71.

?fi%

. El:

ma... 3.ng

.Wuzuh: With

54

 

 

a.‘ our. ”a,“

1.4..

... i 1...... .1mm... .

lunrgft. , is...”

fit?- .
5...

.1
. 3215914.... .
t5!|¥.l):.lfi7 c;

””94 i/VH... .- l

 

 

 

THEE:

acag
5V§M§a3

This is to certify that the
thesis entitled

COMPARISON OF GASOLINES USING GAS
CHROMATOGRAPHY/MASS
SPECTROMETRY AND TARGET ION RESPONSE

presented by

 

Aisha Tamara Barnes

. EBRERY
Michigan State
University

L

has been accepted towards fulfillment
of the requirements for the

 

 

 

 

Master of degree in

lingual

[Major‘P L/I/fepéflo3 rs Signature

Date

Forensic Science

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECAU.ED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIFlC/DateDuopGS—p. 15

 

COMPARISON OF GASOLINES USING GAS CHROMATOGRAPHY/MASS
SPECTROMETRY AND TARGET ION RESPONSE

By

Aisha Tamara Barnes

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Criminal Justice

2003

ABSTRACT
COMPARISON OF GASOLINES USING GAS CHROMATOGRAPHY/MASS
SPECTROMETRY AND TARGET ION RESPONSE
By

Aisha Tamara Barnes

Gas chromatography/mass spectrometry was used to compare gasoline samples
obtained from different sources based on the difference in response of the ion detector to
the target ions of certain components found in gasoline. Many suspected arson cases
involve comparing an accelerant extracted from fire debris to an ignitable liquid found in
a suspect’s possession to determine if it could have been used in the fire. These types of
comparisons are currently based on pattern recognition and component identification and
do not take into account the variation that exists in some commonly used accelerants such
as gasoline. Fifty and seventy-five percent-evaporated gasoline samples were both found
to contain similar ratios of certain components when compared to the same source
gasoline unevaporated. This research proposes ratios to be used to determine if an
unevaporated gasoline sample could have originated from the same source as an
evaporated gasoline extract from fire debris. The results of the comparisons in this study
demonstrate that for cases involving gasoline as the accelerant, that has been evaporated
up to 50% and extracted from pine, it is possible to eliminate comparison samples as
originating from the same source. The results of the 75% comparisons suggest it may be

possible to apply the same type of comparison to cases involving 75% evaporated

gasoline.

ACKNOWLEDGEMENTS

The author wishes to thank those who have offered their time and resources to ensure this
project was successful. First, the Bureau of Alcohol, Tobacco, and Firearms of
Rockville, MD must be acknowledged for the use of instrumentation, supplies, and
chemicals used in this study. Ray Kuk and Julia Dolan of the Bureau of Alcohol,
Tobacco, and Firearms are deserving of specific gratitude for their ideas regarding the
study, their guidance, and support. Also the author would like to thank Troy Ernst for his

advice and assistance in the writing process, and Dr. Jay Siege] for his guidance and

patience.

iii

TABLE OF CONTENTS

List of Tables
List of Figures

Chapter One: Introduction
Background Information
The Problem to be Studied
Purpose of the Study
Hypothesis
Limitations
Refining Gasoline from Crude Oil

Chapter Two: Instrumentation
Capillary Gas Chromatography
Electron Ionization Mass Spectrometry

Chapter Three: Forensic Methods of Accelerant Separation and Concentration
from Fire Debris

Headspace Sampling

Solvent Extraction

Static Adsorption/Elution

Chapter Four: Development of Experiment Parameters
Instrumental Conditions
Sofiware Considerations
Selection of an Extraction/Concentration Method
Selection of Substrates
Determination of Amount of Gasoline to Use
Experimental Setup
A. 50% Comparisons
B. 75% Comparisons
Comparison Mechanics

Chapter Five: Results and Conclusions, Discussion, and Further Research
Results and Conclusions
A. 50% Comparisons
B. 75% Comparisons
Discussion and Further Research

Appendices

Appendix I: Data for 50% Comparisons

iv

vi

viii

'—'\OOO\)MNH

19
19
20
21

22
22
22
25
29
29
3O
31
31
32

35
35
36
37

4O

40

TABLE OF CONTENTS, continued

Appendix II: Data for 75% Comparisons

Appendix III: Demonstration of Comparison Ratios and Where They Lie
on a Typical Chromatograrn

Notes

Bibliography

58

65

67

69

LIST OF TABLES

Table 1.1

Gasoline Information for 50% Comparisons (Left) and 75% (Right) 11
Table 4.1

Components Chosen by Software 22-24
Table 4.2

Example of How Similar Ratios Were Determined 33

Appendix Table 1.1
Values Obtained for Ratios 1-6 for Gasolines #1-16 (Neat=Unevaporated) 40

Appendix Table 1.2
Comparison of 50% Evaporated Gasoline #1 (Debris Extract) to Gasolines #1-16
Neat; Gas=Abbreviation for Gasoline 42

Appendix Table 1.3

Comparison of 50% Evaporated Gasoline #2 (Debris Extract) to Gasolines #1-16
Neat 45

Appendix Table 1.4

Comparison of 50% Evaporated Gasoline #3 (Debris Extract) to Gasolines #1-16
Neat 48

Appendix Table 1.5

Comparison of 50% Evaporated Gasoline #4 (Debris Extract) to Gasolines #1-16
Neat . 51

Appendix Table 1.6
Comparison of 50% Evaporated Gasoline #5 (Debris Extract) to Gasolines #1-16

Neat 54
Appendix Table 2.1
Value Obtained for Ratios 1-4 for Gasolines #01-10 (N eat=Unevaporated) 58
Appendix Table 2.2
Comparison of 75% Evaporated Gasoline #01 to Gasolines #01-10 Neat 59
Appendix Table 2.3

Comparison of 75% Evaporated Gasoline #02 to Gasolines #01-10 Neat 60

vi

LIST OF TABLES, continued

Appendix Table 2.4
Comparison of 75% Evaporated Gasoline #03 to Gasolines #01-10 Neat 62

Appendix Table 2.5
Comparison of 75% Evaporated Gasoline #04 to Gasolines #01-10 Neat 63

vii

LIST OF FIGURES
Figure 2.1
Quadrupole Mass Spectrometer: Ion Source, Mass Analyzer, and Detector

Appendix Figure 3.1
Ratios Used in 50% Comparisons

Appendix Figure 3.2
Ratios Used in 75% Comparisons

viii

17

65

66

Chapter One: Introduction

In many suspected arson cases, the presence of an ignitable liquid may be
detected in fire debris. Often times a comparison ignitable liquid, possibly found in a
suspect's possession, is submitted to the testing laboratory to determine if it is of the same
type as that found in the fire debris. The strongest associations fire debris chemists can
make regarding the debris and ignitable liquid are based on pattern recognition and
component identification through use of gas chromatography coupled to a mass
spectrometer (GC/MS).

Since gasoline is the accelerant in a large percentage of arson cases, more
definitive methods of determining whether a suspected gasoline possibly originated from
the same source as that found in the fire are necessary. It is widely known that different
gasoline companies add patented performance markers to their gasoline that distinguish it
from other gasoline, but that information is proprietary and the markers are present in
only minute concentrations.l However, there are many other measurable ways in which
gasoline produced by oil refineries can differ as a result of chemical conversion methods,
treatment and reformulation processes, blending, and storage. Moreover, once a new
shipment is sent out to a gasoline station, the newly refined gasoline is mixed with the
residual gasoline still in the pumps, further lending to the possibility of distinguishing
gasoline from different sources.

Nevertheless, comparing gasoline found in debris to unevaporated gasoline is
more complicated than analyzing gasoline alone. Fire debris does not contain an
uncontaminated ignitable liquid, but rather many compounds that can contribute to (or

interfere with) the chromatogram and make comparison to the same gasoline

unevaporated very difficult. Materials such as synthetic carpets, found in debris from
residential fires, produce pyrolysis products of the same type as some components
normally found in the chromatogram of gasoline alone, such as ethylbenzene, m- and p-
xylenes, and isopropylbenzene. Wood products such as pine are also commonly found in
residential firesz, yet their pyrolysis products have retention times that do not interfere
with those of gasoline, leaving many components of gasoline unadulterated and useful for
comparison purposes.

Research demonstrates that gasoline from different sources will vary in the
abundance of its components when injected into the gas chromatograph (GC).3“1
Furthermore, by calculating ratios of sequential peak components in gasoline, differences
can be detected. This study proposes a method of differentiating gasoline by comparing
the ratios of certain components present in the headspace of unevaporated and evaporated
gasoline to gasoline from different sources. The results of this study demonstrate it is
possible to eliminate gasoline samples as originating from the same source and should
not be used to suggest with certainty that two gasoline samples originated from the same
source. In addition, the results indicate it is feasible to include a sample as a possible
source of an evaporated gasoline sample.

Background Information

Previous work by Dale Mann demonstrated that by using capillary GC, the peak
ratios and chromatograrns obtained from the C5 to C8 region of gasoline vary among
different sources. 4 In his first paper he compared eight corresponding peak ratios by first
overlaying relevant peaks to see if they were superimposable. Next, he did a quantitative

comparison on non-superimposable peaks by dividing the peak area of the peaks of

interest by reference compounds to get ratios such that gasoline from the same source
yielded similar ratios.

His second paper looked at difficulties in applying his comparison methods to
cases involving gasoline headspace.5 He used his method to compare the headspace of
unevaporated gasoline to that of gasoline up to 40% evaporated, using heated headspace
as his sampling method. He noted that samples involving gasoline evaporated more than
40% resulted in a lack of enough usable peaks in the C5-C8 region of the chromatogram
from GC. He also found that certain materials yielded pyrolysis products in the region of
interest when a sample of fire debris headspace was injected, interfering with making
gasoline headspace to fire debris headspace comparisons. Nevertheless, he established
that comparing sequential peak ratios of the components in gasoline using GC is useful in
discriminating gasoline samples through application of his method to case studies.

Although fundamental to the comparison of gasoline, Mann’s methods had many
limitations. Mann had not incorporated a mass spectrometer into his gasoline analysis
scheme. With GC/MS, not only is the gasoline separated into its individual components,
but the mass spectrometer allows for the examiner to identify and classify the individual
hydrocarbons present in gasoline, moving away from simple pattern recognition and
retention time as an indication of where individual components elute. Also, Mann did his
ratio calculations by hand, which was very time consuming and increased the likelihood
of human error. Therefore, he could not analyze a wider range of possible ratios available
from the gasoline components.

Many of these limitations were evaluated in the research done by Julia Dolan and

Christopher Ritacco.6 They compared liquid gasoline from different sources using

GC/MS with an autosampler and computer software to calculate sequential peak ratios
based on target compound response. Using a Target Compound Program available from
Hewlett Packard, peaks of interest and the relative abundanCe of their target ions were
input based on parameters identified from a liquid gasoline standard. The relative
abundance data was then imported into Microsoft Excel, which was set up to compute
sequential peak ratios, standard deviation, and percent standard deviation of replicate
runs of the same gasoline.

They originally started outwith 87 possible ratios and narrowed the scope by
eliminating ratios that remained similar among different gasolines, ratios that were not
reproducible among replicate runs of the same gasoline, and ratios that did not
reproducibly integrate or resolve among replicate runs. With this data, they were able to
narrow their focus to 20 peak ratios that yielded distinguishable ratios among gasoline
from different sources, yet remained similar among gasoline of the same source when
compared to unevaporated, 25%, and 50% evaporated gasoline from the same source.
The research of Dolan and Ritacco established that ratios used in the comparison of
gasoline could be generated using components throughout the chromatogram of gasoline,
not limited to the C5 to C8 region as proposed by Mann. This research is an ongoing
project, and thus far they have been able to discriminate 44 different gasolines from
various states in the US. through liquid-to-liquid comparisons (straight injections)
evaporated up to 50%. This method has been validated and used by the Bureau of
Alcohol, Tobacco, and Firearms (BATF) in cases involving liquid-to-liquid gasoline

comparisons.

The Problem to be Studied

Regardless of the ability to differentiate liquid gasoline samples, some arson cases
will involve the comparison of collected debris in a sealed paint can to an ignitable liquid
believed to have been used in the fire. Currently, the most common way to analyze fire
debris involves extracting a sample of the volatile material onto an adsorbent surface,
which can be concentrated with a solvent and analyzed using GC/MS.7 This
chromatogram is then compared to an injection of a comparison ignitable liquid (when
available) or to the laboratory's standards of ignitable liquids. If the debris is found to
contain gasoline through pattern and component recognition and the comparison liquid is
determined to be gasoline, a relationship can be established between the two regardless of
the possibility of different origins. These current methods do not take into account the
variation that exists among gasoline.

In this study, using GC/MS and the sequential peak ratio method developed by
Dolan and Ritacco, the headspace of evaporated gasoline extracted from fire debris will
be associated with the headspace of the same unevaporated gasoline. Higher evaporated
gasoline (not extracted from debris) will be compared to unevaporated gasoline from the
same source to test out the possibility of using these methods on highly evaporated
gasoline. If these ratios are statistically similar, then they can be compared to different
gasoline sets of unevaporated and evaporated samples such that gasoline from varying
sources can be discriminated. Therefore, the research questions proposed are: does the
headspace of gasoline from the same source unevaporated and 50% evaporated on debris
contain statistically similar ratios of its components, and are these ratios different when

the gasoline is from a different source? The same questions will be addressed using 75%

evaporated gasoline without debris addition. In researching this question, the presence of
the wood substrate will be analyzed to see if it complicates the association. Lastly, the
comparison methods will be utilized to possibly exclude unevaporated gasoline samples
as originating from the same source as an evaporated extract from fire debris (50%
evaporated gasoline) or clean substrate (75% evaporated gasoline).

Considering that there are four major methods of headspace sampling carried out
by fire debris chemists before injection into GC/MS, a decision of which method yields
the most reproducible data must first be established. In this study, headspace and static
adsorption/elution methods are shown to be appropriate methods whereby adequate
representation of the components in gasoline can be obtained for comparison using
GC/MS.

In order to answer the research questions, optimal sample size and operating
conditions must be established to be able to analyze gasoline headspace using GC/MS.
Furthermore, the sampling methods used must demonstrate highly reproducible ratios of
the same components during replicate runs. Enough replicate runs must be run to display
can to can variation of the same sample, sampling error, and reproducibility as a result of
the method chosen. It is also necessary to deduce the best way to simulate fire debris.
Moreover, a suitable substrate on which to put unevaporated gasoline that will not
interfere in its analysis must be chosen.

Subsequently, it must also be established that even with pyrolysis products
present in the chromatogram, the ratios of the useful components in gasoline from the
same source must remain Similar. Therefore, a material commonly found in arson cases

that does not interfere with the comparison of useful components in gasoline headspace

must be used. Moreover, the useful components in gasoline headspace must be
determined and a mathematical analysis performed that demonstrates their utility for
comparison purposes. All optimization parameters will be demonstrated on 50%
evaporated gasoline. Upon confirmation that 50% evaporated gasoline can be associated
to its unevaporated counterpart, 75% evaporated gasoline will be correlated to gasoline of
the same source unevaporated and compared among different sources without the
introduction of fire debris. This last step serves to exhibit the possibility that highly
evaporated gasoline contains enough components that a useful comparison can be done.
Purpose of the Study

As mentioned, the ability to discriminate among gasoline sources even though
evaporated and present in fire debris would be highly beneficial in cases where a
comparison gasoline sample has been obtained. It can serve to rule out samples as
originating from the same source as the gasoline found in the debris, or include a gasoline
sample as a possible source.

This research will prove very advantageous to the investigation of arson. Since
this research was performed in conjunction with the BATF, it could result in the
implementation of a new protocol in the analysis of arson evidence and possibly
explosion cases. Lastly, this study offers instrument parameters, sampling methods, and a
simple mathematical approach that can be successfully utilized to determine whether a

sample of evaporated and unevaporated gasoline could have or could not have originated

from the same source.

Hypothesis

Upon development of suitable sampling methods to use for 50% and 75%
evaporated gasoline that yield reproducible results, it is expected that correlating 50%
evaporated gasoline extracted from debris to the same source gasoline unevaporated is
possible. Mann’s work demonstrated that such an association is possible using the
headspace of gasoline up to 40% evaporated extracted from debris.4’5 Since he only
looked at a limited part of the gasoline chromatogram, the 50% comparisons using peaks
throughout the chromatogram should also work. Also, the debris type chosen for this
study has been found not to interfere with components normally found in gasoline.
Secondly, the research done by Dolan and Ritacco using liquid injections of gasoline
demonstrated that liquid gasoline up to 50% evaporated could correctly be associated
with the same source unevaporated gasoline, further lending confidence that the same
association between such gasoline’s headspace is possible.‘5

None of the previous research was able to correctly relate gasoline evaporated up
to 75% to the same unevaporated gasoline, and thus the decisions on the result of such
tests are based upon the reproducibility of the sampling method. It is expected that the
sampling method used on 50% evaporated gasoline will not work efficiently on 75%
evaporated gasoline because there will be greater evaporative loss of the highly volatile
components for higher evaporated gasoline. Therefore, using a method that is
reproducible but still yields a sufficient amount of peaks for comparison purposes should
result in an association between same source gasolines, especially considering that in this
study, the 75% evaporated gasoline will not be extracted from debris (no interference). It

is also expected that the peaks used for comparison purposes of the 50% evaporated

gasoline will not all be the same peaks useful in linking 75% evaporated gasoline to
unevaporated gasoline from the same source. This is because the chromatogram of 75%
evaporated gasoline will have a higher concentration of hydrocarbons in the later eluting

region and thus will be shifted to the right in comparison to that of 50% evaporated

gasoline (see Appendix III).

Limitations

The burning process used to simulate fire debris was performed under a hood
until each piece appeared charred. Since the burning took place under a hood, it was
understood that some pyrolysis products would constantly be lost between burning and
sealing the can the debris were placed in.

This study uses gasoline undergoing controlled environmental evaporation up to
75% to represent evaporated gasoline in fires. This does not take into account the
evaporative nature of gasoline if present in a real fire, which burns in an uncontrolled
nature. Also, in a real fire, gasoline is evaporated mostly by burning and can be

evaporated more than 90%.8

Another limitation is the available knowledge about the background of the
gasoline used in the study. It is known from what station and city the gasoline was
purchased, but the terminal from where the gasoline was picked up before it was
deposited to particular stations is unknown. Consequently, it is not known whether a
COrnparison is being attempted between gasoline that has come from the same production
batch, refinery, or terminal.

The substrate choice for the simulation of debris is another limitation in that the

information from this study is only applicable to cases involving pine in the debris, or

other softwoods that yield the same pyrolysis products. There are many other common

debris types found in residential and commercial fires in addition to pine, which are not

addressed in this study.

Refining Gasoline from Crude Oil

The ability to differentiate gasoline has its basis in the process that produces
gasoline—the refinery of crude oil. Gasoline, a product derived from crude oil, is a
mixture of volatile hydrocarbons composed of paraffins (alkanes), naphthenes
(cycloalkanes), olefins (alkenes), and aromatics. Crude oil is naturally made material
found in the ground that contains a mixture of oxygen, carbon, hydrogen, nitrogen, sulfur,
metals, and salts. Since no two crude oils are exactly identical in composition and nature,
variation exists between batches of gasoline produced from different crude oil.
Furthermore, the refinery process varies among refineries, adding more variation to the
final products that result from their processes.l

Gasoline is produced in refineries and sent via pipeline or barge to various
terminals in cities throughout the US. depending on where that particular refinery sends
their gasoline. The gasoline used in this study was obtained from various cities in the
State of Maryland along with a sample from Virginia and Pennsylvania (see Table 1.1).
The refineries responsible for supplying gasoline to these states are located in Delaware,

Pennsylvania, New Jersey, and Texas.9

10

Table 1.1 Gasoline Information for 50% Comparisons (Left) and 75% (Right)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Station Name Station Name
Gasoline and Octane Gasoline and Octane
# Number Obtained From # Number Obtained From
1 Citgo 89 Randallstown, MD 1 Citgo 87 Wheaton, MD
2 Amoco 89 Wheaton, MD 2 Exxon 87 Rockville, MD
3 Sunoco 89 Wheaton, MD 3 Sunoco 89 Wheaton, MD
4 (m; 87 Glenwood, MD 4 Citgo 87 Manassas Park, VA
5 Exxon 89 Randallstown, MD 5 Exxtra Mart 87 Upper Marlboro, MD
6 Exxon 87 Rockville, MD 6 Mobil 87 Glenmont, MD
7 Exxtra Mart 87 Upper Marlboro, MD 7 Amoco 87 Randallstown, MD
8 Exxon 93 Gaithersburg, MD 8 Exxon 89 Randallstown, MD
9 Exxon 87 Bethesda, MD 9 Exxon 87 Gaitherserg, MD
10 Citgo 87 Manassas Park, VA 10 Exxon 93 Gaithersburg, MD
11 Amoco 87 Owings Mills, MD
12 Amoco 93 Harrisburg, PA
13 Exxon 87 Baltimore, MD
14 Mobil 87 Glenmont, MD
15 Amoco 87 Randallstown, MD
15 Amoco 87 Randallstown, MD
16 Exxon 87 Gaithersburg, MD

 

 

 

The first process involved in refining crude oil into gasoline is a separation
process in which crude oil is heated, vaporized, and partitioned according to boiling
point.10 Through such processes as distillation (separation based on boiling point),
crystallization (separation based on melting point and solubility), solvent extraction
(aromatic compounds removed), adsorption (removal of heavier hydrocarbons), and

absorption (purifies lighter hydrocarbons), the crude oil will separate into light, medium,
and heavy fractions and be sent to different areas of the refinery for further processing.
Because these separation processes vary among refineries, variation can be seen in
Products like gasoline that result from a blending of these materials. Moreover, because

the materials at this step are highly volatile, the low molecular weight components of

11

gasoline will contain majority of the differences when compared to gasoline produced by
a different refinery.

The next step involves a chemical conversion of the components. There are three
major processes involved: cracking, polymerization/alkylation, and reforming.ll Cracking
breaks large hydrocarbons into lighter ones and can involve the use of catalysts,
hydrogen, and pressure to aid the breakdown process. For example, fluid catalytic
cracking uses high temperature (up to 538° Celsius), low pressure, and a powdered
catalyst to produce gasoline. Hydrocracking uses lower temperatures, higher pressure,
hydrogen, and a catalyst to convert medium to heavy gas oils into gasoline and jet fuel.12

Polymerization and alkylation involve taking the byproducts of the cracking
process and combining them to make gasoline. Polymerization refers to the combination
of molecules to form higher molecular weight molecules. Alkylation results in production
of high-octane hydrocarbons that are blended with gasoline to improve octane rating
(reduce knocking) and involves combining olefins and paraffins. Octane rating refers to

how much gasoline can be compressed before it causes knocking, gasoline ignited due to

compression. '3

Reforming refers to the process of forming higher-octane hydrocarbons (mostly
aromatics) by rearranging molecular structures. The higher octane number is a result of
using higher reforming temperatures to convert paraffins to olefins. Dehydrocyclization,
dehydrogenation, hydrocracking, and isomerization (all with the aid of hydrogenation-
dehydrogenation catalysts) are the chemical processes occurring during reforming.
Dehydrocyclization involves the removal of hydrogen (dehydrogenation) from straight-

Chain paraffins to form rings in the hydrocarbon structure. Hydrocracking involves using

12

high pressure and temperature to break long chain paraffins into smaller chains in the
presence of hydrogen. Isomerization is the conversion of straight-chain paraffins to
branched chain paraffins. Isomerization is utilized to produCe some of the hydrocarbons
used in the alkylation process. Isomerization also forms products that can be heated under
high pressure to form products with a boiling point in the gasoline range.1
The next step involves treating the chemically processed fractions to remove
impurities such as oxygen, nitrogen, sulfur, water, dissolved metals, some alkenes, and
inorganic salts. The use of drying agents removes water from the processed fractions.
Sulfuric acid is used to remove alkenes, nitro and oxygenated compounds, tars, and
asphalt if present. Most of the contaminating sulfur compounds are converted to
hydrogen sulfide and mercaptans (R-SH) during the processing of crude oil to assist in
their removal. There are numerous methods available to remove sulfur compounds and
can involve the use of chemicals, catalysts, or adsorption onto materials."12
The next step involves blending the fractions to form different grades of gasoline.
What to blend is based on specifications such as octane level, where the gas will be used,
the season the gas is used (i.e. summer versus winter), vapor pressure ratings, and other
Specifications as determined by the government. Lastly, performance additives and dyes

are added, which can be used to distinguish the different types of gasoline from those of

Other gasoline stations.1

13

WW

Capillary Gas Chromatography

Gas chromatography (GC) is used to separate complex mixtures into their
individual components. The method relies on the principle of partition chromatography,
which refers to the competition between the mobile phase and stationary phase for the
substance to be separated, the analyte. In GC, the mobile phase is a gas and the stationary
phase is a liquid adsorbed onto an inert solid inside a column. The separation takes place
because the analyte partitions between the mobile and stationary phase to varying degrees
such that constituents of the analyte that are weakly retained in the stationary phase will
move quickly through the column.” The difference in mobility throughout the column
results in a difference in times of the components of the analyte leaving the column and
being detected. The time it takes for each component of the analyte to reach the detector
upon the analyte being injected into the GC is called its retention time. The output of a
GC is a chromatogram. It contains peaks, which represent the components of the analyte,
their abundance, and their retention time.

The instrument’s main components include an injection port, column, and
detector. The sample is introduced to the GC at the injection port, where it must be
Vaporized because, as mentioned previously, the mobile phase is a gas, and the analyte
must also be a gas also to be carried through the column. The mobile phase therefore is a
Carrier gas and thus must be chemically inactive so it does not react with the substance to
be Separated. In this study, helium was used as the mobile phase. Because the analyte is

VaDorized in the injection port, it must be set to a temperature slightly higher than the

14

boiling point of the least volatile component in the sample to ensure everything becomes
volatile before entering the column.

Depending on how much sample is necessary to be carried through the column
and reach the detector, the injector port can be operated in split or splitless mode. In this
study, Split mode was used to avoid flooding the column. Split mode only allows a
portion of the analyte injected to enter the column while the rest is removed via a valve in
the injection port. This mode is designated by a ratio input by the user. The splitless
method allows all of the analyte to enter the column.

The term capillary GC refers to the type of column used in the GC. Capillary
columns have a small diameter and can have their walls covered with liquid stationary
phase or a layer of support material and then stationary phase. Capillary columns are
normally found wound into coils in the GC. The column Sits in an oven which can be
programmed to increase in temperature over time. This is useful for instances in which
the analyte is composed of components with a wide boiling point range, such as gasoline.
Smaller components will vaporize at a lower temperature and consequently move through
the column quicker compared to heavier components.

Electron Ionization Mass Spectrometry

The mass spectrometer (MS) is used as a detector for the GC. The components of
the analyte that have been separated by GC will each pass into the MS to be detected and
a mass spectrum will be generated for each component. The mass spectrum is a series of
peaks representing the ionized molecule (in some cases) and fragmentations of it, plotted

as relative abundance (y-axis) versus mass-to-charge ratio (x-axis).

15

The mass spectrometer is composed of an ionization source, mass analyzer, ion
detector, and a computer to process the data. The ionization source, mass analyzer, and
ion detector all must be housed in a low pressure environment to prevent the ionized
fragments that are formed from colliding with other gas phase molecules on the way to
the ion detector.

In order for the analyte, that has been separated in the GC, to be detected, it must
first be ionized (acquire a net electric charge). The ionization source in electron
ionization is electrons coming from a heated filament, which are accelerated in a
direction perpendicular to the inlet of the gas phase analyte. These high-energy electrons
can pass close enough to the gaseous analyte molecules that they impart some of their
energy, causing the gaseous analyte to become ionized. There is usually enough energy
remaining from the interaction with the electrons that the ionized analyte molecule can
lose more electrons or undergo fragmentation, which causes numerous peaks in the mass
spectrum that can be interpreted to give structural information about the original
molecule.15

The ionized analyte molecule and its fragments are then sorted on the basis of
their mass-to-charge ratio by a quadrupole mass analyzer. The quadrupole mass analyzer

is comprised of four parallel rods and focusing lenses which aim the ions toward the rods.

16

Figure 2.1 Quadrupole Mass Spectrometer: Ion Source, Mass Analyzer, and Detector.l5

 

There is frequency modulated current (VI-f) and a voltage of direct current (Vdc)
being passed through the poles such that the two poles in the z-axis have a Vrf and Vdc
signal that is 180° out of phase with the Vrf and Vdc signal in the x-axis.15 The potential
at any point inside the quadrupole is constantly changing such that an ion within a desired
mass range undergoes stable oscillations as it moves through the poles. When the
voltages change between the quadrupoles, only ions with a mass that falls within the
stability region can pass through the poles. Those ions outside the selected mass range
will undergo unstable oscillations as a result of an increased force on the molecule,
causing them to hit the quadrupoles as they attempt to pass through to the detector.
Therefore, by changing the voltages passing through the rods, selected masses are
allowed to pass through the quadrupoles and be detected.

Once the ions pass through the poles they reach the ion detector. There are
various types of detectors available, but in general, their goal is to transform the sorted
ions into a usable signal. The detector used in this study was an electron multiplier. The

electron multiplier works by first emitting secondary particles as a result of ions hitting a

17

curved plate known as a dynode. There are several dynodes in the electron multiplier and
the secondary particles hit another dynode with enough energy causing electrons to be
released. The electrons then hit a series of dynodes causing more and more electrons to
be discharged. The amplification factor can reach up to 107 electrons produced each time
a dynode is struck.ls In the end, numerous electrons are created, resulting in a
computable current, whose signal is a representation of the ions leaving the quadrupole

mass analyzer. An analog-to-digital converter changes the analog voltages into digital

 

voltages that can be read by the computer and transformed into usable output. The
computer will display a mass spectrum for each component that has entered the MS and

usually contains a library of spectra to which to compare it for identification purposes.

18

 

Chapter Three: Forensic Methods of Accelerant Separation and

W

The detection of an ignitable liquid at a fire scene is a strong indicator for arson
when the liquid is not normally found at the fire scene. If an accelerant was used, it is
often highly evaporated by the time fire debris evidence is collected and must be
extracted out of the debris for analysis. There are several methods available to fire debris
chemists to separate and concentrate ignitable liquid residues from fire debris such as
steam distillation, headspace sampling, solvent extraction, and static adsorption/elution.
Each varies in preparation time, extraction efficiency for certain petroleum distillates, and
use in the field. Therefore, extraction and separation methods must be dependent upon
the type of accelerant suspected.

One of the most common extraction techniques used on a variety of petroleum
distillates due to its high sensitivity is the static adsorption/elution technique.7 Other
techniques utilized by fire debris chemists that are less sensitive to a wide hydrocarbon
range include headspace sampling and solvent extraction. Headspace sampling, solvent
extraction, and static adsorption/elution are discussed below and were chosen for the
study because they are commonly used techniques by the BATF.

Headspace Sampling

Headspace analysis requires that fire debris be contained within a sealed vessel
for a sufficient amount of time such that a representative sampling of what is contained
within the debris is in the vapor above the debris. Once this equilibrium is established, a
sample of the headspace is removed and analyzed. This method tends to favor the low

molecular weight, highly volatile components in the debris because they have a high

19

vapor pressure and vaporize readily. The high molecular weight components have a
lower vapor pressure due to an increased strength of the intermolecular forces between
the molecules comprising them and do not vaporize as easily.16 Headspace sampling is
ideal in situations where a highly volatile accelerant, such as gasoline, is present and a
high concentration of the low molecular weight components is sought.

To look at some of the less volatile components of an accelerant in a fire debris
sample, the vessel can be heated. Vapor pressure increases with temperature, so by
heating the container the less volatile molecules become more energetic and can escape
into the vapor phase easier. Nevertheless, both headspace and heated headspace methods
Show a higher concentration of the low molecular weight, highly volatile components in
the debris headspace, and thus should only be used when accelerants with a generally
high vapor pressure are suspected. Moreover, both methods are non-destructive, allow for
repeat analysis, have fairly reproducible results, and are commonly used screening
techniques.

Solvent Extraction

Since all vapor concentration methods tend to under represent the less volatile,
heavier components, solvent extraction is a useful technique when heavier petroleum
ignitable liquids are involved. Solvent extraction is primarily used in instances when a
nonporous sample has been collected to determine the presence of an accelerant. In
solvent extraction, the debris is washed with a solvent such as pentane or carbon
disulfide, and the solvent extract concentrated by evaporating to a small volume and
analyzed by a chromatographic technique. Since solvent extraction can be destructive, the

sample may not be useful for further analysis. Therefore, this method should be done last

20

in an analysis scheme and some of the original material should be preserved for other
analytical techniques.

A problem with solvent extraction is that other contaminants can also be
extracted, interfering with pattern recognition of an ignitable liquid. Also, during the
extract concentration step, some ignitable liquid can also evaporate. Nevertheless, solvent
extraction extracts heavy and light components with the same efficiency so it is useful to
differentiate between heavy petroleum distillates.l7
Static Adsorption/Elation

Static adsorption/elution using activated charcoal is one of the most widely used
separation techniques due to its increased sensitivity over the other methods, non-
destructive nature, and ease of preparation. Activated charcoal is created by grinding up
charcoal, washing it with carbon disulfide, and drying and activating (at 50° and 120° C)
the charcoal under vacuum.7 Static adsorption/elution is a technique in which fire debris
is sealed in an airtight container with an activated charcoal strip suspended from the lid.
The container is then heated in an oven for a sufficient amount of time (from 4-24 hours).
It is important not to overheat the container because the less volatile components tend to
displace the more volatile components from the charcoal strip over time. The strip is then
removed and washed with a solvent, usually carbon disulfide, and the solvent extract
injected into the GC/MS. The typical chromatogram generated by this method will be
shifted to the right compared to headspace sampling chromatograms due to more
complete recovery of the heavy components. During incubation, replacement of high
volatiles occurs by heavier, less volatile components over time, so incubation time can be

varied depending on the results expected. ‘7

21

Chapter Four: Development of Experiment Parameters

Instrumental Conditions

All analyses were performed on an Agilent 6890 GC' coupled to a 5973 MS
operated at 20/1 split ratio. The GC/MS was equipped with a 60m DB-l column with a
.25mm internal diameter and 1pm film thickness. The oven temperature was programmed
from 35°C for 2 minutes, followed by a temperature ramp of 5°C/minute to 110°C, then
to 250°C at 12° C/minute. The scanning mode for the MS was 29 to 200 amu’s.

The instrument also had an autosampler that could be used to automatically inject
liquid samples. This was useful for solvent extraction and static adsorption/elution
techniques in which solvent extracts needed to be analyzed.

Software Considerations

Once the instrument parameters were designed, software had to be employed that
could identify components in gasoline so that ratios of components could be analyzed in
the comparison. Hewlett Packard’s Chemstation Target Compound Program came with
the GC/MS and was useful for this study. Since this project was performed in
conjunction with the BATF, the software had already been designed to pick the following
components that elute in the five to twenty-nine minute region of a typical gasoline
chromatogram nm on a 60m column.

Table 4.1 Components Chosen by Software

 

 

 

 

 

 

 

 

 

 

Peak # Retention Time (min) Compound Name
1 5.10 isobutane
2 5.50 n-butane
3 5.62 butene
4 6.37 2-pentene
5 6.70 isopentane
6 6.98 1-pentene
7 7.27 n-pentane
8 7.62 (Z)-2-pentene

 

 

 

 

22

Peak # Retention Time min nd Name
9 7.78
10 8.23
11 8.77
12 9.04
13 9.23
14 9.72
15 9.91
16 10.33 n-hexane
17 10.47 2-hexene
18 10.51
19 10.67
20 10.78
21 11.00
22 11.38
23 12.19
24 12.69
25 12.86
26 13.25
27 13.62
28 13.81 isooctane
29 14.03
30 14.17
31 14.69
32 15.19
33 15.26
34 15.50
35 15.62
36 15.98
37 16.37
38 16.58 toluene
39 16.84
40 16.92
41 16.99
42 17.12
43 17.21
44 17.24
45 17.38
46 17.53
47 17.65

1-ethyl-3-

48 17.88
49 17.93
50 17.99
51 18.13

52 18.32

53 18.53

 

23

Peak # Retention Time min
54 20.01
55 20.27
56 21.82
57 22.55
58 22.70
59 25.39

60 25.86

61 25.92
62 26.30
63 26.49
64 27.08
65 27.98
66 28.68
67 28.93

 

To set up the software to pick these compounds, their name, retention time, and
target ions were entered to be used for selection criteria. The target ions were the base
peak for each component, and their abundance in the mass spectrum was normalized to
100 percent. In instances where more than one component in gasoline had the same base
peak, the abundance data of other distinguishing peaks in the mass spectrum was entered
into the program. With this information, as each sample was run through the program,
data was generated containing the compound name, retention time of the compound, and
the target response of the base peak. The target response refers to a number representative
of the response generated by how many target ions are detected by the ion detector.

It was found that a few changes had to be made to the program such as updating
the retention times and entering in qualifier ions (other MS peaks in addition to the base
peak) to ensure the correct peaks were chosen for purposes of this study. With the mass
spectral information, the mass spectrum of each compound chosen was checked against a
library of known mass spectra to ensure the peak chosen by the program was the correct

gasoline component. Once this was established, the target ion responses were imported

24

into a Microsoft Excel® spreadsheet that was designed to compute ratios of sequential
components for three runs, average them, and determine standard deviation and percent
standard deviation. This data was used to analyze replicate runs to ensure good
reproducibility within a sample.

Selection of an Extraction/Concentration Method

Before any gasoline comparisons could be done, four common extraction methods
used by the BATF were tested to determine which yielded the best reproducibility:
headspace, heated headspace, solvent extraction, and static adsorption/elution. To do this,
a sample of 89-octane gasoline from Citgo located in Randallstown, MD was used.

First, an appropriate injection volume had to be established that would yield
sufficient information for the headspace methods. Using a sample containing three drops
of unevaporated gasoline on a Kimwipe and sealed in an airtight, quart-sized paint can, it
was found that 2ml of headspace was an appropriate injection volume.

Next, it had to be decided if the same syringe could be used for replicate runs of
the same sample for manual headspace injections. Following a headspace injection of
unevaporated gasoline, another injection consisting only of air was made using the same
syringe. The resulting chromatogram displayed carryover, so a different syringe was
used for every injection, except when a blank was performed; the same syringe used for a
blank was then used for the first injection following it and then discarded.

To evaluate the headspace technique, 50ul of unevaporated gasoline was placed
on a Kimwipe inside a quart-sized paint can and immediately sealed. This was done three
times such that there were three sealed cans containing SOul each of the same gasoline. A

can containing only a Kimwipe was also sealed to see if the Kimwipe contributed any

25

chromatographic peaks. The cans were allowed to sit undisturbed for approximately 24
hours. Afier 24 hours, a hole was placed in the lid of the first can, covered with Scotch®
tape, and three (2ml) headspace draws were injected into the'GC. This was repeated for
the other two cans such that a total of nine injections (3 per can) had been done upon
completion. One injection was made from the can that contained only a Kimwipe and it
was found that the Kimwipe alone did not contribute to the chromatogram.

The next method evaluated was heated headspace. The cans were prepared in the
same manner as for the headspace at room temperature, but were placed in a 65° C oven
for fifteen minutes prior to injection and the hole was placed in the lid and covered with
tape prior to heating. For both headspace methods, blanks of room air were injected
between triplicates.

For solvent extraction, 50ul of gasoline was placed on a Kimwipe and sealed in a
can as mentioned above, but after 24hrs, the Kimwipe was rinsed with 5ml of pentane in
the can and the solution poured into a vial. A pentane blank consisting of pentane in a
vial was included in the experiment to make sure the pentane was not contaminated.
Also, a control involving pentane rinsed in a can and then poured into a vial was included
to make sure the method itself was yielding accurate results. The sample vials, the blank,
and the control were allowed to evaporate under the hood to approximately lml and were
then run using the autosampler which was set up to make three injections per vial and one
for the blank.

Lastly, static adsorption/elution was performed using 25 ul of gasoline on a

Kimwipe sealed in a can and allowed to sit overnight. A system blank of just a Kimwipe

sealed in a can was included in the analysis. The following day, the lids from the four

26

cans (3 samples, 1 blank) were removed and a paper clip containing a 1cm long charcoal
strip (Abrayco Lab Inc.) was suspended from each lid using a magnet placed on the
outside of the lid. The cans containing the suspended charcoal strips were then placed in
a 65° C oven for 16 hours. After 16 hours, the charcoal strips were removed and each
placed in separate vials to which 250p] of carbon disulfide was added. The solvent
extracts were then run in triplicate (except for the system blank) using the autosampler.

The percent standard deviation of the ratios (nine runs) generated from each
method was used to assess reproducibility. This was done by calculating the standard
deviation and percent standard deviation of sequential ratios generated using only
compounds eluting in the nine to twenty-nine minute region for each method with
triplicate runs. This time region was chosen because it was found by the BATF to contain
ratios of components that were reproducible and useful in liquid-to-liquid gasoline
comparisons. Headspace and heated headspace had similar reproducibility and displayed
all the peaks of interest based on the peak selection method (from 3-methylpentane to the
methylnaphthalenes). Static adsorption/elution also had good reproducibility but lacked
the first three peaks for which the program looks: 3-methylpentane, 2methyl-l-pentene,
and n-hexane (present in 50% evaporated gasoline). Solvent extraction was found not to
be useful because twenty of the thirty-four components being evaluated were absent from
the chromatogram because they had presumably evaporated during the concentration step
of the technique.

Headspace and static adsorption/elution were further compared by introducing

50% evaporated gasoline spiked on debris to see if the presence of the wood substrate

would interfere with reproducibility of the method. Gasoline 50% evaporated (SOul) was

27

placed on a piece of charred pine and sealed in a can and allowed to sit for 24 hours.
Then headspace and static adsorption/elution techniques were used to extract the gasoline
and the data was run through the Target Compound Program and ratios generated using
the Excel worksheet. This was repeated using the same source gasoline unevaporated.
The averages of the ratios from the unevaporated gasoline were compared to the 50%
evaporated gasoline extract ratio averages using standard deviation and percent standard
deviation. The headspace technique had lower percent standard deviations between the
evaporated and unevaporated ratios compared to static adsorption/elution. Since static
adsorption/elution yields a chromatogram shifted to the right compared to the headspace
technique, it was found that for static adsorption/elution many of the early eluting
components in gasoline were not reproducibly present. Therefore, for the 50%
comparison part of the study, headspace was chosen because it picked up the early
eluting components in the chromatogram and was the most reproducible regardless if the
wood substrate was present.

A similar comparison was done using 50ul of 75% evaporated and unevaporated
gasoline from the same source placed on a Kimwipe and sealed in separate cans. The
ratio averages obtained from both cans were compared using the headspace and static
adsorption/elution methods. Since the chromatogram for 75% evaporated gasoline is
shifted to the right compared to unevaporated gasoline (see Appendix 111), static
adsorption/elution was an ideal method. Headspace was not useful for looking at
components in 75% evaporated gasoline because the target response for the heavier

components was very low to nonexistent. Moreover, 75% evaporated gasoline contains

28

an abundance of the heavier compounds compared to 50% evaporated gasoline, so static
adsorption/elution was chosen for the 75% comparisons.
Selection of Substrates

Two substrate types were selected to compare for the simulation of fire debris:
pine and nylon carpet. Both are common substrates found in fire scenes that ignitable
liquids can be extracted from. The carpet used was obtained from a local carpet store and
was cut into a 6cm x 6cm square. The pine was purchased from Home Depot in
Rockville, MD and was cut into the same dimensions as the carpet. To simulate fire
debris, both materials were burned on all sides under a hood with a 14.1L propane torch
and then allowed to catch flame and dropped in a quart-sized paint can. After the
substrates appeared charred, the lids were placed on the cans and sealed. After
approximately 2 hours, a 2m] sample of each material’s headspace was run on GC/MS
and the results evaluated with the Target Compound software. This process was repeated
several times to see if the pyrolysis products were consistent from run to run.

The pine was selected to be used in the debris experiments because its pyrolysis
products did not interfere with the detection of gasoline (see Appendix IV) and were
fairly consistent among each run.

Determination of Amount of Gasoline to Use

Once the instrument parameters were established, the extraction methods and
substrate types evaluated, and comparison software optimized, a determination of an
appropriate amount of gasoline to use for each part of the study had to be ascertained.
Since 50p] of gasoline on Kimwipes gave decent abundance values in the determination

of an extraction method, that amount was used for all samples consisting of unevaporated

29

gasoline on a Kimwipe for the 50% comparisons. For the 50% evaporated gasoline that
was going to be placed on charred pine, 501.11 and lOOul amounts were compared by
placing each on the charred pine in a can and sealing for 24 hours. After 24 hours, a
sample of the headspace of each can was run and the abundance values compared. Both
50p] and IOOul gave strong abundances, but the can with IOOul placed on the pine gave
stronger abundance values and was the amount chosen to use on all simulated debris.

The comparison of the headspace of 75% evaporated gasoline on a Kimwipe to
the headspace of unevaporated gasoline on a Kimwipe required a smaller amount of
gasoline because static adsorption/elution is a more sensitive extraction technique.
Amounts of Sul, lOul, and 20ul for both the 75% evaporated and unevaporated gasoline
were evaluated. Using 10p] of evaporated and unevaporated gasoline gave sufficient
abundance data such that a comparison could be made, and this volume was used for this
part of the study.
Experimental Setup

The gasoline used for the experiment was collected from the pump of several
gasoline stations and stored in Quorpack® 4oz bottles with Teflon®-lined lids to prevent
evaporation. For each sample of gasoline collected, an aliquot was removed for the
unevaporated samples and the rest was evaporated by 50% and 75% of the original
volume underneath a hood. Gasoline from the same source refers to gasoline collected
from the same station and derived from the same bottle for the study. Sixteen gasoline
samples were used for the 50% comparisons and 10 gasoline samples were used for the

75% comparisons.

3O

A. 50% Comparisons

Three cans were set up for each gasoline sample so that nine runs total, three from
each can, would be made for both the evaporated and unevaporated gasoline (six cans
total per gasoline). This number of replicates was chosen so reproducibility and variation
from can to can could be assessed and to provide enough replicate information such that
outlying data could be discarded if necessary. To represent simulated fire debris where
gasoline was used as an accelerant, first a 6x6cm piece of pine was charred and sealed in
a paint can. Thirty minutes later, IOOul of 50% evaporated gasoline was injected through
a hole (.64cm in diameter) in the lid and covered with tape. This was repeated with the
two other cans, and all three cans were allowed to sit for 24 hours before headspace
samples were run on GC/MS. Before any samples were injected and between each can, a
blank (2ml of air) was run.

Next, three cans containing one Kimwipe each were spiked with SOul of
unevaporated gasoline from the same source as the 50% evaporated gasoline and
immediately sealed. These cans were also allowed to sit 24 hours before any headspace
samples were run on GC/MS.

B. 75% Comparisons

For this part of the study, only six runs per sample were done (four cans total per
gasoline sample). The 50% comparisons demonstrated that two cans per sample were just
as effective as three cans, so 10ul of 75% evaporated gasoline was placed on Kimwipes
and sealed in two separate cans. This was repeated using 10p] of unevaporated gasoline
from the same source. The four cans were allowed to sit for 24 hours and then a 1cm long

charcoal strip was attached to a paper clip and suspended from the inside of the lids using

31

a magnet on the outside. All cans were then placed in a 65° C oven for 16 hours. A
system blank containing a kimwipe and charcoal strip, but no gasoline, was included
anytime cans were placed in the oven. The system blank demonstrated that, if properly
sealed, no cross-contamination occurred between cans in the oven.

Afier all cans were removed from the oven, they were allowed to cool for 30-45
minutes before the charcoal strips were removed and placed into separate glass
autosampler vials using tweezers that were rinsed in carbon disulfide before touching
each charcoal strip. To each vial, 250ul of carbon disulfide was added and the vials were
run (three injections per vial) using the autosampler.

Comparison Mechanics

For each run, the data was run through the target compound program, which
identified the peaks of interest, and provided retention time and the target response for all
the compounds listed in Table 4.1. The target response information was copied into the
Excel templates which were already set up to calculate sequential ratios, average them,
and determine standard deviation. The process resulted in triplicate values for ratios,
averages, and standard deviations per can for the 50% and 75% data.

In selecting valuable ratios for comparison, the ratio must remain similar when
gasoline is compared to gasoline from the same source but evaporated. In the 50%
comparisons, unevaporated gasoline ratios were compared to the ratios obtained from the
50% evaporated gasoline extracted from charred pine. This was done for sixteen different
gasoline sets. The same was done for the 75% evaporated gasoline and unevaporated
gasoline from the same source using ten different gasoline samples. However, the 75%

evaporated gasoline was not extracted from pine. To clarify, for each gasoline sample

32

there were two values: an average (from all runs) of each ratios’ values for evaporated
and an average for the unevaporated gasoline ratio values. The average was of all the
values obtained for a particular ratio after nine runs for the 50% comparisons and after
six runs for the 75% comparisons. The standard deviation and percent standard deviation
were calculated for each ratio and used to compare the unevaporated gasoline to the
evaporated gasoline. Percent standard deviations less than five percent were considered
similar.

Table 4.2 Example of How Similar Ratios Were Determined

Gasoline 1- Percent
75% Gasoline 1 Standard Standard Ratios
Ratio evaporated unevaporated Average Deviation Deviation Similar?
A 0.639 0.630 0.635 0.006 0.997 Yes
B 0.133 0.136 0.134 0.003 1.881 Yes

For all the ratios that had a less than five percent standard deviation when
unevaporated gasoline averages were compared to evaporated averages, their
reproducibility among the three runs per can was assessed. Random error is evident in
every scientific measurement because there is some error in the reproducibility of the
instrument used. Reproducibility was determined by how close the values for all runs of a
sample were (see Appendix VH1). If, for a particular ratio, the values remained under five
percent standard deviation when evaporated and unevaporated gasoline from the same
source were compared and the ratio was reproducible within a sample, it was considered
a useful ratio.

Since a goal of the study was to be able to eliminate unevaporated gasolines as
possible sources of a sample of evaporated gasoline, another comparison was performed.

For all the ratios that were found to be reproducible and remain similar between

33

unevaporated and evaporated gasoline from the same source, the evaporated ratio
averages from one gasoline were compared to all other unevaporated gasoline’s same
ratio averages. The purpose of this comparison was to see if the standard deviations of the
ratios considered useful were greater when the comparison involved a different source
gasoline (see Appendix Table 1.2). If they were, the ratios were useful to discriminate
among different gasolines.

In this study, pattern recognition based on chromatograms was not used to

compare the gasoline samples.

34

Chapter Five: Results and Conclusions, Discussion, and Further Research

Results and Conclusions
A. 50% Comparisons

The sixteen gasolines used in the 50% comparisons were readily distinguished
using the six ratios outlined in Appendix Figure 5.1. Ratio one corresponds to the target
response of the target ion in methylcyclohexane: dimethylcyclopentane, ratio two to 2,4-
dimethylhexane: 2,5-dimethylhexane, ratio three to 1,2,4-trimethylcyclopentane: 2,4-
dimethylhexane, ratio four to 2,3,4-trimethylpentane: 1,2,4-trimethylcyclopentane, ratio
five to dimethylcyclohexane: 1,2,4-trimethylcyclopentane, and ratio six to 1,4
dimethylcyclohexane: 1,2-dimethylcyclohexane.

Furthermore, the headspace of unevaporated gasoline and 50% evaporated
gasoline (extracted from debris) from the same source contain similar ratios of their
components and these six ratios differ among gasoline from different sources. Appendix
VI shows multiple comparisons of all sixteen gasolines for each of the six ratios and they
clearly are different from one another when all six ratios are taken into account. It must
be noted though that in order to have an effective comparison, all six ratios must be used.
For example, gasolines #1 and #2 are very similar for all ratios except ratio four based
upon the comparison graphs, which is very interesting considering they are from stations
over 50 miles away from each other. Nevertheless, by comparing gasoline #1, 50%
evaporated extract, to gasoline #2’s values for all ratios (Appendix Table 1.2), the percent
standard deviations are all higher than a comparison of only gasoline #1 ’s values. For the
comparison of gasoline #1, 50% evaporated to the same unevaporated gasoline, the

values for ratios four and five were the highest standard deviations of all the ratios, but

35

still were under the five percent cut-off. It must also be noted that there are instances,
where for one of the six ratios, two samples with different sources have a lower standard
deviation than five percent when compared to two samples from the same source.
However, most ratios will be significantly higher than five percent compared to two
samples from the same source and thus all ratios must be considered when determining
whether a sample can be eliminated as originating from the same source.

Appendix Figures 6.7-6.12 demonstrate that when using the data from an
evaporated gasoline sample and comparing it to all other gasolines unevaporated, only
one unevaporated sample remains similar when compared among all ratios. This
demonstrates that it is possible to rule out unevaporated samples as originating from the
same source as a 50% evaporated sample.

Pine was chosen for the substrate because the pyrolysis products produced (see
Appendix IV) do not interfere with the compounds used in the comparison ratios.
However, once the gasoline was added, it was not known whether the pine would absorb
some of the gasoline, interfering with the reproducibility of the comparisons.
Nevertheless, the data demonstrates the presence of the wood substrate does not interfere
in the comparisons. The sources of five samples selected from the sixteen gasoline
samples were correctly identified in a blind study.

B. 75% Comparisons

The ten gasolines used in this study were distinguished using the four ratios
outlined in Appendix Figure 5.2. Ratio one corresponds to the target response of the
target ion in 2,4-dimethylhexane: 2,5-dimethylhexane, ratio two to 1,2,4-

trimethylcyclopentane: 2,4-dimethylhexane, ratio three to methylindane: 1,2,3,5-

36

tetramethylbenzene, and ratio four to l-methylnaphthalene: 2-methylnaphthalene. Ratios
one and two correspond to ratios two and three of the 50% comparisons, indicating it may
be possible to use those two ratios for cases involving gasoline up to 75% evaporated.

In this study, 75% evaporated gasoline was found to contain similar values to the
same unevaporated gasoline using the four ratios mentioned above and clean substrates
(see Appendix VII). Appendix Figure 7.5 clearly demonstrates that all ten gasolines are
different using the data obtained from the unevaporated gasoline samples for the four
ratios. Addition of the data obtained using the 75% evaporated gasoline made it easy to
eliminate unevaporated samples as originating from the same source as the 75%
evaporated sample. Appendix Figures 7.6-7.9 compare 75% evaporated gasoline to all ten
unevaporated gasolines for each ratio, and only gasoline from the same source show
consistently similar data for all ratios. As mentioned previously, the comparison is only
valid when all four ratios are used because as Appendix Tables 2.2-2.5 demonstrate, there
are instances where, for a particular ratio, a sample from a different source may have a
percent standard deviation under five percent. However, inspection of the percent
standard deviation information for a comparison of unevaporated and 75% evaporated
gasoline from the same source shows it will be the only comparison in which all ratios
have a percent standard deviation under five percent. The sources of three samples
selected from the ten gasoline samples were correctly identified in a blind study.
Discussion and Further Research

The reproducibility for the 75% values was very good as evidenced by the
example of reproducibility shown in Appendix VIII. This is probably due to using the

autosampler versus manual injections, which were used in the 50% comparisons. Using

37

an autosampler automates the process and removes inherent human error present when
making manual injections. Also, because clean substrates were used in the 75%
comparisons, addition of a complex substrate like wood may interfere with how good the
reproducibility of the data is or the ability to do the comparisons at all. The 75%
evaporated gasoline was not extracted from debris because it was only an attempt to see
if such comparisons could be done on highly evaporated gasoline. Further research can
examine what happens when 75% evaporated gasoline is extracted from debris. The
method looks most promising for cases involving softwoods such as pine because its
presence in debris does not interfere with the four ratios found to be useful for such a
comparison. Also, because ratio three and four of the 75% comparisons involve
compounds eluting in the late region of the chromatogram, attempts to do such a
comparison using higher than 75% evaporated gasoline may prove feasible.

. The results of the comparisons in this study demonstrate that for cases involving
gasoline as an accelerant that has been evaporated up to 50% and extracted from pine, it
is possible to eliminate gasoline comparisons as originating from the same source. The
results of the 75% comparisons suggest it may be possible to apply the same type of
comparison to cases involving 75% evaporated gasoline. Because there were two ratios
that were applicable to both comparisons, it may be possible to use them in cases
involving gasoline evaporated up to 75% and extracted from pine.

The ability to carry out the comparisons using gasoline from a small distribution
area is very beneficial. The samples used in this study were obtained from places that
could possibly have received gasoline from the same refinery and/or production batch, or

terminal. Because they were distinguished from one another, doing larger scale

38

comparisons involving gasoline from outside the distribution area of one another should
result in even more levels of comparison and the ability to possibly classify gasoline.

It was not known at the beginning of this research if this would be an attempt to
distinguish gasolines that are not technically from different sources because specific
distribution information was not available. Nevertheless, the fact that the gasoline
samples were distinguished from one another suggests that even without such
information, a comparison can still be made, lending more credibility to evidence of this
type. Not knowing the distribution information may be the reason why only a few ratios
were useful in the comparison, but the combination of new shipments of gasoline, storage
conditions of the gasoline, and the gasoline residue still in the tank add to detecting the
types of differences found in this study.

In this study, all gasoline samples were collected in the summer and thus do not
have as many light end components because they cause vapor lock in car engines. Future
research should attempt to incorporate samples from winter as well, because winter
gasoline has more light end compounds, and see how such changes affect the ratios
useful for gasoline comparisons.

Before application of this method to actual arson cases, a ‘controlled’ arson
scenario should be carried out where a certain amount of gasoline is added to the pine and
then burned until the desired evaporation state is obtained. The evaporation amount of the
gasoline should then be determined and the components remaining identified to see if
they change under combustion or what effect combustion has at all on using the specified
ratios. If the comparison method is still found to be useful, it should then be applied to

actual arson cases.

39

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

vaé Nome NmmN KEN wFN mmNN m can”.
wand rmwd mooN NmmN NNvN mmvN m 03mm
wmnér mvN.vF 53.3” N®N.®m Nu F .N oNF .N v Ozmm
2nd mde 02.0 mmwd Nnd mmnd m can”.
mmod meo 30.0 30.0 8.0 womd N can”.
vmom NONm mmd Nvmd mnmm mom.m _. 03mm
.mflfi and: .mm>e «no: .mm>e «no:
3cm m 0.50qu a 05.0950 $3 a 0c=owa0 a 05.0m50 30m N 250099 N o:=owa0
owoN mmoN wwoé wwoé mmnd mknd 0 03mm
3mm m—Nd mmNV NmN.v $90 onwé m away.
Nvd nwmd mum. 5. won. 3 med Em v 05mm
Rind rmvd End 086 mwmd mwmd m 05mm
Nkwd mnod and mud N550 and N 05mm
warm meow oNvd mnvé N_.N.: m 5. Z. w ovum
.mm>o «no: dye «no: .mm>o «we:
{can u eczemamu @ 05—Oma0 .30.» m Orion—«0 m 25.300 {com v 057.550 v oc=oau0
mmod awkd EON :vN mNmN NmmN m can”.
omNm mama.“ wwwN othN momN mmvN m 251
Nmm and medF none? mmmNr 0mm. E. v 03mm
mde ommd ode Nde vomd nomd m can”.
086 med 00nd m {.0 Nmnd 35° N 03mm
wwwN 09L. mmné need .256 Bad F 0.3mm
Ilium; use: .mm>e use: .mm>o «so:
.xbm n eczowuw n ec=ow00 30m N o:=0aa@ N o:__oau0 $0». w oc=oua0 r oc=ooa0

 

 

 

8388983887: 2-; 85330 5 3 89m 5 8:38 85> 2 2.3

meat—«@800 $3 new Sun a 5259?.

40

AwoumHoqgocbufiozv 0.-..“ 35.800 8.. 0-. mosey. 8.. 35030 mo:.m> €035.50 ... 63m...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3:.” mme 0 30¢
vamN anN m 2.0m.
NVme mmmer v 2.01
med mwmd n 0:0”.
wood 050.0 N 0:0”.
N36 mmmd .. 2.0m
.mm>0 .00:
{can 09 05.0000 0.. 05.0000
mmve mvmé oNFN mONN mum... mmmé 0 030m.
mmNm mmvd wNmN erN nmwN N KN m 030m.
Nmm.m mvmm med wad mmNe ané v 0:0”.
mmvd Nomd Nwmd wand 90.0 thd m 2.0m
mmwd mmwd mmmd 08.0 Pvmd mmmd N 30¢
0.56 rm...“ mmrfi 3. 3 tvé Nam.“q w 050m.
. «>0 «no: .mm>e «no: .mm>e .00:
{com m... 05.0000 0.. 05.0000 $00 3 05.0000 0.. 05.0000 $00 2 05.0000 2 05.0000
3mm...” mm god 806 mmmN NnmN 0 020m.
3 3 vmmd mN.o mme NrmN mmmN m 0:0”.
50.9 m2. : 83 «SN 38 on; a
mNNd mNNd mowd :90 mwvd mmvd m 030w.
3.0.? wvoé vmmd wNoe mwhd and N 0:01
m omwN Novd reed vnd 3.5m F 030m
immm. «no: .mm>o «no: .mm>e ~00:
$00 Nr 05.0000 NF 05.0000 $00.. 3 05.0000 5 05.0000 $00 0.. 05.0000 2 05.0000

 

 

 

 

41

Table 1.2 Comparison of 50% Evaporated Gasoline #1 (Debris Extract) to Gasolines

#1-16 Neat; Gas=Abbreviation for Gasoline

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard %std dev =
Gas 1 50% Gas 1 neat Averagi deviation (stdevlanfii 00
Ratio 1 4.047 3.904 3.976 0.101 2.543
Ratio 2 0.762 0.764 0.763 0.001 0.185
Ratio 3 0.304 0.307 0.306 0.002 0.694
Ratio 4 12.335 11.596 11.966 0.523 4.367
Ratio 5 2.606 2.458 2.532 0.105 4.133
Ratio 6 2.325 2.332 2.329 0.005 0.213
Standard
Gas 1 50% Gas 2 neat Average deviation %std dev
Ratio 1 4.047 4.665 4.356 0.437 10.032
Ratio 2 0.762 0.713 0.738 0.035 4.698
Ratio 3 0.304 0.282 0.293 0.016 5.309
Ratio 4 12.335 14.787 13.561 1.734 12.785
Ratio 5 2.606 2.779 2.693 0.122 4.543
Ratio 6 2.325 2.417 2.371 0.065 2.744
Standard
Gas 1 50% Gas 3 neat Average deviation %std dev
Ratio 1 4.047 7.756 5.902 2.623 44.441
Ratio 2 0.762 0.948 0.855 0.132 15.383
Ratio 3 0.304 0.336 0.320 0.023 7.071
Ratio 4 12.335 9.742 11.039 1.834 16.610
Ratio 5 2.606 5.339 3.973 1.933 48.648
Ratio 6 2.325 0.719 1.522 1.136 74.613
Standard
Gas 1 50% Gas 4 neat Avegqe deviation %std dev
Ratio 1 4.047 11.313 7.680 5.138 66.899
Ratio 2 0.762 0.780 0.771 0.013 1.651
Ratio 3 0.304 0.389 0.347 0.060 17.346
Ratio 4 12.335 9.700 11.018 1.863 16.912
Ratio 5 2.606 6.538 4.572 2.780 60.812
Ratio 6 2.325 0.773 1.549 1.097 70.848
Standard
Gas 1 50% Gas 5 neat Average deviation %std dev
Ratio 1 4.047 6.475 5.261 1.717 32.634
Ratio 2 0.762 0.760 0.761 0.001 0.186
Ratio 3 0.304 0.336 0.320 0.023 7.071
Ratio 4 12.335 11.703 12.019 0.447 3.718
Ratio 5 2.606 4.292 3.449 1.192 34.566
Ratio 6 2.325 1.088 1.707 0.875 51.256

 

42

 

Table 1.2 Continued: Comparison of 50% Evaporated Gasoline #1 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ea. 1 Gas 5 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 5.083 4.565 0.733 16.047
Ratio 2 0.762 0.673 0.718 0.063 8.771
Ratio 3 0.304 0.481 0.393 0.125 31.887
Ratio 4 12.335 8.987 10.661 2.367 22.206
Ratio 5 2.606 3.213 2.910 0.429 14.752
Ratio 6 2.325 2.039 2.182 0.202 9.268
Gas 1 Gas 7 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 5.369 4.708 0.935 19.855
Ratio 2 0.762 0.968 0.865 0.146 16.840
Ratio 3 0.304 0.733 0.519 0.303 58.505
Ratio 4 12.335 2.126 7.231 7.219 99.839
Ratio 5 2.606 2.435 2.521 0.121 4.797
Ratio 6 2.325 2.253 2.289 0.051 2.224
Gas 1 Gas 8 Standard
50% neat Averag deviation %std dev
Ratio 1 4.047 3.842 3.945 0.145 3.675
Ratio 2 0.762 0.634 0.698 0.091 12.967
Ratio 3 0.304 0.135 0.220 0.120 54.442
Ratio 4 12.335 36.262 24.299 16.919 , 69.630
Ratio 5 2.606 2.557 2.582 0.035 1.342
Ratio 6 2.325 2.747 2.536 0.298 11.767
Gas 1 Gas 9 Standard
50% neat Avegge deviation %std dev
Ratio 1 4.047 5.702 4.875 1.170 24.008
Ratio 2 0.762 0.653 0.708 0.077 10.894
Ratio 3 0.304 0.329 0.317 0.018 5.585
Ratio 4 12.335 14.245 13.290 1.351 10.162
Ratio 5 2.606 3.851 3.229 0.880 27.268
Ratio 6 2.325 1.867 2.096 0.324 15.451
Gas 1 Gas 10 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 3.714 3.881 0.235 6.068
Ratio 2 0.762 0.790 0.776 0.020 2.551
Jatio 3 0.304 0.489 0.397 0.131 32.992
Jatio 4 12.335 6.136 9.236 4.383 47.462
Jatio 5 2.606 2.359 2.483 0.175 7.035
Ratio 6 2.325 2.572 2.449 0.175 7.133

 

43

 

Table 1.2 Continued: Comparison of 50% Evaporated Gasoline #1 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 1 Gas 11 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 8.461 6.254 3.121 49.907
Ratio 2 0.762 1.021 0.892 0.183 20.543
Ratio 3 0.304 0.871 0.588 0.401 68.243
Ratio 4 12.335 2.482 7.409 6.967 94.042
Ratio 5 2.606 6.295 4.451 2.609 58.612
Ratio 6 2.325 0.689 1.507 1.157 76.764
Gas 1 Gas 12 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 2.896 3.472 0.814 23.445
Ratio 2 0.762 1.048 0.905 0.202 22.346
Ratio 3 0.304 0.228 0.266 0.054 20.203
Ratio 4 12.335 11.705 12.020 0.445 3.706
Ratio 5 2.606 3.994 3.300 0.981 29.741
Ratio 6 2.325 3.900 3.113 1.114 35.781
Gas 1 Gas 13 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 4.392 4.220 0.244 5.782
Ratio 2 0.762 0.953 0.858 0.135 15.750
Ratio 3 0.304 0.927 0.616 0.441 71.572
Ratio 4 12.335 1.217 6.776 7.862 - 116.021
Ratio 5 2.606 2.712 2.659 0.075 2.819
Ratio 6 2.325 1.593 1.959 0.518 26.422
Gas 1 Gas 14 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 4.114 4.081 0.047 1.161
Ratio 2 0.762 0.838 0.800 0.054 6.718
Ratio 3 0.304 0.388 0.346 0.059 17.167
Ratio 4 12.335 6.882 9.609 3.856 40.130
Ratio 5 2.606 2.521 2.564 0.060 2.345
Ratio 6 2.325 2.209 2.267 0.082 3.618
Gas 1 Gas 15 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 5.910 4.979 1.317 26.461
Jatio 2 0.762 0.899 0.831 0.097 1 1.664
Jatio 3 0.304 0.502 0.403 0.140 34.741
Ratio 4 12.335 5.545 8.940 4.801 53.705
iatio 5 2.606 3.459 3.033 0.603 19.890
Ratio 6 2.325 1.545 1.935 0.552 28.504

 

44

 

Table 1.2 Continued: Comparison of 50% Evaporated Gasoline #1 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

Gas 1 Gas 16 Standard
50% neat Average deviation %std dev
Ratio 1 4.047 3.959 4.003 0.062 1.554
Ratio 2 0.762 0.670 0.716 0.065 9.086
Ratio 3 0.304 0.389 0.347 0.060 17.346
Ratio 4 12.335 11.983 12.159 0.249 2.047
Ratio 5 2.606 2.782 2.694 0.124 4.620
Ratio 6 2.325 2.989 2.657 0.470 17.671

 

 

 

 

 

 

Table 1.3 Comparison of 50% Evaporated Gasoline #2 (Debris Extract) to Gasolines

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

#1 -1 6 Neat
Gas 2 Gas 1 Standard
50% neat Avergge deviation %std dev
Ratio 1 4.785 3.904 4.345 0.623 14.339
Ratio 2 0.706 0.764 0.735 0.041 5.580
Ratio 3 0.286 0.307 0.297 0.015 5.008
Ratio 4 14.829 11.596 13.213 2.286 17.302
Ratio 5 2.888 2.458 2.673 0.304 11.375
Ratio 6 2.361 2.332 2.347 0.021 0.874
Gas 2 Gas 2 Standard
50% neat Average deviation %std dev
Ratio 1 4.785 4.665 4.725 0.085 1.796
Ratio 2 0.706 0.713 0.710 0.005 0.698
Ratio 3 0.286 0.282 0.284 0.003 0.996
Ratio 4 14.829 14.787 14.808 0.030 0.201
Ratio 5 2.888 2.779 2.834 0.077 2.720
Ratio 6 2.361 2.417 2.389 0.040 1.658
Gas 2 Gas 3 Standard
50% neat Avera e deviation %std dev
Ratio 1 4.785 7.756 6.271 2.101 33.503
Ratio 2 0.706 0.948 0.827 0.171 20.692
Ratio 3 0.286 0.336 0.311 0.035 11.368
Ratio 4 14.829 9.742 12.286 3.597 29.279
Ratio 5 2.888 5.339 4.114 1.733 42.132
Ratio 6 2.361 0.719 1.540 1.161 75.394
Gas 2 Gas 4 Standard
50% neat Avera e deviation %std dev
Ratio 1 4.785 11.313 8.049 4.616 57.349
Ratio 2 0.706 0.780 0.743 0.052 7.043
Ratio 3 0.286 0.389 0.338 0.073 21.580
Ratio 4 14.829 9.700 12.265 3.627 29.571
Ratio 5 2.888 6.538 4.713 2.581 54.762
Ratio 6 2.361 0.773 1.567 1.123 71.658

 

 

 

 

 

 

 

45

 

Table 1.3 Continued: Comparison of 50% Evaporated Gasoline #2 (Debris Extract) to

Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 2 Gas 5 Standard
50% neat Average deviation %std dev
Ratio 1 4.785 6.475 5.630 1.195 21.226
Ratio 2 0.706 0.760 0.733 0.038 5.209
Ratio 3 0.286 0.336 0.311 0.035 11.368
Ratio 4 14.829 11.703 13.266 2.210 16.662
Ratio 5 2.888 4.292 3.590 0.993 27.654
Ratio 6 2.361 1.088 1.725 0.900 52.198
Gas 2 Gas 6 Standard
50% neat Avergge deviation %std dev
Ratio 1 4.785 5.083 4.934 0.211 4.271
Ratio 2 0.706 0.673 0.690 0.023 3.384
Ratio 3 0.286 0.481 0.384 0.138 35.955
Ratio 4 14.829 8.987 11.908 4.131 34.690
Ratio 5 2.888 3.213 3.051 0.230 7.534
Ratio 6 2.361 2.039 2.200 0.228 10.349
Gas 2 Gas 7 Standard
50% neat Average deviation %std dev
Ratio 1 4.785 5.369 5.077 0.413 8.134
Ratio 2 0.706 0.968 ' 0.837 0.185 22.134
Ratio 3 0.286 0.733 0.510 0.316 62.037
Ratio 4 14.829 2.126 8.478 8.982 105.955
Ratio 5 2.888 2.435 2.662 0.320 12.035
Ratio 6 2.361 2.253 2.307 0.076 3.310
Gas 2 Gas 8 Standard
50% neat Avera e deviation %std dev
Ratio 1 4.785 3.842 4.314 0.667 15.458
Ratio 2 0.706 0.634 0.670 0.051 7.599
Ratio 3 0.286 0.135 0.211 0.107 50.724
Ratio 4 14.829 36.262 25.546 15.155 59.327
Ratio 5 2.888 2.557 2.723 0.234 8.597
Ratio 6 2.361 2.747 2.554 0.273 10.687
Gas 2 Gas 9 Standard
50% neat Average deviation %std dev
Ratio 1 4.785 5.702 5.244 0.648 12.366
Ratio 2 0.706 0.653 0.680 0.037 5.515
Ratio 3 0.286 0.329 0.308 0.030 9.888
Ratio 4 14.829 14.245 14.537 0.413 2.841
Ratio 5 2.888 3.851 3.370 0.681 20.209
Ratio 6 2.361 1.867 2.114 0.349 16.524

 

46

 

Table 1.3 Continued: Comparison of 50% Evaporated Gasoline #2 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 2 Gas 10 Standard
50% neat Average deviation %std dev
Ratio 1 4.785 3.714 4.250 0.757 17.821
Ratio 2 0.706 0.790 0.748 0.059 7.941
Ratio 3 0.286 0.489 0.388 0.144 37.043
Ratio 4 14.829 6.136 10.483 6.147 58.639
Ratio 5 2.888 2.359 2.624 0.374 14.258
Ratio 6 2.361 2.572 2.467 0.149 6.049
Gas 2 Gas 11 Standard
50% neat Average deviation %std dev
Ratio 1 4.785 8.461 6.623 2.599 39.247
Ratio 2 0.706 1.021 0.864 0.223 25.795
Ratio 3 0.286 0.871 0.579 0.414 71.505
Ratio 4 14.829 2.482 8.656 8.731 100.868
Ratio 5 2.888 6.295 4.592 2.409 52.469
Ratio 6 2.361 0.689 1.525 1.182 77.527
Gas 2 Gas 12 Standard
50% neat Average deviation %std dev
Ratio 1 4.785 2.896 3.841 1.336 34.780
Ratio 2 0.706 1.048 0.877 0.242 27.575
Ratio 3 0.286 0.228 0.257 0.041 15.958
Ratio 4 14.829 11.705 13.267 2.209 16.650
Ratio 5 2.888 3.994 3.441 0.782 22.728
Ratio 6 2.361 3.900 3.131 1.088 34.762
Gas 2 Gas 13 Standard
50% neat Avera}; deviation %std dev
Ratio 1 4.785 4.392 4.589 0.278 6.056
Ratio 2 0.706 0.953 0.830 0.175 21.056
Ratio 3 0.286 0.927 0.607 0.453 74.733
Ratio 4 14.829 1.217 8.023 9.625 119.969
Ratio 5 2.888 2.712 2.800 0.124 4.445
Ratio 6 2.361 1.593 1.977 0.543 27.469
Gas 2 Gas 14 Standard
50% neat Avegge deviation %std dev
Ratio 1 4.785 4.114 4.450 0.474 10.663
Ratio 2 0.706 0.838 0.772 0.093 12.090
Ratio 3 0.286 0.388 0.337 0.072 21.402
Ratio 4 14.829 6.882 10.856 5.619 51.765
Ratio 5 2.888 2.521 2.705 0.260 9.595
Ratio 6 2.361 2.209 2.285 0.107 4.704

 

47

 

Table 1.3 Continued: Comparison of 50% Evaporated Gasoline #2 (Debris Extract) to

Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 2 50% Gas 15 neat Average Standard deviation %std dev
Ratio 1 4.785 5.910 5.348 0.795 14.876
Ratio 2 0.706 0.899 0.803 0.136 17.006
Ratio 3 0.286 0.502 0.394 0.153 38.765
Ratio 4 14.829 5.545 10.187 6.565 64.443
Ratio 5 2.888 3.459 3.174 0.404 12.723
Ratio 6 2.361 1.545 1.953 0.577 29.544

Gas 2 50% Gas 16 neat Average Standard deviation %std dev
Ratio 1 4.785 3.959 4.372 0.584 13.359
Ratio 2 0.706 0.670 0.688 0.025 3.700
Ratio 3 0.286 0.389 0.338 0.073 21.580
Ratio 4 14.829 11.983 13.406 2.012 15.011
Ratio 5 2.888 2.782 2.835 0.075 2.644
Ratio 6 2.361 2.989 2.675 0.444 16.600

 

Table 1.4 Comparison of 50% Evaporated Gasoline #3 (Debris Extract) to Gasolines

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

#1-16 Neat

Gas 3 50% Gas 1 neat Average Standard deviation %std dev
Ratio 1 7.818 3.904 5.861 2.768 47.221
Ratio 2 0.926 0.764 0.845 0.115 ' 13.556
Ratio 3 0.329 0.307 0.318 0.016 4.892
Ratio 4 9.920 11.596 10.758 1.185 11.016
Ratio 5 5.256 2.458 3.857 1.978 51.296
Ratio 6 0.699 2.332 1.516 1.155 76.193

Gas 3 50% Gas 2 neat Avera 9 Standard deviation %std dev
Ratio 1 7.818 4.665 6.242 2.230 35.721
Ratio 2 0.926 0.713 0.820 0.151 18.379
Ratio 3 0.329 0.282 0.306 0.033 10.879
Ratio 4 9.920 14.787 12.354 3.441 27.858
Ratio 5 5.256 2.779 4.018 1.752 43.597
Ratio 6 0.699 2.417 1.558 1.215 77.972

Gas 3 50% Gas 3 neat Avera e Standard deviation %std dev
Ratio 1 7.818 7.756 7.787 0.044 0.563
Ratio 2 0.926 0.948 0.937 0.016 1.660
Ratio 3 0.329 0.336 0.333 0.005 1.489
Ratio 4 9.920 9.742 9.831 0.126 1.280
Ratio 5 5.256 5.339 5.298 0.059 1.108
Ratio 6 0.699 0.719 0.709 0.014 1.995

 

 

 

 

 

 

48

 

 

 

Table 1.4 Continued: Comparison of 50% Evaporated Gasoline #3 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard
Gas 3 50% Gas 4 neat Average deviation %std dev
Ratio 1 7.818 11.313 9.566 2.471 25.836
Ratio 2 0.926 0.780 0.853 0.103 12.103
Ratio 3 0.329 0.389 0.359 0.042 11.818
Ratio 4 9.920 9.700 9.810 0.156 1.586
Ratio 5 5.256 6.538 5.897 0.907 15.372
Ratio 6 0.699 0.773 0.736 0.052 7.109
Standard
Gas 3 50% Gas 5 neat Average deviation %std dev
Ratio 1 7.818 6.475 7.147 0.950 13.288
Ratio 2 0.926 0.760 0.843 0.117 13.924
Ratio 3 0.329 0.336 0.333 0.005 1.489
Ratio 4 9.920 11.703 10.812 1.261 11.661
Ratio 5 5.256 4.292 4.774 0.682 14.278
Ratio 6 0.699 1.088 0.894 0.275 30.785
Standard
Gas 3 50% Gas 6 neat Average deviation %std dev
Ratio 1 7.818 5.083 6.451 1.934 29.981
Ratio 2 0.926 0.673 0.800 0.179 22.376
Ratio 3 0.329 0.481 0.405 0.107 26.538
Ratio 4 9.920 8.987 9.454 0.660 6.979
Ratio 5 5.256 3.213 4.235 1.445 34.115
Ratio 6 0.699 2.039 1.369 0.948 69.213
Standard
Gas 3 50% Gas 7 neat Avera e deviation %std dev
Ratio 1 7.818 5.369 6.594 1.732 26.264
Ratio 2 0.926 0.968 0.947 0.030 3.136
Ratio 3 0.329 0.733 0.531 0.286 53.799
Ratio 4 9.920 2.126 6.023 5.511 91.502
Ratio 5 5.256 2.435 3.846 1.995 51.872
Ratio 6 0.699 2.253 1.476 1.099 74.447
Standard
Gas 3 50% Gas 8 neat Avera e deviation %std dev
Ratio 1 7.818 3.842 5.830 2.811 48.224
Ratio 2 0.926 0.634 0.780 0.206 26.471
Ratio 3 0.329 0.135 0.232 0.137 59.129
Ratio 4 9.920 36.262 23.091 18.627 80.666
Ratio 5 5.256 2.557 3.907 1.908 48.854
Ratio 6 0.699 2.747 1.723 1.448 84.048
49

 

Table 1.4 Continued: Comparison of 50% Evaporated Gasoline #3 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard
Gas 3 50% Gas 9 neat Average deviation %std dev
Ratio 1 7.818 5.702 6.760 1.496 22.134
Ratio 2 0.926 0.653 0.790 0.193 24.451
Ratio 3 0.329 0.329 0.329 0.000 0.000
Ratio 4 9.920 14.245 12.083 3.058 25.311
Ratio 5 5.256 3.851 4.554 0.993 21.818
Ratio 6 0.699 1.867 1.283 0.826 64.373
Standard
Gas 3 50% Gas 10 neat Average deviation %std dev
Ratio 1 7.818 ' 3.714 5.766 2.902 50.329
Ratio 2 0.926 0.790 0.858 0.096 11.208
Ratio 3 0.329 0.489 0.409 0.113 27.662
Ratio 4 9.920 6.136 8.028 2.676 33.329
Ratio 5 5.256 2.359 3.808 2.048 53.801
Ratio 6 0.699 2.572 1.636 1.324 80.979
Standard
Gas 3 50% Gas 11 neat Averafi deviation %std dev
Ratio 1 7.818 8.461 8.140 0.455 5.586
Ratio 2 0.926 1.021 0.974 0.067 6.900
Ratio 3 0.329 0.871 0.600 0.383 63.875
Ratio 4 9.920 2.482 6.201 5.259 84.816
Ratio 5 5.256 6.295 5.776 0.735 12.721
Ratio 6 0.699 0.689 0.694 0.007 1.019
Standard
Gas 3 50% Gas 12 neat Averggg deviation %std dev
Ratio 1 7.818 2.896 5.357 3.480 64.969
Ratio 2 0.926 1.048 0.987 0.086 8.740
Ratio 3 0.329 0.228 0.279 0.071 25.644
Ratio 4 9.920 11.705 10.813 1.262 11.673
Ratio 5 5.256 3.994 4.625 0.892 19.294
Ratio 6 0.699 3.900 2.300 2.263 98.432
Standard
Gas 3 50% Gas 13 neat Average deviation %std dev
Ratiot 7.818 4.392 6.105 2.423 39.681
Ratio 2 0.926 0.953 0.940 0.019 2.032
Ratio 3 0.329 0.927 0.628 0.423 67.333
Ratio 4 9.920 1.217 5.569 6.154 110.514
Ratio 5 5.256 2.712 3.984 1.799 45.153
Ratio 6 0.699 1.593 1.146 0.632 55.162

 

50

 

Table 1.4 Continued: Comparison of 50% Evaporated Gasoline #3 (Debris Extract) to

Gasolines #1 -1 6 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 3 50% Gas 14 neat Average Standard deviation %std dev
Ratio 1 7.818 4.114 5.966 2.619 43.901
Ratio 2 0.926 0.838 0.882 0.062 7.055
Ratio 3 0.329 0.388 0.359 0.042 11.637
Ratio 4 9.920 6.882 8.401 2.148 25.571
Ratio 5 5.256 2.521 3.889 1.934 49.735
Ratio 6 0.699 2.209 1.454 1.068 73.434

Gas 3 50% Gas 15 neat Average Standard deviation %std dev
Ratio 1 7.818 5.910 6.864 1.349 19.656
Ratio 2 0.926 0.899 0.913 0.019 2.092
Ratio 3 0.329 0.502 0.416 0.122 29.442
Ratio 4 9.920 5.545 7.733 3.094 40.008
Ratio 5 5.256 3.459 4.358 1.271 29.161
Ratio 6 0.699 1.545 1.122 0.598 53.317

Gas 3 50% Gas 16 neat Avergqg Standard deviation %std dev
Ratio 1 7.818 3.959 5.889 2.729 46.340
Ratio 2 0.926 0.670 0.798 0.181 22.684
Ratio 3 0.329 0.389 0.359 0.042 11.818
Ratio 4 9.920 11.983 10.952 1.459 13.320
Ratio 5 5.256 2.782 4.019 1.749 43.528
Ratio 6 0.699 2.989 1.844 1.619 87.813

 

 

 

Table 1.5 Comparison of 50% Evaporated Gasoline #4 (Debris Extract) to Gasolines

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

#1 -16 Neat
Standard
Gas 4 50% Gas 1 neat Avera e deviation %std dev
Ratio 1 11.212 3.904 7.558 5.168 68.372
Ratio 2 0.772 0.764 0.768 0.006 0.737
Ratio 3 0.389 0.307 0.348 0.058 16.662
Ratio 4 9.825 11.596 10.711 1.252 11.692
Ratio 5 6.584 2.458 4.521 2.918 64.533
Ratio 6 0.765 2.332 1.549 1.108 71.555
Standard
Gas 4 50% Gas 2 neat Average deviation %std dev
Ratio 1 11.212 4.665 7.939 4.629 58.316
Ratio 2 0.772 0.713 0.743 0.042 5.619
Ratio 3 0.389 0.282 0.336 0.076 22.552
Ratio 4 9.825 14.787 12.306 3.509 28.512
Ratio 5 6.584 2.779 4.682 2.691 57.472
Ratio 6 0.765 2.417 1.591 1.168 73.422

 

 

 

 

 

 

 

51

 

 

Table 1.5 Continued: Comparison of 50% Evaporated Gasoline #4 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard
Gas 4 50% Gas 3 neat Average deviation %std dev
Ratio 1 11.212 7.756 9.484 2.444 25.767
Ratio 2 0.772 0.948 0.860 0.124 14.471
Ratio 3 0.389 0.336 0.363 0.037 10.338
Ratio 4 9.825 9.742 9.784 0.059 0.600
Ratio 5 6.584 5.339 5.962 0.880 14.767
Ratio 6 0.765 0.719 0.742 0.033 4.384
Standard
Gas 4 50% Gas 4 neat Average deviation %std dev
Ratio1 11.212 11.313 11.263 0.071 0.634
Ratio 2 0.772 0.780 0.776 0.006 0.729
Ratio 3 0.389 0.389 0.389 0.000 0.000
Ratio 4 9.825 9.700 ‘ 9.763 0.088 0.905
Ratio 5 6.584 6.538 6.561 0.033 0.496
Ratio 6 0.765 0.773 0.769 0.006 0.736
Standard
Gas 4 50% Gas 5 neat Average deviation %std dev
Ratio 1 11.212 6.475 8.844 3.350 37.876
Ratio 2 0.772 0.760 0.766 0.008 1.108
Ratio 3 0.389 0.336 0.363 0.037 10.338
Ratio 4 9.825 11.703 10.764 1.328 12.337
Ratio 5 6.584 4.292 5.438 1.621 29.803
Ratio 6 0.765 1.088 0.927 0.228 24.651
Standard
Gas 4 50% Gas 6 neat Avegge deviation %std dev
Ratio 1 11.212 5.083 8.148 4.334 53.192
Ratio 2 0.772 0.673 0.723 0.070 9.689
Ratio 3 0.389 0.481 0.435 0.065 14.955
Ratio 4 9.825 8.987 9.406 0.593 6.300
Ratio 5 6.584 3.213 4.899 2.384 48.661
Ratio 6 0.765 2.039 1.402 0.901 64.255
Standard
Gas 4 50% Gas 7 neat Average deviation %std dev
Ratio 1 11.212 5.369 8.291 4.132 49.836
Ratio 2 0.772 0.968 0.870 0.139 15.930
Ratio 3 0.389 0.733 0.561 0.243 43.359
Ratio 4 9.825 2.126 5.976 5.444 91.106
Ratio 5 6.584 2.435 4.510 2.934 65.058
Ratio 6 0.765 2.253 1.509 1.052 69.727

 

 

 

 

 

 

52

 

Table 1.5 Continued: Comparison of 50% Evaporated Gasoline #4 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 4 50% Gas 8 neat Average Standard deviation %std dev
Ratio 1 11.212 3.842 7.527 5.2.11 69.236
Ratio 2 0.772 0.634 0.703 0.098 13.881
Ratio 3 0.389 0.135 0.262 0.180 68.552
Ratio 4 9.825 36.262 23.044 18.694 81.124
Ratio 5 6.584 2.557 4.571 2.848 62.302
Ratio 6 0.765 2.747 1.756 1.401 79.811
Gas 4 50% Gas 9 neat Average Standard deviation %std dev
Ratio 1 1 1.212 5.702 8.457 3.896 46.070
Ratio 2 0.772 0.653 0.713 0.084 11.810
Ratio 3 0.389 0.329 0.359 0.042 11.818
Ratio 4 9.825 14.245 12.035 3.125 25.969
Ratio 5 6.584 3.851 5.218 1.933 37.039
Ratio 6 0.765 1.867 1.316 0.779 59.212
Gas 4 50% Gas 10 neat: Average Standard deviation %std dev
Ratio 1 11.212 3.714 7.463 5.302 71.042
Ratio 2 0.772 0.790 0.781 0.013 1.630
Ratio 3 0.389 0.489 0.439 0.071 16.107
Ratio 4 9.825 6.136 7.981 2.609 32.686
Ratio 5 6.584 2.359 4.472 2.988 66.813
Ratio 6 0.765 2.572 1.669 1.278 76.580
Gas 4 50% Gas 11 neat Average Standard deviation %std dev
Ratio 1 11.212 8.461 9.837 1.945 19.776
Ratio 2 0.772 1.021 0.897 0.176 19.640
Ratio 3 0.389 0.871 0.630 0.341 54.099
Ratio 4 9.825 2.482 6.154 5.192 84.379
Ratio 5 6.584 6.295 6.440 0.204 3.173
Ratio 6 0.765 0.689 0.727 0.054 7.392
Gas 4 50% Gas 12 neat Avergagm Standard deviation %std dev
Ratio 1 11.212 2.896 7.054 5.880 83.361
Ratio 2 0.772 1.048 0.910 0.195 21.446
Ratio 3 0.389 0.228 0.309 0.114 36.902
Ratio 4 9.825 11.705 10.765 1.329 12.349
Ratio 5 6.584 3.994 5.289 1.831 34.627
Ratio 6 0.765 3.900 2.333 2.217 95.039

 

 

 

53

 

 

Table 1.5 Continued: Comparison of 50% Evaporated Gasoline #4 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard
Gas 4 50% Gas 13 neat Average deviation %std dev
Ratio 1 11.212 4.392 7.802 4.822 61.811
Ratio 2 0.772 0.953 0.863 0.128 14.839
Ratio 3 0.389 0.927 0.658 0.380 57.815
Ratio 4 9.825 1.217 5.521 6.087 110.248
Ratio 5 6.584 2.712 4.648 2.738 58.905
Ratio 6 0.765 1.593 1.179 0.585 49.659
Standard
Gas 4 50% Gas 14 neat Average deviation %std dev
Ratio 1 11.212 4.114 7.663 5.019 65.497
Ratio 2 0.772 0.838 0.805 0.047 5.797
Ratio 3 0.389 0.388 0.389 0.001 0.182
Ratio 4 9.825 6.882 8.354 2.081 24.912
Ratio 5 6.584 2.521 4.553 2.873 63.108
Ratio 6 0.765 2.209 1.487 1.021 68.666
Standard
Gas 4 50% Gas 15 neat Average deviation %std dev
Ratio 1 11.212 5.910 8.561 3.749 43.793
Ratio 2 0.772 0.899 0.836 0.090 10.748
Ratio 3 0.389 0.502 0.446 0.080 17.936
Ratio 4 9.825 5.545 7.685 3.026 39.381
Ratio 5 6.584 3.459 5.022 2.210 44.005
Ratio 6 0.765 1.545 1.155 0.552 47.753
Standard
Gas 4 50% Gas 16 neat Average deviation %std dev
Ratio 1 11.212 3.959 7.586 5.129 67.611
Ratio 2 0.772 0.670 0.721 0.072 10.003
Ratio 3 0.389 0.389 0.389 0.000 0.000
Ratio 4 9.825 11.983 10.904 1.526 13.994
Ratio 5 6.584 2.782 4.683 2.688 57.408
Ratio 6 0.765 2.989 1.877 1.573 83.783

 

 

 

 

 

 

Table 1.6 Comparison of 50% Evaporated Gasoline #5 (Debris Extract) to Gasolines
# 1 ~ 16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard
Gas 5 50% Gas 1 neat Avera e deviation %std dev
Ratio 1 6.426 3.904 5.165 1.783 34.527
Ratio 2 0.761 0.764 0.763 0.002 0.278
Ratio 3 0.331 0.307 0.319 0.017 5.320
Ratio 4 11.673 11.596 11.635 0.054 0.468
Ratio 5 4.236 2.458 3.347 1.257 37.563
Ratio 6 1.068 2.332 1.700 0.894 52.575

 

54

 

Table 1.6 Continued: Comparison of 50% Evaporated Gasoline #5 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 5 50% Gas 2 neat AVfige Standard deviation %std dev
Ratio 1 6.426 4.665 5.546 1.245 22.455
Ratio 2 0.761 0.713 0.737 0.034 4.605
Ratio 3 0.331 0.282 0.307 0.035 11.304
Ratio 4 11.673 14.787 13.230 2.202 16.643
Ratio 5 4.236 2.779 3.508 1.030 29.373
Ratio 6 1.068 2.417 1.743 0.954 54.742
Gas 5 50°/e Gas 3 neat Average Standard deviation %std dev
Ratio 1 6.426 7.756 7.091 0.940 13.263
Ratio 2 0.761 0.948 0.855 0.132 15.474
Ratio 3 0.331 0.336 0.334 0.004 1.060
Ratio 4 11.673 9.742 10.708 1.365 12.752
Ratio 5 4.236 5.339 4.788 0.780 16.291
Ratio 6 1.068 0.719 0.894 0.247 27.620
Gas 5 50% Gas 4 neat Average Standard deviation %std dev
Ratio 1 6.426 11.313 8.870 3.456 38.961
Ratio 2 0.761 0.780 0.771 0.013 1.744
Ratio 3 0.331 0.389 0.360 0.041 11.392
Ratio 4 11.673 9.700 10.687 1.395 13.055
Ratio 5 4.236 6.538 5.387 1.628 30.216
Ratio 6 1.068 0.773 0.921 0.209 22.661
Gas 5 50% Gas 5 neat Average Standard deviation %std dev
Ratio 1 6.426 6.475 6.451 0.035 0.537
Ratio 2 0.761 0.760 0.761 0.001 0.093
Ratio 3 0.331 0.336 0.334 0.004 1.060
Ratio 4 11.673 11.703 11.688 0.021 0.181
Ratio 5 4.236 4.292 4.264 0.040 0.929
Ratio 6 1.068 1.088 1.078 0.014 1.312
Gas 5 50% Gas 6 neat Averege Standard deviation %std dev
Ratio 1 6.426 5.083 5.755 0.950 16.503
Ratio 2 0.761 0.673 0.717 0.062 8.679
Ratio 3 0.331 0.481 0.406 0.106 26.125
Ratio 4 11.673 8.987 10.330 1.899 18.386
Ratio 5 4.236 3.213 3.725 0.723 19.422
Ratio 6 1.068 2.039 1.554 0.687 44.197

 

 

 

 

 

55

 

 

Table 1.6 Continued: Comparison of 50% Evaporated Gasoline #5 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard
Gas 5 50% Gas 7 neat Average deviation %std dev
Ratio 1 6.426 5.369 5.898 0.747 12.673
Ratio 2 0.761 0.968 0.865 0.146 16.931
Ratio 3 0.331 0.733 0.532 0.284 53.432
Ratio 4 11.673 2.126 6.900 6.751 97.844
Ratio 5 4.236 2.435 3.336 1.273 38.180
Ratio 6 1.068 2.253 1.661 0.838 50.462
Standard
Gas 5 50% Gas 8 neat Average deviation %std dev
Ratio 1 6.426 3.842 5.134 1.827 35.589
Ratio 2 0.761 0.634 0.698 0.090 12.875
Ratio 3 0.331 0.135 0.233 0.139 59.482
Ratio 4 11.673 36.262 23.968 17.387 72.544
Ratio 5 4.236 2.557 3.397 1.187 34.955
Ratio 6 1.068 2.747 1.908 1.187 62.240
Standard
Gas 5 50% Gas 9 neat Avera e deviation %std dev
Ratio 1 6.426 5.702 6.064 0.512 8.442
Ratio 2 0.761 0.653 0.707 0.076 10.802
Ratio 3 0.331 0.329 0.330 0.001 0.429
Ratio 4 11.673 14.245 12.959 1.819 14.034
Ratio 5 4.236 3.851 4.044 0.272 6.733
Ratio 6 1.068 1.867 1.468 0.565 38.499
Standard
Gas 5 50% Gas 10 neat Average deviation %std dev
Ratio 1 6.426 3.714 5.070 1.918 37.824
Ratio 2 0.761 0.790 0.776 0.021 2.644
Ratio 3 0.331 0.489 0.410 0.112 27.249
Ratio 4 11.673 6.136 8.905 3.915 43.969
Ratio 5 4.236 2.359 3.298 1.327 40.250
Ratio 6 1.068 2.572 1.820 1.063 58.433
Standard
Gas 5 50% Gas 11 neat Averag deviation %std dev
Ratio 1 6.426 8.461 7.444 1.439 19.332
Ratio 2 0.761 1.021 0.891 0.184 20.634
Ratio 3 0.331 0.871 0.601 0.382 63.534
Ratio 4 11.673 2.482 7.078 6.499 91.826
Ratio 5 4.236 6.295 5.266 1.456 27.650
Ratio 6 1.068 0.689 0.879 0.268 30.506

 

56

 

 

Table 1.6 Continued: Comparison of 50% Evaporated Gasoline #5 (Debris Extract) to
Gasolines #1-16 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Standard
Gas 5 50% Gas 12 neat Average deviation %std dev
Ratio 1 6.426 2.896 4.661 2.496 53.553
Ratio 2 0.761 1.048 0.905 0.203 22.437
Ratio 3 0.331 0.228 0.280 0.073 26.058
Ratio 4 11.673 11.705 11.689 0.023 0.194
Ratio 5 4.236 3.994 4.115 0.171 4.158
Ratio 6 1.068 3.900 2.484 2.003 80.617
Standard
Gas 5 50% Gas 13 neat Average deviation %std dev
Ratio 1 6.426 4.392 5.409 1.438 26.590
Ratio 2 0.761 0.953 0.857 0.136 15.842
Ratio 3 0.331 0.927 0.629 0.421 67.001
Ratio 4 11.673 1.217 6.445 7.394 114.717
Ratio 5 4.236 2.712 3.474 1.078 31.020
Ratio 6 1.068 1.593 1.331 0.371 27.902
Standard
Gas 5 50% Gas 14 neat Averagg deviation %std dev
Ratio 1 6.426 4.114 5.270 1.635 31.021
Ratio 2 0.761 0.838 0.800 0.054 6.810
Ratio 3 0.331 0.388 0.360 0.040 11.211
Ratio 4 11.673 6.882 9.278 3.388 36.516
Ratio 5 4.236 2.521 3.379 1.213 35.894
Ratio 6 1.068 2.209 1.639 0.807 49.241
Standard
Gas 5 50% Gas 15 neat Average deviation %std dev
Ratio 1 6.426 5.910 6.168 0.365 5.915
Ratio 2 0.761 0.899 0.830 0.098 11.757
Ratio 3 0.331 0.502 0.417 0.121 29.031
Ratio 4 1 1.673 5.545 8.609 4.333 50.333
Ratio 5 4.236 3.459 3.848 0.549 14.280
Ratio 6 1.068 1.545 1.307 0.337 25.816
Standard
Gas 5 50% Gas 16 neat Averagg deviation %std dev
Ratio 1 6.426 3.959 5.193 1.744 33.595
Ratio 2 0.761 0.670 0.716 0.064 8.993
Ratio 3 0.331 0.389 0.360 0.041 11.392
Ratio 4 11.673 11.983 11.828 0.219 1.853
Ratio 5 4.236 2.782 3.509 1.028 29.300
Ratio 6 1.068 2.989 2.029 1.358 66.963

 

57

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 34.0 0 3.0 v
000.0 050.0 0
00 _. .0 00 r .0 N
000.0 000.0 _.
.m0>0 000.. a 20m
$05 0r 05.0000 0.. 05.0000
00N.0 00N.0 30.0 034.0 N000 000.0 .4
N000 000.0 03.0 v3.0 N000 00v0 0
500.0 00v0 0N0.0 :00 000.0 .000 N
0000 500.0 005.0 005.0 000.0 000.0 ..
a «00: .005 .00: .mm>0 «00: a» 20m.
$05 00 05.0000 00 05.0000 $05 00 05.0000 00 05.0000 $05 50 05.0000 50 05.0000
30.0 034.0 000.0 000.0 000.0 00v.0 v
505.0 005.0 50Ne 50N.w 000.0 050.0 0
N000 500.0 055.0 505.0 N0v.0 00v.0 N
Nv0.0 500.0 0.0.0 000 000.0 V050 r
i...M>0 «00: .m0>0 “00: .0105 00: 0 2.0”.
$05 00 05.0000 00 05.0000 $05 00 05.0000 00 05.0000 $05 00 05.0000 3 05.0000
0 5.4.0 0Nv0 00N.0 00N.0 00v.0 V0v0 .4
va0 N00 00.0 000.0 500.0 0.0.0 0
000.0 500.0 050.0 500.0 N000 000.0 N
5v00 000.0 500.0 000.0 N050 V550 r
iimmfl .00: . 0>0 ~00: .0M>o .00: # 2.0m
$05 00 05.0000 00 05.0000 $05 N0 05.0000 N0 05.0000 $05 P0 05.0000 00 05.0000

 

 

 

50065850320 2- as 65630 5 I 850 50 68500 25> 3 2.3

28.000800 $0 5 .8.“ Sam. A. 5002.0...

58

Table 2.2 Comparison of 75% Evaporated Gasoline #01 to Gasolines #01-10 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 01 75% Gas 01 Standard %std
Ratio # evap. neat Average deviation dev=(stdevlaeg.)'100
1 0.782 0.774 0.778 0.005 0.682
2 0.382 0.398 0.390 0.011 2.922
3 0.557 0.546 0.551 0.007 1.347
4 0.436 0.434 0.435 0.001 0.244
Gas 01 75% Gas 02 Standard
Ratio # evap. neat Average deviation %std dev
1 0.774 0.669 0.721 0.074 10.298
2 0.398 0.497 0.447 0.070 15.731
3 0.546 0.858 0.702 0.221 31.489
4 0.434 0.293 0.363 0.100 27.545
Gas 01 75% Gas 03 Standard
Ratio # evap. neat Average deviation %std dev
1 0.774 0.930 0.852 0.110 12.955
2 0.398 0.357 0.377 0.029 7.690
3 0.546 0.421 0.483 0.088 18.217
4 0.434 0.429 0.431 0.004 0.902
Gas 01 75% Gas 04 Standard
Ratio # evap. neat Averag deviation %std dev
1 0.774 0.794 0.784 0.014 1.805
2 0.398 0.490 0.444 0.065 14.668
3 0.546 0.675 0.610 0.092 15.005
4 0.434 0.400 0.417 0.024 5.854
Gas 01 75% Gas 05 Standard
Ratio # evap. neat Averege deviation %std dev
1 0.774 0.964 0.869 0.135 15.505
2 0.398 0.737 0.567 0.240 42.320
3 0.546 1.257 0.901 0.503 55.823
4 0.434 0.489 0.462 0.039 8.427
Gas 01 75% Gas 06 Standard
Ratio # evap. neat Avergge deviation %std dev
1 0.774 0.837 0.805 0.045 5.534
2 0.398 0.387 0.392 0.008 1.984
3 0.546 0.706 0.626 0.113 18.087
4 0.434 0.449 0.442 0.011 2.402
Gas 01 75% Gas 07 Standard
Ratio # evap. neat Average deviation %std dev
1 0.774 0.900 0.837 0.089 10.665
2 0.398 0.491 0.444 0.066 14.876
3 0.546 0.488 0.517 0.040 7.802
4 0.434 0.486 0.460 0.036 7.921

 

59

 

Table 2.2 Continued: Comparison of 75% Evaporated Gasoline #01 to Gasolines

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

#01-10 Neat
Gas 01 75% Gas 08 Standard
Ratio # evap. neat Average deviation %std dev
1 0.774 0.760 0.767 0.010 1.243
2 0.398 0.341 0.369 0.040 10.844
3 0.546 0.444 0.495 0.072 14.491
4 0.434 0.443 0.438 0.006 1.403
Gas 01 75% Gas 09 Standard
Ratio # evap. neat Average deviation %std dev
1 0.774 0.667 0.720 0.075 10.464
2 0.398 0.405 0.401 0.005 1.271
3 0.546 0.908 0.727 0.257 35.293
4 0.434 0.256 0.345 0.126 36.555
Gas 01 75% Gas 10 Standard
Ratio # evap. neat Average deviation %std dev
1 0.774 0.630 0.702 0.101 14.447
2 0.398 0.136 0.267 0.185 69.211
3 0.546 0.576 0.561 0.022 3.863
4 0.434 0.418 0.426 0.012 2.730

 

 

 

 

 

 

 

Table 2.3 Comparison of 75% Evaporated Gasoline #02 to Gasolines #01-10 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Gas 02 75%
Ratio # evap. Gas 01 neat Avegg Standard deviation %std dev
1 0.687 0.774 0.731 0.062 8.421
2 0.478 0.398 0.438 0.057 12.986
3 0.845 0.546 0.696 0.211 30.399
4 0.293 0.434 0.364 0.100 27.428
Gas 02 75%
Ratio # evap. Gas 02 neat Average Standard deviation %std dev
1 0.687 0.669 0.678 0.013 1.930
2 0.478 0.497 0.488 0.013 2.756
3 0.845 0.858 0.852 0.009 1.080
4 0.293 0.293 0.293 0.000 0.121
Gas 02 75%
Ratio # evap. Gas 03 neat Average Standard deviation %std dev
1 0.687 0.930 0.808 0.171 21.215
2 0.478 0.357 0.417 0.086 20.590
3 0.845 0.421 0.633 0.300 47.364
4 0.293 0.429 0.361 0.096 26.559
Ratio # Gas 02 75% evap. Gas 04 neat Average Standard deviation %std dev
1 0.687 0.794 0.740 0.075 10.173
2 0.478 0.490 0.484 0.008 1.681
3 0.845 0.675 0.760 0.120 15.817
4 0.293 0.400 0.346 0.075 21.749

 

6O

 

 

 

Table 2.3 Continued: Comparison of 75% Evaporated Gasoline #02 to Gasolines

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

#01-10 Neat
Gas 02 75% Gas 05 Standard
Ratio # evap. neat Average deviation %std dev
1 0.687 0.964 0.826 0.196 23.727
2 0.478 0.737 0.608 0.183 30.147
3 0.845 1.257 1.051 0.291 27.719
4 0.293 0.489 0.391 0.139 35.446
Gas 02 75% Gas 06 Standard
Ratio # evap. neat Average deviation %std dev
1 0.687 0.837 0.762 0.106 13.878
2 0.478 0.387 0.432 0.065 14.968
3 0.845 0.706 0.775 0.099 12.724
4 0.293 0.449 0.371 0.110 29.733
Gas 02 75% Gas 07 Standard
Ratio # evap. neat Avera e deviation %std dev
1 0.687 0.900 0.793 0.150 18.956
2 0.478 0.491 0.484 0.009 1.891
3 0.845 0.488 0.667 0.252 37.814
4 0.293 0.486 0.389 0.136 34.970
Gas 02 75% Gas 08 Standard
Ratio # evap. neat Average deviation %std dev
1 0.687 0.760 0.724 0.052 7.137
2 0.478 0.341 0.409 0.097 23.681
3 0.845 0.444 0.645 0.283 43.981
4 0.293 0.443 0.368 0.106 28.776
Gas 02 75% Gas 09 Standard
Ratio # evap. neat Averege deviation %std dev
1 0.687 0.667 0.677 0.014 2.097
2 0.478 0.405 0.441 0.052 11.742
3 0.845 0.908 0.877 0.045 5.106
4 0.293 0.256 0.274 0.026 9.608
Gas 02 75% Gas 10 Standard
Ratio # evap. neat Average deviation %std dev
1 0.687 0.630 0.659 0.040 6.108
2 0.478 0.136 0.307 0.242 78.674
3 0.845 0.576 0.711 0.190 26.755
4 0.293 0.418 0.355 0.088 24.791

 

 

 

 

 

 

 

61

 

 

Table 2.4 Comparison of 75% Evaporated Gasoline #03 to Gasolines #01-10 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ratio # Gas 03 75% Gas 01 neat Average Standard %std dev
evap. deviation
1 0.947 0.774 0.861 0.122 14.216
2 0.343 0.398 0.370 0.039 10.529
3 0.421 0.546 0.484 0.088 18.281
4 0.418 0.434 0.426 0.011 2.656
Ratio # Gas 03 75% Gas 02 neat Average Standard %std dev
evap. deviation
1 0.947 0.669 0.808 0.197 24.380
2 0.343 0.497 0.420 0.109 26.027
3 0.421 0.858 0.640 0.309 48.320
4 0.418 0.293 0.355 0.089 24.980
Ratio # Gas 03 75% Gas 03 neat Average Standard %std dev
evap. deviation
1 0.947 0.930 0.938 0.012 1.319
2 0.343 0.357 0.350 0.010 2.832
3 0.421 0.421 0.421 0.000 0.000
4 0.418 0.429 0.423 0.007 1.754
Ratio # Gas 03 75% Gas 04 neat Average Standard %std dev
evap. deviation
1 0.947 0.794 0.870 0.109 12.472
2 0.343 0.490 0.416 0.104 24.987
3 0.421 0.675 0.548 0.180 32.775
4 0.418 0.400 0.409 0.013 3.200
Ratio # Gas 03 75% Gas 05 neat Average Standard %std dev
evap. deviation
1 0.947 0.964 0.956 0.012 1.258
2 0.343 0.737 0.540 0.279 51.682
3 0.421 1.257 0.839 0.591 70.458
4 0.418 0.489 0.454 0.050 11.070
Ratio # Gas 03 75% Gas 06 neat Average Standard %std dev
evap. deviation
1 0.947 0.837 0.892 0.078 8.762
2 0.343 0.387 0.365 0.031 8.536
3 0.421 0.706 0.563 0.201 35.716
4 0.418 0.449 0.434 0.022 5.057
Ratio # Gas 03 75% Gas 07 neat Average Standard %std dev
evap. deviation
1 0.947 0.900 0.923 0.033 3.624
2 0.343 0.491 0.417 0.105 25.190
3 0.421 0.488 0.455 0.048 10.489
4 0.418 0.486 0.452 0.048 10.566

 

62

 

Table 2.4 Continued: Comparison of 75% Evaporated Gasoline #03 to Gasolines
#01-10 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ratio # Gas 03 Gas 08 neat Average Standard %std dev
75% evap. deviation
1 0.947 0.760 0.854 0.132 15.490
2 0.343 0.341 0.342 0.001 0.335
3 0.421 0.444 0.433 0.016 3.776
4 0.418 0.443 0.430 0.017 4.058
Ratio # Gas 03 Gas 09 neat Average Standard %std dev
75% evap. deviation
1 0.947 0.667 0.807 0.198 24.542
2 0.343 0.405 0.374 0.044 11.774
3 0.421 0.908 0.665 0.345 51.843
4 0.418 0.256 0.337 0.115 34.065
Ratio # Gas 03 Gas 10 neat Average Standard %std dev
75% evep. deviation
1 0.947 0.630 0.789 0.224 28.415
2 0.343 0.136 0.239 0.146 60.915
3 0.421 0.576 0.499 0.110 22.003
4 0.418 0.418 0.418 0.000 0.074

 

 

 

 

 

 

Table 2.5 Comparison of 75% Evaporated Gasoline #04 to Gasolines #01-10 Neat

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ratio # Gas 04 75% Gas 01 Average Standard %std dev
evap. neat deviation
1 0.800 0.774 0.787 0.018 2.336
2 0.482 0.398 0.440 0.059 13.497
3 0.685 0.546 0.616 0.098 15.969
4 0.393 0.434 0.413 0.029 7.101
Ratio # Gas 04 75% Gas 02 Average Standard %std dev
evap. neat deviation
1 0.800 0.669 0.734 0.093 12.664
2 0.482 0.497 0.489 0.011 2.240
3 0.685 0.858 0.772 0.122 15.856
4 0.393 0.293 0.343 0.071 20.645
Ratio # Gas 04 75% Gas 03 Average Standard %std dev
egg. neat deviation
1 0.800 0.930 0.865 0.092 10.589
2 0.482 0.357 0.419 0.088 21.095
3 0.685 0.421 0.553 0.187 33.757
4 0.393 0.429 0.411 0.025 6.201

 

63

 

 

Table 2.5 Continued: Comparison of 75% Evaporated Gasoline #04 to Gasolines

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

#01-10 Neat
Ratio # Gas 04 75% evap. Gas 04 neat Average Standard deviation %std dev
1 0.800 0.794 0.797 0.005 0.577
2 0.482 0.490 0.486 0.006 1.165
3 0.685 0.675 0.680 0.007 1.040
4 0.393 0.400 0.396 0.005 1.250
Ratio # Gas 04 75% evap. Gas 05 neat Average Standard deviation %std dev
1 0.800 0.964 0.882 0.116 13.148
2 0.482 0.737 0.609 0.181 29.654
3 0.685 1.257 0.971 0.404 41.654
4 0.393 0.489 0.441 0.068 15.482
Ratio # Gas 04 75% evap. Gas 06 neat Average Standard deviation %std dev
1 0.800 0.837 0.818 0.026 3.154
2 0.482 0.387 0.434 0.067 15.478
3 0.685 0.706 0.695 0.014 2.085
4 0.393 0.449 0.421 0.040 9.495
Ratio # Gas 04 75% evap. Gas 07 neat Average Standard deviation %std dev
1 0.800 0.900 0.850 0.070 8.293
2 0.482 0.491 0.486 0.007 1.375
3 0.685 0.488 0.587 0.139 23.687
4 0.393 0.486 0.439 0.066 « 14.980
Ratio # Gas 04 75% evap. Gas 08 neat Average Standard deviation %std dev
1 0.800 0.760 0.780 0.028 3.624
2 0.482 0.341 0.411 0.099 24.182
3 0.685 0.444 0.565 0.170 30.173
4 0.393 0.443 0.418 0.035 8.500
Ratio # Gas 04 75% evap. Gas 09 neat Average Standard deviation %std dev
1 0.800 0.667 0.733 0.094 12.830
2 0.482 0.405 0.443 0.054 12.254
3 0.685 0.908 0.797 0.158 19.821
4 0.393 0.256 0.324 0.097 29.841
Ratio # Gas 04 75% evap. Gas 10 neat Average Standard deviation %std dev
1 0.800 0.630 0.715 0.120 16.799
2 0.482 0.136 0.309 0.244 79.029
3 0.685 0.576 0.631 0.077 12.207
4 0.393 0.418 0.405 0.018 4.375

 

 

 

 

 

 

 

 

 

.39. 68$

  
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00.0. 00.0. 00.5. 00.5. 00.0. 00.0. 00.0.
H _
m C
m. a 2
a N
9 u. . n
0 B
S H D. rJiL
B 0 ,.
n. c. K
0 / v m
.7 H
IL M."
0
H I
m
0 0.
Z

38.000800 $00 5 .38.. 8.03. ..m 050.”.

8000908650 .029? 0 so 0.1. 52... 82.3 000 8.33. 08.000800 ..o 5505000809 H... 5002.93.

00.0.

 

..
W
_.

585:
800.0550.

65

35 25

00.0.

. D ’

 

 

00.0w 00.5N 00.0w 00.0N 00.0.
b _ m
0
r1
H
B
m. m
.7
v
IL \
H
m m
u. 0
0 z
Co

 

 

 

38.80800 $05 a. 033 moan”. N0 2%....

.0

 

 

 

j

 

 

NE.—
v-n=__—_

109sz {

 

 

000.03
028055<

 

66

Notes

Chapter One: Introduction

1 Speight, J .G. 1999. The Chemistry and Technology of Petroleum, 4’” ed. New York:
Marcel Dekker, Inc.

2 Trimpe, MA. 1991. Turpentine in Arson Analysis. Journal of Forensic Sciences. 36(4):
1059-1073.

3 Friedman, A.J. 2002 Hydrocarbon Pattern Analyses: Elucidating Deeper Truths or Just
Finding the Pope in the Pizza? Or the Sorcerers. . .And Their Apprentices.
Presentation from the American Academy of Forensic Sciences Meeting in Atlanta,
GA.

4 Mann, DC. 1987. Comparison of Automotive Gasolines Using Capillary Gas
Chromatography I: Comparison Methodology. Journal of Forensic Sciences 32(3):
606-615.

5 Mann, DC. 1987. Comparison of Automotive Gasolines Using Capillary Gas
Chromatography 11: Limitations of Automotive Gasoline Comparisons in
Casework. Journal of Forensic Sciences 32(3): 616-628.

6 Dolan, J .A. and C. Ritacco. 2002. Gasoline Comparison by Gas Chromatography-Mass
Spectrometry Utilizing an Automated Approach to Data Analysis. Presentation
from the American Academy of Forensic Sciences Meeting in” Atlanta, GA.

7 Smith, CB. and J. Macy. 1991. Methods of Fire Debris Preparation for Detection of
Accelerants. Forensic Science Review. 3: 57-68.

8 Kuk, R. Fire Debris Chemist, Bureau of Alcohol, Tobacco, and Firearms. Personal
Communications.

9 Klontz, Bill. Petroleum Products Supervisor, Exxon-Mobil. Personal Communications.

‘0 Thornton, J .I. and B. Fukayama. The Implications of Refining Operations to the
Characterization and Analysis of Arson Accelerants: Part I Physical Separation.
Arson Analysis Newsletter 3(3): 1-16.

” Thornton, J .I. and B. F ukayama. The Implications of Refining Operations to the
Characterization and Analysis of Arson Accelerants: Part 11 Chemical Conversions,

Treating Processes, and Subsidiary Processes. Arson Analysis Newsletter 3(3): 1-
16.

'2 San Joaquin Geological Society. 2002. What is a Refinery? A Lesson in How to Make
Gasoline. www.sjgs.com/refinery.html

67

Iu'lm '

 

 

‘3 Freudenrich, C. 1998. How Oil Refining Works. http://www.howstuffworks.com/oil-
refiningShtm.

Chapter Two: Instrumentation

l4 Skoog, D., F.J. Holler, and T. Nieman. 1998. Principles of Instrumental Analysis, 5m
ed. Florida: Harcourt Brace & Co.

‘5 Hoffman, E. and V. Stroobant. 2002. Mass Spectrometry: Principles and Applications,
2"d ed. England: John Wiley & Sons.

Chapter Three: Forensic Methods of Accelerant Separation and Concentration
from Fire Debris

16 Atkins, P.W. 1989. General Chemistry. New York: Oxford University Press.

17 American Society for Testing and Materials (ASTM). 2002. Standard test methods E
1386 and E 1412: West Conshohocken, PA.

68

BIBLIOGRAPHY

American Society for Testing and Materials (ASTM). 2002. Standard test methods E
1386 and E 1412: West Conshohocken, PA. -

Atkins, P.W. 1989. General Chemistry. New York: Oxford University Press.

Dolan, J .A. and C. Ritacco. 2002. Gasoline Comparison by Gas Chromatography-Mass
Spectrometry Utilizing an Automated Approach to Data Analysis. Presentation
from the American Academy of Forensic Sciences Meeting in Atlanta, GA.

F reudenrich, C. 1998. How Oil Refining Works. http://www.howstuffworks.com/oil-
refiningShtm.

Friedman, A.J. 2002 Hydrocarbon Pattern Analyses: Elucidating Deeper Truths or Just
Finding the Pope in the Pizza? Or the Sorcerers. . .And Their Apprentices.

Presentation from the American Academy of Forensic Sciences Meeting in Atlanta,
GA.

Hoffman, E. and V. Stroobant. 2002. Mass Spectrometry: Principles and Applications,
2"d ed. England: John Wiley & Sons.

Klontz, Bill. Oil Chemist, Exxon-Mobil. Personal Communications.

Kuk, R. Arson Chemist, Bureau of Alcohol, Tobacco, and Firearms. Personal
Communications.

Mann, DC. 1987. Comparison of Automotive Gasolines Using Capillary Gas

Chromatography I: Comparison Methodology. Journal of Forensic Sciences 32(3):
606-615.

Mann, DC. 1987. Comparison of Automotive Gasolines Using Capillary Gas
Chromatography 11: Limitations of Automotive Gasoline Comparisons in

Casework. Journal of Forensic Sciences 32(3): 616-628.

San Joaquin Geological Society. 2002. What is a Refinery? A Lesson in How to Make
Gasoline. www.sjgs.com/refinery.html

Skoog, D., F .J . Holler, and T. Nieman. 1998. Principles of Instrumental Analysis, 5:}: ed.
Florida: Harcourt Brace & Co.

Smith, CB. and J. Macy. 1991. Methods of Fire Debris Preparation for Detection of
Accelerants. Forensic Science Review. 3: 57-68.

69

Speight, J .G. 1999. The Chemistry and Technology of Petroleum, 4” ed. New York:
Marcel Dekker, Inc.

Thornton, J .I. and B. Fukayama. 1979. The Implications of Refining Operations to the
Characterization and Analysis of Arson Accelerants: Part I Physical Separation.
Arson Analysis Newsletter 3(2): 1-16.

Thornton, J .I. and B. F ukayama. The Implications of Refining Operations to the
Characterization and Analysis of Arson Accelerants: Part 11 Chemical Conversions,
Treating Processes, and Subsidiary Processes. Arson Analysis Newsletter 3(3):1-16.

Trimpe, MA. 1991. Turpentine in Arson Analysis. Journal of Forensic Sciences. 36(4):
1059-1073.

70

     

    

       

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

llllll’lllml/Will lllll’lllllllllll
3 1293 02469 2323