EPA-R2-73-275
lune 1972
Environmental Protection Technology Series
II
li
i • • •
:•:•:«•>:•:•:•:•:.:«•:•:*:;:
-------
EPA-R2-73-275
ANALYSIS
OF THE ODOROUS COMPOUNDS
IN DIESEL ENGINE EXHAUST
by
P.L. Levins
Arthur D. Little, Inc.
15 Acorn Park
Cambridge, Massachusetts 02140
Contract No. 68-02-0087
Program Element No. 1A1010
EPA Project Officer: John E. Sigsby, Jr.
Chemistry and Physics Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
COORDINATING RESEARCH COUNCIL, INC.
30 ROCKEFELLER PLAZA
NEW YORK, N.Y. 10020
CRC PROJECT CAPE-7-68
and
OFFICE OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
June 1972
-------
This report has been reviewed by the Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Agency, nor does
mention of trade names or commercial products constitute endorsement
or recommendation for use.
11
-------
ACKNOWLEDGMENT
It has been especially rewarding during this contract to pro-
ceed from the knowledge of the odorous diesel exhaust species to devel-
oping means for their measurement. We would like to thank the CAPE-7
Committee members whose guidance and contributions of engine and fuel
technology have been especially helpful:
Dr. Roderick S. Spindt - Chairman, Gulf Research and Development
Roger C. Bascom - Cummins Engine Company
Dr. 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
Mr. Timothy C. Belian's efforts on behalf of the CRC office
are also gratefully acknowledged.
PROJECT TEAM
The research program at ADL has been under the overall manage-
ment of Dr. Philip L. Levins. Technical direction has been provided jointly
by Dr. Levins (analytical) and Mr. David A. Kendall (sensory). Miss Alegria
B. Caragay and Dr. James E. Oberholtzer have assumed major responsibilities
for the analytical program, and Mr. Gregory Leonardos for the sensory eval-
uation.
We would also like to acknowledge the technical assistance of
Mr. James L. Stauffer, Mr. Rafael A. Cruz-Alvarez, and Mr. Ralph C. Steeves.
Odor panel members were David A. Kendall, Gregory Leonardos, Shirley A.
Raymond, Loren B. Sjostrom and Frederick Sullivan.
Ill
Arthur D Little Inc
-------
TABLE OF CONTENTS
Page
ACKNOWLEDGMENT m
PROJECT TEAM iii
LIST OF TABLES vi
LIST OF FIGURES yiii
GLOSSARY X
I. SUMMARY 1
A. Introduction 1
B. Results 1
C. Recommendations 3
II. BACKGROUND 5
III. CHEMICAL ANALYSIS OF ODOR FRACTIONS 9
IV. SAMPLE COLLECTION 11
A. Trapping 11
B. Extraction 11
C. Analytical and Odor Results 14
V. LIQUID CHROMATOGRAPHY 19
A. Methylene Chloride Modification 19
• B. Routine Preparative Liquid Chromatography
Procedure 21
VI. DOSE/RESPONSE MEASUREMENT OF ODOR INTENSITY 25
VII. DEVELOPMENT OF A SIMPLIFIED ANALYTICAL METHOD 31
A. Method Development 31
B. Quantitative Response 38
C. Sensitivity to Fuel Composition 38
D. Response to Exhaust Odor Fractions 45
VIII. STUDY OF ENGINE AND FUEL VARIABLES 47
IV
Arthur D Little hie
-------
Table of Contents cont.
IX. CORRELATION OF ANALYTICAL AND ODOR DATA 55
A. Odor Interaction of Kerosene and Smoky-Burnt
Fractions 55
B. Correlation of Odor Intensity and Sample Mass .. 55
X. REFERENCES 65
APPENDIX A - Experimental Test Facility
APPENDIX B - Odor Profile Measurements
APPENDIX C - Chemical Composition of the Odor
Components in Diesel Exhaust
APPENDIX D - Sample Collection Procedures
APPENDIX E - Liquid Column Chromatography (LCC)
Procedures
APPENDIX F - Gas Chromatographic Mass Analysis
of Exhaust Samples
APPENDIX G - Mass Spectrometric Analysis of
Diesel Exhaust Odor Fractions
Arthur DLitttelnc
-------
VI
LIST OF TABLES
Table No. Page
1 Typical Chemical Analysis Information for Odor
Fractions 10
2 Extraction of Diesel Exhaust Samples from
Chromosorb 102 13
3 Comparison of Diesel Exhaust Collected by Various
Sampling Techniques 15
4 Summary Chromosorb 102 Analytical and Odor
Data 16
5 Static Test Room Odor of Chromosorb Samples 17
6 Chromosorb 102 Detailed Odor Fraction Mass
Spectrometric Analyses 18
7 Methylene Chloride LCC Elution Scheme for
Diesel Exhaust Sample CHROM 32 20
8 Major Species Observed in CHROM-27 LCC-5 22
9 R+DB Oxygenate Analysis of CHROM-27 LCC-5 23
10 Preparative Liquid Chromatography of Diesel
Exhaust Sample CHROM-33 24
11 Dynamic Odor Test Chamber Results - CHROM-33
Odor Fractions 26
12 Comparison of ALC and GC Analysis Methods 43
13 Physical Characteristics and ALC Response of
Various Fuels 44
14 ALC Response Sensitivity to Exhaust Variables
Studied 46
Arthur D Little Inc
-------
List of Tables Cont,
Table No.
Vll
15 Diesel Exhaust Variables Analytical Data
Summary 49
16 Diesel Exhaust Variables Odor Summary 50
17 Mass Spectrometric Composition Analysis of
Odor Fractions (%) 54
18 Odor of Mixtures of LCA and LCO (TIA) 57
19 Summary of Correlatable Odor and Analytical
Data i 60
Arthur D Little Inc
-------
LIST OF FIGURES
Figure No.
1 Liquid Chromatography Diesel Exhaust
Fractionation Procedure 7
2 Chromosorb 102 Total Hydrocarbon Trapping
Efficiency 12
3 Dose/Response Data for CHROM 33 LCA Odor
Fraction 2?
4 Dose/Response Data for CHROM 33 LCO Odor
Fraction 28
5 Analytical Liquid Chromatograph Schematic 32
6 ALC of Chrom 33 - TOE, 0.4 £ 34
7 ALC of Chrom 33 - LCP, 0.75 A 35
8 ALC of Chrom 33 - LCA, 0.36 t 36
9 ALC of Chrom 33 - LCO, 0.75 £ 37
10 Calibration Curve for LCC 4 of No. 1 Diesel Fuel . 40
11 Calibration Curve for Diesel Exhaust LCC 4 41
12 Calibration Curves for Diesel Exhaust LCC 10 42
13 Variation of ALC Response with Aromatic Content
of Fuel 45
14 Comparative Effect of Fuel Type and Injector
Design on Diesel Exhaust Odor (Static Test
Room) 51
15 TIA of LCA Fractions: Effect of Fuel Type and
Injector Design 52
viii
Arthur D Little Inc
-------
Figure No.
List of Figures Cont,
16 TIA of LCD Fractions: Effect of Fuel Type and
Injector Design 53
17 Study OF LCA-LCO Interaction Effects 58
18 Comparison of Exhaust and LCO Odor Intensities 59
19 Correlation of LCA Odor Intensity and Mass 61
20 Correlation of LCO Odor Intensity and "Mass 62
21 Correlation of Exhaust Odor Intensity with LCO 63
IX
Arthur D Little Inc
-------
GLOSSARY
The following terms have been used frequently in the text
and are summarized here with their definitions, for the convenience of
the reader.
GC - Gas Chromatography, used for sample comparison and
quantitative mass measurement.
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 cali-
bration with fuel oil.
HEMS - High Resolution Mass Spectrometry, used for chemical
identification and quantitative mixture analysis of
the oxygenate fraction.
R+DB - Rings plus Double Bonds, a representation of chemical
structure type by expressing the degree of hydrogen
unsaturation (see Appendix G).
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
LCG-10 contain the aromatic and oxygenate exhaust
odor complexes.
ALC - Analytical Liquid Chromatography, the basis for the
instrumental method under development.
LCP - The paraffin fraction isolated from the preparative
liquid Chromatography procedure.
LCA - The aromatic (oily-kerosene) fraction isolated from
the preparative liquid Chromatography procedure.
LCO - The oxygenate (smoky-burnt) fraction isolated from
the preparative liquid Chromatography procedure.
x
Arthur D Little Inc
-------
Glossary cent.
UV - Ultraviolet, the detection mode for the ALC method.
TIA - Total Intensity of Aroma, see Appendix B for details.
Value at 1 H/m3 calculated from dose/response data.
TOE - Total Organic Extract, the total organic exhaust species
isolated from the sample collection by solvent extraction,
yg/S, - Concentration of LCO or LCA exhaust fractions per liter
of exhaust.
mg/k£ - Equal to ug/fc.
£/m3 - Liters of exhaust or exhaust odor fraction per m3 of
air. Reciprocal of dilution.
XI
Arthur D Little Inc
-------
I. SUMMARY
A. Introduction
The chemical species responsible for the major characteris-
tics of diesel exhaust odor have been identified in the previous effort
(1-3) on the CAPE-7 program sponsored jointly by the Coordinating Research
Council and the Environmental Protection Agency. This report represents
the beginnings of an effort to obtain appropriate means for measuring the
odorous species in diesel exhaust and develop the quantitative relation-
ships between these measurements and the exhaust odor.
Through the integrated application of analytical chemistry
and sensory methods, the major odorous species have been identified.
Diesel exhaust odor can be described as having two major odor groups—
oily-kerosene and smoky-burnt. The oily-kerosene odors are due princi-
pally to the alkyl-substituted indans, tetralins and alkyl benzenes in
the aromatic portion of the unburned hydrocarbons. The smoky-burnt char-
acter is due primarily to the partial oxidation products of these same
aromatic species, plus a smaller contribution from paraffin oxidation
products—specifically; alkyl, hydroxy, and/or methoxy-substituted in-
danones, phenols, benzaldehydes, and alkenones.
Our efforts during this past year have been to translate the
large amount of detailed basic information obtained on diesel exhaust
odor into a simpler base from which the final objectives could be real-
ized.
B. Results
To obtain a more precise determination of exhaust odor in-
tensity, new odor measurement techniques were developed in the form of
a dose/response relationship. In this technique the sample odor inten-
sity is reported by an odor panel using a dynamic test chamber and the
presentation of a range of controlled concentrations. This method has
allowed us to obtain data satisfactory for use in developing the quanti-
tative odor/analytical relationships.
Knowledge of the exhaust odor chemistry has led to simpler
means for the measurement of those species. Correspondingly, the sample
size requirements for analysis have decreased. These results have en-
abled us to develop a simple, convenient means for collecting the re-
quired 500 - 1000 liter exhaust samples using Chromosorb 102 adsorbent
traps at sampling rates of about 10 Jl/min. This size sample is sufficient
-1-
Arthur D Little Inc
-------
for the detailed odor and chemical analyses. For the proposed new anal-
ysis method, only 5-50 liters are required and can be collected in
smaller traps at the rate of 1 &/min.
A routine preparative scale liquid chromatographic procedure
has been described which separates the diesel exhaust sample into its
three major chemical fractions: paraffin (LCP), aromatic (LCA), and
oxygenates (LCD). The LCA and LCO fractions represent the oily-kerosene
and smoky-burnt odor groups. The LCP fraction does not contain species
contributing to the exhaust odor.
Mass spectrometry methods have been completed for the aromatic
and oxygenate fraction which enable us to compare the relative amounts
of odorous components in the two isolated odor fractions. The methods
involve a matrix analysis using low resolution data for the aromatics
fraction and high resolution data for the oxygenate fraction.
A range of fuel and injector variables were examined using
the 4-71 engine in order to generate a set of data to test odor-ana-
lytical correlations and to determine the adequacy of the available chem-
ical data to describe the odor. The data available from the odor charac-
ter definition studies did not cover a wide enough dynamic range for cor-
relation studies.
The study included No. 1 and No. 2 diesel fuels and a para-
ffin fuel, and the S-60, N-60, and LSN-60 injectors. The exhaust odor
intensity was the same for all conditions except for the S-60 injectors
which was higher. The No. 2 fuel exhaust had new tarry character terms
to modify the normal description of exhaust. The paraffin fuel exhaust
was significantly different in character, best described as smoky candle-
sour oxidized, although of equal intensity.
These studies have shown that the intensity of the oily-kero-
sene and smoky-burnt exhaust odors can be represented by the total quan-
tity of the aromatic (LCA) and partial oxidation (LCO) exhaust fractions.
Further, we have found that the total exhaust odor intensity appears to
be accurately represented by the abundance of the partial oxidation pro-
ducts as determined by liquid chromatographic procedures.
We have begun to develop the basis for an instrumental approach
to the simple measurement of the exhaust odorants based on analytical liq-
uid chromatography procedures.
Detroit Diesel Allison Division, General Motors Corporation, Model
4154N
-2-
Arthur D Little; Inc
-------
C. Recommendations
1. Definition of the quantitative relationship between
odor and the odorous chemicals should be completed.
This would be done through further dose/response
studies of exhaust fractions and their analysis.
2. Support of the above study should be derived from
exhaust samples obtained from a wide range of engine
types and conditions and fuels.
3. Research should be completed on the proposed ana-
lytical liquid chromatographic method for the in-
dication of exhaust odor intensity through measure-
ment of the odorous fractions.
-3-
Arthur D Little Inc
-------
II. BACKGROUND
This section briefly reviews the approach which was developed
during our previous research on the CAPE-7 project and describes some of
the methods which carry over into the research described in this report.
The main emphasis in our program is to use the sensory re-
sponse of experienced odor chemists to guide and determine the effect
of the various analytical procedures developed for the identification
and measurement of the odorous species in diesel exhaust.
The test facility is described in detail in Appendix A.
Standard operating conditions have been 1800 rpm, 33% load, N-60 injec-
tors, and No. 1 diesel fuel. The facilities for determining the odor
of exhaust or its analytical fractions are described in Appendix B.
Using the odor profile method, the exhaust produced under
the standard operating conditions has been described at a 600/1 dilu-
tion as:
Total Intensity of Aroma (TIA) 2
Oily 2
Burnt 2
Kerosene 1^2
Eye irritation /
Nose irritation /
(See Appendix B for Explanation of Odor Method)
Our integrated experience in the perception of diesel exhaust
odor has led us to describe the odor quality and intensity in terms of
two groups—oily-kerosene and smoky-burnt—each of which contributes
about equally to the overall exhaust odor.
The oily-kerosene odors are due principally to the alkyl-sub-
stituted indans, tetralins and alkyl benzenes in the aromatic portion
of the unburned hydrocarbons. The smoky-burnt character is due primarily
to the partial oxidation products of these same aromatic species, plus a
smaller contribution from paraffin oxidation products—specifically:
alkyl, hydroxy, and/or methoxy-substituted indanones, phenols, benzalde-
hydes, and alkenones. These findings are summarized in greater detail
in Appendix C.
In the earlier stages of the program, it was necessary to
collect large volumes of exhaust in order to obtain sufficient quantity
-5-
ArthurD Little Inc
-------
of odorous material to carry through the detailed analytical procedures
required for identification of specific odorous compounds. However, now
that the odorous species are known, it has been possible to develop sim-
pler procedures for their measurement. In addition, more efficient means
have been found for collection of the odorous exhaust species. The new
analytical and odor procedures require 500 - 1000 liters of exhaust which
are collected in the manner shown below.
Exhaust
Heated
Particulate
Filter
^
Metal
Bellows
Pump
•—
Adsorbent
Trap
Volumetric
Meter
Silica gel was first used in the adsorbent trap, but the results in this
report describe the currently preferred adsorbent, Chromosorb 102.
The exhaust isolated from the adsorbent trap by solvent elu-
tion is separated into a major paraffinic, non-odorous fraction and aro-
matic and oxygenate fractions containing the oily-kerosene and smoky-
burnt exhaust odors respectively. This silica liquid chromatography pro-
cedure is shown schematically in Figure 1, and it is also described in
detail in Appendix E. The original procedure involved the collection of
eleven fractions, of which two, LCC-4 and LCC-10, contained the aromatic
and oxygenate odorous materials.
In developing procedures for the analysis of the odorous
exhaust species, we have followed a generalized scheme which involves
Collection and
Isolation of
Exhaust Sample
Liquid Chromatographic
Separation of
Odor Fractions
Analysis of
Odor Fractions
-6-
ArthurD Little Inc
-------
LCC-1
Paraffins
No Odor
Adsorbent
Trap
Solvent
Elution
Silica Gel
Liquid
Chromatography
LCC-4
Aromatics
Oily-Kerosene
LCC-10
Oxygenates
Smoky-Burnt
FIGURE 1 LIQUID CHROMATOGRAPHY DIESEL EXHAUST
FRACTIONATION PROCEDURE
-7-
ArthurD Little Inc
-------
III. CHEMICAL ANALYSIS OF ODOR FRACTIONS
Previously (3), we were able to show that a matrix analysis
of the low resolution mass spectrum of the LCC-4 fraction containing the
oily-kerosene odor would provide a measure of the amount of alkyl benzenes
and indans/tetralins in that sample, as well as the more abundant naphtha-
lenes. Further, an indication was obtained that the amount of indans/
tetralins injected into the odor test room, as computed from the analytical
data, did relate to the kerosene odor intensity. To a lesser extent, the
alkyl-benzene concentration also correlated with the oily odor intensity
of those fractions.
We also suggested that, although the oxygenates smoky-burnt
odor fraction was much more complicated, a comparable type of analysis
might be possible utilizing data obtained from the complete high resolu-
tion mass spectrum of the LCC-10 liquid chromatography fraction. It is
important to remember also that we do not know with the certainty that was
established for the oily-kerosene odor fraction just which groups of oxy-
genated species we prefer to measure to reflect the odor of that sample.
Therefore, an analytical method for the oxygenate fraction should summa-
rize chemical species concentration data where possible but leave a maxi-
mum possibility for examining the data in several combinations.
An analytical method has been developed for a detailed repre-
sentation of the smoky-burnt oxygenate fraction, A detailed explanation
of the analysis methods for both the aromatic and oxygenate fractions is
given in Appendix G. The type of information obtained from each analysis
is given in Table 1.
The aromatics analysis gives the abundance of the indans and
tetralins related to the kerosene odor and the abundance of the other
chemical groups in the fraction. The oxygenates analysis is a means of
representing, on an internally consistent basis, the different types of
chemical groups present in the fraction which were identified as odorous
species. The matrix does not represent the percent abundance of these
groups in the sample but does provide an efficient means of comparing
samples and searching for odor significant differences. The R+DB value
is a definition of chemical class based on hydrogen unsaturation, while
the columns Oi, 62, and 63 indicate the relative amount of each class
found with that number of oxygens. Thus, R+DB l/Oi could be aldehydes
and ketones, R+DB 2/Oi unsaturated aldehydes, R+DB 4/Oj phenols, R+DB
6/02 hydroxy indanones, etc. (See Appendix G for a more complete expla-
nation) .
ArthurD Little Inc
-------
Table 1
Typical Chemical Analysis Information for Odor Fractions
Composition of Aromatics Oily-Kerosene Fraction (LCC-4 or LCA)
Group Weight %
Alkylbenzenes 62
Indans/tetralins 23
Naphthalenes 17
Acenapthenes -
Phenanthrenes -
Representation of Oxygenates Smoky-Burnt Fraction (LCC-10 or LCO)
R+DB Oi 02 03
1
2
3
4
5
6
7
8
1
15
13
10
16
13
4
5
0
1
2
2
3
3
2
1
0
0
0
0
0
1
0
0
-10-
ArthurD Little; Inc
-------
IV. SAMPLE COLLECTION
A. Trapping
Previous studies (3) have shown that silica gel adsorbent
traps were a much more efficient means of collecting the odorous com-
ponents in diesel exhaust than the condensation methods. Further studies
were conducted to quantitatively define the performance of these traps.
We found that their efficiency was particularly susceptible to the water
content in the exhaust, and this factor limited the utility of silica
gel. Experiments were then conducted with two Johns-Manville Chromosorb
adsorbents.
Chromosorb 101 and 102 are crosslinked polystyrene porous
adsorbents frequently used in gas chromatography studies. They were
attractive for our purposes because, like charcoal, they would have low
water retention but also had a higher probability of quantitative sample
recovery. In previous studies with charcoal we had been unable to re-
cover the odorous fractions of exhaust. Chromosorb was also shown to be
effective in the IIT Research Institute (5) diesel odor program.
Comparison of the Chromosorb 101 and 102 which have, respec-
tively, surface areas of 15 - 30 m2/g and 300 - 400 m2/g showed that the
Chromosorb 102 had the greatest overall efficiency. Exhaust samples were
collected by the scheme diagrammed in "Background" using stainless steel
collection tubes and the procedures described in Appendix D. The over-
all collection efficiency of the Chromosorb 102 traps can be seen in
Figure 2 where total hydrocarbons were monitored at the entrance and
exit from the trap. No odor was observed to penetrate the trap. We
believe that the 30 - 40 ppm C level out of the trap represents the light
ends, CHtt, etc., of the exhaust hydrocarbons. We have designated samples
collected using the Chromosorb procedure by a number preceded by the code
CHROM.
B. Extraction
The sample is extracted from the Chromosorb bed by backflowing
solvent through the trap. Using this method, studies have been conducted
to determine the optimum solvent volume for extracting the exhaust compo-
nents from the Chromosorb. The data are summarized in Table 2 for 1000-
liter collections of exhaust under standard engine operating conditions.
By maintaining the solvent flow through the adsorbent bed at 0.5 m£ per
minute, the trapped sample is virtually totally eluted in the first 10 mfc
of pentane. Only negligible amounts of material are further extracted
with more pentane and with the more polar methylene chloride.
-11-
ArthurD Little Inc
-------
1000
900
800
700
0 600 -
a
a.
Sg 500 -
l-
9
"- 400 -
300 -
200 -
100 -
Exhaust
wyiwiNwnyirnwv
-
Stop
{ Trap Exit
300
Start
Std
0
Std
«^^^
Std
0
-
~
60
-Time (Minutes)
f
D
<-f'
«•»
ft>
FIGURE 2 CHROMOSORB 102 TOTAL HYDROCARBON TRAPPING EFFICIENCY
-------
TABLE 2
Extraction of Diesel Exhaust Samples from Chromosorb 102
(a)
mg/K£ of Exhaust Extracted from Various Samples
Solvent Sequence
Pentane, 10 ml
Pentane, 10 ml
Pentane, 25 ml
CH2C12, 50 ml
(b)
Chrom. 32
260.0
1.8
1.3
(c)
Chrom. 33
223.0
1.4
0.7
t
(c)
Chrom. 34
231.0
2.0
1.1
(c)
Chrom. 35
236.0
1.7(d)
_
0.7(e)
(a) Represents a collection of 1000 i from an engine operating
at 33% load with No. 1 Diesel Fuel and N-60 injectors.
(b) Represents a collection of 600 £.
(c) No detectable material observed.
(d) 50 mi of pentane extract.
(e) 10 mi of CH2C12 extract.
-13-
ArthurD Little Inc
-------
Each solvent extract has been evaluated for odor character-
istics on blotter strips. The results of these analyses consistently
show that all the characteristic odor of diesel exhaust is concentrated
in the first 10 mfc of pentane. The other fractions do not have any rec-
ognizable odor.
C. Analytical and Odor Results
A comparison of the composition of the diesel exhaust samples
collected by various sampling methods is summarized in Table 3. All
samples were collected under the standard engine operating conditions.
The total organic extracts (TOE) of the samples collected by aqueous
condensation, which we have reported before (3) to be equivalent to
only about 10% of the total exhaust odor, is also only 10% of the total
organic extract compared to the silica gel and the Chromosorb 102 samples.
The data also suggest an expected preferential trapping of the oxygenated
compounds in the aqueous condensate. The silica gel traps suffer by com-
parison in the apparent loss of aromatics. Both Chromosorb 101 and 102
appear to trap the aromatics and oxygenated compounds efficiently and
equally, but as reported earlier, the Chromosorb 101 showed early break-
through of hydrocarbons and, therefore, provided less margin of safety.
The samples collected on the Chromosorb 102 were subjected
to the standard liquid chromatography (LCC) fractionation and examined
in the odor test room (See Appendix B for procedure). The results of
this work are summarized in Table 4. The detailed odor profiles are
listed in Table 5, and the mass spectrometry analyses of selected aro-
matic and oxygenated fractions are given in Table 6. (See Appendix G
for details on analysis procedure).
The reproducibility of the LCC-4 and LCC-10 oily-kerosene
and smoky-burnt odor fractions is quite encouraging and appears to be
within the measurement error. The sensory evaluation of the two intense
odor fractions, TOE and LCC-10, is similar for all samples, but the LCC-
4 fraction shows some variation. We believe that in part the variability
observed is due to use of the static test room facility as previously re-
ported. Results reported in Section VI demonstrate the value of the
dynamic test chamber for such observations and a greater reproducibility
in results.
Table 6 summarizes the detailed analyses of the odor-containing
fractions. The aromatics fractions of the diesel exhaust samples show a
composition similar to that of the No. 1 fuel. (These results should be
taken on a comparative basis only, because of periodic changes in the spec-
trometer calibration.) The oxygenate analysis of the LCC-10 fractions for
Samples Chrom 32 and Chrom 33 shows a remarkably good agreement.
-14-
ArthurD Little Inc
-------
TABLE 3
Comparison of Diesel Exhaust Collected by Various
Sampling Techniques
(Concentrations in mg/K2.)
Sampling Method
Aqueous Condensate
Gel
Chromosorb 101
Chromosorb 102
TOE
23
240
138
258
LCC 4
4.0 (20%)
13.0 (6%)
36.4 (28%)
38.2 (15%)
LCC 10
2.7 (14%)
7.3 (3%)
6.5 (5%)
5.7 (2%)
(c)
a. Engine conditions - 33% load, No. 1 Diesel Fuel, N-60 injector.
b. All values derived from GC data described in Appendix F.
c. This value is slightly lower than the other LCC-10 fractions by
an amount equivalent to the oxygenated compounds that elute in
earlier fractions in the methylene chloride system (see Section v)
-15-
ArthurD Little Inc
-------
TABLE 4
Summary Chromosorb 102 Analytical and Odor Data
TOE LCC-4(a) LCC-10(a)
CHROM Mg/Kfc TIA(b) Mg/K& TIA(b) Mg/KJt TIA(b)
28 259 1*5-2 36 ? 5.0 1-1*5
29 316 2 39 h 7.6 1
30 260 2
31 270 Us - -
32 260 Us-2 38 ? 4.5 1%
33(c) 223 - 38 1 6.4
34 231 - 40 - 5.0
Avg 258 1*5-2 38 % 5.7 1%
(a) Using methylene chloride LCC elution modification (see
Section V).
(b) 21 I aliquot in odor test room, 600/1 dilution.
(c) Three fraction routine separation; LCP, LCA, LCO (see
Appendix A).
-16-
ArthurD Little, Inc
-------
TABLE 5 : STATIC TEST ROOM ODOR OF CHROMOSORB SAMPLES
Chrom 27
Chrom 28
Chrora 30
Chrom 31
Chrom 32
Chrom 33
TOE
600/18
LCC-4
600/1
LCC-4
300/1
LCC-10
600/1
TIA 1
Oily 4-1
Musty 4
TIA 14
Smoky 14
Ox. oily 1
Musty
Naph.
Nose Irr.
-------
TABLE 6
Chromosorb 102 Detailed Odor Fraction Mass Spectrometric Analyses
Aromatics Analysis
% Relative Composition
Alkyl Benzenes
Indans/Tetralins
Naphthalenes
Oxygenate R+DB Analysis
Chrom 32
LCC-10
R+DB
1
2
3
4
5
6
7
8
0
16
20
13
9
11
3
0
0
2
3
3
4
5
2
2
03
0
0
0
0
0
0
0
0
Chrom 34
LCC-4 No. 1 Fuel
Chrom 33
LCO
Ol 02
1 0
15 1
13 2
10 2
16 3
13 3
4 2
5 1
57 57
26 27
19 18
Relative Abundance
Chrom 34
LCC-10
03 QI Q£
0 00
0 13 2
0 18 3
0 13 3
0 94
1 14 5
0 42
0 02
°3
0
0
0
0
0
1
0
0
a. See Appendix G for details of analysis
-18-
ArthurD Little; Inc
-------
V. LIQUID CHROMATOGRAPHY
A. Methylene Chloride Modification
The standard liquid chromatography procedure used earlier
in our research (Appendix E-l) had several disadvantages. The solvent
choices for the gradient elution were more complex than necessary for
the separation, and the hexane used was frequently found to contain
higher aromatic impurities. Also, benzene absorbed strongly in the
ultraviolet spectrum and prevented us from developing simpler analysis
methods based on UV spectrophotometry with this solvent.
Accordingly, the normal procedure was modified to a smaller
scale to match the new analytical requirements, and a simpler solvent
gradient was developed using the series pentane, methylene chloride,
and 10% methanol in methylene chloride. The procedure is described'in
detail in Appendix E-2.
This solvent system gives sample resolution comparable to
that of the former method in that the paraffins wh^.ch constitute the
main bulk of the sample still elute in fraction LCC-1, the aromatics
in fraction LCC-4, and the oxygenates in fraction LCC-10.
The results of a typical separation using this scheme are
shown in Table 7. The data shown represents the fractionation of a
diesel exhaust sample (Chrom 32) from the collection of 1000 liters of
exhaust on Chromosorb 102. The diesel engine was operated under standard
conditions.
The values noted are closely analogous to our previously re-
ported data (3) using the benzene solvent gradient. Note that the LCC-3
fraction contains a slichtly higher mass than the preceding fraction indi-
cating partial elution of the aromatics. Indeed, UV analysis of this
fraction at 254 nm shows the presence of UV absorbers indicative of aro-
matic compounds. There is also some mass observed in fractions LCC-5 to
LCC-8 which was not observed in the benzene solvent gradient.
Blotter strip odor evaluation of each of these fractions re-
vealed no odor for fractions LCC-1 and LCC-2, a trace of oily-kerosene
in fraction LCC-3 with most of that characteristic odor note in fraction
LCC-4. Fraction LCC-10 contained the smoky-burnt odor notes. The inter-
vening fractions contained the oxidized-oily notes that at times have
also been observed in corresponding fractions in experimental gradient
systems.
In order to determine the composition of the species eluting
in fractions LCC-5-8, a high resolution mass spectrometric analysis was
-19-
Arthur D Little Inc
-------
TABLE 7
Methylene Chloride LCc'3' Elution Scheme for Diesel Exhaust Sample CHROM 32
LCC Fraction Solvent and Effluent Volume
1 Pentane , . 5 . 0 mfc
2 Pentane, 3.0
3 Pentane, 3.0
4 CH2C12, 5.0
5 CH2C12, 2.5
6m PI 9 ^
\jtLj\sJ-f\j *• * 3
7 CH2C12, 5.0
8 10%MeOH/CH2Cl2, 2.5
9 10%MeOH/CH2Cl2, 2.5
10 10%MeOH/CH9Cl9, 2.0
mg/0(c> of
Exhaust
201.0
1.3
12.1
37.7
0.3
0.2
0.5
0.2
4.5
Paraffins
Aromatics
Aldehydes
and
Ketones
Oxygenate
(a) On micro-column of activated silica Grade 950, 0.6 cm ^ 12 cm
The column volume is 2.1 m£.
(b) Engine operating conditions 33% load, No. 1 diesel fuel, N-60
injector.
(c) Based on GC analysis of each fraction on 1 1/2' x 1/8" 10% OV-1
column.
-20-
ArthurDLittklnc
-------
made on fraction LCC-5 from sample CHROM-27, which contained sufficient
material for analysis. The species observed and their relative abundance
are listed in Table 8. After examination of the entire spectrum, it ap-
pears that most of these species are aldehydes and ketones as follows:
R+DB 5 Benzaldehydes
R+DB 6 Indanone
R+DB 8 Naphthaldehydes
The R+DB matrix presentation of this fraction is shown in
Table 9. The indication from this sample is that oxidized-oily odor
notes are represented at least in part by R+DB 5/0^ and perhaps to some
extent by the DI R+DB 6 and 8 groups.
B. Routine Preparative Liquid Chromatography Procedure
The liquid chromatographic separation of the primary diesel
exhaust fractions has been simplified so that a routine procedure could
be described for isolation of the two odor fractions. Since the exhaust
species consist only of the paraffin, aromatic, and oxygenate components,
a simple procedure would give rise to only these three fractions—rather
than the 10-12 fractions normally collected in the LCC procedures used
to date.
Such a procedure has been developed and is described in detail
in Appendix E-3. Typical results using this method are shown in Table 10.
The silica column is as used in the normal LCC procedure. The gradient
consists of three solvents added in the order and volumes shown. Three
fractions containing the paraffins (LCP), aromatics (LCA), and oxygenates
(LCO) are collected.
Odor evaluation of these fractions reveals no odor in LCP, the
oily-kerosene odor in LCA, and the smoky-burnt odor in LCO. Mass spectral
analysis of the LCA and LCO fractions gives results comparable to those
of LCC-4 and LCC-10 fractions as described in the preceding section. Av-
erage mass results for LCC-A and LCC-10 were 38 mg/kS, and 5.7 mg/k£ respec-
tively, while the LCA and LCO fractions analyzed for 38 mg/k2, and 6.4
mg/kJl.
-21-
Arthur D Little Inc
-------
TABLE 8
Major Species Observed in CHROM-27 LCC-5
% Abundance0
MW
H 0
R+DB 5
R+DB 6
R+DB 8
14
38
5
2
2
8
3
4
106.0403
134.0718
148.0877
162.1036
118.0417
132.0564
156.0564
170.0726
7
9
10
11
8
9
11
12
6
10
12
14
6
8
8
10
1
1
1
1
1
1
1
1
1 (132 fragment)
a. Ions greater than 1% abundance
-22-
ArthurD Little, Inc
-------
TABLE 9
R+DB Oxygenate Analysis of CHROM-27 LCC-5
R+DB
1
2
3
4
5
6
7
8
°1
0
3
1
1
61
15
2
8
°2
0
0
0
0
3
2
0
0
°3
0
0
0
0
0
0
2
0
-23-
ArthurD Little: Inc
-------
Table 10
Preparative Liquid Chroraatography of Diesel Exhaust
Sample Chrom 33( ^
mg/kfc of
FractionSolvent Exhaust Present
LCp Pentane, 20 mi 165.0
LCA CH2C12, 10 mA 38.5
LCO 50% MeOH/CH2Cl2, 10 mA 6.4
(a) Column of activated silica Grade 950, 1.1 cm x 12 cm.
Column volume = 4.5 ml. (See Appendix E-3)
(b) Engine operating conditions = 33% load, No. 1 diesel fuel,
N-60 injector.
(c) P - Paraffins, A - Arotnatics, 0 - Oxygenates.
-24-
ArthurD Little, Inc
-------
VI. DOSE/RESPONSE MEASUREMENT OF ODOR INTENSITY
As we began to explore quantitative relationships between
the analytical measures of exhaust fractions and odor intensity, a care-
ful review was made of all of the available odor data. The summation
revealed that virtually all of the odor data fell in the range of TIA =
1.25 + 0.25. Clearly, this range is too narrow to test the validity of
correlations which we had developed. Most of the data had been obtained
when the emphasis was still primarily on qualitative odor descriptions
of exhaust and its analytical fractions. It was felt that the precision
of the odor data may have been unnecessarily restricted due to observing
a single concentration in the static test room. Preliminary studies in
the static chamber showed that the precision of the odor measurement
could be improved by observing the odor intensity at a series of con-
trolled dilutions covering a wide dynamic range.
As a result of the success of these and subsequent experiments,
this method of presenting a series of doses and measuring the odor response
has been used routinely to describe the odor intensity of exhaust fractions.
These measurements are carried out in a dynamic odor chamber which has been
described in detail in Appendix B. This facility was constructed under
programs carried out jointly for the Manufacturing Chemists Association
and the Environmental Protection Agency. The primary purpose of the cham-
ber is to instrumentally measure the involuntary physiological response
to odorants. For this study, we have used only the sample presentation
portions of the facility, relying on the normal panel member response to
record odor intensity and character.
Typical results using this approach for the LCA and LCO oily-
kerosene and smoky-burnt odor fractions are listed in Table 11 and are
shown graphically in Figures 3 and 4. Using this approach, the odor in-
tensity of these fractions is described as
TIA (LCA) = 1.45 + 1.35 log (CLCA)
TIA (LCO) = 2.07 + 1.80 log (CLCO)
where C is in 2,/m3
-25-
ArthurD Little: Inc
-------
Table 11
Dynamic Odor Test Chamber Results - CHROM 33
Odor Fractions
TIA (Ave.)
Concentration (&/m3)a
0.8
0.05
0.1
0.2
0.4
0.8
0.4
0.8
1.6
3.2
6.4
LCAb
1.25
0
0
0.75
1.0
1.5
0.5
1.25
1.5
2.25
2.5
LCOC
1.5
0.25
0
1.0
1.5
1.5
1.25
1.75
2.5
3.0
3.0
a. Concentration of .exhaust fraction
in the odorized air.
b. Average of five panel members, + 0.25.
c. Average of four panel members, +_ 0.25.
-26-
ArthurD Little Inc
-------
to
•^i
i
D
TIA
3
2
1
„
--
—
0.01
. _r|_
,
i
I
•
.
i
1
'
— ( —
.
r ; i
|
i
-..
-
_
i
!
i
. i [...
I
- |
I
• |
I -
I
j
i
j
[
i
|
i
I"
_„ i —
j
,
i
..i
1 _
"1 !
— i -"" -r
i
-
1
_ i
... .
~
.
! -
_i_ ._: j._ -^
\ J^A ^
•£' j-
0.1
.__
>
—
; • i
. \
'•
1 1 ;
1
\ ' i
i ' '
I i— r
I ' j
i : i
i ^
. j^\ \
t_
~
1 i
! |
*\
w3
• • • i
--H !rj.
j
J
\
f-
_
^
'
^
— f- -
:
^
S1
|
-
i
i
i
i
_i j
!
ft.
S
s*
j^ i-k .
^ »
jgy
i
•^
effih
1
i
!
1.0
\ : ! ;
j ^\
1-T f._.pl_^] . l -,
+ ^1
tT^
«**\
M
1-
1
•
r
"'
^<
c
„
i
i
~~1
!
-
i
1
...
-I!
.*
10
Concentration of LCA Exhaust Fraction
Figure 3:
DOSE/RESPONSE DATA
FOR CHROM 33 LCA
ODOR FRACTION
-------
TIA
0.01
C/M3
Concentration of LCD Exhaust Fraction
Figure 4: DOSE/RESPONSE DATA FOR CHROM 33 LCD ODOR FRACTION
-------
The constants and slopes of the least squares lines are thus
sufficient to describe the intensity of any particular sample. Statis-
tical analysis of a large set of data obtained in this manner revealed
an odor predictability of + 0.4 TIA within 95% confidence limits (a =
0.2 TIA). These procedures have enabled us to describe odor intensity
more precisely than the older methods.
For the study of exhaust samples, the dose/response data were
obtained using the static test room, which is adjacent to the engine fa-
cility.
-29-
ArthurD Little Inc
-------
VII. DEVELOPMENT OF A SIMPLIFIED ANALYTICAL METHOD
A. Method Development
The basis for a simple, routine method for the analysis of
the odorous fractions present in diesel exhaust has been developed.
The method is based on modern analytical liquid chromatographic (ALC)
methods using ultraviolet detection. In preliminary studies, we found
that the mass of the LCA and LCO fractions could be measured by ultra-
violet spectroscopy using the correlation between their ultraviolet
absorbance at 254 nm and their mass as determined by the GC method.
The method is based on the use of newly available commercial
components assembled as shown schematically in Figure 5. The following
components are used:
1). Two solvent reservoirs
2). Positive displacement pump (Milton Roy-Milroyal
Model HDB-1-30R)
3). A silicon septum sealed injector
4). A 30 cm x 0.6 cm O.D. glass column packed with
a silica gel type support - Corasil II (Waters
Associates, 37-50 microns)
5). Ultraviolet absorbence detector and amplifier,
detecting wavelength 254 nm (Laboratory Data
Control Model 1205)
6). Recorder (a Hewlett-Packard X-Y recorder Model
7030A was available and was used with the time
base).
The unit operates essentially the same as a gas chromatograph,
except in the liquid phase. A solvent flow is established through the
column (we have used 1 m£/min in most of our work). A sample is injected
with a syringe, and the separated components, which absorb ultraviolet
light at 254 nm, are recorded as they elute from the column. To achieve
the proper elution and detection, several factors were considered. The
first was that a solvent gradient was needed to resolve the odor groups,
-31-
ArthurD Little Inc
-------
40%
Hexane
in
Methylene
Chloride
10%
Methanol
in
Methylene
Chloride
Syringe
Septum
.Seal
Corasil II
Packing
Methylene
Chloride
Reference
Positive
Displacement
Pump
Photo-
Multiplier
Detector
254 mm
Drain
JUi
o
o
Amplifier
Recorder
Figure 5
ANALYTICAL LIQUID CHROMATOGRAPH SCHEMATIC
-32-
ArthurD Little Inc
-------
and the second was that the paraffin fraction did not absorb UV at any
wavelength above 240 run. Thus, an elution scheme which would allow the
paraffins and aromatics to elute as a group would only record the aro-
matic portion. Therefore, a two-step sequence (vs. three) could be
developed which required only a separation of aromatics (plus paraffins
and oxygenates).
The procedure finally developed is as follows:
1). Condition column with 40% hexane in methylene
chloride at a flow rate of 1.0 mH/min for at
least 15 minutes.
2). Inject 1 - 10 yfi. of sample (containing 0.1 -
10 i of exhaust fraction.)
3). After 1 minute, switch solvent selection valve
to allow 1 mSt, of 10% methanol in methylene
chloride to enter solvent stream and then switch
back to the 40% hexane solution. '
The results obtained on the TOE of sample CHROM 33 are shown
in Figure 6. Peak 1 corresponds to the aromatics portion (LCC-4 or LCA)
of the sample while peaks 2 and 3 correspond to LCO. In this system,
using methylene chloride as an eluent, the oxygenate fraction is par-
tially resolved into the two components as discussed earlier in Section
V. Peak 2 contains the aromatic aldehydes and ketones, while peak 3
contains the more polar species. The elution volume of peak 3 repre-
sents the time required for the injected slug of methanol to equilibrate
with the column and elute while displacing the polar oxygenated compo-
nents. The paraffins are, as discussed, transparent to the 254 run UV
and not observed while eluting coincidentally with the aromatics in
peak 1.
Confirmation of the peak assignments can be seen in Figures
7-9 which show the analysis of the LCP, LCA, and LCO fractions isolated
using the preparative liquid chromatography procedure. The paraffin,
fraction LCP (Figure 7), shows no UV absorbance; fraction LCA (figure 8)
contains the aromatics; and fraction LCO (Figure 9) contains the total
oxygenates.
At times a trace component elutes in the region of peak No.
3 due to septum and solvent background. This is considerably minimized
by using only distilled solvents and conditioning septa. In the instances
when this background peak is observed, the quantity of peak No. 3 is ad-
justed accordingly.
-33-
Arthur D Little Inc
-------
I
to
D
q
d
6 8
Retention Volume
FIGURE 6 ALCOFCHROM33-TOE,0.4e
10
12
14
-------
rt>
s
CM
q
d
J- -I -'-H-!-1
0
Retention Volume
FIGURE 7 ALC OF CHROM 33 - LCP, 0.75C
10
12
-------
I
i
D
I
3
I
Us
§ i
468
Retention Vplume
FIGURE 8 ALC OF CHROM 33 - LCA, 0.368
10
-------
D
cr
<-*
»•»
—
R
6 8
Retention Volume
10
12
14
FIGURE 9 ALC OF CHROM 33 LCD, 0.758
-------
The ALC method represents a unique way of quantitatively
measuring the total aromatics (oily-kerosene) and oxygenates (smoky-
burnt) from a single injection of a small amount of total exhaust.
B. Quantitative Response
The Laboratory Data Control ultraviolet (UV) detector is
linear in absorbance from 0.002 to 0.64 optical density (O.D.) units.
Using several reference samples, the linearity of the O.D. response
as a function of sample size was determined over a period of time and
using fractions derived from exhaust and fuel.
Figures 10-12 show the results obtained from these studies.
Each of the samples shows a linear O.D. response vs. yg up to an O.D.
of about 0.5. One can expect to readily relate the O.D. response from
an exhaust sample to the odor fraction quantity, given the appropriate
calibration curve. The slope of the curves for the LCC-4 fractions
from No. 1 fuel and its exhaust are similar, as expected, and the LCC-
10 exhaust fraction shows the expected higher response.
A comparison of the results obtained from the analysis of
several samples by the GC method and the new ALC method is given in
Table 12. It is clear that the UV method provides data matching the
GC data provided the instrument has been suitably calibrated. The ALC
method is preferred for reasons of convenience, but it also has the
potential for much more precise measurement of the exhaust fractions
than the 20% error found with the GC method.
C. Sensitivity to Fuel Composition
The UV response per unit of mass to exhaust fractions derived
from different fuels was expected to vary with the composition of the
fuel, since the absorbtivities of the various aromatic species cover a
wide range, we have analyzed a number of fuels to determine their ALC
response. These fuels included those used in the variables study (Sec-
tion VIII) and others representing the range of fuels suitable for use
in diesels which are commonly available. The results, along with certain
characteristics of the fuels, are given in Table 13. The ALC response
varies in a regular manner with fuel composition. The response is di-
vided into two main categories due to the No. 1 and No. 2 fuel types.
Fortunately there is a wide separation between these two main differences.
The variation in response with aromatic composition is shown in Figure
13 (symbols code given in Table 13). The only grossly irregular point
is the one for Soltrol 200 which is a special narrow cut—aromatics free
fuel. By suitable calibration procedures, we believe it will be a simple
matter to select the correct ALC sensitivity curve from an analysis of
the fuel used in the exhaust study.
-38-
ArthurD Little; Inc
-------
D. Response to Exhaust Odor Fractions
In examining the sensitivity of the ALC response to compo-
sition, the odor fractions isolated from the samples obtained in the
variables study were also studied. These data are summarized in Table
14. The response for each of the LCA or LCO fractions derived from No.
1 fuel are the same, while the differences due to fuel composition are
apparent in each. These sensitivity factors will have to be defined in
detail to obtain appropriate calibration curves for each fraction.
—39—
Arthur D Little Inc
-------
>
3
D
tr
0.60 -
0.50
0.40
Q
0
10
15
20 25
M9 of LCC 4
35
40
FIGURE 10 CALIBRATION CURVE FOR LCC 4 OF NO. 1 DIESEL FUEL
-------
0.6
0.5
0.4
0
0
0.3
0,2
0.1
D
IT
r?
*
10 15 20 25 30
/ig of LCC 4
FIGURE LI CALIBRATION CURVE FOR DIESEL EXHAUST LCC 4
35
40
-------
0.1
10
15 20 25
;jgof LCC 10
FIGURE 12 CALIBRATION CURVES FOR DIESEL EXHAUST LCC
35
40
10
-------
Table 12
Comparison of ALC and GC Analysis Methods
Sample
CHROM
32
33
34
Analysis
Method
GC
ALC
GC
ALC
GC
ALC
Peak No. 1
LCA
38
31
38
40
40
40
Peak No. 2
1.2
0.6
1.2
1.9
1.2
Peak No. 3
LCO
4.5
4.4
6.4
6.3
5.0
6.3
-43-
ArthurD Little Inc
-------
Table 13
X
X
X
O
*
X
0
0
0
0
0
0
D
>
-i
D
CT
Fuel Type
Kerosene (61611)
Kerosene (60064)
Kerosene (61338)
Soltrol 200*
No. 1 Diesel*
Kerosene (61835)
No. 2 Diesel (East Coast)
No. 2 Diesel (61836)
No. 2 Diesel (Midwest)*
No. 2 Diesel (61339)
No. 2 Diesel (61610)
No. 2 Diesel (60063)
Heating Oil (East Coast)
Heating Oil (Midwest)
[1] - The
Specific
Gravity
0.812
0.808
0.820
0.780[2
0.832
0.811
0.826
0.849
0.852
0.863
0.852
0.837
0.854
mass was
[2] - These values
%Aromat:
16.5
16.0
16.5
11
20.8
16.5
24.1
32.0
34.7
37.5
30.5
28.0
35.4
estimated
are based
:ics and ALC Response
ALC Response
Based on Aromatics
Composition
"V.
C\ f\{\t
-------
40
30
o
20
10
O
4-
0.01
0.02
O.D.254per/ig
0.05
FIGURE 13 VARIATION OF ALC RESPONSE WITH AROMATIC CONTENT OF FUEL
-45-
ArthurD Little Inc
-------
Table 14
ALC Response Sensitivity to Exhaust Variables Studied
Fuel
No.
No.
No.
1
1
1
Injector
S-60
N-60
LSN-60
— ALL; Response
Arotnatics (LCA)
0
0
0
.013^
•020( 0.017
.019 I
wu/yg;
Oxygenates
0
0
0
->
.041 I
.046
.043 ,
(LCO)
0.043
No. 1
Midwest No. 2
Soltrol
Fuel
N-60 exhaust
Fuel
N-60 exhaust
Fuel
N-60 exhaust
Avg excluding Midwest
0.051
0.017
0.051
0.050
0.012
0.012
0.015
0.038
0.043
0.069
0.066
0.025
0.043
-46-
Arthur D Little, Inc
-------
VIII. STUDY OF ENGINE AND FUEL VARIABLES
Several test parameters including load, fuel, and injectors
have been selected for study in order to examine the adequacy of our
identification data. These variables were also chosen to provide the
range of data for establishing the quantitative relationships between
exhaust odor and chemical composition. Results on the effect of engine
load have been previously reported (3). The exhaust odor intensities at
10%, 33%, and 90% load were all quite similar.
In these recent tests the old style S-60 injectors and the
new low sac LSN-60 style were compared with the standard N-60 injectors
which have been used in most of our studies. A high aromatics content
Midwest No, 2 diesel fuel and a special aromatics-free paraffin fuel,
Soltrol 200, were also studied to examine these variables. The pertinent
characteristics of these fuels were summarized in Table 13.
Two to four sets of experiments were run for each of the
variables studied. The average quantitative results of these studies
are summarized in Tables 15 and 16 for the analytical and odor data.
The 4-71 engine was run at 33% load and 1800 rpm for each of these studies,
The dose/response odor data are also shown graphically in Figures 14-16
for the exhaust and LCA and LCO odor fractions. The detailed composition
analysis of the odor fractions from each of these conditions is given in
Table 17.
The S-60 injectors condition is the only one which gives
clearly different exhaust results. The odor intensity of all of the
other conditions is quite similar (Figure 14). Use of No. 2 fuel led
to an exhaust odor character change with a substitution of terms such
as the use of a tarry descriptor for the kerosene and smoky character
notes. This may be consistent with the higher molecular weight aromatics
such as acenaphthenes (Class 5 in the aromatics mass spec analysis) pre-
sent in the fuel. Exhaust from the Soltrol 200 fuel has a completely
different odor character than normal and is best described as smoky-candle
and sour-oxidized.
Perhaps the most surprising part of these data, both from
chemical analysis and sensory observation, is the closer similarity of
results with the S-60 injectors to the data for the No. 2 fuel—instead
of being similar to the No. 1 reference fuel. The exhaust odor character
with the S-60 injectors was somewhat similar to that observed with Midwest
No. 2 fuel in the heaviness of the odor notes as compared with the refer-
ence conditions.
-47-
ArthurD Little Inc
-------
The total hydrocarbon levels for the three injectors is con-
sistent with the sac volumes (S-60 = 175 mm3, N-60 =3.5 mm3, LSN-60 =
0.5 mm3) as reported by Ford, Merrion, and Hames (6). The hydrocarbon
emission level also drops with the No. 2 fuel again as noted by Ford et
al.
The odor intensity of the LCO fractions is generally greater
than that of the LCA fraction (Table 16) and is about the same as the
exhaust odor intensity.
The relationships between the data gathered in these experi-
ments will be explored in the next section.
-48-
ArthurD Little; Inc
-------
Table 15
Diesel Exhaust Variables Analytical Data Summary
LCA
Variable
Fuel Injector
No. 1
No. 1
No. 1
No. 2 - Midwest
Soltrol 200
S-60
N-60
LSN-60
N-60
N-60
THC
(ppm)
3100
620
300
550
590
TOE Aromatics
mg/KJl ms/.KJ. % of TOE
1220 210
230 39
86 16
145 38
230 8
18%
17%
18%
26%
4%
14
LCO
Oxygenates
/KE. % of TOE
1%
3%
5%
4%
3%
-49-
ArthurD Little Inc
-------
Table 16
Diesel Exhaust Variables Odor Summary
Variable
Fuel
No. 1
No. 1
No. 1
No. 2
Soltrol
Injector
S-60
N-60
LSN-60
N-60
N-60
Exhaust
m
2.15 1.50
1.45 1.72
1.50 1.50
1.60 2.00
1.50 1.36
Odor'
LCA
Oily-Kerosene
b m
1.80 1.60
1.50 1.66
1.15 1.30
1.70 1.65
0.65 0.87
LCO
Smoky-Burnt
b
m
1.95 1.65
1.60 1.57
1.50 1.52
1.65 1.65
1.00 1.84
a. TIA = b + m log C; C in Jl/m3
at 1 £/m3 (1000/1 dilution) TIA = b
-50-
ArthurD Little, Inc
-------
Ln
-C
X
LJU
No. 1 Fuel; N-60
Soltrol 200; N-60
Midwest 2; N-60
No. 1 Fuel; LSN-60
No. 1 Fuel;S-60
0.01
0.1
1.0
CT
FIGURE 14 COMPARATIVE EFFECT OF FUEL TYPE AND INJECTOR DESIGN
ON DIESEL EXHAUST ODOR (STATIC TEST ROOM)
-------
I
Ui
<
o
Soltrol 200
Midwest 2
No. 1 Fuel, N-60
LSN-60
S-60
(.01
0.1
1.0
FIGURE 15 TIA OF LCA FRACTIONS: EFFECT OF FUEL TYPE AND INJECTOR DESIGN
-------
I
Ul
U)
--J-- ~ Soltrol 200
Midwest 2
No. 1 Fuel. N-60
LSN-60
S-60
0.01
0.1
1.0
FIGURE 16 TIA OF LCD FRACTIONS: EFFECT OF FUEL TYPE AND INJECTOR DESIGN
C
-------
Table 17
Ul
r
LCA. Aromatic
AlkyIb enz enes
Indans/Tetralins
Naphthalenes
Class 4
Class 5
LCO, Oxygenate
No.
R+DB OT
Fuel
54
28
20
c
1
2
3
4
5
6
7
8
11
9
7
10
21
8
8
S-60
0
2
3
3
4
4
2
1
0
0
0
0
0
1
1
1
ipectrometric Composition Analysis of Odor
• No. 1
S-60
61
24
16
No.
«L
1
15
13
10
16
13
4
5
•pi,pi
N-60
62
23
17
1, N-60
£2 £3_
0 0
1 0
2 0
2 0
3 0
3 1
2 0
1 0
— __ M-MT
LSN-60 Fuel
50 17
28 22
21 52
8
2
No. 1, LSN-60
Oi 02 03
200
12 6 0
15 4 0
12 3 0
840
13 4 0
621
710
Fractions (%)
jest
N-60
30
23
43
3
1
Midwest,
Oi 02
1 0
12 1
8 2
8 2
14 6
10 6
9 3
10 2
Solt
Fuel
19
43
14
24
N-60
£3
0
0
0
0
0
0
0
0
rol
N-60
23
28
12
37
Soltrol.
£L Jk
3 1
18 4
17 8
11 5
5 A
8 3
3 1
2 1
N-60
£3
0
0
0
1
1
0
0
0
-------
IX. CORRELATION OF ANALYTICAL AND ODOR DATA
A. Odor Interaction of Kerosene and Smoky-Burnt Fractions
In attempting to correlate observed engine emissions with
odor intensity, it was necessary to determine the odor interaction be-
tween the two isolated odor fractions, LCA and LCO. If either one alone
best represented the exhaust odor intensity, then it would be sufficient
to measure all or some part of that group to determine the exhaust odor.
If, on the other hand, there were a synergistic effect between the groups,
then it would be necessary to measure both groups in some combined manner.
This matter was studied by measuring the odor intensity of the
total organic extract (TOE) and the LCA and ECO fractions and then com-
bining % of each LCA and LCO fraction with the whole of the other. The
odor results obtained from the study in the dynamic test chamber are
given in Table 18 and are depicted graphically in Figure 17.
It appears from these data that there are no unusual effects
of one odor fraction on the other. Also, the exhaust odor intensity
(TOE) is accurately represented by the odor of the LCO fraction.
This matter was explored further by comparing the odor in-
tensities of the exhaust and the LCO odor fractions. A comparison of
the odor intensities calculated at 1 fc/m3 from the dose/response data
for these two samples (exhaust and LCO) in the variables study is shown
in Figure 18. These data clearly show the accuracy with which the ex-
haust odor may be represented by the LCO smoky-burnt odor fraction for
normal exhaust conditions. The only poor correlation is for the exhaust
from the unusual Soltrol fuel.
B. Correlation of Odor Intensity and Sample Mass
The ultimate test of the approach we have developed comes in
examining how well the quantity of the two primary odor fractions repre-
sents the odor of those fractions and the exhaust. The data from the
variables study is summarized in Table 19. We have represented the odor
intensity of the exhaust and its fractions by the value computed for the
1000/1 dilution (1 £/m3). This concentration is in the centroid of our
data and represents a realistic roadway dilution.
The correlation of LCA odor intensity and abundance is shown
in Figure 19. The data correlate over this limited range within the
Arthur D Little Inc
-------
precision of the odor data with the exception of the Soltrol Fuel. A
similar plot for the LCO fraction is shown in Figure 20 with comparable
results. The final test of these approaches comes in the correlation
of the exhaust odor intensity with abundance of the LCO fraction as
shown in Figure 21. The correlation is promising, but premature, con-
sidering the small amount of data available for evaluation.
These approaches appear to offer a means of predicting ex-
haust odor intensity based on the simple measurement of the primary odor
fraction abundance using the ALC procedure. Further data will be required
representing a wider range of emissions to fully test this approach.
It should be remembered that the success of this simplified
approach has been based on a detailed understanding of the odorous chem-
ical species present in each fraction and their relationship to total ex-
haust odor.
-56-
ArthurD Little Inc
-------
Table 18
Odor of Mixtures of LCA and LCD (TIA)
(a)
Dose («./M3)
l.O1
TOE LCA LCA + LCO/4 LCA + LCO LCA/4 + LCD LCD
0.2
0.4
0.8
0.8
1.6
3.2
W If \<
**Z «fr ^
2 h 4
21 1
* i M,
2% 14 14
23/42 2
3/4 4
1 1
c
1 3/4 1 3/4
14 2
2 2
2% *
*
14
2
1 3/4
2k
24
1.9 1.2
1.3
1.7
1.8
1.8
(a) Sample Chrom 35
(b) Calculated from least squares data, TIA = b + m log C
-57-
ArthurD Little Inc
-------
0.1
0.5 1.0
Dose(C/M3)
5.0
FIGURE 17 STUDY OF LCA-LCO INTERACTION EFFECTS
-58-
ArthurDLittklnc
-------
I
Ul
X
1/1
Correlation
•H-
. _J ,4
j
'• T
I •:
r . ..
i
i
' . ;
i
" ~ '\--j2
t'l>M
l?T ; .^.;,
•
f
TTTT
C-." '
-.
..:.
--
#
—
'
/
F
--•
» —
R ':
i
*~
- ;^
1Z
2T
i
j_
f
Ll£
_..!_
i
: _-j
Pm*9ji\
'"i '
i
•.!- |
..... .|._r
1 ! '
! i i
i i ' |
-: i :i ! ;
: rr • ! ;
-i •t-ji-^f---
i ! ;
! i , • I
l ' ! i
i i i i
L '
-
L
_l
i
1
L , ,
.
'
i
;
•|-
|||
1
TIA-LCO
FIGURE 18 COMPARISON OF EXHAUST AND LCO ODOR
INTENSITIES
-------
Table 19
Summary__of_ Correlatable Odor and Analytical Data
TIA (1 £/M3ja Concentration
Exhaust LCA LCO LCA LCD
S-60, No. 1 2.15 1.8 1.95 210 14
N-60, No. 1 1.45 1.5 1.6 39 7
LSN-60, No. 1 1.5 1.15 1.5 16 5
Midwest No. 2 1.6 1.7 1.65 38 6
Soltrol 1.5 0.65 1.0 8 6
a. Computed from the dose/response observations.
b. yg/£ E mg/k£
-60-
ArthurD Little Inc
-------
10
50
100
LCA (ng/2)
FIGURE 19 CORRELATION OF LCA ODOR INTENSITY AND MASS
D
-------
o
o
10
LCO (jug/C)
100
FIGURE 20 CORRELATION OF LCO ODOR INTENSITY AND MASS
-62-
ArthurD Little: Inc
-------
ra
J=
X
rffr
MJ-ri
IEIL
4+H-
11
-i-
rr
rh
flf
ill
Ht
10
LCO (jug/C)
50
100
FIGURE 21 CORRELATION OF EXHAUST ODOR INTENSITY
WITH LCO
-63
Arthur D Little Inc
-------
X. 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. Chemical Identification of the Odor Components in Diesel
Engine Exhaust, final report June 1971, CRC Project CAPE
7-68, EPA Contract No. EHSD 71-18.
4. F. W. McLafferty, "Interpretation of Mass Spectra," Benjamin
New York, 1966.
5. Chemical Species in Engine Exhaust and their Contributions
to Exhaust Odors, Report No. IITRI C6183-5, final report
November, 1970.
6. H. S. Ford, D. F. Merrion, and R. J. Hames, "Reducing
Hydrocarbons and Odor in Diesel Exhaust by Fuel Injector
Design", SAE Paper 700734 presented in Milwaukee, Wisconsin,
September, 1970.
-65-
ArthurD Little, Inc
-------
APPENDIX A
EXPERIMENTAL TEST FACILITY
The test facility used for these studies is shown schemat-
ically in Figure A-l and consists of engine, odor, and sample areas.
The engine used is a 71 series 4-cylinder diesel . Operating at a con-
stant speed of 1800 rpm, most of the work has been done at 33% load using
No. 1 diesel fuel. The three component resistance load (10 KW and two
30 KW) allows us to operate at 10-90% load. ,
The static odor chamber is described in detail in Appendix
B. Sample collection for analysis had initially been done using various
condenser systems, and these are still required when sufficient sample
(10,000 - 100,000 liters) is needed to run a complete analytical iden-
tification series. However, now that the composition is known, we find
that all of the required analyses can be achieved on the material iso-
lated from 500 - 1000 liters of exhaust. The sorbent tube collection
method described in Appendix D is preferred for these purposes.
Detroit Diesel Allison Division of General Motors Corporation
Model 4154N
A-l
Arthur DLittklnc
-------
WATER
ACOUSTIC WALL
FLEXIBLE
SEGMENT
| I
ENGINE EXHAUST
ANTECHAMBERJ
DOORS'
ODOR TEST ROOM
CIRCULATING
^ FANS
INJECTION PQRTS
POR1
—\
PRE AND
MUFFLER
HEATED/'
FILTER
POST MUFFLER SAMPLING POINTS
METAL
BELLOWS
PUMP
-
COLLECTION
TUBE
EXHAUST
TO
ROOF
>
I
FIGURE A1 EXPERIMENTAL ARRANGEMENT OF DIESEL ENGINE, ODOR TEST ROOM,
EXHAUST SAMPLING AND COLLECTION SYSTEM
-------
APPENDIX B
ODOR PROFILE MEASUREMENTS*
1. DIESEL EXHAUST ODOR
The standard diesel exhaust sample has normally been obtained
from a 4-71 diesel engine operating at 33% load and 1800 rpm on No. 1
diesel fuel. The profile analysis of the standard sample diesel exhaust
was consistent from day to day, but the odor of diesel exhaust did show
some differences with variations in engine operation. 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 char-
acterization is consistent with the odor observed when traveling behind
a bus, which confirms our belief that the mode of engine operation pro-
vided a representative sample for analytical studies.
The description of the diesel exhaust odor in the static test
room with a dilution ratio of 600:1 can be described by three character
notes; oily, represented by technical grade hexadecane among other stan-
dards; burnt, which, although similar to a low dilution of propionaldehyde,
phenol, and cresol, is produced in fuels with partial oxidation at ele-
vated temperatures, and kerosene, which is the top odorous component of
the fuel and may be described as having sweet, sharp, sour, tarry, and
solvent components. In addition to these odor characteristics which ap-
peared 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 charac-
terization of a 150-y£ aliquot of fuel, which (by computation) is equiv-
alent to the amount of fuel burnt to produce the 21-liter sample of ex-
haust, produces an odor in the test room at least as strong as the exhaust
odor. The dominant odor characteristic is kerosene, with the oily note
being less intense and the burnt aroma barely detectable. With the diesel
exhaust, the oily and burnt aromas are primary character notes and kerosene
a supplementary factor.
* Taken in part from Final report of third year's work, June 1971, "Chem-
ical Identification of the Odor Components in Diesel Engine Exhaust."
B-l
Arthur D Little Inc
-------
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 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 appear-
ance of odor character notes indicates the other odor qualities pre-
sent 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 experience 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)
J5 Very slight
1 Slight
1^ slight to moderate
2 Moderate
2% 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 pre-
sented 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 ap-
pearance, and their intensity. After the observations in the test room,
the panelists 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.
B-2
Arthur D Little, Inc
-------
The panel's results were then composited into an odor pro-
file 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. STATIC ODOR TEST ROOM
The static 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 carbon (C-42 cannister from Dorex) and provided a low-odor
background diluting medium and was also used to flush odorized air from
the chamber and acclimate the four panel members to a low odor back-
ground. 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 pri'or 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. Most of the initial odor studies
were done using a 600/1 dilution of exhaust obtained
from a 21-liter sample. 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 back-
ground air.
e. The panel then enters the odor chamber to make ob-
servations.
B-3
Arthur D Little Inc
-------
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. DYNAMIC ODOR TEST CHAMBER
A dynamic odor test chamber has been constructed at ADL in
conjunction with other research progrms being conducted on the basic
characteristics of odor. This facility, shown schematically in Figure
B-l, provides an excellent means for efficiently conducting the dose/
response odor studies required for more precise determination of the
odor intensity of exhaust samples.
The facility has been constructed based on the same general
guidelines used for the static ahmber. It consists of a carbon filter
to provide normally a 100 Jl/sec odor-free air flow which is controlled
by the exhaust fan. A remote controlled syringe drive introduces TOE,
LCA, or LCO exhaust samples which are then thoroughly mixed in an ADL
static mixer to insure homogeneous sample presentation in a laminar
flow through the test chamber. The chamber (approximately 6 x 6 x A1)
comfortably seats a single observer. Communication for reporting odor
intensity and character is achieved through a microphone/headset com-
bination.
During a test the odor analyst is not informed as to the
time or order in which samples are presented. The analyst is instructed
to report observations when perceived. A typical procedure would be as
follows:
1. Analyst enters chamber, adjusts microphone,
and adapts to odor-free background for three
minutes.
2. A sample is injected for a period of 30 sec.
at a predetermined concentration. Odor ob-
servations when perceived are reported verbally
where they are recorded by the technician operating
the facility from an adjacent room.
3. An odor-free background is presented for two
minutes.
4. Sample presentation is conducted at a new con-
centration.
5. The sequence is repeated for the number of points
required.
B-4
Arthur D Little, Inc
-------
f
Ul
Static Mixer
Sample
Introduction
(Harvard
Syringe Pump)
Carbon Filter
Total
Hydrocarbon
Analyzer
FID
Plenum
Chamber
In
Test Chamber
Air Shunt
Exhaust Fan
D
^4
f+
—
R
FIGURE B-l DYNAMIC ODOR TEST CHAMBER
-------
Each sample is examined by four panelists, and the average TIA value
is reported. Typical results for a TOE total exhaust extract would
be:
Dose Concentration (&/m3) TIA
0.2 0.5
0.4 1.0
0.8 1.5
1.6 1.75
3.2 2.5
break
0.8 1.25
0.2 0.5
3.2 2.25
1.6 1.75
0.4 0.75
When these data are treated by the established semi-logarithmic relation-
ship of sensory response, a dose/response curve such as shown in Figure
B-2 is obtained. From the least squares correlation of these data, the
exhaust odor intensity can be described as
TIA =1.50+1.50 log Cone.
The odor intensity at 1 £/m3, corresponding to a 1000/1 dilution is
mathematically represented by the constant in the equation above. It
is thus convenient to report the observed odor intensity in terms of
the constant and slope of the line.
B-6
Arthur D Little, Inc
-------
w
IT
-1
D
c:
TIA
0.1
0.5 1.0
EXHAUST CONCENTRATION - «/m3
5.0
10
Figure B-2 DOSE/RESPONSE DETERMINATION OF DIESEL EXHAUST ODOR
-------
APPENDIX C
CHEMICAL COMPOSITION OF THE ODOR COMPONENTS IN DIESEL EXHAUST
A detailed chemical description of the odorous compounds
responsible for the characteristic diesel exhaust odor is given in the
second and third year final reports of the CAPE-7-68 project (refs. 2
and 3). The conclusions of these studies are presented here for review
purposes. In the sensory evaluation of diesel exhaust odor, we have
described the odor as having two primary odor character groups, oily-
kerosene and smoky-burnt, each contributing about equally to the exhaust
odor intensity. The composition of each of these groups is summarized
below.
1. Oily-kerosene
The oily-kerosene odor charcter group is isolated from
diesel exhaust samples in the aromatic (LCA) fraction using the liquid
chromatographic separation procedure described in Appendix E. The pri-
mary chemical species identified in this fraction are summarized in
Table C-l. Using a No. 1 diesel fuel, the quantitative composition of
this fraction was found to be typically
Species Weight
Alkyl benzenes 24
Indans/tetralins/indenes 14
Naphthalenes 57
Completion of the two-stage gas chromatography-odor-high resolution mass
spectrometry analysis of this fraction revealed that the main odor con-
tributor responsible for the kerosene type odor was the series of alkyl
indans and tetralins. The alkyl benzenes were found to contribute to
the oily odor character assisted by the indenes. No specific odor cor-
relation was found with the naphthalenes, although they may contribute
to the overall odor perception through some form of synergism.
C-l
Arthur D Little Inc
-------
Table C-l
ODOR-STRUCTURE CORRELATION FOR THE OILY-KEROSENE ODOR COMPLEX
ODOR TYPE
1
D
c:
r^
i-+
—
B
o
to
OILY
KEROSENE
ASSOCIATED STRUCTURES
ALKYL BENZENES1
ALKYL INDENES2
ALKYL INDANS3
ALKYL TETRALINS4
SENSATION (FEEL, IRRITATION) METHYL NAPHTHALENES5
-------
Further studies showed that the composition of the aromatic
fraction was the same as that isolated from the 'fuel. The principal
source of this odor fraction is, therefore, believed to be in the aro-
matic fraction of the unburned fuel in the exhaust hydrocarbons.
2, Smoky-Burnt
The chemical species responsible for the smoky-burnt exhaust
odor are much more numerous and complex than was the case for the oily-
kerosene odor group. The primary species found to be associated with
the characteristic odor are listed in Table C-2.
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.
Summarized, 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 benzal-
dehydes.
• The oxidized oily character is usually ascribed to
alkenones, dienones, hydroxy cyclocarbonyls, and
indanones.
• Irritation factors seem most frequently to be associated
with the lower molecular weight phenols. Some benzaldehydes
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 wi£h exhaust odors
The major source of the most significant odor contributors
in this group appears to be the aromatic portion of the fuel, although
some contribution is also seen from the paraffin portion. Confirmation
of this observation was obtained with the study of a paraffin fuel
(Soltrol 200) where the odor character was seen to change markedly —
the odor was not eliminated, however.
C-3
Arthur D Little Inc
-------
-«
D
Table C-3
ODOR-STRUCTURE CORRELATION
SMOKY-BURNT ODOR COMPLEX
STRUCTURE TYPE
ODOR
ALKENONE
DIENONE
OXIDIZED OILY
OXIDIZED OILY
FURAN
FURFURAL
IRRITATION
BURNT
n
BENZENES
METHOXY
PHENOL
ALDEHYDE
SMOKY, PUNGENT
BURNT, IRRITATION
BURNT, PUNGENT
BENZOFURAN
PARTICLE SIZE
INDANONE
INDENONE
SMOKY METALLIC
LEATHERY TARRY BURNT
NAPHTHALDEHYDE
PARTICLE SIZE
-------
APPENDIX D
SAMPLE COLLECTION PROCEDURES
This appendix describes the original silica gel and the new
Chromosorb 102 sample collection procedures.
1. 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 hydro-
carbon breakthrough is noted after the collection of 720 liters 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
the silica in the collection tubes successively with 50 m£ of pentane
and 50 m£ of 10% MeOH/H20 solution which is,0.01N in H2SOit. The aqueous
acid methanol extract is reextracted twice with 5 m£ aliquots of CHC13.
Analysis of the pentane and chloroform extracts suggests that 90% of the
organic sample collected is extracted by the pentane. This extract con-
sists 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.
2. CHROMOSORB 102 SAMPLE COLLECTION AND ISOLATION
A. Summary
Sufficient sample may be collected from 500 - 1000 liters
of exhaust for all of the analyses (odor and analytical) now required
using Chromosorb 102 adsorbent traps. The collected exhaust is then
isolated by pentane elution.
B. Supplies
Stainless steel collection trap (Figure D-l) - the drawing
is to scale (1/1). The Chromosorb is contained between two 100 mesh
screens held in slight compression by the coil spring. Complete blue-
prints are available on request.
Arthur D Little Inc
-------
Glass funnels - 10 cm diameter with stem dimensions the
same as the trap (1/4 in. O.D.) so that they may be close coupled with
Teflon tubing.
n-Pentane (Phillips Petroleum, Chromatoquality) - distilled
at atmospheric pressure, B.P. 36°C.
Methylene chloride (Fisher Scientific, Spectroquality) -
distilled at atmospheric pressure, B.P. 38°C.
Methanol (Fisher Scientific, Spectroquality)
Chromosorb 102 (Johns-Manville, 60/80 mesh)
C. Procedure
1 . Chromosorb Preparation
The adsorbent is prewashed as follows: Plug the end of a
large 10 cm O.D. glass column with silanized glass wool and transfer in-
to it 150 g of Chromosorb. Wash the Chromosorb in the column successively
with 600 mi of MeOH, 600 mi of methylene chloride, 600 mi of n-pentane.
Pour these solvents into the column in that order and allow one solvent
to drain completely from the column before adding the next solvent. The
solvent flow is ca. 5-10 mi per minute.
The first portions of the solvent wash have a yellow tint
and a characteristic odor. The intensity of the color and odor decreases
as the prewashing cycle proceeds. The absence of this color and odor
should be used as a guide to determine completeness of the prewashing
process.
When the prewashing cycle is completed, the adsorbent is
left in the column at room temperature for 24 hours to get rid of most
of the residual pentane. It is then activated at 70°C for 24 hours.
2. Trap Preparation
Ten grams of Chromosorb 102 is weighed out and packed in
the stainless steel trap between the wire screens. Glass wool may also
be required if the screen mesh is too large. The direction of exhaust
flow is marked and the trap weighed to at least + 10 mg.
3. Exhaust Collection
The trap is connected to the exit side of a Diapump Teflon
diaphragm pump with a Swagelok nut and Teflon ferrules. Exhaust (hot,
fiber glass filter filtered) is then passed through the trap at about
10 Z/min until the desired 500 - 1000 liters have been collected. After
collection, tiie Swagelok fastenings are removed and the trap reweighed.
D-2
Arthur D Little Inc
-------
4. Sample Isolation
The trap is fastened upright on a ring stand above a
graduated receiver with the trap inlet side facing down to the receiver.
The funnel is then attached to the trap exit end with Teflon tubing.
A separatory funnel or solvent reservoir is positioned
just above the funnel and filled with about 10 mi of pentane. The sol-
vent is allowed to slowly percolate through the trap. Flow should be
about 0.5 mil/min. The solvent must be added to what had been the exit
end of the trap so that the trapped odor species may be eluted as easily
as possible. If the solvent flow is reversed, the trapped species chroma-
tograph on the bed packing with resultant lower recovery.
The first 10 mfc of pentane effluent is collected and con-
tains all of the total organic extract (TOE).
D-3
Arthur D Little Inc
-------
bushing
Cap
r-i
V,
Teflon
washer
a
i I
i
- 1/4" O.D.
Teflon washer
Cap
7/8" O.D. body
100 mesh S.S. screens
coil spring
SCALE = 1/1
Figure D-l Chromosorb Diesel Exhaust Sampler
D-4
Arthur D Little Inc
-------
APPENDIX E
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 E-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 76,5008, of exhaust.
The fractionation of the sample, along with the elution scheme and
qualitative odor is given in Table E-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.
E-l
Arthur D Little Inc
-------
Table E-l
SILICA LCC ELUTION SCHEME AND ODOR OBSERVATIONS FOR SAMPLE 25
Fraction
1
2
3
4
5
6
7
8
9
10
11
12
Solvent
Pentane, 150 ml c
Pentane, 100 ml
Pentane, 100 ml
Comment
Colorless effluent.
Colorless effluent
Colorless effluent
FOE (mg)
3,500
Odor
Odorless
Benzene, 100 ml The yellow component starts
moving down upon addition of
benzene; collected effluent
was still colorless
Benzene, 100 ml
Benzene, 100 ml
CHCI3, 150ml
700
Oily, kerosene
150
Smoky-burnt, oily
Greenish yellow effluent .
Greenish yellow
Light greenish yellow
i, 100 ml Very light greenish yellow
10% MeOH/CHCI3, 100 ml Very light greenish yellow
25%MeOH/CHCI3, 100ml Brown
50% MeOH/CHC I3 100 ml 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.
E-2
Arthur D Little Inc
-------
2. MICRO COLUMN PROCEDURE FOR QUANTITATIVE ANALYSIS
A new micro LCC procedure was developed to simplify the anal-
ysis 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 char-
acteristics are shown in Table E-2.
3. PREPARATIVE ROUTINE PROCEDURE
A. Summary r
Diesel exhaust total organic extract (TOE) is separated into
the three primary chemical groups (paraffin, aromatic, and oxygenate)
by solvent gradient elution on silica gel.
B. Supplies
Chromatographic column - 1.1 cm I.D. x 30 cm with Teflon stop-
cock.
Graduated test tubes or volumetric flasks - 10 mil and 25 m£
capacity.
Graduated cylinders.
Separatory funnels for use as solvent reservoirs.
n-Pentane (Phillips Petroleum, Chromatoquality) - distilled at
atomospheric pressure and representing fraction boiling at 36° C.
Methylene chloride (Fisher Scientific, Spectroquality) -
tilled at atmoshperic pressure and representing fraction boiling at 38°C,
Methanol (Fisher Scientific, Spectroquality).
Silanized glass wool.
Teflon tubing - 1/4 inch.
E-3
Arthur D Little Inc
-------
Table E-2
Micro-LCC Fractionation of Gel Trap Samples
(b)
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/CHCl3, 2.5
10% MeOH/CHCl3, 2.5
10% MeOH/CKCl3, 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.
E-4
Arthur D Little, Inc
-------
Silica Gel (Fisher Scientific Company 60 - 200 mesh, Grade
950) - this adsorbent is activated at 110°C for two hours just prior to
use.
C. Procedure
1. Dry pack the chromatographic column, plugged at
one end with glass wool, with 6.2 grams of freshly
activated silica gel. For properly activated silica
gel, a portion weighing 6,3 +_ 0.2 g occupies 8 m£
graduated cylinder.
2. Pour pentane into the solvent reservoir
positioned above the column and let the
pentane flow into the silica gel bed until
the column is homogeneous throughout and
free of any break and trapped air bubbles.
Gently tapping the column (with a. short
piece of rubber pressure tubing), aids in
tight packing of the column. The total
height of the silica bed in this packed
column should be 10 cm.
3. As soon as column is ready, allow the pentane
level in the column to reach the level of the
top of the silica bed. This top portion of
the silica bed will be referred to as the
origin.
4. Position a 25 mi graduated receiver at column
end.
5. Quantitatively transfer the 10 mi of the TOE
directly into the column and maintain solvent
flow at 1 mi per minute. Add more pentane as
required and do not allow column to run dry.*
6. Collect 20 mi of pentane effluent. At the
end of this collection, the pentane level
in the column must be at the origin. Slowly
pour 10 mi of methylene chloride into the
column and collect 4 mi more of effluent.
This is fraction LCP. (The 24 mi of effluent
may be made up to 25 mil with distilled pen-
tane) .
* After every change in solvent, make the necessary stopcock adjustment,
so that the solvent flow is maintained at 1 mi per minute.
E-5
Arthur D Little Inc
-------
7. Replace receiver with a 10 m& volumetric
flask and collect 10.0 mS, of methylene
chloride effluent. At the end of this
collection, solvent level in the column must
be at the origin. This is fraction LCA.
(Carefully, control the last portion of
methylene chloride added into the column to
avoid exceeding the required amount).
8. Replace receiver with another 10.0 mJl volumetric
flask and pour 10 mS, of 50% methanol/methylene
chloride into the column. Collect 10.0 mH of
effluent. This is fraction LCO.
This procedure can be represented schematically as shown
in Figure E-l, It has been possible to automate this procedure with
pumps, closed columns, detectors, etc., but the overall efficiency is
not especially improved. The manual procedure is quite satisfactory
for the analytical needs and, with a small amount of practice, simple
and efficient to use.
E-6
Arthur D Little Inc.
-------
Pentane
Methylene
Chloride
N /
50%
Methanol
in
Methylene
. Chloride .
TOE
V >
BSj-— Glass Woo I
SiO-
Glass Wool
Graduated
" Receiver
Figure E-l Preparative LCC Schematic
E-7
Arthur Dbttlelnc
-------
APPENDIX F
GAS CHROMATOGRAPHIC MASS ANALYSIS OF EXHAUST SAMPLES
Most of the relationships developed between composition and
odor have been based on the mass (weight) of the exhaust sample. A
gas chromatography simple procedure was developed to provide data on
the total mass of the sample present in the solutions of the various
total exhaust organic extract (TOE) and liquid chromatography fractions
(LCA, LCO).
1. PROCEDURE
Equipment - Temperature programmable gas chromatograph with
FID detector (P-E 900). .
Column - 1/8" O.D. x 1% ft stainless steel packed with
10% OV-1 on 100/120 mesh Gas Chrom Q.
Temperature Program - Initial isothermal operation at room
temperature for three minutes after sample in-
jection to allow solvent elution - then ballistic
heating to 250°C in three minutes. A typical
chromatogram is shown in Figure F-l.
Calibration - Achieved by measuring the area FID response
from a 1% or 0.1% solution of diesel fuel in
pentant. The mass in a sample thus related to
fuel is measured as a fuel oil equivalent (FOE).
2. RESULTS
The reproducibility of the procedure can be seen from the
data in Table F-l which was obtained by repeated analysis of the same
samples on the dates shown.
F-l
Arthur D Little Inc
-------
Time (min)
FIGURE F1 TOTAL ORGANIC MASS GAS CHROMATOGRAPHIC
ANALYSIS OF AN LCO EXHAUST SAMPLE
F-2
Arthur D Little Inc
-------
Table F-l
Reproducibility of Total Organic GC Method
Date
12/18/70
12/23/70
12/29/70
1/5/71
1/14/71
Fuel Oil
Response
(sq. in/yg)
27.1
30.7
26.6
27.1
24.5
locai.
LCP
50.4
45.7
52.8
49.4
-
ur games uoseirv
LCA
9.1
8.0
11.1
8.6
11.5
LCO
1.40
1.40
1.34
1.05
1.24
Average
27.213.5
49.6±4.0
9.7±1.8
1.29+0.24
a. pg/Jt as FOE values
F-3
Arthur DLittldnc
-------
APPENDIX G
MASS SPECTROMETRIC ANALYSIS OF DIESEL EXHAUST
ODOR FRACTIONS
Computer programs have been developed for the detailed com-
position analysis of the LCA and LCO oily-kerosene and smoky-burnt odor
fractions isolated from diesel exhaust by the liquid chromatography pro-
cedures. A complete Fortran program listing is available from our lab-
oratories on request. The basic characteristics of each program are
described briefly in this appendix.
1. AROMATICS—OILY-KEROSENE FRACTION
The procedure for the LCA fraction is a modification of the
usual matrix analysis of aromatic fractions conducted by the petroleum
industry. The version developed by the Mobil Research and Development
Corporation was kindly supplied to us for our studies. The program was
modified to enable us to process data on our laboratory Hewlett-Packard
2116B computer.
The input requirements are a listing of the unit mass peak
height intensities from m/e 80 obtained from a low resolution (m/Am =
500) mass spectrum of the LCA fraction. A typical input is shown in
Table G-l.
The resultant final analysis from that set of data is given
in Table G-2. The exhaust and fuel samples from No. 1 fuel do not nor-
mally have a significant amount of the acenaphthene and phenanthrene
classes, but these classes are more significant in higher boiling fuels.
Data in this form provide a means for determining the indan/tetralin
concentration in the exhaust related to the kerosene odor.
2. OXYGENATES—SMOKY-BURNT FRACTION
A procedure, similar in concept to that used for the aromatics
fraction, has been developed for comparative analysis of the oxygenate
(LCO) odor fraction. The method is based on a Fortran program analysis
of the high resolution mass spectrum of the LCO sample.
G-l
Arthur D Little Inc
-------
TABLE G-l
INPUT DATA FOR AROMATICS MATRIX ANALYSIS
Listing of observed peak heigh intensities, in arbitrary
units, listed in order of increasing mass.
m/e 80
^0
0
180
123
60
477
270
238
156
306
1555
41
75
58
25
1
2
0
360
1860
168
63
96
633
4950
240
369
339
155
26
36
4
1
0
0
189
747
336
54
51
309
3720
807
330
135
305
24
84
7
2
1
0
480
156
304
168
62
1044
717
891
114
462
114
21
15
3
1
0
0
510
178
261
111
24
888
201
711
59
249
210
20
29
9
3
0
0
240
153
1008
2028
62
198
714
2550
348
147
84
30
6
5
2
0
0
150
81
329
429
299
39
336
4350
166
225
105
74
5
12
1
0
0
210
117
120
531
738
66
489
742
546
63
156
27
24
2
1
0
0
87
300
54
294
2802
59
420
168
639
84
177
42
6
6
2
0
0
453
108
102
1866
915
750
102
657
516
59
57
30
6
1
i
1
N
m/e 249
G-2
Arthur D Little Inc
-------
TABLE G-2
AROMATICS COMPOSITION ANALYSIS OF AN
OILY-KEROSENE LCA ODOR FRACTION
Species Composition
Alkylbenzenes
Tetralins, Indans, Indenes
Naphthalenes
Acenaphthenes, etc.
Phenanthrenes, etc.
G-3
Arthur D Little Inc
-------
In the analysis a complete high resolution spectrum (m/Am =
10,000) is first obtained on the sample. The first and last page of the
output from a typical sample is listed in Table G-3. The full spectrum
typically contains data on 500 - 1000 spectral lines. This spectrum con-
tains all of the basic mass (DET), R+DB value (DB), and composition (C12/
13 H 0). The analysis procedure takes advantage of all of this chemically
resolved data and that desirable and undesirable data (hydrocarbon inter-
ferences, fragmentation, isotopes) are resolved and can be easily separated.
This output is also provided on punched cards as input for
the analysis procedure. A Fortran IV program is written for use on an
IBM 360/65 computer. The data from the entire sample is then searched,
and the program is instructed to list all of the desired chemistry. In
arriving at criteria and a format for listing the data, we utilized as
much of the previously determined specific chemical identification data
as possible. We also chose groupings which satisfied our initial feeling
of the odor significance of certain chemical classes. We have found it
convenient to discuss the odor significant chemical classes in terms of
their R+DB (rings plus double bonds—a representation of chemical un-
saturation) values and have shown that the species in a particular R+DB
class have primarily one type of chemistry. For instance, R+DB 2 with
1 oxygen is primarily unsaturated aldehydes or ketones, R+DB 4 with 2
oxygens is primarily methoxy and hydroxy substituted phenols.
Further, we have determined that unsaturated aldehydes are
more closely related to oxidized oily odors than those with 2 oxygens.
Similarly, indanols and indanones (R+DB 5 and 6) with 2 oxygens are more
odor significant than those with 1 oxygen for the burnt odor character.
A review of the meaning of R+DB, as given in the last final
report (ref. 3) is appropriate at this point. One of the most useful
first interpretative aids in restricting the possible structural assign-
ments 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. (ref. 4) Values of R+DB
are basically arrived at by a simple analysis of the degree of unsatu-
ration in a molecule with a particular composition. For species con-
taining only carbon, hydrogen, and oxygen, the values are arrived at
numerically from the formula
R+DB = No. C atoms - ^(No. H. atoms) +1
G-4
Arthur D Little Inc
-------
Several examples will serve to demonstrate the utility of the values:
for an n-paraffin
R+DB =6-7+1=0; i.e., the n-paraffin has no
rings nor double bonds
for a hexenone
R+DB =6-7+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.
The Fortran program has been written to organize all of the data by
R+DB class and number of oxygen atoms and also includes the following
exclusion criteria in order to minimize information unique to the mass
spectrometric fragmentation process, and match the identification data:
1) Only even mass ions are considered (fragment ions
have too many origins).
2) Only ions above mass 94 are searched (no odorous species
were found below this MW).
3) No 13C isotope peaks are included.
4) Peaks with large composition computation error (greater
than 0.002 arau) are excluded.
5) Ions without oxygen (hydrocarbons) are excluded.
G-5
Arthur D Little Inc
-------
6) An ion in a particular R4-DB class must have at least
the following number of carbon atoms
R+DB Lower C Limit
0 5
1 5
2 6
3 4
4 6
5 7
6 8
7 9
8 10
A complete listing of the output obtained from the full spectral data is
given in Table G-4. A listing is obtained within an R-fDB and oxygen
restriction of all of the observed species including their relative
abundance (HGT, normalized to a base of 1000 for the total data set,
molecular weight (DET.MASS), and composition (12C,H,0). A total sum
of each R+DB/oxygen group is thus also listed (SUM HGT). From such an
analysis for all R+DB classes, the final analysis matrix is then assembled
as shown in Table G-5.
The data in this table has been normalized to total 100% and
therefore represents the mass spectrometric abundance of any particular
R+DB/oxygen combination. These abundances have not yet been adjusted
for the relative differences inherent in the parent ion intensity due
to differences in fragmentation patterns between chemical classes. An
approximation of the relative amount of any particular combination of
chemical classes can be computed from the product of the summed (or
individual) percentages and the concentration of total sample (FOE value)
in mg/KA. These values can then be used for odor correlations.
Presently we feel that the oxidized oily character note of
these odor fractions should be due to the unsaturated aldehydes and
furans represented in the R+DB 2/Oi, R+DB 3/0^ and 02 classes. Some
portion of the smoky character should be due to the phenols, acetophenones,
and indanones in R+DB 4,5,6/02, and perhaps some irritation is due to the
simpler phenols in R+DB 4/Oi class. At the present time we do not feel
that the chemistry represented in R+DB classes 1, 7, and 8 or the 03
classes contribute significantly to the odor from most operating con-
ditions; however, they may become important under some test conditions.
G-6
Arthur D Little; Inc
-------
Table G-3
High Resolution Mass Spectrum
SERIAL NUMBER 829
CHROM - 33-1A LCC 5C(HEOH PP 9-27 EXP 11./
MAXHET= 4 C-13= 1 C-16* 4
INT(ESTM) DIST OETM CALC ERRCR OB X C12/13 H 0
19 C.5377 7C.C3146 *** (No Composition Assigned for C, H, 0)
96 0.5487 7C.04116 70.04186 -0.71 2.00 2 4/0 6 1
337 0.5889 7C.C7662 70.07825 -1.63 1.00 0 5/0 10 0
6 1.1070 70.53448 ***
7 1.6271 7C.SS561 +**
18 1.6470 71.01328 71.C1330 -0.02 2.50 3 3/0 3 2
18 1.6674 71.C314C ***
13 1.6772 71.04C11 ***
204 1.6875 71.04926 71.04969 -0.43 1.50 I 4/0 7 1
228 1.7278 71.C85C7 71.08608 -I.01 0.50 -1 5/0 11 0
3 2.6815 71.935C7 ***
3 2.7772 72.02C64 72.02113 -0.49 2.00 2 3/0 4 2
2 2.7856 72.C2815 ***
53 2.8C15 72.04238 *** '
127 2.8171 72.C5633 72.05751 -1.18 1.00 0 4/081
9 2.8533 72.C8672 72.08943 -0.71 0.50 -1 4/1 11 0
40 3.8786 73.CCS15 73.C0783 1.33 6.50 11 6/0 1 0
25 3.9213 73.C4761 ***
3 3.9352 73.C6C13 73.06087 -0.74 1.00 0 3/181
38 3.9403 73.C6473 73.06534 -0.61 0.50 -1 4/0 9 1
G-7
-------
Table G-3 cont.
<, 1C7.6402 1S6.181S2 I S6 . 18271 -0.00 2.00 2 13/0 24 1
3 108.2070 1S7.13467 197.13303 1.84 7.50 13 15/0 17 0
2 108.3251 1C,7.1C-1C7 lS7.iy054 C.54 1.50 1 13/0 25 1
3 IC8.9697 1S8.14322 1<58.140B:5 2.37 7.00 12 15/0 13 0
2 10S.64S5 lSS.14«;e4 199.14868 1.16 6.5C 11 15/0 19 0
6 110.3C31 2CC.12CC7 2CO.12011 -0.04 7.00 12 14/0 16 1
5 110.9797 2C1.126<;5 2C1.12794 -0.99 6.50 11 14/0 17 I
13 111.6557 2C2.13546 2C2.13576 -0.30
5 112.2975 2C3.CS52C, ***
3 112.3261 2C3.14111 2C3.14359 -2.48
203.13912 1.99
6.00 10 14/0 18 1
5.50 " 9 14/0 19 1
6.00 10 13/1 18 1
2 112.9773 2C4.11442 2C4.11503 -0.61 6.00 10 13/0 16 2
7 113.CC12 2C4.15C3C 2C4.15141 -1.12 5.00
14/0 20 1
3 113.0264 204.16613 2C4.18780 0.33 4.00 6 15/0 24 0
3 113.6717 205.158C2 2C5.15924 -1.22 4.50 7 14/0 21 1
10 114.3415 2C6.L6666 2C6.167C6 -0.18 4.00
14/0 22 1
6 115.0C81 2C7.1733*; 2C7.17489 -1.50 3.50 5 14/0 23 I
8 115.6747 2C8.18234 208.18271 -0.37 3.00 4 14/0 24 1
2 117.0CC9 21C.1S6S5 210.19836-1.4? 2.00 2 14/0 26 1
1 119.575C 214.13444 214.13576 -1.33 7.00 12 15/0 18 1
3 120.883fa 216.15C27 216.15141 -1.15 6.00 10 15/0 20 I
5 121.5C3c! 217.1C64f. ***
G-8
-------
Table G-4
R+DB Analysis of art LCO Smoky-Burnt Odor Fraction
SERIAL NUMBER 829
CHROM - 33-1A LCC 5C(MECH PP 9-27 FXP II./
ANALYSIS FOR DB= 0
D£.TJ_MAiS. £12 H Q
SUM HOT = .. Q/J.QQQ. ... .NUM6EA Qf__CXYQ.ENS =
SUM HGT = 0/1QQO NUMBER QF CXYQENS = 2
SUM HGT = 0/100Q NUMBER OF CXYGENS = 3
G"9 Arthur D Little Inc
-------
SERIAL NUMBER _&29_
CHROM - 33-1A LCC 50(MECH PP 9-27 EXP 11./
ANALYSIS FOR DB=
HGT PET. MASS
C12 H 0
12
100.08688
142.13385
6 12 1
9 18 1
_1UM.HGT = 12/1000
NUMBER OF OXYGENS =
SUM HGT =
0/1000
NUMBER OF OXYGENS =
SUM HGT =
0/1000
NUMBER OF OXYGENS =
G-10
Arthur D Little Inc
-------
SERIAL NUMBER dZ9_ . .
CHROM - 33-1A LCC 50(MECH PP 9-27 EXP 11./
ANALYSIS FOR DB= 2
HOT QE.J.,. MAS.S C12 H 0
85
29
19
io
4
1
1
1
0
98.07167
112.C8760
116,10295
140.11863
154,13436
168.15037
LaZ.16655-.
1S6. 18192
210.19695
6
7
8
9
10
11
._.. 12.
13
14
10
12
14
16
18
20
22
24
2.6
1
1
1
1
1
1
I ....
1
1
SUM HGT 150/1000 NUMBER OF OXYGENS =
114.06759 6 10 2
U8,08a38 7 12 2
142.09859 8 14 2
SUM HGI = 8/10J10_. WJtfBER OF CXYGENS j!
SUM HGT = 0/1000 NUMBER OF CXYG6NS =.. .3 _.
G—11
Arthur DUttklnc
-------
SERIAL NUMBER 829
CHROM - 33-1A LCC 50(MECH PP 9-27 E XP 11./
ANALYSIS FOR OB=
HGT PET. MASS C12 H 0
50
25
22
10
7
6
5
4
3
2
96.05617
110.07166
124. 08696
138.103C8
152.11942
166.13447
180.15070
194.16581
208.18234
222.19783
6
7
8
9
10
11
12
13
1*
15
8
10
12
14
16
18
20
22
24
26
1
1
1
1
1
1
1
1
1
1
SUM HGT = 134/1000 NUMBER OF CXYGENS =
10 9B.03619
5
7
2
1
112.05168
126.06641
140.08263
154.09831
6 8
7 10
8 12
9 14
2
2
2
2
168.11423 10 16 2
SUM HGT = 25/1000 NUMBER OF OXYGENS =
SUM. HGT = 0/1000 N.UMBEK OF OXYGENS = 3
. e_12 Arthur D Little Inc
-------
SERJAL NUMBER _JJ29
CHROM - 33-1A LCC 50
-------
SERIAL NUMBER .829 _
CHROM - 33-1A LCC 5C(MECH PP 9-27 EXP 11./
ANALYSIS FOR DB=
HGT PET. MASS C12 H Q_
3
7
12
11
5
2
0
130.04132
144.05714
158.07230
172.08738
186.10351
200.12007
214.13444
9
10
11
12
13
14
15
6
8
10
12
14
16
18
1
1
1
1
1
1
SUM HGT = 40/1000 NUMBER CF CXYGENS =
6
8
3
0
146.03622
160.05224
174.06781
188.C8240
9 6
10 8
11 10
12 12
2
2
2
2
SUM HGT = 17/1000 NUMBER OF CXYGENS =
2 162.03177 963
2 176.04667 10 8 3
SUM HGT = 4/1000 NUMBfcR OF CXYGENS =
G-1.6 . ..... _.. ..... ArthurDLittleInc
-------
SERIAL .NUMBER 829
CHROM - 33-1A LCC bCXMECH PP 9-27 EXP IL. /
ANALYSIS FOR DB= 8
_HGT PET. MASS CL2 H 0
21 . 156.55671 _ _U _8
23 170.07220 L2 10 1
6 184.08764 __ 13 12 1
SUM HGT 50/1000 NUMBER OF OXYGENS =
2 158.03632 10 6 2
4 172.05179 1182
5 186.06853 12 10 2
SUM HGT = 11/1000 NUHBER OF OXYGENS
SUM HGT = 0/iOOO NUMBER CF OXYGENS _= 3__
17 Arthur D Little Inc
—i /
-------
TABLE G-5
R4-DB MATRIX REPRESENTATION OF CHROM 33 LCD FRACTION
R+DB
0
1
2
3
4
5
6
7
8
°I
0
1
15
13
10
16
13
4
5
£2
0
0
8
2
2
3
3
2
1
£l
0
0
0
0
0
0
1
0
0
G-18
Arthur DLitthlnc
-------
BIBLIOGRAPHIC DATA
SHEET
1. Report No.
EPA-R2-73-275
3. Recipient's Accession .
PB-220 392
4. Title and Subciclc
Analysis of the Odorous Compounds in Diesel Engine
Exhaust
5. Report Date
June 1972
6.
7. Author(s)
8. Performing Organization Rept.
No- ADL 73686-5
9. Performing Organization Name and Address
Arthur D. Little, Inc.
Cambridge, Massachusetts
10. Pro|ect,'Task/Work Unit N'c
11. Contract/Grant No.
68-02-0087
12. Sponsoring Organization Name and Address
Coordinating Research Council, New York, N. Y.
and
ENVIRONMENTAL PROTECTION AGENCY, Research Triangle Park, N. C.
13. Type of Report & Period
Covered
Fi nal
14.
15. Supplementary Notes
CRC Project CAPE-7-68
to obtain appropriate means for
in diesel exhaust and to develop the
16. Abstracts Tne repOrt describes the effort undertaken
measuring the previously identified odorous species
quantitative relationships between these measurements and the exhaust odor. Diesel ex-
haust odor can be described as having two major odor groups: oi ly-kerosine and smoky-
burnt. To obtain a more precise determination of exhaust odor intensity, new odor mea-
surement techniques were developed in the form of a dose/response relationship. In this
technique the sample odor intensity is reported by an odor panel using a dynamic test
chamber and the presentation of a range of controlled concentrations. A routine prepara-
tive scale liquid ch romatograph i c procedure is described which separates the dieseT ex-
haust sample into its three major fractions: paraffin, aromatic, and oxygenates. Mass
spectrometry methods have been completed for the aromatic and oxygenate fraction which
allow the comparison of the relative amounts of odorous components in the two isolated
odor fractions. A range of fuel and injector variables were examined using the k-J\
engine to generate data to test odor-analytical correlations and to determine the adequa
cy of available chemical data. Results and conclusions are presented.
17. Key Words and Document Analysis. 17o. Descriptors
Ai r pollut ion
Odors
Exhaust emissions
Diesel engines
Chemical analysis
Sampling
Ch romatog raphy
Dos imetry
Fuels
Kerosene
17b. Identifiers/Open-Ended Teems
Air pollution detection
Smoky-burnt fractions
17e. COSATI Field/Group
13B
18. Availability Statement
Unlimited
19. Security Class (This
Report)
UNCLASSIFIED
20. Security Class (This
Page
UNCLASSIFIED
21- No. of Pages
127
22. Price
F"RM NTIS-35 (REV. 3-72)
THIS FORM MAY BE REPRODUCED
USCOMM-DC 14952-
-------
INSTRUCTIONS FOR COMPLETING FORM NTIS-35 (10-70) (Bibliographic Data Sheet based on COSATI
Guidelines to Format Standards for Scientific and Technical Reports Prepared by or for die Federal Government
PB-180 600).
1. Report Number. Each individually bound report shall carry a unique alphanumeric designation selected by che performing
organization or provided by the sponsoring organization. Use uppercase letters and Arabic numerals only. Examples
FASEB-NS-67 and FAA-RD-68-09.
2. Leave blank.
3- Recipient's Accession Number. Reserved for use by each report recipienc.
4- Title and Subtitle. Title should indicate clearly and briefly the subject coverage of the report, and be displayed promi-
nently. Set subtitle, if used, in smaller type or otherwise subordinate it to main title. When a report is prepared in more
than one volume, repeat the primary title, add volume number and include subtitle for the specific volume.
5- Report Dote. F.ach report shall carry a date indicating at least month and year. Indicate the basis on which it was selected
(e.g., date of issue, date of approval, date of preparation.
6. Performing Organization Code. Leave blank.
7. Authors). Give name(s) in conventional order' (e.g., John R. Doc, or J.Robert Doe). List author's affiliation if it differs
from the performing organization.
8> Performing Organization Report Number. Insert if performing organization wishes to assign this number.
9. Performing Organization Name and Address. Give name, street, city, state, and zip code. List no more than two levels of
an organizational hierarchy. Display the name of the organization exactly as it should appear in Government indexes such
as USGRDR-I.
10. Projeet/Tosk/Work Unit Number. Use the protect, task and work unit numbers under which the report was prepared.
11. Contract/Grant Number. Insert contract or grant number under which report was prepared.
12* Sponsoring Agency Nome and Address. Include zip code.
13. Type of Report and Period Covered. Indicate interim, final, etc., and, if applicable, dates covered.
14. Sponsoring Agency Code. Leave blank.
15. Supplementary Notes. Enter information not included elsewhere but useful, such as: Prepared in cooperation with . . .
Translation of ... Presented at conference of ... To be published in ... Supersedes . . . Supplements . . .
16. Abstract. Include a brief (200 words or less) factual summary of the most significant information contained in the report.
If the report contains a significant bibliography or literature survey, mention it here.
17. Key Words and Document Analysis, (a). Descriptors. Select from the Thesaurus of Engineering and Scientific Terms the
proper authorized terms that identify the major concept of the research and are sufficiently specific and precise to be used
as index entries for cataloging.
(b). Identifiers and Open-Ended Terms. Use identifiers for project names, code names, equipment designators, etc. Use
open-ended terms written in descriptor form for those subjects for which no descriptor exists.
(c). COSATI Field/Group. Field and Group assignments are to be taken from the 1965 COSATI Subject Category List.
Since the majority of documents are multidisciplinary in nature, the primary Field/Group assignmcnt(s) will be the specific
discipline, area of human endeavor, or type of physical object. The application(s) will.be cross-referenced with secondary
Field/Group assignments that will follow the primary posting(s).
18. Distribution Statement. Denote releasability to the public or limitation for reasons other than security for example "Re-
lease unlimited". Cite any availability to the public, with address and price.
19 & 20. Security Classification. Do not submit classified reports to the National Technical
21. Number of Pages. Insert the total number of pages, including this one and unnumbered pages, but excluding distribution
list, if any.
22. Price. Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
FORMNTIS-33 (REV. 3-7Z) USCOMM-DC 1.44S2-P72
-------
|