within
the temperature range of interest
* Compounds identified in diesel exhaust
-145-
ArthurD Little, Inc
-------
Table H-2
Test Room Odor Evaluation of Oxygenated Reference Compounds
Compound
m-cresol
2,3 dimethylphenol
3,5 dimethylphenol
2,6 dimethylphenol
3,4 dimethylphenol
2,6 dimethoxyphenol
tetramethylphenol
1 - methyl-4-isopropylphenol
2 - methyl-5-isopropylphenol
allyldimethylphenol
allylphenol
5-indanol
tropolone
1 - naphthaldehyde
2 - hydroxynaphthaldehyde
terephthaldehyde
4 - methylsalicylaldehyde
salicyladldehyde
3,4 - dimethoxybenzaldehyde
p - methoxybenzaldehyde
2,4 - dihydroxybenzaldehyde
2 - methyltetralone
5 - methoxytetralone
1 - methylcyclopentene
ol-2-one-3(MCP)
nonylalchohol
2 - nonenal
Cone, in Test
Room (mg/M3)
TIA
12.0
0.4
0.4
0.4
0.4
1.0
1.0
1.0
0.1
1.0
1.0
1.0
1.0
0.4
1.0
0.4
1.0
0.4
0.4
0.4
0.4
0.5
1.0
0.1
0.1
0.1
h
h
h
h
)(
1
1
h
%
)(
l
h.-l
)(
)(
)(
%
h
)(
Js-i
h
)(
h
^
h
h
h
Odor Character
phenolic, irritation
tarry, solventy
burnt, sooty
medicinal, phenobic
irritation
smoky, caramel
hot wood
metallic, burnt sweet
sour, burnt, oily
non recognizable
metallic
tarry naphthenate
sweet, musty
pungency
irritation
sour, irritation
metallic, burnt wax
sweet, irritation
burnt, pungent
burnt, pungent
irritation
particle feel, sooty
particle feel, scorched
burnt, sweet
oxidized oily
sour, oxidized oily
-146-
------- |