A .4.

yumofig

«4* .~.uv_....u.

mfiwfi: _ . a .

62w? . .kflfifn‘mflwnmu.
. . s

, firlmgw

2...» “125.9%. .
. .a‘ (a. ¢fitF¢VJ§nu 4
.3 C .-
.. .fluufi

r”

. s,

a é umflmqfi re. - . .
. Junk s. “was... 5

3b? 1%.. .. . .

.2;

nmfip...%mmma .42..

‘u: v .. .n

, 2.5. ‘ fir

.
I

.2. 4.1;? 1
“3.53 u. :Rfifiv: \
. . ,. , Jan.»
A . fin"? . . 1
, V r, . a “as...“m,
. . “up.z~ (fa’
, .xfiufieniu. g?
A
BA. _‘ g
.1.»

ix:
v 0.3..

in

\ I; Ll». V .I
; ”Tar! i;
1 i; :3.

2!. u

.. H. 2.... y :.
z 2....
um... r.

 

 

 

 

 

 

 

 

 

 

 

 

THESE

11“

fl

’3 :3 1--

.LIBRARY
Michigan State
University

This is to certify that the

thesis entitled

Elemental Analysis of Glass Via
Variable Pressure Scanning Electron Microscopy
Energy Dispersive Spectroscopy (SEM—EDS)

presented by
Elizabeth Reeves

has been accepted towards fulfillment
of the requirements for

Master's degreein Criminal Justice

F .
:1 15450: professd:
[ 0 l

l 3.2: 0

Date

 

0.7639 MS U 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 RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuepSS-pJS

ELEMENTAL ANALYSIS OF GLASS VIA VARIABLE PRESSURE SCANNING
ELECTRON MICROSCOPY—ENERGY DISPERSIV E SPECTROSCOPY (SEM-EDS)

By

Elizabeth Reeves

A THESIS
Submitted to
Michigan State University

In partial fulfillment of the requirements
For the degree of

MASTER OF SCIENCE
Department of Criminal Justice

2001

 

 

cor
enl
var
Iran
Phi

USE

aux.

RUE:

Star

ABSTRACT

ELEMENTAL ANALYSIS OF GLASS VIA VARIABLE PRESSURE SCANNING
ELECTRON MICROSCOPY-ENERGY DISPERSIVE SPECTROSCOPY (SEM-EDS)

By

Elizabeth Reeves

The forensic examination of glass samples has traditionally relied upon
comparison of the physical and optical properties between two samples. However, the
enhancement of quality control measures in the manufacture of glass has led to decreased
variability of these properties between samples of different origins. The comparison of
trace elements in glass has been shown to be advantageous over the comparison of
physical and optical properties alone. SEM-EDS is an analytical technique that may be
used to analyze elemental composition on the surface of a sample of glass.

A LEO 435 VP SEM coupled with an EDAX Phoenix EDS system was
automated for elemental analysis of glass. The automated method analyzed tempered
automotive glass at variable pressure (10 Pa) for 100 live seconds in five iterations.
Elemental ratios (Sn/Na, Na/Al, Mg/Al, Na/Mg, Ca/N a, Ca/K) were averaged and
standard deviations calculated. A statistical model for inclusion was developed. Fifty
samples of tempered automotive glass were examined by refractive index measurement
and by the automated SEM-EDS method and statistical model for inclusion. When used
alone, both the comparison of glass based on refractive index measurement and SEM-
EDS comparison yielded a false inclusion in 7.6% of the cases. However, when
elemental analysis was used in tandem with the measurement of refractive index, only

0.8% of the cases were falsely included.

 

 

WC

ins

 

 

 

 

ACKNOWLEDGMENTS

The author wishes to thank those who have ensured the success of this project through
their support. First, the Michigan State Police Forensic Laboratory of Lansing, Michigan
must be acknowledged for the use of the instrumentation and chemicals used in this
study. Chris Bommarito of the Michigan State Police Laboratory is deserving of specific
gratitude for his concept, guidance, patience, and constructive criticism. Also, the author
would like to thank Dr. Jay Siege] of Michigan State University for his advice and

instruction.

iii

TABLE OF CONTENTS

List of Tables ........................................................................................................ v
List of Figures .................................................................................................... vii
Chapter 1
Introduction .......................................................................................................... 1
Forensic Importance and Analysis of Glass ................................................... l
The Problem to be Studied ............................................................................. 2
Purpose of the Study ....................................................................................... 3
Hypothesis ....................................................................................................... 4
Limitations ...................................................................................................... 4
Methodology ................................................................................................... 4
Chapter 2
Scanning Electron Microscopy and the Analysis of Glass ................................. 7
Scanning Electron Microscopy ....................................................................... 7
Manufacture of Glass ...................................................................................... 8
Elemental Analysis of Glass ........................................................................... 9
Summary ....................................................................................................... 10
Chapter 3
Review of Literature .......................................................................................... 11
Comparison of Glass Based Upon Refractive Index Measurement ............ 11
Elemental Analysis ....................................................................................... 12
Summary ....................................................................................................... 13
Chapter 4
Development of Suitable SEM-EDS Operating Parameters ............................ 15
Automation of the SEM with EDAX ........................................................... 15
Procedure .................................................................................................. 18
Parameters for Analysis ........................................................................... 20
Statistical Model for Inclusion ................................................................ 23
Summary .................................................................................................. 27
Examination of the Effect of Sample Tilt Angle ......................................... 27
Procedure .................................................................................................. 28
Summary .................................................................................................. 29
Application of Variable Pressure SEM-EDS to Glass Analysis ................. 30
Procedure .................................................................................................. 30
Elemental Analysis ....................................................................................... 32
Overall Results .............................................................................................. 32
Summary ....................................................................................................... 33

TABLE OF CONTENTS, continued

Chapter 5

Conclusions and Discussion .............................................................................. 34
Conclusions ................................................................................................... 34
Discussion ..................................................................................................... 35
Future Research ............................................................................................ 37
Beyond SEM-EDS ........................................................................................ 39

Appendices ......................................................................................................... 40

Appendix I

Glossary .............................................................................................................. 41

Appendix II

Elemental Data: high vacuum, low vacuum, and sputter coated samples ....... 43

Appendix III

Elemental Data: tilt angle effects ....................................................................... 49

Appendix IV

Summary of Elemental Ratio Comparisons in the Establishing of

a Statistical Model for Inclusion ........................................................................ 53

Appendix V

Refractive Index Comparisons .......................................................................... 55

Appendix VI

186 Elemental Comparisons of Samples not Distinguished by

Refractive Index ................................................................................................. 58

Appendix VII

Summary Data for Refractive Index and Elemental Analysis Comparisons. 105

Notes ................................................................................................................. 107

Bibliography ..................................................................................................... 109

LIST OF TABLES

Table 1.1
Elemental ratios constructed for all elemental comparisons .............................. 5

Table 4.1

Sample spreadsheet used to calculated elemental ratios,

averages and standard deviations from data obtained

in the automated analysis ................................................................................... 17

Table 4.2
Sample calculation for Sample 1,
Na/Al ratio examined under low vacuum ......................................................... 19

Table 4.3
Comparison of standard deviation as a percent of the average
ratios for high vacuum, low vacuum, and sputter coated samples ................... 20

Table 4.4
Comparison of spot mode vs. reduced raster analysis of glass
from three sources .............................................................................................. 21

Table 4.5
Comparison between three iterations and five iterations of each sample ........ 22

Table 4.6
Number and percent of same source automotive glass included
and excluded based on statistical model for inclusion ...................................... 25

Table 4.7
Number and percent of same source samples included, excluded,
or deemed inconclusive using the statistical model for inclusion .................... 26

Table 4.8
Summary of Data ............................................................................................... 32

Table 5.1
Comparison of the discriminating power of refractive index
measurement and SEM-EDS when used alone and in combination ................ 36

Appendix Table 2.1:
Elemental Data: high vacuum, low vacuum, and sputter coated samples ....... 43

LIST OF TABLES, continued

Appendix Table 3.1
Elemental Data: Tilt Angle Effects ................................................................... 49

Appendix Table 4.1
Summary of Elemental Ratio Comparisons in the Establishing
of a Statistical Model for Inclusion ................................................................... 53

Appendix Table 5.1:
Refractive Index Comparisons .......................................................................... 55

Appendix Table 6.1:
186 Elemental Comparisons of Samples
not Distinguished by Refractive Index .............................................................. 58

Appendix Table 7.1:
Summary of Refractive Index Data ................................................................. 105

Appendix Table 7.2:
Summary of Elemental Comparisons for Samples with
Different Refractive Indices ............................................................................ 105

Appendix Table 7.3:
Summary of Elemental Comparisons for Samples with
Indistinguishable Refractive Indices ............................................................... 105

Appendix Table 7.4:
Summary of Elemental and Refractive Index Comparisons .......................... 106

Appendix Table 7.5:
Summary of All Comparisons ......................................................................... 106

LIST OF FIGURES

Figure 3.1
The mathematical relationship between the speed of light in
a vacuum and the speed of light in a transparent medium

determines the refractive index of the sample ................................................... 11
Figure 4.1

The tilt angle of the surface of a sample may range

from perfectly flat to highly sloped ................................................................... 28

miner 1: Introduction

Forensic Importance and Analysis of Glass

It has long been recognized that the analysis of glass is an important tool of the
forensic laboratory. In many cases, glass can be a valuable piece of evidence that links a
suspect to a crime. For example, the glass fragments found in the clothing of a hit-and-
run victim may be compared to glass from the broken window of the suspect’s car. Such
a comparison might traditionally have been accomplished by comparing the optical and
physical properties of the two samples to determine a possible common origin. Thus, the
forensic analysis of glass has traditionally relied upon the variability of optical and
physical properties amongst glass samples of different origins. However, the
enhancement of quality control measures in the manufacture of glass has led to a decrease
in the variability of these properties between glass samples, even those from different
sources. This decrease in variability of optical and physical properties has highlighted
the advantages of also comparing the trace elements present in glass samples, rather than
relying on optical and physical properties alone.

The elemental analysis of glass may be accomplished by such instrumental
methods as neutron activation analysis, inductively coupled plasma mass spectrometry,
and atomic absorption. Also, scanning electron microscopy coupled with energy
dispersive spectroscopy (SEM-EDS) may be used to detect the presence of trace elements
in glass samples.

This study focuses on the elemental analysis of glass by variable pressure SEM-
EDS as a means to distinguish between samples from different sources that share similar
optical properties. Some components of this study include the demonstration of
appropriate instrument operating parameters and the implementation of a statistical model

for inclusion.

 

pt

88]

sar
616
this
firs

suit

First

for 11

The Problem to be Studied

The analysis of glass samples in the forensic laboratory traditionally relies on
comparison of the optical and physical properties of the samples. Optical comparison of
glass is made by comparing the refractive index of the questioned sample to that of the
known sample. Physical comparison is made by comparing the color and thickness of the
two samples, and also by comparing the densities of the two samples. However, it has
been shown that the refractive index of a piece of glass is directly related to its density
and the correlation between the two properties is 0.93.i With such a relationship, forensic
scientists generally measure only refractive index as the measurement of density is much
more difficult and does not add to any conclusions formed after comparison of refractive
index.

The forensic scientist would thus compare two samples based on color, thickness
and optical properties. Samples with similar properties are concluded to have possibly
arisen from the same source. It has been demonstrated that samples with similar optical
properties may be better distinguished by comparing the elemental composition of the
samples.ii

Thus, the fundamental question posed and answered by this study is: may two
samples of glass with similar optical properties be distinguished by analyzing for specific
elements using variable pressure SEM-EDS? However, in order to arrive at an answer to
this primary question, a method by which elemental analysis may be accomplished must
first be demonstrated. In this study, variable pressure SEM-EDS will be shown to be a
suitable method by which elemental composition can be examined in a substance such as
glass.

Therefore, the problem to be studied is best answered in a step-wise fashion.
First, the appropriate operating conditions of the SEM-EDS system must be established

for the analysis of glass. Subsequently, the ability of SEM-EDS to distinguish between

samples of glass originating from different sources but sharing similar Optical properties

will be demonstrated.

Purpose of the Study

As described, it is becoming increasingly important to procure an analytical
method that is able to supplement the traditional optical analyses. Such a method should
be able to detect variation and discriminate between pieces of glass that did not come
from the same source. However, with hundreds of thousands of glass formulationsi“, it is
desirable that the proposed method be suitable for analysis of every type of glass
encountered in the forensic laboratory. SEM-EDS is a method that is potentially able to
distinguish between pieces of glass from different sources, and is applicable to a large
number of elemental compositions. In this study, the combination of SEM—EDS and
analysis of optical preperties will be researched in order to highlight the enhanced
discriminating power of associating optical analysis with SEM-EDS as opposed to optical
analysis alone. Also, this study will focus upon the development of a statistical model for
inclusion based on elemental ratios, and the demonstration of optimal SEM instrument
parameters for glass analysis.

Research of this nature will have several beneficial outcomes. First, as this
research is being conducted in conjunction with the Michigan State Police, this project
will serve to implement a protocol for the comparison of glass samples based upon the
combination of measurement of optical properties and elemental analysis. This protocol
will be used by the Michigan State Police Forensic Science Division in casework
situations. Secondly, this study will highlight the superlative power of elemental analysis
to discriminate between glass samples from different sources that share the same
refractive index. Finally, this study will demonstrate effective instrumental parameters

for variable pressure SEM-EDS analysis of glass, and present a statistical model for

inclusion based on elemental ratios. In combination, these benefits will enhance the

ability of the forensic scientist to discriminate between samples of glass with similar
optical properties thereby reducing the possibility of a suspect being falsely implicated in

a crime.

Hypothesis

Upon development of appropriate instrumental operating parameters and
construction of a statistical model for inclusion, it is predicted that performing elemental
analysis subsequent to Optical analysis will result in the increased ability of the forensic
scientist to distinguish between samples of glass not able to be distinguished by optical
analysis alone. This is predicted to be so, as the manufacturing of glass (discussed in
Chapter 2) creates a product with uniform optical properties but introduces trace elements
into the glass in a random fashion. This creates the variability of elemental composition
that is predicted to assist the forensic scientist in distinguishing between similar glass
samples. Furthermore, this hypothesis is supported by the work of Reeve, Mathieson and
F ong, which demonstrated that 100% of 81 glass samples analyzed could be
distinguished by combining optical analyses with traditional high vacuum SEM-EDS.”

Limitations

This study focuses exclusively on the use of variable pressure SEM-EDS as a
method for elemental analysis of glass and thus will not examine other instrumentation
such as neutron activation or mass spectrometry techniques. Furthermore, the entirety of

the samples examined in this study will be tempered automotive glass.

Methodology
This study contains primary research in elemental analysis by variable pressure

SEM-EDS, and optical analysis by refractive index measurement.

The research was conducted at the Michigan State Police Laboratory in East
Lansing, Michigan, using a LEO 435 variable pressure SEM with EDAX Phoenix x-ray
detector and software. Refractive index measurements were taken using the Emmond's
double variation method using equipment and techniques already in place at the East
Lansing Laboratory.

Elemental analysis was conducted by quantitating the amount of tin, sodium,
aluminum, magnesium, potassium and calcium on the surface of each glass sample. The
amount of each element in each piece of glass was used in constructing specific elemental

ratios (Table 1.1) that were calculated for each sample.

 

in:Sodium Sn/Na lMagnesiumzAluminum Mg/Al
SodiumzAluminum Na/Al Sodiuszagnesium Na/Mg
Calcium:Sodium Ca/N a Calcium:Potassium Ca/K

 

 

 

 

 

 

Table 1.1: Elemental ratios constructed for all elemental comparisons.

Suitable operating parameters for the variable pressure SEM were established by
comparing the standard deviation of the elemental ratios for each sample. The criteria
that were examined to establish appropriate operating parameters include: size of raster
area on the sample, the number of repeat measurements for each sample, the pressure of
the sample chamber, and if the sample requires conductive coating. Furthermore, the
effect of tilt angle on the sample was investigated.

Upon determination of appropriate operating parameters, a statistical model for
inclusion was developed. A statistical model was employed to distinguish between the
natural variation inherent in a piece of glass and the variation that exists between pieces
of glass from different sources. This model established a statistical range applicable to
each sample so that when comparing two samples of glass, the analyst is able to
determine if the results are inclusive (the samples could have come from the same source)

or exclusive (the samples did not come from the same source).

Finally, 50 tempered automotive glass samples were collected from local
junkyards and subjected to analysis by variable pressure SEM-EDS and refractive index
measurement. Subsequent pair-wise comparisons yielded 2550 comparisons, upon which

the number of false inclusions by refractive index and by SEM-EDS was calculated.

Chapter 2: Scanning Electron Microscopy and the Analysis of Glag

Scanning Electron Microscopy

The scanning electron microscope is an instrument that is primarily used to
produce magnified images of a sample. Fundamentally, the scanning electron
microscope (SEM) uses a beam of electrons to excite the electrons on the surface of the
sample. These excited electrons, called secondary electrons, are emitted from the surface
of the sample and are captured by a detector that enables a magnified image to be
projected to a television or computer monitor. Along with the secondary electrons, x-
rays are also emitted from the surface of the sample. Each x-ray emitted has a specific
energy characteristic of the element that produced it. An x-ray detector may be coupled
with the SEM permitting detection and spectroscopic analysis of the emitted x-rays. This
spectroscopic capability is referred to as Energy Dispersive Spectroscopy or SEM-EDS.

SEM-EDS is traditionally performed in a high vacuum environment in order to
maintain the electron beam, and generally with samples that are conductive or semi-
conductive. Non-conductive samples, such as glass, are problematic to analyze under the
traditional SEM vacuum conditions. This is because a nonconductor tends to build up
electrical charge when bombarded with the beam of electrons. This built up electrical
charge deflects the secondary electrons resulting in inaccurate quantification of elemental
concentrations. In order to prevent charging artifacts, non-conductive samples are
commonly coated with a conductive substance, such as carbon, to prevent the build up of
negative charge. Usually, the sample is coated with carbon in a process called sputter
coating. However, in a forensic case, it is not desirable to sputter coat evidentiary
samples as this process adds foreign material to the evidence, thereby altering the
composition of the evidence prior to analysis.

In a high vacuum environment it may be possible to perform SEM-EDS on a non-

conductive sample without a conductive coating provided the proper accelerating voltage

 

 

 

 

the

pn
to

M

Spl

ch:
the
dis

are
cle:

ma

Ma

is selected. Charging effects may be reduced if the accelerating voltage is chosen so that
the electron emission coefficient is greater than or equal to 1.0. However, selection of a
proper accelerating voltage is complicated by the fact that the electron emission
coefficient varies between samples and between different points on the same sample,
which makes the technique inordinately difficult to employ in an automated analysis.

Recently, non-conductive samples have been able to be analyzed without being
sputter coated by use of a variable pressure SEM. The variable pressure SEM is better
able to handle glass and other non-conductive samples as it maintains the sample
chamber at a much higher pressure (approximating that of the outside environment) than
the traditional SEM. The higher pressure in the variable pressure SEM is able to
dissipate the built up electrical charge on the sample, thereby reducing charging effects
and improving the accuracy of elemental analysis. Thus, it is possible to accurately
analyze a non-conductive sample such as glass by using a variable pressure SEM.

The analysis of glass by SEM-EDS is performed by examining a point or small
area on the surface of the glass for trace elements such as tin and sodium. These
elements are deposited randomly on the surface of the glass as a result of the

manufacturing process.

Manufacture of Glass

A process first developed in 1959 is used in the manufacture of flat glasses such
as those found in windowpanes." This process begins with the mixing of silica, metals
(such as sodium and magnesium), metal oxides (such as calcium oxide), and recycled
glass. These raw materials are passed from a mixing silo into a melting furnace where
they are heated to become molten. The molten glass is then poured continuously from
the fumace onto a shallow bath of molten tin. As the glass is highly viscous and the tin is
very fluid, the two do not mix and the glass spreads out to form a flat surface. The

molten glass floats on top the bath of tin, and is gradually cooled as it floats down the

tank. Glass prepared in this way is termed "float glass". The float glass forms a
continuous ribbon over three meters wide and three- to nineteen millimeters thick.Vi The
thickness of the glass is controlled by the speed at which the solidifying glass ribbon is
drawn off the bath.Vii The glass is then cooled, subjected to quality checks, and cut into
sheets. As mentioned in the previous chapter, manufacturers assure the quality of their
product by checking optical and physical properties, such as the density and refractive
index of the glass.

This process cheaply and quickly produces glass with virtually parallel surfaces
and uniform thickness. However, it is by this same process, from the raw materials to the
float bath, that elemental variation is introduced to the glass, providing forensic scientists
with a useful characteristic for analysis.

Although the main composition of the glass is carefully controlled, metallic
elements may be randomly deposited on the glass as it passes over the bath of molten tin.
The float tank containing the molten tin is operational 24 hours a day for approximately
ten years."iii Over this period of time, metals are randomly introduced to the molten tin as
the melting furnace corrodes, and metals from the glass ribbon are retained in the tin
bath. As the molten glass ribbon floats over the tin bath, tin and the other random metals
present in the float tank adhere to the glass, giving the ribbon of glass a coating of
metallic elements on the side of the glass that was in contact with the molten tin (known
as the "float side"). As the concentrations of the metals present in the tin bath is
consistently changing, the concentration of trace metals on the float side of the glass will

vary between different areas of the glass ribbon.

Elemental Analysis of Glass
As the float side of the glass now contains specific amounts of trace metals, SEM-
EDS may be used to determine which metallic elements are present on the surface of the

glass, and how much of each element there is. For example, some metallic elements

commonly found on glass include aluminum, calcium, sodium, tin and magnesium. It has
been demonstrated that trace element concentrations are relatively constant across a sheet
of glass, and vary between different sheets of glass.ix Thus, the concentrations of these
elements on one piece of glass can be compared to the concentrations found on a second

piece to determine if the two pieces could have originated from the same sheet of glass.

Summary

The process by which float glass is manufactured results in variation in the
concentration of trace metals on the float side of the glass. These metals can be analyzed
by SEM-EDS, and the concentration of the metals present on different pieces of glass can

be compared to determine if the glasses could have come from the same source.

10

 

 

Q

 

re fr

spe

ligl
ind

pas

con
Pro

C011

Chapter 3: Review of L_it_er_a_tu£
Comparison of Glass Based Upon Refractive Index Measurement

Optical comparison of glass is most commonly based on the measurement of the
refractive index of the glass. Refiactive index is defined by the relationship between the
speed of light in a vacuum and the speed of light in a transparent medium, such as glass.
In practical terms, the speed of light in a vacuum is considered to be equal to the speed of
light in air. The refractive index is calculated to be a numerical value, such as 1.5, which
indicates that for a specific sample, the speed of light is decreased by 1.5 times when it

passes from the air into the sample.

S eed of li tin a vacuum
Refractive _ p gh

Index Speed of light in
a transparent medium

 

Figure 3.1: The mathematical relationship between the speed of light in a vacuum
and the speed of light in a transparent medium determines

the refiactive index of the sample.

As discussed in Chapter 2, refractive index is monitored for purposes of quality
control in the manufacturing process. The increasing sophistication of the manufacturing
procedures and the effect on refractive index is demonstrated by GA. Brown, who
compared the distribution of refractive indices of older (greater than 30 years old) and
newer (less than 30 years old) glass samples" Brown found that the unimodal distribution
of sheet glass had a refractive index range of 1.5155 to 1.5159. However, the total spread
of refractive indices of older glass was greater than that of newer glass. According to
Brown, the decreased range of reflective indices in newer glass is due in part to the
change in manufacturing procedures over time. This observation has been made by other

scientists, including M.D.G. Dabbs who agrees with Brown's conclusion that the

11

"difference in refractive index distribution with the age of the glass is caused by
improved quality control of raw materials and composition" of the glass.Xi

In the forensic laboratory, the narrow distribution of refractive indices was
observed by D. Hickman who stated that the most often occurring glass in casework was
colorless glass with a refractive index of 1.5125 to 1.5250."ii

The observations of Brown and Hickman indicate that the range of refractive
indices for colorless glass is exceedingly narrow, with differences occurring only at the
hundredths-, or thousandths-decimal place. As improvements in the manufacturing
process serve to narrow this range even further, it is becoming increasingly important to

use additional means of comparison to discriminate between two glass samples.

Elemental Analysis

The significance of elemental composition was demonstrated by Koons and
Buscaglia in 1999. Their research measured elemental concentrations of several elements
by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). In pair-wise
comparison of the elemental compositions of over 200 samples, most samples could be
distinguished by their elemental composition."iii The researchers concluded that the
measurement of elemental concentration and refractive index provide a "high source-
discrimination capability" between glass fragments.“

Also operating on the premise that trace elements arising from the manufacturing
process can characterize glass samples, J. Andrasko and AC. Maehly sequentially used
refractive index, SEM-EDS, and emission spectroscopy to distinguish fragments of
glass."v The researchers found that SEM-EDS demonstrated good reproducibility of
results when specific elemental ratios (Na/fMg, Na/Al, Mg/Al, Ca/N a, Ca/K and Ba/Ca)
were used for comparison. When these elemental ratios were compared between 29

samples with indistinguishable refractive indices, only 4 remained indistinguishable.

Emission spectroscopy was performed subsequent to SEM-EDS and 2 samples remained

12

indistinguishable. Based on this result, Andrasko and Maehly recommend that SEM-
EDS be performed on samples of glass that are indistinguishable by refractive index
alone.“ Emission spectroscopy may be used subsequent to SEM-EDS, however the
researchers caution that the technique is destructive and requires at least one milligram of
sample -rest1ictions that may preclude analysis of many forensic samples.

Andrasko and Maehly also analyzed the effect of sample tilt angle on the
calculation of elemental concentrations."Vii The sample tilt angle is the angle to which the
surface of the sample is tilted in relation to the stage or surface of another sample. At a
sample tilt angle of sixty degrees, the Na/Mg ratio remained relatively constant, although
the signal intensity (and elemental concentration) varied highly. Therefore, the
researchers concluded that in order to obtain reproducible results carefirl attention must
be paid to surface geometry and the manner in which the sample is mounted for analysis.

In similar research, V. Reeves, J. Mathiesen, and W. Fong investigated the ability
of SEM-EDS to distinguish between samples of glass based on a series of elemental
concentrations ratioed to the concentration of calcium in the sample.“ii Using high-
vacuum SEM-EDS, the researchers report that two out of eighty-one samples were
indistinguishable by SEM-EDS alone. However, when coupled with refractive index
measurement, 100% of the samples were able to be distinguished.

From these results, and the results of the previous research, it has been
demonstrated that elemental analysis can lead to discrimination of glass samples. Most
important is the observation of the increased discrimination capability that is achieved

when refractive index measurement is coupled with SEM-EDS in the analytical scheme.

Summary
While refractive index measurement is a widely accepted, convenient, and
sensitive technique for glass comparisons, its power to discriminate between glass

samples from different sources is diminished. Improved quality control in the

13

manufacture of glass has decreased the distribution of refractive indices to a range that
differs only in the thousandths place.“ Consequently, forensic scientists must turn to
more discriminating analytical techniques, specifically elemental analysis.

Comparative analysis of the elemental composition of glass samples is
appropriate for forensic work as the elemental composition of the glass is more variable
which may permit the discrimination of two otherwise indistinguishable samples.
Various methods of elemental analysis have been investigated, including SEM-EDS.

It has been demonstrated that when refi'active index measurements are combined
with elemental analysis by SEM—EDS, a high proportion of physically indistinguishable

1

samples may be distinguished.xx xx

14

 

sup
ED
|
qua.
the 1
anal.

expe

glass

First.
Seton
for inc
autonu

the sun

SEM-E1

 

Automa
T
an31131.8 C
ficflhak,;
mksmdy
All
be Perform

requires 1h;

Chapter 4: Development of Suitable SEM-EDS Operating Parameters

As it is becoming increasingly important to procure an analytical method to
supplement the traditional optical analyses of glass, the use of a variable pressure SEM-
EDS system was investigated. This system allows for the identification and
quantification of trace elements, and so is suitable for the analysis of elements found on
the float surface of a sample of glass. However, before SEM-EDS is introduced into the
analytical scheme, a protocol for appropriate use must be prescribed. Therefore, through
experimentation, a suitable set of operating parameters was developed for the analysis of
glass by variable pressure SEM-EDS.

The issues addressed in the method developed by this research are four-fold.
First, the appropriate sample chamber pressure level for this analysis was investigated.
Second, electron beam scanning parameters were established. Next, a statistical model
for inclusion was prepared and evaluated using repeat analyses of single-source
automotive glass. Finally, the effect of sample tilt-angle was investigated to determine if
the surface orientation of the sample limited the precision of elemental analysis.

However, to facilitate the investigation of these issues, the basic operation of the

SEM-EDS system was automated.

Automation of the SEM with EDAX

The EDAX Phoenix EDS system allows for the automation of the SEM-EDS
analysis of multiple samples. The automation of the system was set up in order to
facilitate and expedite the analysis of the large volumes of samples that were analyzed in
this study and in some casework situations.

Although the system may be automated, there are multiple requirements that must
be performed by the operator in order for proper operation of the system. The system

requires that the operator choose and prepare his own samples, and insert them into the

15

stage in the SEM sample chamber. The operator must then choose the parameters of
pressure, working distance and elements to be quantitated. Also, the operator must
indicate the number of repetitions of the analysis and the size of the area on the sample to
be analyzed. Upon completion of these tasks, the collection of data is performed
automatically by the EDAX system.

Sample preparation for the automated analysis is quite straightforward. First, a
small cube of tempered automotive glass was washed in several milliliters of acetone in a
sonic bath. The sample was sonicated for five minutes and then removed from the bath.
The sample was then observed under short wave ultra violet light (254 nm) to determine
which side was the float side of the glass. The float side fluoresced a bright yellow and
the sample was mounted float-side up on an aluminum SEM stub with double-sided
carbon tape.

The sample was then placed in the stage of the sample chamber in the SEM. The
stage may hold up to seven samples, and so the automation program was set up to
analyze fi'om one to seven samples.

The program was arranged to analyze for the elements tin (Sn), sodium (Na),
magnesium (Mg), aluminum (Al), potassium (K), and calcium (Ca). These elements
were chosen for analysis because they are commonly used in float glass manufacture as
raw materials or components of the float bath, and so would be expected to be found in
samples of float glass. Each sample was scanned for 100 live seconds at a count rate of
approximately 2500 counts per second. In a run of multiple samples, the analysis begins
at sample one. Sample one is scanned for 100 live seconds and then the stage
automatically moves to sample two so that sample two is scanned next. For a second
iteration, the samples are scanned alternately so that no sample is scanned twice in a row.
This alternating analysis was designed to minimize potential problems such as charge
build up on the sample, and to ensure that instrumental variation does not accrue between

the analysis of sample one and sample two. Upon completion of analysis, the elements

16

are quantitated automatically and concentrations are tabulated. The elemental data is
then manually transferred to a spreadsheet template that calculates the elemental ratios
listed in Table 1.1. The spreadsheet also calculates the average elemental ratio for each
ratio in a sample, and the standard deviation for multiple analyses is also calculated

(Table 4.1).

 

 

 

Sample lNa/Al A1
14.01652 4.206948

Std Dev= 1.300133 Std Dev= 0.458536
Lo= 10.11612 Lo= 2.83134
Hi= 17.91692 Hi= 5.582556

4.283033 1.484822

Std Dev= 0.206776 Std Dev= 0.101284
Lo= 3.662706 Lo= 1.180971
Hi= 4.90336 Hi= 1.788674

 

 

 

 

 

 

Table 4.1: Sample spreadsheet used to calculate elemental ratios, averages

and standard deviations from data obtained in the automated analysis.

Upon construction of the automated analysis program and calculations template,
the operating parameters for the analysis of glass were established. First, suitable sample

chamber pressure conditions were established.

Selection of Suitable Sample Chamber Pressure Conditions

As described in Chapter 2, SEM-EDS is generally performed in a vacuum with
samples that are conductive. When a non-conductor such as glass is analyzed under high
vacuum conditions in the SEM, it will build up a negative charge if it is not coated with a
conductive coating. Rather than using a conductive coating, non-conductive samples
may also be analyzed under a low vacuum (higher pressure) to reduce the charging

effects present at higher vacuum levels.

17

 

El

wi

Prt

auto
sam
scan
the e
autor
samp

and st

 

Pa A
and Sta
was Sp]

high \a

To determine if analysis under a low vacuum is appropriate for SEM-EDS
analysis of glass, fourteen tempered automotive glass samples were analyzed by SEM-
EDS at high vacuum, then at low vacuum, and finally at high vacuum after being coated

with carbon.

Procedure

The fourteen samples, divided into two groups of seven, were analyzed by the
automated elemental analysis program previously implemented on the SEM-EDS. The
samples were first analyzed at high vacuum (0.01 Pa). Each sample was alternately
scanned in one area five separate times for 100 seconds each time. The quantitation of
the elements tin, sodium, magnesium, aluminum, potassium, and calcium was performed
automatically by the EDAX system. The elemental concentrations present in each
sample were then transferred to the spreadsheet where the appropriate ratios, averages
and standard deviations were calculated.

Next, the same samples of glass were re-analyzed under a low vacuum level of 10
Pa. Again, the concentrations of the six elements were calculated and the ratios, averages
and standard deviations were tabulated in the spreadsheet. Finally, each sample of glass
was sputter coated with carbon. The carbon-coated samples were then analyzed under
high vacuum (0.01 Pa) and elemental ratios were again calculated.

This process resulted in three sets of elemental ratios, averages and standard
deviations: one set for each pressure level. Within each data set were elemental ratios,
averages and standard deviations that were calculated individually for each of the
fourteen samples. For example, the average of the five measurements of the Na/Al ratio
for sample one were calculated at each pressure level. The value of one standard
deviation from this average was also calculated at each pressure level. Then, to

determine the standard deviation as a percent of the average, the standard deviation was

18

divided by the corresponding elemental ratio average and the result multiplied by 100%

(Figure 4.2). This calculation was performed at each pressure level.

 

Ratio Avg. Std Dev Std Dev as %
Na/Al 6.571 0.3113 (0.3113/6.571) x 100%
= 4.737

 

 

 

 

 

 

 

 

Figure 4.2: Sample calculation for Sample 1, Na/Al Ratio

Examined under Low Vacuum.

The same calculations were performed for the other elemental ratios of sample
one, and also for each of the other samples. The procedure was repeated for fourteen
glass samples, each analyzed at 3 different pressure levels.

Upon completion of the calculations for each of the fourteen samples, there were
252 total standard deviations calculated as a percentage of the elemental ratio average.
Eighty-four of these standard deviations belonged to the data collected at low vacuum.
These eighty-four values were averaged to reach the average standard deviation from the
elemental ratios collected at low vacuum. The averages for the data collected at high
vacuum and with sputter coated samples were also calculated.

The results of these calculations were compared to determine the most precise
method of analysis. As shown in Table 4.3, the lowest standard deviation as a percent of
the average was shown by those samples analyzed at low vacuum. This indicates that the
most precise method of analysis of glass by SEM-EDS requires that the samples be
analyzed under low vacuum. The data collected fi'om elemental analysis at high vacuum
was less precise than the other two methods. And while the use of sputter-coated samples
was more precise than analysis of uncoated samples at high vacuum, it was less precise

than analysis of uncoated samples at the low vacuum level.

19

 

 

 

 

 

Standard Deviation
as
Mode of Operation Percent of Ratio
Low Vacuum 11.45
igh Vacuum 19.83
Sputter Coated 16.2

 

 

 

 

Table 4.3: Comparison of Standard Deviation as a Percent of the Average Ratios

for High Vacuum, Low Vacuum, and Sputter Coated Samples.

Therefore, it is concluded that a low vacuum level is most favorable to use in SEM-EDS

analysis of glass samples, as it is more precise.

Parameters for Analysis

The two parameters that must be established for all analyses of glass by this
automated SEM-EDS method are the number of iterations for each sample, and the size
of the area of analysis.

To determine the appropriate size of the area of analysis, seven samples of glass
were obtained fi'om a single source. Each piece had a float surface area of approximately
25 m2. Each of the samples was mounted, float side up, on an aluminum SEM stub
with double-sided carbon tape. The samples were then analyzed by automated SEM-
EDS method, using two different sized areas of analysis.

First, each of the seven samples was analyzed in spot mode using a single point as
the area of analysis. Then, the same seven samples were analyzed using a small area
across which the electron beam is scanned or "rastered". This area, called a reduced
raster, measured 2.25 x 10.2 um.

Upon quantification of the elements tin, sodium, magnesium, aluminum,
potassium, and calcium, elemental ratios were calculated. The average for each

elemental ratio, along with the corresponding standard deviation, was calculated for each

20

sample. The elemental ratio averages for the first sample were compared to the standard
deviations from the ratio averages of the other six samples in order to determine if the
first average fell within one, two, three, or greater than three standard deviations from the
average of the second sample. This comparison was performed with each of the six ratios
for each of the seven samples being compared to the standard deviation of the other six
samples, which resulted in 252 comparisons. These comparisons were performed for the
data obtained using the reduced raster analysis, and then for the analysis done in spot
mode.

This procedure was repeated with fourteen more samples (two sets of seven
samples from each of two separate sources) resulting in 504 more comparisons of data
obtained using spot mode, and 504 comparisons of data obtained using a reduced raster.
According to the theory of the normal curve, 99% of the average ratios for a sample
should fall within 3 standard deviations of the average ratio of another sample obtained
from the same source.”ii The results of the comparison of spot mode versus reduced
raster mode (Table 4.4) indicate that for two out of the three sources examined, the
distribution of elemental ratios more closely approximates a normal distribution when the

analysis occurs over a reduced raster area, rather than at a single point.

 

% Within 3 Std Dev. Dodge Chrysler Mercury
Spot Mode 90.87 92.86 97.62
Reduced Raster 92.46 95.24 95.4

 

 

 

 

 

 

 

 

Table 4.4: Comparison of Spot Mode vs. Reduced Raster Analysis

of Glass from Three Sources.

Therefore, in the analysis of glass by SEM-EDS it is more precise to analyze a
small area of the sample, rather than just a single point.
The other parameter for analysis that must be examined is the number of

iterations of the analysis that must be performed in order to assure a precise result. While

21

the requirement of precision must be satisfied, it must also be balanced with limitations
of time. Although performing hundreds of repeat measurements on each sample would
demonstrate the reliability of the measurements, these tests would take days to complete
under ideal circumstances. While precision is the goal, practicality is the guide.

Under this principle, fourteen samples of glass from different sources were
examined by SEM-EDS under a low vacuum, and using a reduced raster analysis. The
samples were prepared by washing each with acetone in a sonic bath, and then mounting
each one, float side up, on an aluminum stub with carbon tape.

The samples were scanned for 100 live seconds each, and then alternately
rescanned twice more, for a total of 3 iterations. For each sample, elemental ratios were
calculated for each of the three iterations. Then, for each of the six ratios, the average of
the three iterations was calculated along with the corresponding standard deviation.
Finally, the average standard deviation for each ratio was calculated.

The procedure was then repeated with the same samples, only the samples were
alternately scanned for a total of five iterations. The same calculations were performed,

and the results compared in Table 4.5.

 

 

 

Standard Deviation as
Percent of the Average
Ratio ,1); ix
Na/Al 16.18 12.42
Mg/Al 15.53 1 1.28
Na/Mg 11.36 3.50
Ca/Na 11.47 2.79
Ca/K 27.87 15.66
Sn/Na 30.81 19.46
Total 113.22 65.11

 

 

 

 

Table 4.5: Comparison between Three Iterations and Five Iterations of Each Sample

These results indicate that analysis done with five iterations shows a lower

average standard deviation for all elemental ratios. F urtherrnore, when the average

22

standard deviations are totaled, the analysis done with three iterations shows a higher
total than that done with five iterations. As standard deviation is a measure of precision,
these results indicate that precision of analysis is improved when five iterations are
performed. Therefore, analysis done with five iterations has an overall greater precision
than does an analysis done with three iterations.

As increasing the number of iterations from three to five adds only approximately
200 seconds per sample to the analysis time, analysis done in five iterations remains
practical. Therefore, it is recommended that by this method, samples be analyzed in an

alternating manner for a total of five iterations per sample, rather than three.

Statistical Model for Inclusion

Upon determination of appropriate operating parameters, a statistical model for
inclusion was developed using the designated parameters. This model establishes a
statistical range applicable to each sample so that when comparing two samples of glass,
the analyst will be able to detemrine if the results are inclusive (the samples could have
come from the same source) or exclusive (the samples did not come from the same
source). The inclusion or exclusion of a sample is based on a statistical range rather than
an absolute number because of random error. Random error is introduced into every
scientific measurement as each measuring device itself has a limit to the reproducibility
of its measurements. This results in slightly different elemental concentrations being
calculated for different analyses of the same piece of glass. Thus, although the calculated
numbers are not exactly the same, they may fall within a range that statistically permits
the calculated values to be considered to be consistent.

For this reason, a statistical range must be established which will allow the analyst
to determine if two samples of glass with different elemental concentrations should be
considered to have come from the same source. This particular statistical range has been

termed the "statistical model for inclusion" in this study.

23

The statistical model for inclusion was established based on the normal curve
described earlier. The normal curve predicts that repeated measurements of a sample,
taken under the same conditions will yield results that cluster around some central value.
The normal curve predicts that 99.7% of the measurements taken for a particular sample
will fall within 3 standard deviations from the mean, or central value. Thus, the first part
of the statistical model for inclusion says the average elemental ratio of the questioned
sample must fall within 3 standard deviations from the mean of the known sample.
However, the normal curve is a theoretical distribution, and so an additional factor of 5%
was included to account for natural sample and instrumental irregularities. With this
consideration, the statistical model for inclusion was stated as follows: if the average
elemental ratio of the questioned sample falls within the greater of the range of plus-or-
minus three standard deviations or 5% of the average ratio of the known sample, then the
samples are concluded to have possibly come from the same source.

To test the statistical model for inclusion, seven glass samples were taken from
each of four separate sources. Each set of seven samples was analyzed by the automated
method previously developed. The elemental ratios described in Table 1.1 were
calculated, and then compared.

The elemental ratio averages were compared in the following manner. The
average of the first elemental ratio, Na/Al, for the first sample was compared to the
average Na/Al ratio for the second sample from the same source. The object of the
comparison was to determine if the average elemental ratio of the first sample would fall
within the greater of the range of plus-or-minus three standard deviations or 5% of the
average ratio of the second sample.

This comparison was carried out for all ratios of sample 1 being compared to the
statistical ranges for samples 2 through 7. Then, the comparison was performed between
the elemental ratios of sample 2 and the statistical range for the ratios of the other 6

samples (1,3,4,5,6,7). This was carried out so that each elemental ratio average for each

24

sample was compared against the statistical range for every other corresponding
elemental ratio average for every other sample from the same source. The procedure was
repeated for each of the 7 samples originating from each of the 4 sources.

This resulted in 252 comparisons for each source. Each comparison resulted in a
designation of "included" if the elemental ratio average fell within the range of 3 standard
deviations or 5% of the elemental ratio average it was being compared against.
Conversely, if the average did not fall within the prescribed range, it was given a
designation of "excluded". A summary of these results is given in Appendix V.

Of all 1008 comparisons, 96% were "included" while 4% were "excluded" (Table

 

 

 

 

 

 

4.6).
Source
1 2 3 4 Total Percent
cluded 240 238 240 25 1 969 96.1%
Excluded 12 14 12 1 39 3.9%

 

 

 

 

 

 

 

Table 4.6: Number and Percent of Same Source Automotive Glass Included and

Excluded Based on Statistical Model for Inclusion.

As described earlier, in a standard normal distribution, 99% of all results will fall
within 3 standard deviations of the average. In this study, it was shown that 96% of all
results fell within the larger of 3 standard deviations or 5% of the average. The range of
three standard deviations is then appropriate. If the statistical range had been narrowed to
two standard deviations, a larger portion of the measurements of the same sample would
have been falsely excluded. Similarly, if the range had been widened to include results
that fell within 4 standard deviations from the average, a number of samples from
different sources would have been falsely included. The statistical model using a range
of three standard deviations or 5% of the average as a basis for inclusion limits the extent
of false inclusions and false exclusions. Thus, this model was accepted as an appropriate

statistical range for inclusion and exclusion.

25

Therefore, the statistical model for inclusion establishes that for all ratios
compared, if the average of the five measurements of elemental ratios of the questioned
glass fit within the greater of a three standard deviation or 5% range of the five
measurements of the known glass, it can be concluded that the samples could have
originated from a common source. Conversely, if the average elemental ratios of the
questioned glass do not fall within the range of three standard deviations or 5% of the
average of the known, it is concluded that the samples could not have originated fiom the
same source. If only one or two of the six ratios fall outside the prescribed range, the
samples must be reanalyzed before exclusion. If, after reanalysis, one or two ratios still
fall outside of the prescribed range, the result is inconclusive. That is, it can neither be
said that the questioned sample originated from the known, nor can it be said that it did
not originate fi'om the known. The result in this case must be inconclusive because the
elemental concentration of the questioned falls outside the prescribed range of the known,
but within the range of variation within the known. This variation was demonstrated
above as 4% of known same source elemental ratios fell outside the range demonstrated
by other samples from the same source.

Using these criteria, upon reanalysis of the samples used to establish the statistical
model for inclusion, no sample was falsely excluded (Table 4.7). A summary of the

results can be found in Appendix V.

 

 

 

 

 

Number Percent
eluded 135 80.4%
Excluded 0 0.0%
Inconclusive 33 19.6%
Total 168 100.0%

 

 

 

 

 

Table 4.7: Number and Percent of Same Source Samples Included,

Excluded or Deemed Inconclusive Using the Statistical Model for Inclusion.

26

Therefore, based on data obtained from elemental analysis and by using the
statistical model established above, the analyst has the capability to determine if a

questioned piece of glass could have originated from a known source or not.

Summary

A method for the elemental analysis of glass was established. This method
requires that the samples be analyzed under a low vacuum of 10 Pa. Each sample was
analyzed five separate times using a reduced raster (2.25 x 10.2 um) for 100 live seconds
each time. The elements quantitated were tin, sodium, magnesium, aluminum,
potassium, and calcium. The elemental concentrations were paired in the following ratios
for each of the five iterations: Sn/Na, Na/Al, Mg/Al, Na/Mg, Ca/Na, Ca/K. The average
of each ratio (taken over the five iterations per sample) was calculated and the standard
deviation for each average was recorded. Comparison between the elemental
concentrations of different samples was performed using the statistical model of
inclusion. Using this model, if the average of the five measurements of elemental ratios
of the questioned glass fit within the greater of a three standard deviation or 5% range of
the five measurements of the known glass, it was concluded that the samples could have
originated from a common source. However, if the elemental ratios of the questioned
glass fall outside of the statistical range for the known, the questioned sample is excluded
as originating from the same source as the known. Using this method and the statistical
model, an analysis of same source automotive glass resulted in 96% of the comparisons

yielding an inclusive conclusion.

Examination of the Effect of Sample Tilt Angle
An important component to consider in the development of this methodology was
if the shape of the sample could have any effect on the accuracy of the analysis. Forensic

glass samples arise from many different sources and scenarios and so do not demonstrate

27

uniformity in their shape or surface characteristics. An important characteristic to
consider when conducting SEM-EDS analysis is the tilt angle of the surface of the

sample. The tilt angle of the surface of a sample may range from 0° (perfectly flat) to

-1— k

Flat sanrple Sloped sample

highly sloped (Fig. 4.1).

Figure 4.1: The tilt angle of the surface of a sample may

range from perfectly flat to highly sloped.

The tilt angle of the surface of the sample is an important consideration in SEM-
EDS analysis because the quantification program assumes that the sample is flat and
smooth (see Appendix ID). A sample that is indeed flat and smooth, satisfies this
assumption and quantification is performed accurately. However, if a sample is tilted or
has rugged topography the assumption made by the quantification is not correct and may
lead to inaccurate quantification results.

To examine this effect and to determine if the inaccuracy in the quantification of
elements from tilted samples affects their inclusion in the statistical model, one sample

was examined at tilt angles ranging from 0° to 5°.

Procedure

A sample of glass with parallel sides was mounted float side up on an aluminum
SEM stub. As the sample had parallel sides, it was assured to have been mounted so that
the surface for analysis was parallel to the face of the aluminum stub. In other words, the
sample was assured to have been mounted at a tilt angle of 0°. The sample was then

analyzed under low vacuum, with the prescribed elements being

28

 

 

angl

 

tilt a
mg:

mme

 

vam'
comp
Ckme
deter-n
suitab;
With 0

mmon

Summ.

Van-able
Ulster ar
SPECifiC
inCIUSior

another 5

quantitated. This analysis was repeated for a total of five iterations. The appropriate
elemental ratios and standard deviations were calculated.

The stage of the SEM, which holds the sample, was then mechanically tilted to an
angle of 1°, effectively changing the sample tilt angle from 0° to 1°. The elemental
analysis was completed as before, with five iterations and the calculation of elemental
ratio averages and standard deviations. The procedure was repeated with the stage being
tilted to 2°, 3°, 4°, and 5°, with all analyses being conducted on the same piece of glass.

Upon completion of the analyses, the results from the samples analyzed at tilt
angles of 1°, 2°, 3°, 4°, and 5° were compared to the results obtained from the analysis at
tilt angle 0°. Using the statistical model of inclusion, the analyses performed at a tilt
angle of 1° or greater resulted in the sample being excluded as having come from the
same source as the sample analyzed as 0°. However, as the samples analyzed at the
varying tilt angles and the sample analyzed at 0° were one and the same, it is evident that
comparison of samples with a tilt angle difference of 1° will result in inaccurate
elemental quantification and inaccurate exclusion of the sample. Therefore, it was
determined that only samples that can be mounted and analyzed at a tilt angle of 0° are
suitable for analysis by this method. As it would be difficult to mount glass fragments
with only one flat side to a tilt angle of 0° with any measure of certainty, it is concluded

that only samples with two parallel sides are permitted to be analyzed by this method.

Summary

The elemental analysis of glass may be easily and accurately accomplished by
variable pressure SEM-EDS. The method herein established provides for the reduced
raster analysis of glass samples under a low vacuum with the quantification of six
specific elements. The six elements are ratioed and under the statistical model of
inclusion the average ratio of one sample may be compared to the statistical range for

another sample to determine if the two samples could have originated fi‘om the same

29

source. This method must be performed on glass samples that demonstrate two parallel
sides, one of which must be the float side. Using this method, forensic samples may be
accurately analyzed and questioned samples may be compared to known samples to
determine common origin.

While elemental analysis demonstrates a greater ability to distinguish between
similar samples, it is not a replacement for the traditional optical analysis of glass.
Rather, it is a supplement used to distinguish between glasses that cannot be
distinguished by optical and physical properties alone. This supplementary role was
investigated and the results were demonstrative of the necessity and utility of elemental

analysis as a component of the analytical scheme in forensic cases.

Application of Variable Pressure SEM-EDS to Glass Analysis

Upon the development of suitable operating parameters for the SEM-EDS system,
this method was applied to the analysis of tempered automotive glass samples. The
elemental data obtained by this method for each sample was used to distinguish between

samples that were indistinguishable by refractive index alone.

Procedure

Fifty samples of tempered automotive glass were collected, each from a different
source. Refractive index measurements were taken for each sample.

The refractive index was measured by way of Emmons double variation method.
A small piece of glass was crushed into fine pieces using a mortar and pestle. These tiny
pieces were transferred to a clean microscope slide and a drop of Dow Corning 810 Oil
was applied to cover the glass. A cover slip was applied over the sample, and the slide
was placed inside the stage of a Mettler FPS hot stage microscope. A piece of glass was

located and situated in the center of the field of view. The temperature of the hot stage

was set so that the Becke line around the sample disappeared at, or around, a wavelength

30

of 610 nm. The monochromator was returned to a wavelength of less than 310 nm and
then rotated through the spectrum, from 310 nm, to the wavelength at which the Becke
line disappeared. This wavelength was recorded. Without changing the temperature of
the hot stage, this procedure was repeated for a total of 6 iterations. The temperature of
the hot stage was then increased by one degree Celsius. The measurement of the
wavelength at which the Becke line disappeared was again performed with six iterations.
The procedure was repeated for ten temperatures, the last temperature being nine degrees
above the starting temperature. For example, if the starting temperature was 62 degrees,
the final temperature would have been 71 degrees.

The data was then entered into a computerized calculation template that
calculated the dispersion and the refractive index of the glass at 656 nm, 589 nm, and 486
nm (C-, D-, and F - lines respectively). A graph of wavelength versus temperature was
also plotted, which, if not linear prompted the re-analysis of the sample. This procedure
was repeated for each of the fifty samples of glass collected from different sources.

The refractive indices were recorded and each compared to those of the other 49
samples, yielding a total of 2450 comparisons. The criteria for inclusion used by the
Michigan State Police in comparing refiactive index measurements were used as the
criteria for comparison in this study. These criteria state that a sample is to be excluded if
the variation between its refractive index and that of the known sample is greater than
0.0001 at 589 nm, or 0.0002 at 486 nm or 656 nm, or if the difference in dispersion is
greater than 2.0000.

Using these criteria, the refractive indices of all 50 samples were compared. Of
the 2450 comparisons, 186 were inclusive. That is, for these samples there was little or
no difference in refractive index measurements between the two samples being
compared. These 186 comparisons resulted in a false inclusion, as each of the samples
compared originally came from a different source. Therefore, in comparing refractive

index alone, a false inclusion resulted in 7.6% of the comparisons made.

31

Elemental Analysis

All 50 samples were also analyzed by SEM-EDS. 1092 pair-wise
comparisons were made resulting in 7.6% being falsely included when SEM-EDS was
used without regard to refractive index (Table 4.9). SEM-EDS was also performed on
those 186 samples not distinguished by refractive index, and the elemental data was
compared using the statistical model for inclusion. 131 of these comparisons resulted in
exclusion, 20 resulted in inclusion and 35 yielded an inconclusive result. Therefore, in
the cases that could not be discriminated by refractive index, 10.8% of the comparisons

made resulted in a false inclusion after SEM-EDS analysis.

Overall Results

When the elemental analysis comparisons were combined with the results of the
refractive index measurements a marked improvement in discriminating power was
observed. Of the 2450 comparisons made between samples from different sources,
97.8% of the samples could be distinguished when elemental analysis was used to
distinguish between samples with similar refi'active indices. Using only refractive index,
a false inclusion resulted in 7.8% of the cases, while combining refractive index and

elemental analysis resulted in a false inclusion in less than 1% of all cases (Table 4.8).

 

Refractive Index Elemental Analysis Elemental Analysisi Combined

 

Inconclusive 0 0 35 18.8 182 16.7 35 1.4

 

 

(Same R1) (R1 not considered)
# % # % # % # %
Inclusive 186 7.6 20 10.8 83 7.6 20 0.8

Exclusive 2264 92.4 131 70.4 827 75.7 2395 97.8

 

 

 

 

 

Total 2450 100 186 100 1092 100 2450 100

 

Table 4.8: Summary of Data

32

Summary

In the manufacturing process, the quality of glass is often checked by controlling
the refractive index of the product. As such, it is not unusual to encounter glasses from
different sources that possess similar or indistinguishable refractive indices. This has
been demonstrated by this study. When such uniformity is encountered in the forensic
laboratory a second means of distinguishing samples from different sources must be
utilized in order to reach an acceptable conclusion regarding the sources of the samples.
The use of SEM-EDS to distinguish between glasses fi'om different sources with
indistinguishable refractive indices was investigated.

The investigation was preceded by development of suitable SEM-EDS operating
parameters and identification of sample limitations. Upon implementation of the
appropriate analytical method, the refractive indices of glasses from 50 different sources
were compared.

In the analysis of the 50 samples of tempered automotive glass, the refractive
index of each sample was measured and compared to the refractive index of each of the
other 49 samples. By comparing only the refractive indices of the glass samples, 7.6% of
the comparisons were inclusive. That is, although the samples originated from different
sources, they were indistinguishable when comparison was based only on the
measurement of their respective refractive indices. Similarly, when analyzed by SEM-
EDS without regard to refractive index, 7.6% were falsely included. However, when the
results of the refractive index comparisons were coupled with the results of the elemental

analysis, only 0.8% of the comparisons resulted in a false inclusion.

33

Chapter 5: Conclusions and Discussion

This study developed a method by which the elemental analysis of glass may be
accomplished by variable pressure SEM-EDS. The method was established through
repeat analyses and statistical interpretation of the accumulated data to determine the
operating parameters that yielded the best precision of results. When combined with
refractive index measurement, the utility of this SEM-EDS method was then

demonstrated.

Conclusions

It was first concluded that variable pressure SEM-EDS is' applicable to the
analysis of glass. The analysis of glass under a low vacuum of 10 Pa resulted in greater
precision in repeat analyses than both the traditional high vacuum analysis (0.01 Pa) and
analysis after coating the sample with carbon. The analysis performed under a low
vacuum yielded greatest precision when samples were scanned with a reduced raster of
2.25 x 10.2 um2 in five iterations.

Also, it was concluded that the tilt angle of the sample affects the accuracy of the
elemental data. It was determined that in order to accurately compare two samples of
glass, they must both be mounted to a tilt angle of 0°. As it is difficult to mount jagged
glass fragments to a tilt angle of 0° with any degree of certainty, only those samples with
two parallel sides are suitable for analysis by this method.

Using these parameters, it was concluded that an appropriate statistical model for
inclusion requires the calculation of elemental ratios (Sn/N a, Na/Mg, Mg/Al, Na/Al,
Ca/Na, Ca/K) and the comparison of the average of each ratio from the questioned
sample to the greater of the range of three standard deviations or 5% of the average ratio
of the known sample. It was determined that if the questioned average ratio fell within

the prescribed statistical range, the result was inclusive.

34

Using the method developed by this study, 50 samples of tempered automotive
glass were examined. Refractive index measurements were performed on each sample,
and the results were compared. Those samples that were indistinguishable by refractive
index were compared based on the statistical model for inclusion developed for the SEM-
EDS method. These comparisons demonstrated the value of SEM-EDS in the analysis of
glass.

It was demonstrated that if refractive index measurement was the sole comparison
criterion, 7.6% of all comparisons resulted in a false inclusion. Similarly, if elemental
analysis data were compared without regard to refractive index measurement, 7.6% of all
comparisons resulted in a false inclusion. However, when elemental analysis was used in
tandem with the measurement of refiactive index, only 0.8% of all comparisons were
falsely included.

Therefore, it is concluded that by applying variable pressure SEM-EDS, the
forensic scientist is better able to distinguish samples of glass that are indistinguishable

by refractive index alone.

Discussion

As technology advances and traditional methodology is replaced, the capabilities
of the scientist are often increased. However, in this study, there proved to be little
advantage in completely replacing the traditional refi'active index measurement with
SEM-EDS in the forensic analysis of glass.

Under the parameters established, SEM-EDS by itself did not demonstrate an
increase in discriminating power over that of refractive index measurement. However,
the combination of the two techniques resulted in a dramatic increase in discriminating

power over either technique when used alone (Table 5.1).

35

 

Discriminating Power Population # Discrinrinated

 

 

 

 

 

 

 

RI alone 0.924 2450 2264
SEM-EDS alone 0.909 910 827
R1 & SEM-EDS 0.992 2414 2395

 

Table 5.1. Comparison of the Discriminating Power of Refractive Index

measurement and SEM-EDS when used alone and in combination.

Thus, elemental analysis by this SEM-EDS method does not serve to replace
traditional refractive index measurement of glass samples in forensic cases. Rather, this
SEM-EDS method plays a supplementary role by providing the ability to analytically
distinguish between samples of glass that are indistinguishable by refi'active index
measurement alone.

In this supplementary role, elemental analysis by SEM-EDS decreased the
number of false inclusions. A false inclusion results when the analysis concludes an
unknown sample could have arisen from known source, when in fact it did not. For
example, a window in a suspect's car may be falsely included as a source of glass found
on the victim. This study demonstrated that when analyzed by refractive index alone, a
false inclusion resulted in 7.6% of all comparisons. However, when these glass samples
were compared using SEM-EDS and the statistical model for inclusion, less than 1%
were falsely included. Therefore, by supplementing refractive index measurement with
elemental analysis by SEM-EDS, the number of false inclusions was greatly reduced.

For this analytical scheme, elemental analysis by SEM-EDS is logically
performed subsequent to refractive index measurement. However, with automated
techniques such as this method, elemental analysis may serve as an efficient screening

procedure.

36

 

 

 

C01

suit
OPP
(16ml

ma)

geoni

berm

Flltui

innru]

mmds
MdSE
SOUICe.

’egardi.

In cases with multiple known and unknown samples, refractive index
measurement of each sample may be tedious and time consuming. In such a case,
automated elemental analysis may allow efficient examination of the many samples by
eliminating samples with exclusive elemental comparisons from further analytical
comparison. In this way, a large number of samples may reduced to a smaller portion
that may then be analyzed by refractive index.

However, this SEM-EDS method is not without certain limitations. Of primary
concern is that fact that sample geometry plays an important role in elemental analysis,
and may be prohibitive in certain forensic cases.

As demonstrated by this study, only those samples with two parallel sides are
suitable for analysis. In general, this requirement restricts analysis to window glass as
opposed to container glass. Furthermore, fragments of window glass do not always
demonstrate parallel sides, and so the proportion of samples for which this SEM-EDS
analysis is applicable is greatly limited.

While this SEM-EDS method is applicable to only samples with certain
geometries, it has proven to increase the ability of the forensic scientist to distinguish

between samples of glass that are indistinguishable by refractive index alone.

Future Research

Upon conclusion of this study, two additional areas of future research are
suggested: research furthering the SEM-EDS method developed and research into other
instrumental methods for elemental analysis of glass.

The SEM-EDS method developed herein might benefit from examination in a
blind study. Such a study would involve analysis of multiple samples by refi'active index
and SEM-EDS to determine which samples could have originated from a common
source. In such a blind study, the analyst would not possess any knowledge beforehand

regarding the actual sources of the samples. The results of the analysis could then be

37

interpreted without bias, and compared to determine if the correct associations had been
made.

The current method should also be investigated to determine if it is usefirl for the
analysis of other types of flat glass. At present, only float glass has been analyzed by this
method. Two other types of flat glass, rolled- and sheet- glass, should be subjected to
this method. Rolled glass is compressed between two rotating rollers as it leaves the
melting fumace and so would not demonstrate the tin fluorescence or other characteristics
of float glassmii Similarly, sheet glass would not demonstrate float characteristics as it is
prepared by drawing a sheet of glass out of a bath of molten glassm" As both rolled- and
sheet- glass have not yet been analyzed by this method, it would be beneficial to
determine if the method requires modification for analysis of non-float glasses.

Also, the implementation of a third technique might be studied in future research.
This third technique would be used subsequent to refractive index measurement and
elemental analysis in order to further differentiate those samples that are not
discriminated by the methods of the current study. Emission spectroscopy has been
previously linked to a refractive index - SEM-EDS method,"xv no third technique has
been investigated in conjunction with the variable pressure SEM-EDS procedure
developed in this study. By adding a third instrumental technique to the method, the
number of inconclusive results and false inclusions could be reduced even further.

However, the major limitation of the current study is the requirement that the
sample must have a tilt angle of 0° (or parallel sides). The most beneficial research
related to this SEM-EDS method would be the investigation of tilt angle measurement
and correction for samples that do not demonstrate parallel sides. As this sample
requirement is severely limiting, successfirl research in this area would greatly increase

the applicability of the current SEM-EDS technique.

38

Beyond SEM-EDS

Since SEM-EDS methods are limited by the requirement of flat samples, other
instrumental techniques must be investigated. Recent research has focussed on
inductively coupled plasma mass spectrometry (ICP-MS) consequently making ICP-MS
the fastest growing trace element techniquemi The popularity of ICP-MS is primarily
based upon the advantages of increased sensitivity and the ability to analyze for multiple
elements.

The ICP-MS requires a liquid sample, which is converted into a fine aerosol
spray. A plasma is formed when argon gas in a magnetic field is sparked with a high
voltage source of electrons. The plasma is then used to generate positive ions from the
sample. The ions are introduced into the mass spectrometer and separated based on their
mass to charge ratio (m/z).

ICP-MS has been used to analyze small samples of glass (100 pg) but the glass
must be dissolved in a series of acids prior to introduction into the ICP-MS.""Vii More
recently however, solid samples have been able to be analyzed without having to undergo
acid digestion. In a process called laser ablation, a laser is used to vaporize an area on the
surface of the sample. The particles that result from the laser ablation are then swept into
the ICP-MS.

This method has distinct advantages. First, the method can be used to detect ultra
trace levels directly in a solid without requiring acid digestion. Secondly, any solid may
be used, irrespective of its electrical conductivity, size, shape or surface topography. As
shape and surface topography are limiting factors for SEM-EDS analysis of glass, future
research should logically attempt to eliminate these restrictions. In this way laser

ablation ICP-MS might be a promising avenue of future research.

39

Appendices:

I.
II.
III.
IV.

VI.
VII.

Glossary

Elemental Data: high vacuum, low vacuum, and sputter coated samples

Elemental Data: Tilt Angle Effects

Summary of Elemental Ratio Comparisons in the Establishing of a Statistical Model
for Inclusion

Refractive Index Comparisons

186 Elemental Comparisons of Samples not Distinguished by Refractive Index

Summary Data for Refractive Index and Elemental Analysis Comparisons

40

Appendix I

Glossary
Becke Line
The bright halo that outlines a transparent particle immersed in a
liquid of different refractive index.
Discriminating Power
The probability that two distinct samples, taken at random from the
population of interest, would be discriminated.
Dispersion
The relationship between the refractive indices of a substance at
different wavelengths. For glass, dispersion is represented by the relationship
V=(Nd - 1)/(Nf-Nc).
Electron
A negatively charged subatomic particle.
Hot Stage Microscope
A microscope with an enclosed, oven-like sample holder that allows the sample to
be heated to precise temperatures.
Iteration
A repetition of a measurement or procedure.
Monochromator
An instrument that selectively allows only one wavelength (color) of light to be
emitted.
Physical Properties
Properties of a substance that can be measured and observed without changing the
composition of the substance. For example, color, density, and thickness are

physical properties.

41

 

 

 

 

Te

Optical Properties
Those physical properties specifically associated with light. For example,
refi'active index and dispersion are optical properties.

Refi'active Index
The ratio of the speed of light traveling in a vacuum to the speed of light traveling
in a transparent medium such as glass.

Tempered Glass
Glass that has been thermally or chemically toughened to resist breakage.
Tempered glass is most often found in side and rear windows of automobiles,

shower stalls, and doors to business places.

42

Appendix II

Elemental Data: high vacuum, low vacuum, and sputter coated samples

43

 

III

 

.CLEFFDNLLNM.

 

 

 

 

 

 

 

 

Sample uzuki Esteem 1 Ford Probe Fonda Accord
RI - C-line 1.5162 1.5175 1.5162

R1 - D-line 1.5188 1.5200 1.5188

R1 - F-line 1.5257 1.5267 1.5256

LOW VAC

Na/Al 9.107 0.3855 6.207 0.7520 3.352 0.1117
Mg/Al 3.135 0.1418 2.002 0.1715 1.243 0.0566
Na/Mg 2.905 0.0476 3.093 0.1314 9.561 0.1286
Ca/Na 1.275 0.0185 1.211 0.0223 1.085 0.0289
Ca/K 30.036 2.9976 18.831 1.2337 6.891 0.6884
Sn/Na 0.036 0.0051 0.036 0.0059 0.045 0.0089
HI VAC

Na/Al 34.453 8.0338 10.012 0.8174 6.400 0.9989
Mg/Al 8.752 3.0053 2.423 0.1353 1.935 0.0994
Na/Mg 4.217 1.3547 4.132 0.2557 9.982 0.3870
Ca/Na 0.261 0.0422 0.146 0.0154 0.196 0.0404
Ca/K 49.379 27.7909 11.125 2.3856 5.263 0.9946
Sn/Na 0.000 0.0333 0.084 0.0257 0.087 0.0197
SPUTTER CT.

Na/Al 10.880 0.5249 5.956 0.3231 3.550 0.0403
Mg/Al 3.794 0.1114 1.977 0.0808 1.369 0.0259
Na/Mg 2.868 0.1160 3.013 0.1180 9.468 0.0352
Ca/Na 1.348 0.0358 1.255 0.0339 1.149 0.0229
Ca/K 49.756 11.0820 19.642 0.8619 7.205 0.3235
Sn/Na 0.027 0.0090 0.019 0.0034 0.020 0.0053

 

 

 

 

Appendix Table 2.1: Elemental Data for high vacuum, low vacuum and

sputter coated samples.

 

 

 

 

 

 

Sample b8 Pontiac Gd Prix b8 Toyota Celica b0 Acura Legend
RI-C-line 1.5164 1.5174 1.5137

R1 - D-line 1.5189 1.5202 1.5163

R1 - F-line 1.5257 1.5275 1.5233

LOW VAC

Na/Al 5.927 3.3842 3.784 0.2052 9.279 0.7270
Mg/Al 2.011 1.1154 1.340 0.0000 3.101 0.2091
Na/Mg 2.926 0.1434 2.823 0.1131 2.991 0.0517
Ca/Na 1.152 0.9815 1.295 0.0620 1.162 1.0159
Ca/K 22.191 13.4472 6.507 0.6687 37.930 5.0336
Sn/Na 0.052 0.0086 0.034 0.0190 0.044 0.0016
HI VAC

Na/Al 8.636 3.3073 6.331 0.2687 14.196 1.5225
Mg/Al 2.903 0. 7685 1.713 0.0636 4.253 0.6356
Na/Mg 2.949 0.6072 3.698 0.1468 3.358 0.2055
Ca/Na 0.656 1.2470 0.259 0.0640 0.411 1.5926
Ca/K 20.301 7.4152 5.016 0.6170 36.434 11.1678
Sn/Na 0.054 0.0109 0.056 0.0073 0.054 0.0138
SPUTTER CT.

Na/Al 6.171 3.1422 3.762 0.7843 10.513 2.1192
Mg/Al 2.218 0.9304 1.307 0.2315 3.586 0.6054
Na/Mg 2.727 0.3079 2.866 0.1246 2.918 0.1127
Ca/Na 1.707 0.6411 1.364 0.0205 1.296 0.8667
Ca/K 23.991 14.6425 7.000 0.2695 46.735 12.8838
Sn/Na 0.033 0.0112 0.015 0.0024 0.022 0. 0109

 

 

 

 

Appendix Table 2.1 continued: Elemental Data for high vacuum, low vacuum

and sputter coated samples.

45

 

 

 

 

 

 

Sample 1 Mercu Sable 4 Ford Escort 1 Dad e Shadow
RI-C-line 1.5163 1.5168 1.5164
RI-D-line 1.5189 1.5195 1.5190

R1 - F-line 1.5260 1.5267 1.5259

LOW VAC

Na/Al 9.168 0.5913 8.950 0.4842 14.017 1.3001
Mg/Al 2.754 0.1630 2.585 0.3695 4.207 0.4585
Na/Mg 2.904 0.0867 3.518 0.5249 3.337 0.1260
Ca/Na 1.262 0.0312 0.600 0.0390 0.654 0.0399
Ca/K 33.281 5.4060 17.568 6.6721 32.518 6.1133
Sn/Na 0.039 0.0075 0.034 0.0107 0.042 0.0019
HIVAC

Na/Al 29.826 9.8295 14.954 3.74965 39.773 16.5655
Mg/Al 7.167 2.4469 3.313 0.76215 9.376 3.8568
Na/Mg 3.937 0.2483 4.499 0.24306 4.240 0.0827
Ca/Na 0.276 0.0339 0.119 0.02786 0.168 0.0222
Ca/K 23.746 11.0057 7.075 5.14342 18.805 6.0878
Sn/Na 0.059 0.0210 0.000 0.01329 0.059 0.0072
SPUTTER CT.

Na/Al 11.102 1.2104 5.944 0.2540 11.063 1.1855
Mg/Al 3.425 0.3298 2.129 0.1068 3.705 0.3251
Na/Mg 2.773 0.0762 2.793 0.0480 2.982 0.0733
Ca/Na 1.321 0.0602 1.472 0.0209 1.386 0.0345
Ca/K 51.854 12.2277 25.395 3.1009 48.574 6.0323
Sn/Na 0.017 0.0137 0.014 0.0042 0.020 0.0019

 

 

 

 

Appendix Table 2.1 continued: Elemental Data for high vacuum, low vacuum

and sputter coated samples.

46

 

i .|.Il.1|l .t ,t
.Ccfi R_..L\.\\l.\\lCC

al

C

 

 

 

 

 

 

Sample b0 Datsun 210' Mid? Ford Taurus
R1 - C-line 1.5145 1.5170

R1 - D-line 1.5170 1.5198

R1 - F-line 1.5237 1.5271

LOW VAC

Na/Al 5.780 3.3523 11.539 1.5068
Mg/Al 1.808 0.7279 3.833 0.4410
Na/Mg 3.070 0.4194 9.780 0.0931
Ca/Na 0.643 1.5441 0.742 0.0293
Ca/K 5.285 0.4123 42.942 13.6361
Sn/Na 0.052 0.0191 0.045 0.0115
HI VAC

Na/Al 7.801 2.6892 23.596 5.3467
Mg/Al 2.068 0.6075 6.216 1.4059
Na/Mg 3.739 0.1726 10.435 0.0591
Ca/Na 0.237 2.0386 0.207 0.0140
Ca/K 4.127 0.5431 22.037 9.0185
Sn/Na 0.109 0.0187 0.063 0.0141
SPUTTER CT.

Na/Al 5.025 3.8114 8.586 1.6420
Mg/Al 1.849 1.2230 3.132 0.5222
Na/Mg 2.629 0.1780 9.584 0.1033
Ca/Na 1.385 0.7663 1.583 0.1102
Ca/K 12.101 14.1318 71.369 42.8497
Sn/Na 0.021 0.0035 0.018 0.0048

 

 

 

 

Appendix Table 2.1 continued: Elemental Data for high vacuum, low vacuum

and sputter coated samples.

47

 

 

 

 

 

 

Sample ord Tern 0 GL 9 Chrysler B0 Datsun 210, driv. Rear ,
RI - C-line 1.5166 1.5162 1.5138

R1 - D-line 1.5193 1.5188 1.5163

R1 - F-line 1.5265 1.5256 1.5231

LOW VAC

Na/Al 6.571 0.3113 14.276 0.8490 4.542 0.1416
Mg/Al 2.080 0.0000 4.415 0.1476 1.866 0.0450
Na/Mg 3.162 0.1238 3.231 0.0909 2.904 0.1132
Ca/Na 0.704 0. 0399 0.726 1.3865 0.581 0.0270
Ca/K 18.607 0. 7214 63.822 58.221] 5.443 0.3814
Sn/Na 0.042 0.0022 0.040 0.0030 0.046 0.0044
HI VAC

Na/Al 10.804 1.1509 16.773 7.0892 7.551 0.5507
Mg/Al 2.729 0.2414 4.601 1.5457 2.760 0.1501
Na/Mg 3.955 0.1235 3.565 0.3388 3.623 0.1003
Ca/Na 0.159 0.03 78 0.237 1.6849 0.094 0.0127
Ca/K 12.249 6.8740 31.018 6.5843 3.087 0.5402
Sn/Na 0.063 0.0068 0.087 0.0261 0.098 0.0153
SPUTTER CT.

Na/Al 5.665 0.4386 10.923 0. 7735 3.494 0.05 72
Mg/Al 1.957 0.1199 3.773 0.2541 1.767 0.0418
Na/Mg 2.894 0.0825 2.894 0.0425 2.520 0.0367
Ca/Na 1.440 0.0407 1.419 0.7922 1.342 0.0331
Ca/K 23.238 1.7337 59.869 31.3815 5.729 0.2446
Sn/Na 0.018 0.0012 0.016 0.0031 0.022 0.0033

 

 

 

 

Appendix Table 2.1 continued: Elemental Data for high vacuum, low vacuum

and sputter coated samples.

48

 

Appendix III:

Elemental Data: Tilt Angle Effects

Because of complex interactions between components of a sample, a correction factor
must be applied to the apparent elemental concentration in order to determine the true
concentration. One part of the correction factor is the absorbanee of x-rays by
components of the sample. The quantification program assumes the sample is flat,
and so correction is made for x-ray absorbanee upon this assumption. However,
when the sample is tilted, more x-rays are blocked or absorbed creating an additional
absorption factor. The quantification program does not recognize additional

absorption factors and so quantification in this case would be performed inaccurately.

The following data demonstrate this effect.

49

080m 2?... as ”33 5585 ”3 2.3 288%.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

”Ease cages 32 .o 323 mass Sod 2: 2. 826 286 2: .o «25
03.5.06 cameos 03.3: $8.3 $2.2 macs: $3.3 $2.2. ENE 33% 030
05206 ceases 885 $33 236 sad 83s gas 2% ~86 azao
cameos cameos to?“ £80 £86 2 .m 38.... 4:3 086 2 2 wzaz
cameos cameos ~82 _ 82 R :6 83 2%.... ES“ 326 S3 3»:
cameos cameos was: 28.x as: 34.2 22.2 3%.: 52 83: Zaz
< as m m as < E 2 E 2 2.5 Bo.—
mlowgimi g d 0.4.55

033—23 032:2: 5: :6 wmwcd m. m. 66 6 @606 he: _6 N8: .9 6N 66 6 no: .0 «gum
032:2: 032:2: :VNNdE «3666 N N N 6 .N N mm _ .3. evwwéa mwmm.mm “~me 6~ $066 MED
032:2: 030.328 Smad camwd 6~ N 6 6 mead mmaad :366 V6~ 6 6 mmad «ZED
032:2: 032:2: :bmzm $36 6666 .6 hONfi vmom. m E. : ad 366 6 O: : .m wgaz
038—20 032:2: on mvfi woomd ~63. 6 aamé : mamé m _ hhfi 6M6~ 6 wwo:V :(EE
032:2: 032:2: ©5666: Nomvd: ~36 .~ 6mm. 2 sham. m: 366.: : NMMM 6 mags _<\aZ
< 0:: m m 0:: < 3: A: 3: 0: 04> 39‘—
amflum majoflmfl m 2&an

032:3"0 032:2: @626 cmcod VN N 6 6 oo— 6 he: :6 N8: 6 m: 66 6 me: 6 «2:5
032:2: 032:2: whaaéw momm. : m NEWW vwmdc cvwwém mwmmdm VVKN6~ 39$ MRO
032:2: 032:2: mmoo: :wad N m. N 6 6 wmad mamad 3566 V6 N 6 6 Nmad «ZED
032:2: 032:2: MOMMA amoocm 6X. 6 6 be: .m vmomd vtam N366 O: :.m m§Z
032:2: 032:2: mvwmfi vwhwd N ~66 6 Nwmé :mam.v m : Rum 6M6~ 6 Nwoé 3‘32
033—20 032:2: 6000.2 266.0: 6N66< ohwd: hmacfiu $66.: NMMM 6 macN: 2R2
< 0:: m m 0:: < 3: O: 3: 0_ U<> BOA
010.301...»- wfiam oafiam

 

 

50

floutm: 0:22. 5:. “Sun: 3380:”: 625680 :.m 053. 56:09:44.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

033:020 032:2: 6: :6 6wh66 6666 .6 666.6 56: :6 N66:.6 6N 66 .6 m6: .6 07:65
032:020 032:2: 636.6mv :N66.66:- N36 .V6N :66.NN: 6vww.va mwmm.mm VVNN6~ $666 MRO
032:8“0 032:020 w :666 an : w6 6N N 6 .6 amw6 M6666 3666 V6 N 6 .6 N666 «ZED
032:2: 032:2: mmmm.m 66666 6666 .6 66N.m Vmomd 1:66 Q66 .6 6: :.m wzxmz
032:2: 032:2: 6:. 66. m 666?. m N66N .6 hhmé :m6mé m : Ru m 6m. 6 N .6 Nw6.v :<\w$:
03:04:020 032:2: V6666: 66bm6: 6~6N.~ mwm.m: BN666: 356.: : :66 .6 M666: :<\mz
< 3:: m m: 3:: < 3: o: 3: 0: U<> BOA

a a M 295m
032:020 032:2: mm: :6 mmh66 R66 .6 566.6 56: :6 N66: .6 6 N 66 .6 66: .6 023m
032:2: 032:2: 3666:. :66: .mm 6666 6 hNQmm 6vww.va mwmm.mm 6»va 6~ $666 MRO
032:020 032:020 6wN66 Nmmw6 6m N 6 .6 6:66 mm666 3666 V6 N 6 .6 Nm66 mZRU
032:2: 032:020 56v. m tum : . m 66:6 .6 66m.m VN6M. m Vb : 6.N NV66 .6 6: : . m w§Z
032:020 032:2: wwwmfi @6666 6k 3. .6 9%.? Swami m : 55m 66 6 N .6 Nw6.v :<\w$:
032:020 032:2: 65566: 666w: : 6 6N 6 .~ wwaé: 5N666: 366.: : NMMM .6 m66.N: :<\m7:
< 0:: m m: 3:: < 3: 0: 3: 0: U<> 301:

80.. 0:. m a .6 fl min—am

 

 

 

 

 

 

 

51

Appendix IV

Summary of Elemental Ratio Comparisons in the Establishing of a Statistical Model

for Inclusion

Source # Excluded # Included Total Comparisons
Chrysler 12 240 252

Dodge 14 238 252

Mercury 1 2 240 252

Toyota 1 25 1 252

Total 39 969 1008

Percent 3.9% 96.1% 100%

Reanalysis of Data Under Statistical Model for Inclusion Guidelines

Source Excluded Included Inconclusive
Chrysler 0 32 10

Dodge 0 3O 12

Mercury 0 32 10

Toyota 0 41 1

Total 0 1 3 5 33

Percent O 80.4% 19.6%

Appendix Table 4.1: Summary of Elemental Ratio Comparisons in the Establishing of a
Statistical Model for Inclusion.

52

Appendix V

Refractive Index Comparisons

53

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample 94 Ford Escort 91 Dodge Shadow 80 Datsun 210 M Ford Tempo GL
C-Line 1.5168 1.5164 1.5145 1.5166
D-Line 1.5195 1.5190 1.5170 1.5193
F-Llne 1.5267 1.5259 1.5237 1.5265

V 52.512 54.412 56.067 52.443

Sample 89 Chrysler 80 Datsun 210 R. 94 Ford Taurus 85 Toyota Corolla
C-Line 1.5162 1.5138 1.5170 1.5173
D-Line 1.5188 1.5163 1.5198 1.5199
F-Line 1.5256 1.5231 1.5271 1.5270

55.261 55.585 51.954 53.558
ample Pontiac Gd Prix 86 Buick Century 89 Ford Tempo 88 Pontiac Gr Pr.
C-Line 1.5168 1.5188 1.5161 1.5164

ED-Line 1.5192 1.5125 1.5188 1.5189
F-Line 1.5259 1.5284 1.5260 1.5257

V 56.728 54.436 52.256 55.962

Sample 96 Ford Escort 87 Olds Cutlass 96 Ford Contour Suzuki Esteem
C-Line 1.5189 1.5162 1.5175 1.5162

ED-Line 1.5215 1.5189 1.5202 1.5188
F-Line 1.5284 1.5258 1.5274 1.5257

V 54.537 54.088 52.361 54.282

SamJfle 91 Ford Probe 88 Tgyota Celica 90 Acura Legend 91 Merc. Sable
C-Line 1.5175 1.5174 1.5137 1.5163

E D-Line 1.5200 1.5202 1.5163 1.5189
F-Line 1.5267 1.5275 1.5233 1.5260

V 56.510 51.120 53.502 53.491

Sample Honda Accord 86 Chrys LeBaron 89 Pont Sunbird 95 Ford Contour
C-Line 1.5162 1.5184 1.5174 1.5179
D-Line 1.5188 1.5211 1.5201 1.5205
F-Line 1.5256 1.5281 1.5271 1.5275

55.149 54.015 53.419 53.907

Sample 94 GM 85 Chev Celebrity Skylark 95 Buick
C-Line 1.5161 1.5164 1.5156 1.5171
D-Line 1.5186 1.5190 1.5183 1.5197
F-Line 1.5254 1.5258 1.5257 1.5265

V 55.931 55.425 51.266 55.506

Sample 86 Colt Vista 87 Ford Ford Escort LX 92 Olds Cut. Sp.

RIC-Line 1.5146 1.5164 1.5171 1.5175

R1 D-Line 1.5171 1.5190 1.5197 1.5201

R1 F-Line 1.5238 1.5257 1.5266 1.5270

V 56.123 55.793 54.879 55.183

 

 

 

 

 

 

 

Appendix Table 5.1: Refractive Index Comparisons.

54

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Sample 85 Lincoln TC Pontiac Sunbird 93 Hyundai Sonata 92 Merc Tracer
C-Line 1.5171 1.566 1.5166 1.5171
D-Line 1.5960 1.5193 1.5193 1.598
F-Llne 1.5265 1.5263 1.5265 1.5269

V

Sample Oldsmobile 84 Ford Tempo 91 Oldsmobile 90 Toyota Camry
C-Line 1.5170 1.5166 1.5196 1.5161
D-Line 1.5196 1.5193 1.5223 1.5187
F-Line 1.5266 1.5263 1.5295 1.5256

53.719 53.974 52.366 54.887

Sample 85 Honda Acc'd 95 Toyota Tercel 85 Chev Eurosport 89 Cavalier Z-24
C-Line 1.5160 1.5161 1.5169 1.5173

ED-Line 1.5187 1.5187 1.5195 1.5199
F-Line 1.5256 1.5256 1.5265 1.5268

V 54.317 54.451 54.195 54.773

Sample 93 Chev Cavalier 9O Beretta 85 Olds Brougham 88 K-Car Wagon
C-Line 1.5166 1.5170 1.5165 1.5161
D-Line 1.5192 1.5196 1.5191 1.5186
F-Line 1.5263 1.5265 1.5259 1.5254

53.362 54.791 55.061 55.610

Sample 93 Olds/GM 89 Sunbird/red
C-Line 1.5162 1.5173

R1 D-Line 1.5189 1.5199

R1 F-Line 1.5259 1.5270

V 53.498 53.268

 

 

 

 

 

Appendix Table 5.1 continued: Refractive Index Comparisons.

55

 

Appendix VI

186 Elemental Comparisons of Samples not Distinguished by Refractive Index

56

039:88
039:88
039:2:
030388
039:28
030288

< 2.: m:

039:2:
039:28
039:28
039:2:
039:28
039:88

< 22 m

039:2:
030388
039:2:
039:2:
039:28
039:88

meg: <

0332:
039:2:
039:28
039:88
039:88
033:2:

m02:<

 

 

08:2: 038060: .3 3:932:65 87: moan—mm :0 2098an0 ”:6 0:20.:- 36:02?

 

 

 

 

 

 

 

 

 

 

 

 

 

N956 $8.0 2.8.6 086

$5: 3R2 $2...“ 3.3

32.: :26: 33.6 30:

$2.». 228 8%.: can

~22 $33.- 23: awed

:mmmd: 83.x 32:: 393
E 0:

20%

$2.:

2%.:

022

336 ovmod 3.3.6 Sod

«20.2: 020.2. $3.2. $0.3

32.: 33: 836 $2

286 Seed 2.8.0 www.~

9.8:» $22 Sam-.6 39m

023.2 88.0 3:4 802
E 0:

figm

:59:

m2?

. 2:2

255$. Far.— V6

 

 

 

 

 

 

 

33.0 285 :82 $2. «25
~82 2:0 222 22.0 25
£22 £92 236 :2 226
282 53 232 $2 02222
9:2 922 832 82 302
22.4 832 232 22m :22
2 2 2.5 33
23% >

422 02-2-2

52 22-2-2

922 22-0-2

350 22.5 2 295.0

823 $85 :22 23 «2%
22.2 22.9 222 22.2 28
822 222 282 $2 «25
$22 222 222 $3 0232
282 $32 226 22 302
2%.” 32.4 $3 23 322
2 2 02$ 33
xwfi. >

8.32 22-2-2

222 22-2-2

ES 22-0-2

0.955

 

 

57

039:2:
039:2:
039:2:
039:2:
039:2:
030228

<22 2

030228
039:28
039:2:
032:2:
030228
0302—008

<02: m

039:2:
039:2:
039:2:
039:28
039:2:
039:2:

m02:<

039:2:
039:2:
032:2:
039:2:
030228
0303008

moan: <

 

 

.0805 03008.:3— >2 00202223005 82 00:22am .20 209-82800 6022:0200 _.e 030,—. 0:282?

 

 

 

 

 

 

 

 

 

 

 

 

 

moved ammod uged wmod
@0325 ngo $03.2 N02?
3:”: 2.2.: NNSG omNA
m: mm.m wmvwd 2366 waofi
021v NORA.” MNm-Nd 39m
$2.3 ovfid mam-e mmmA—
3: 0—
22$
03%:
ca: m.—
3;:
1%
cmood name-o mbgd owed
wan 2.5m ammwfi: 3.0.2.6 wmvfim
3mm: 85.: :86 com.—
NmO: .m SVmN 3.36 omwd
who: .v Ema-N Sam-c Nmmd
wan-c: Name-h 22m. 6 m 2 :d
3: 0—
wwoém
wmmm:
ow _ m:
NE W:
000.020 020 ha

 

 

 

 

 

 

 

 

 

008.0 280 208.0 03.0 2220
22.2 22.: 05.0 0%.: 220
22.2 822 22.0 $2 2220
~33 22.3 522 022 0:22
~83 £02 22% ~22 302
08:: N2: 022 0:2 222
2 2 2.5 302
23.2 >

082 22-2-2

222 22-2-2

222 22-0-2

2.230 220 2 22:30

$3.0 $8.0 28.0 2.20 2220
29.2 23.: 2:2 902 020
$22 222 003.0 22 22.20
202 $22 2230 032 022,2
822 200.2 250 :3 302
22: 23.0 022.0 90.0 222
2 2 0<> 302
02.2 >

822 22-2-2

092 22-2-2

322 22-0-2

2.02. 0.3,.— S 0.22%

 

58

aims—8:
aims—88
833—88
2528:
25288
25388

< 33 m

2,838:
2,828:
aims—8:
3828:
2528:
839:8:

< 85 m

2:385
833—88
833—88
3838:
38288
25388

m85<

2,828:
3828:
833—8:
3828:
38205
3828:

m 25 <

 

 

88—85 guns-aux an conmfiwccma 82 338% MO :85qu0 62.5280 fie 2an £989?

 

 

 

 

 

 

 

 

 

 

 

 

 

omwcd $86 836 mmod

m8— .3 p.256 $34 c3};

Son; 324 $36 5:

«coma ofififi m6: 6 SEN

33d awed Sam-6 aw;

88.x $3; VNNN< mflfi
o—

83.3
hmmmg
$3;
$3;

USE—«.5 85.8.— mm
33d 88.: 2.86 305
S 3.5 amt-6- 3R .2 omnov
S2“; con: 3.36 :VNA
Emma mOmfim 836 awed
vommé waned Sums >36
$8M _ mmiuw at 6 33:

0.

Gmdm
can;
334
, NE m.—

'3

 

 

 

 

 

 

$86 28¢ $23 $55 «2%
82.3 $3.: 326 Main 25
an? 85 :86 $2 «25
«82 5.3 586 as.“ msfiz
~83. E2 286 an?” 3»:
88.2 as: 832 23 23A
2 8.5 33

$3.“ >

£2 052-2

3:2 eed-E

~22 25-0-2

22.5 2.5 8 285m

033 $85 $86 Sod «25
82.8 8%.: 32.» 812 $6
an? 85 286 $2 «25
~83 ~32 £85 23 382
~83. and 283 «mum 2E2
832 RS 32 2; 2%
2 SS 33

83% >

£3 052-2

9:2 25-9-2

~22 256-2

man—EU 35 5 8.95%

 

 

59

38202
38288
38288
38202
38288
38288

< 35 m

38202
38202
38202
38202
38202
38202

< 82 m

38202
38288
38208
38202
38288
38288

m 95 <

38202
38202
38202
38202
38202
38202

m82<

 

 

88202 333.28% 23 208828285 82 8.292% .20 2883200 6832200 80 2an 2282?-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wvwod mmmod v.39: avod
$3.3. $3.3 QRN6~ emf?
gem; @822 3.36 m3.—
mmmmd chhfi 2.36 ~mw.~
ommné amt-N- RNm-d v3.6
38.? moms waad 352
2
35mm
hmmmg
cam.—
$3.2
9.0% ha
wvcod 386 3666 Quad
mwoméh ammo-3 QKNGN omvdv
gem-2 cog; 3.86 32.—
mmdQN mmowd 2‘56 _mw.m
cap-v amnfim Emmd and
38.2 38.5 Goad SKA:
2
mafimm
man;
can;
33.2
a

 

 

 

 

 

 

 

885 285 ~28 ~23 «28
82.2 E; ~82 9%.: 280
as? 882 88.8 :3 2,80
888 $22 826 82 8:82
88~ 88¢ 83 88 282
288.8 83; V-~2 82 352
5 2 35 >52
88.8 >

882 25-8-8

882 85-9-8

.22 25-0-8

USE—EU 022::— ww «En—um

82:0 828 88.5 228 «228
82.8 88.: $8.» 828 v80
:82 ~22 :86 $2 «780
~83 ~32 ~88 o~w~ 8:22
~83 £82 £86 -~.~ 282
~82: ~83 253 2; 282
5 2 03 BS
82% >

8.82 25-2-8

882 25-2-8

~23 25-0-8

222.0 2.5 8 28.8

 

6O

38288
38202
38288
38288
38288
38288

< 22 m

38288
38202
38202
38202
38202
38202

< 22 m

38202
38288
38288
38202
38288
38288

m 82 <

38202
38202
38202
38288
38202
38202

m 22 <

 

 

08—02 3:08-82 3 2002828285 002 moi—Ham .20 202-82200 60:20:00 26 033. 2882?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

omwod 28.0 836 $90

32.3 man N384 0323

85m.— cmwmA 35.0.0 22:

gown Sofia 262.0 and

mmgN omwod Smmd a~w2

Sow-w @321 $25 mfld
2

08%?

533

$3.2

$3.2

USE—«.5 0220A mm

300.0 33.0 “.306 $0.0

$8.3 38.: «ENE 03.?

moon; com: avgd a:

mmmmN Sofia $.36 SQN

32.6 ade Rde v3.6

300.2 mmows 83.0 3::
2

35mm

Sun;

022

$3.2

620m 5

 

 

 

 

 

 

83.0 885 88.5 58.0 258
88.8 885 820.2 ~32. 080
52 E: :80 82 2,80
282 828 2.8.8 83 8232
8:8 ~oa~ 880 $3 382
$2.3 2.8.8 880 80: 22.2
2 03 302

~28 >

88.2 25-8-8

222 25-0-8

382 25-0-8

32:25 0 can 3 22,—2.6

83.0 88.0 828 ”8.0 225
88.8 88.8 820.2 ~25. 080
8:... E: :80 082 «25
~83 88~ 2.85 83“ mzaz
8:}. ~22 28.0 $3 282
$2.: 88.” 880 85: 32
2 0<> 302

~38 >

882 25-2-8

88.2 25-0-8

882 25-0-8

22.28 «“25 a 2258

 

61

032:2:
25205
25288
03335
36288
95286

< 35 m

2522:
023.05
25288
2,628.8
3622:
023—2:

<85 m

o>_m=_o=_
2,622:
9522:
2,622:
aims—axe
36288

m 95 <

03335
03335
aims—axe
25388
2522:
023—05

m 95 <

 

 

48?: gauged 3 conmmsmcuma 82 335mm mo :OmEnEoU ”3:538 fie £an xmwcoqaxx

 

 

 

 

 

 

 

 

 

 

 

 

 

mmmod $86 939% 366

mwoado mKONN QNNVN oavév

GEM.— cmmo.— V986 2N;

ammmd wNPN macs wvoa

393. $36. N926 ovad

N302 $51: «336 392
E 2

5.3%

can;

$34

034

F33< 2:3: mm

3.36 386 369% wmod

$8.3 @354» “SEX: cwfiww

2‘va 33; @366 awn.—

nmvwd whavd $.36 NSN

mowv.v Read thmd wmbd

m3fim~ mmmbfi Qbmmd Gad
E 2

35mm

comm;

can;

234

232—320

 

 

 

 

 

 

885 $86 $86 3.3 «2%
:32. 823 SEN 83m V36
3:; $2.2 :86 $2 «25
85m ER 885 RS wznz
89m 23a 3mg 3:. 3m:
63.: 88¢ $3 3.3 3%
E 2 SS 33
an: >

8%; 052-2

3:2 25-92

83; 25-0-2

3|;

:36 $85 286 was «2%
38.3: 53%- V822 so: E6
32 ~83 :86 Q: 25
89% $23 $86 $3 msfiz
2m: 88a 326 S3 3»:
:32 8%; $3 at: 3%
E 2 BS 33
5% >

8%; 052-2

52 25-92

EB 25-0-2

 

5 team Es— «.maam

 

62

2,2282
2,2288
25202
25288
2,2288
22288

< 25 2

25202
2,2288
22.2202
22.2288
2,2288
25288

< 32 m

.825 250382 3 2.22.2225 82 .8383 we 2859280 682228 fie 853- 298%?

2,2202
25288
222202
26288
2,2288
2,7288

m 32 <

2,2202
2,2288
25.202
2,2288
22.2288
22.2288

2 25 <

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

336 memod NNeee Que-O
N2 me.o_ meag- vave 3nd
mm— m._ mhwfig eemed Gem;
cmoad wmvmd Seed EKN
Saw; ogmg vmvee Km;
2 ~85.” Rhea gem-ed emu-m
E 2
3.26m
emmm._
$3.—
~e~ m2
1%
wag-c memod N89 .e 390
Seed— memafi quvd vomd
m2 m2 mum—A eegd ommé
cmoad anm.m Seed and
33: 94%.— vaee Km.—
_ fimafi :3.»- Qeeee Ens-m
2 0—
:vém
emmm._
hm _ m2
_e~ m.—

 

.35. .2. a. ma

 

 

 

 

 

23° 2.85. 886 3.3 A2%
28.8 28.2 222 83% 25
822 22: £36 22 «25
222-. ”22 :22 22m 2222
«83. 2.23“ 822 $3 222
22.2 53: 233 822 222
2 8; 33

:22 >

082 25-2-2

2:2 25-5-2

22 25-0 - 2

e38< «ESE mm 295%

R35 $86 222 Sod 22m
:22? 82.2 222 822 28
23.5 2:? :82 22.. «25
882 ES 886 23 322
cam-Z 8:2 2mg 23 322
88.: £26 226 3.3 222
2 35 BS

52% >

032 25-2-2

2:3 25-5-2

$52 25-0-2

:80 3.. £- 8 can—am

 

63

039:2:
03335
36205
95286
$6288
2,6288

<35 m

032:2:
039:2:
023—05
022:2:
03335
033.05

< 35 m

48?: $3933“ 3 uoaflnwauma 82 838mm mo acmEQEoU 6035280 fie Baa-- £32??-

2532:
aims—83
2522:
033—88
36288
36288

m 35 <

3622:
032:2:
033—2:
3632:
03.0.22:
033.05

m 25 <

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

waved mom-ed ~39: mi;
N_ 3.2 $3.5 RES vomd
3:”; mum: 336 Own;
omoad mewN 336 vwfim
$92 $va wmgd Km;
:36 hug-M $36 35m
5 2
33%
can;
bwfi m4
#2 m;
.338 So PH ma
336 3.86 3.36 $56
2 mmfiw 8:6- 3R .2 32:.
~:m._ ooh: 3.36 gm.—
Pmmd 8de 836 awed
vommé wooed gums a:
$8.2 33% at 6 33:
E 2
Swan
cmmmg
wwfig
m2 m4

Ila-«Ilsa .5 a; a.

 

 

 

 

 

 

£23 885 ~32 mac «2%
23.; 8:6- 255 amne- M5
«:2 8:; £86 32 «25
Km? M82 386 an: mzhz
«82- 803 £56 5% 3m:
$8.2 $3.» 3:3 $2: 3.2
2 u<> 33

8mm >

on”? aid-E

$22 053.3

~22 25-0-2

.5...“ :0 an can—am

83¢ #85 a8: 33. seam
88.8 28.2 amt-N 83...- M5
SR4 82: R85 22 «25
am? $2 :36 $3 wsfiz
3a? 35% 826 $3 3w:
«3.2 52: 232 mood 352
2 BS 33

23% >

8%; 052-2

2:2 aid-E

22 25-0-2

282 2:8: ma gag—am

 

64

020205
02020000
0202000
020205
02020000
02020000

.0 02 0

020205
020205
020205
020205
020205
020205

.0 as 0

02020000
020205
02020000
020205
02020000
02020000

m005 <

020205
020205
020205
020205
020205
020205

m35 <

 

 

.0005 020000-030 03 0020225005 002 02058 .20 000203500 609558 2.0 030.0. 50:005-4-

 

 

 

 

 

 

 

 

 

 

 

 

 

owvcd name-o N896 086
2 mwfiw 8:6- 300 .2 amndv
N20”; 82.2 000.36 20.0...—
Kmmd mowhN 2036 $96
veg-v wgqm €0N6 haw-m
vwmcdfi mag-w $000 6 awe-o2
2 2
Smdw
can;
ww _ m2
Ne _ m2
Ila
ego-o name-o mmocd 086
3 3.5 amt-6- 300 .2 mmmdv
N20“; cog-2 0.0.86 3N.—
hhmmd womb-N gas-Q awe-m
comma“. wooed bbnms haw-m
$8.3 gin-w 0300 6 awed—
2 2
Smdm
cam.—
ww _ m2
N3 m.—

. Julia 00 00_ 0

 

 

 

 

 

500 0.0 0,—- 00

0000.0 0000.0 0000.0 000.0 02000
0000.0_ 0000.0 0000.0 03.: 0000
0000.0 0000.0 0000.0 0:1 0200
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.0 0000.. 0000.0 02.0 2002
2 SS 300

0000.00 >

0000.0 80.0.:

0000.0 000-0-E

0000.0 80.0.2

052000 0250.— : 0. 50m

0000.0 0000.0 0000.0 30.0 02000
200.00 0000.00 0000.0 000.00 00000
200.0 0000.0 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 02002
0000.0 200 0000.0 000.0 2002
0000.: 0000.0 0000.0 000.0 2002
2 35 30.0

000.00 >

0000.0 80.0.2

0000; 25.0.2

0002 02-0-2

 

 

0.925

65

020205
0202000
020205
020205
02020000
02020000

<85 m

020205
02020000
020205
020205
0202000
02020000

<35m

.0005: 020000.030 03 0002052505 002 003.5% .00 00002-80500 600558 0.0 0500- 55000900

020205
020205
020205
02020000
02020000
02020000

0 22 <

020205
020205
020205
020205
02020000
02020000

0 22 .0.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0000.0 0000.0 0000.0 000.0

0000.0 0000.0 0000.00 000.00

0000.0 0003 0000.0 000.0

0000.0 0000.0 0000.0 000.0

0000.0 0000.0 0000.0 000.0

0000.00 0000.0 0000.0 000.0
E 2

000.00

0000.0

0000._

0:02

23020.05

0000.0 0000.0 0000.0 000.0

0000.000 0000.0 0000.00 000.00

0000._ 000: 0000.0 000.0

0000.0 0000.0 0000.0 000.0

0000.0 0000.0 0000.0 000.0

0000.0_ 0000.0 0000.0 000.00
E 2

000.00

0000._

0000.0

0000.0

[lg-0.0%

 

 

 

 

 

 

00 0 20.0 0.00

0000.0 0000.0 0000.0 000.0 070000
00:00 0000.0. 0000.0 000.00 0000
0000.0 000: 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.0 0000.0 0000.0 000.0 2002
2 00$ 23

000.00 >

0000._ 22.0.2

0000._ 0200-00

0000._ 22-0-00

“Macaw finch V6 Sufi—am

0000.0 0000.0 0000.0 000.0 02000
_0_0.00 0000.0_ 0000.0 :000 00000
000.0 0002 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.0 0000.0 0000.0 000.0 2002
2 SS 200

000.00 >

0000.0 22.0.0.0

0000._ 22.0.2

0032 22-0-2

 

 

0.0800

66

020205
020205
020205
020205
020205
020205

< 22 0

020205
020205
020205
020205
020205
020205

< 25 m

02020000
02020000
020205
020205
02020000
02020000

m25<

020205
02020000
020205
020205
020205
020205

m 25 <

 

 

.2062 02008-002 03 6020525005 002 02ng .00 0002-80500 60:50:00 26 0509. 5200054.

 

 

 

 

 

 

 

 

 

 

 

 

 

500.06 ommod 60666 686
35.3 wag-v— 0.3.6.0 :vdm
332 222 6:66 0.0m;
$50.0” avgm 6626 0.8.0.
0.300“ goo.— 0026 Sod
33.x 86:0 3.066 626
2 2

mvam

mom-2

mam-2

83.2

lama-nag

ammcd 386 0.0666 3066
Emu-c comm-v N636 3nd
$0602 5066.2 6366 m2;
maomN onmvd 0.6066 wand
:09..— 6822 6366 com;
32.6 23.0“ 60606 on:

E 2

$6.6m

awn-2

60.3.2

903.2

E:— 63 500.:— an

 

 

 

 

 

 

 

0000.0 0000.0- 0000.0 000.0 02000
0000.00 0000.00 0000.0 000.00 0000
0000.0 0000.0 0000.0 000.0 02000
0000.0 0000.0 0000.0 50.0 02002
0000.0 0000.0 0000.0 000.0 0.000:
0000.0_ 0000.0 0000.0 0_0.0 2002
2 2 00$ 23
000.00 >

0000._ 0200.00

000.0 22.0.2

0000._ 22.0.2

6.5255 050.5; 0355

0000.0 0000.0 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 0000
0002 0000.0 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 02002
0000.0 0002 0000.0 000.0 2002
0000.0 0000.0 0000.0 000.0 2002
E 2 00> 202
000.00 >

0000._ 22-0-00

0000.0 22-0-00

0000.0 025.2

32> :00 00 0.05%

 

67

02020000
020205
020205
020205
02020000
02020000

< 22 0

020205
02020000
020205
020205
020205
020205

.00 25 m

020205
020205
020205
020205
020205
020205

m005<

02020000
020205
020205
020205
020205
020205

m 25 <

 

 

.0065 02000-602 03 6020525005 002 00358 .20 5002006500 60:50:00 26 0300- 56:09:00

 

 

 

 

 

 

 

 

  

 

 

 

 

 

3.66.6 m6 2 66- 0 0- 0 6 .6 mmod
Smog-.0 3666— 0.60.6.0- 306.3
awe-2 832 00.066 hem-2
6mmm.m onwafi 6606.6 8— .m
mmmmé :90.— 00006 20w.N
hmhmfiu 33.000 663:0 656
E 2
366m
mcmmé
mam-2
6632
6.0355 022:5
N366 owned 6 N 66 .6 68.6
mamm- 2N~ m6mm.m .0660 60 36.8
wmmmg mam—.2 0266 m8;
mam-.0” 663.6 0- 666 .6 «mo-m
$3.0 chafim 60 0. N .6 mend
0.63.2 20va 0- 666 .6 one-o2
2
036m
mommé
m2 m2
@632

 

. can—0H 6.8..“ 06

 

 

 

 

 

 

0000.0 0000.0 0000.0 000.0 02000
0000.00 0000.00 0000.0 000.00 0000
0000.0 000: 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.0_ 0000.0 0000.0 000.: 2002
2 2 00$ 23
000.00 >

0000.0 22-0-00

0000.0 000-0-E

0000.0 22-0-00

805% 606:: E ma 265%

0000.0 0000.0- 0000.0 000.0 02000
0000.00 0000.00 0000.0 000.00 2000
0000.0 0000.0 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.. 0000.0 000.0 302
0000.0_ 0000.0 0000.0 0_0.0 2002
2 2 00$ 23
000.00 >

0000.0 22-0-00

0000.0 22.0.2

00002 22-0-2

.5550 00500.2 0.0800

 

 

68

830202
830288
830202
830202
830202
830202

< 300 0

830202
830288
830282
830202
830288
830288

.4. 82 m

830202
830202
830202
830202
830202
830202

009030

830202
830288
830202
830202
830288
830288

00 35 <

 

 

 

 

 

 

90006.6 6mm6.6 6066.6 606.6
8.0.0.22 m6mm.m 0.660 .60 wM6.N6
620.0 32.— 0006.6 m3.—
m0mm.m 66m0.~ 0. 666 .6 Nm6.m
$5.0 60.60..~ 60 0. N .6 mom-m
0.600-.2 66006.6 0.666 .6 606.60
2
006.0%
832
mam-0
660 6.0
'3
0306.6 626.6 60 66 .6 68.6
33.6w 6666.3 0066.0 20.8
3.00m; 260.0 600.66 5&2
6630;” 656d 66m 0 .6 006.0.
0.6 SN 6666.0 0000. 0 .6 26d
663.6 666:0 0.0.06.6 6N0.6
2
www.mm
m6mm2
0606.0
6606.0

 

AU ONE—0H. munch

 

 

.0862 8388-800 03 uoammswcuma 82 08—68% .00 0803-86800 ”68:20:05 0.6 260,—- 00328690

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0000.0 0000.0 0000.0 000.0 02000
0000.00 0000.00 0000.0 000.00 0008
0000.0 000: 0000.0 000.0 0206
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.00 0000.0 0000.0 000.: 2002

2 80> Boa
000.00 >
0000.0 25.0.2
0000._ 2.0.9.2
0000._ 80.0.2

00. 328m 206: 00 m6 268%
0000.0 0000.0 0000.0 000.0 02000
0000.00 0000.00 0000.0 000.00 006
0000.0 0003 0000.0 000.0 0205
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.00 0000.0 0000.0 000.: 2002

2 90$ 33

000.00 >

0000._ 25-0-00

00_0._ 000-00-E

0000.0 2.0.6.2

00 300% 3:. m 00 0.600—am

 

69

830202
830288
830202
830202
830288
830288

< 22 .0—

830202
830202
830288
830288
830288
830202

< 22 0

830202
830288
830202
830202
830202
830202

m 22¢.

830202
830288
830288
830202
830288
830202

m022<

 

 

.0862 8380686 03 6820262005 82 026806 00 2003-26200 ”68:20.08 0.6 260.0. 262866.00

 

 

 

 

 

 

 

 

 

 

 

 

 

6006.6 60N6.6 N60 6 .6 N066
6N60 .60 0000.6 N 00. 0. .0 630.3
6060.0 6NON._ N636 000.2
066N.m 606m.N 6600 .6 060.N
0666.N 6666.6 0 660. .6 6Nw.0
0666.0 6060.— 02.0.0 0.2.0
2

6N66.mm

030.0

660 0.0

060 0.0

“$650.00 0225.0 66
0606.6 003.6 0066 .6 006.6
Nm6m.mm 366.0.— 60 6N .0. 60.0.3
060N2 602.0 6606.6 00:
6060.0. 66N6.N 6600 .6 069m
6Nm6.N 60.06.— 6060 .6 wm0.N
6900.6 0660.0 N060 .6 N006
2

0N0.Nm

66Nm._

6660.0

0000.0

.8820. 3.82 N6

 

 

 

 

   

 

0:200 80 0.0. mm

   

0000.0 0000.0 0000.0 000.0 02000
0000.00 0000.00 0000.0 000.00 2000
0000.0 0000.0 0000.0 000._ 020.0.0
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.0 0000.0 0000.0 000.0 2002
2 2 SS 33
000.00 >

0000._ 22-0-2

0000.0 22-2-2

020.0 22-0-2

2822 25040006

0000.0 0000.0 0000.0 _00.0 020.0
0000.000 0000.007 0000.00 000.00 2000
0000.. 0000.0 0000.0 000.0 02000
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.: 0000.0 0000.0 000.0 2002
2 2 0<> 33
000.00 >

0000._ 22-0-2

0000.0 22-2-2

0000.0 22-0.2

 

   

70

030202
030288
830202
030202
030288
030288

< 82 m

030202
030288
030288
030288
030288
030288

< 22 2

830202
030288
830288
030288
030288
030288

008200

030202
030288
030288
030288
030288
830288

0— 82 <

 

 

.0862 03008030 02 6822002005 002 022200 .02 28202220 ”68220220 0.6 220.0- 260822.00

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0606.6 0606.6 0066.6 006.6
0006.60 0606.0 0600.6 060.6
0000.0 000: 6606.6 600.0
6066.0 0000.0 0666.6 000.0
0060.0 6000.0 0006.6 000.0
206.0 0000.0 6606.6 000.0
2 2

000.00

6000.—

0000.0

0600.0

.0980- 30 0.0. 06

6666.6 0606.6 6066.6 006.6
6060.0 0006.0 6600.6 660.6
0660.0 0606.0 0006.6 000.0
0060.0 6000.0 6066.6 606.0
0000.0 6600.0 0006.6 600.—
0000.0 0006.0 6066.6 000.0

2 2

600.00

6000.—

0000.0

0600.0

6.000000 20:00.0

 

 

 

 

 

   

2.080-,0 20000

0000.0 0000.0 0000.0 000.0 02000
0000.00 0000.02 0000.0 000.00 208
0000.0 0000.0 0000.0 000.0 0208
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.0 0000.0 0000.0 000.0 2002
2 00> >52

000.00 >

0000.0 22-2-2

0000.0 22-2-2

0000.0 22-0-2

5030”..— fl—Em o—flaam

0000.0 0000.0 0000.0 000.0 02000
0000.00 0000.: 0000.0 000.00 206
0000.0 0000.0 0000.0 000.0 0.205
0000.0 0000.0 0000.0 000.0 02002
0000.0 0000.0 0000.0 000.0 2002
0000.0 0000.0 0000.0 000.0 20002
2 80> 322

000.00 >

0000.0 22-2-2

0000.0 222.2

0000.0 22.6-2

   

 

   

 

71

9632:
25283
aims—88
26205
36288
95288

< 32 m

0233.:
03335
029:2:
3622:
022:2:
033—05

<85 m

032:2:
3622:
023—88
2522:
033—38
033—88

m 35<

3622:
033—88
033—88
039:2:
36205
9522:

m 35 <

 

 

48?: gauged 3 gamma—magma 82 8.99% we soar—3:80 602550 fie 2an xmvaoaadc.

 

 

 

 

 

 

 

 

 

 

 

 

 

336 $86 @332 $56
33.8 mKon QNEQ oat:
comm; ammo; $36 2N;
ammmd mafiw :36 wvod
33+ vmwvd 32d 034m
~33: $3.2 amend moQS
0—

Cmém

can;

334

Eng

PSS< 2:3: mm

535 $23 $.36 Sad
23.? Smfiwm wBNN cowém
3cm.“ cam; Ides www.—
Nonod «find 336 mmad
oamnd chd and 306
$3.: 386 33.: 3M;

2

233%

wan;

33;

SEA

flan—SQ GHGNGH. Ga

 

 

 

 

 

 

885 $86 £86 8.3 «2%
mm-d~ 88.: ~23 08.2 $6
-_2 ¢~v~2 $36 :2 «25
88.... amwo~ ~83 ~3~ wsfiz
886 #22 ~83 -- 3w:
:3: £86 £53 a; .52
2 BS 33

~33 >

52 8:472

~82 053-2

~22 25-0-2

Euoumfl 335m can—am

823 $85 3.8.: oz; «25
mm-d~ 85.: ~23 oS- 25
~22 v~v~2 323 R2 «25
8o? ~83 ~83 ~93 $32
won: «23 ~83 -~.~ 3w:
33g £86 836 a; 3.2
2 SS 33

~w~$ >

~22 052-2

~22 25-92

~22 25-0-2

E83”.— Eausm «En—am

 

 

72

32222
022288.
022288
032222
032288
033.88

<35 m

2,222:
022:2:
032:2:
032:2:
022288
3222:

< 85 m

3222:
032:2:
032288
3232:
032202
032286

mega <

32222
2,2305
2,2288
222:2:
3222:
032:2:

m 22 <

 

 

.222: 2,3030% an 3:22:55 32 moafiam no 22859280 60:52.8 fie 29d- 228%?

 

 

 

 

 

 

 

 

 

 

 

 

 

888 888 -8.~ ~88

28.2 ~23- 82.2 22:.

~:2 8:2 888 :2

R-~ 8-.~ 888 ~8.~

~83 -8.~ 828 5%

88.2 -v- 828 222
2 2

8.8

~82

~22

, ~22

1%

888 88.0 ~88 288

82.2 $2.: 888 ~32

-~2 ~82 28.8 ~82

~83. ~82 288 -~.~

~83 8-~ 828 -~.~

~82: ~23 ~88 2;
E 2

~88

~82

~22

~22

82.5 85 2

 

 

 

 

 

 

888 $88 88.8 ~88 seam
--- 8a.: 28.» 88.2 25
~22 --2 888 P2 «25
8o~.~ --~ ~88 ~28 ~22
~o8.~ 2:2 ~88 -~.~ 282
v8: -8.~ ~88 ~:.~ 222
2 35 BS

-~+~ >

82 22-2-2

~22 22-2-2

~22 22-0-2

88.2— 23am?

88.8 288 88.8 ~88 28
82% ~23: 28.» 822 28
-22 --2 888 :2 ~25
8o~.~ m-- ~88 ~8~ ~22
~c8.~ 8:2 ~88 -~.~ 222
882 -8.~ 88.8 2; 222
2 35 33

-~.- >

-~2 22-2-2

~22 22-2-2

~22 22-0-2

finlg 2.52m

 

 

73

022222
022288
022288
022222
022288
022288

< 22 2

022222
022288
022282
022288
022208
022288

< 32 m

02.0.2022
022288
022288
022282
022282
022202

m 82 <

022288
022222
022202
022288
022288
022288

2 22 <

 

 

.8222 0288.202 22 2022:2225 82 ~2ng .20 28229200 ”2022280 2.0 035. 22020222..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

omwod 286 88.2 «35
33.2 nut-d ~22-.2 92.3
@322 £52 N236 ~32
voomé gem-N 3:6 25d
maomd omwod 33.6 mg.—
Soww 233.2 ~22-2 ~26
22 02

omoadm

2.322

22m.—

$22

22:23.5 822:5 an

£36 886 N326 286
Emmi-m Ont-6- meta-2 ammdv
2202 82.2 3.3.2 322
Emma Swfim 386 232
39.2 wooed gums 32m
228.2 332 3R6 $23

22 2

Gmdm

232

mm;—

SE2

'3

 

 

 

 

 

 

 

 

--.~ --.~ ~83 ~8.~ 22
~82 --.~ --.~ -~.~ 25
~82 88.2 ~83 82 «25
~82 --~ 283 ~_~.~ ~22
-~2 82.2 --.~ -2 :82
~82 2-.~ --.~ -~.~ 322
2 8; 23

82.8 >

~82 ~22-2

~22 22-2-2

~22 22-0-2

2300< .223! 02923

--.~ --.~ ~83 ~8.~ 22.2
~82 --.~ --.~ -~.~ 25
~82 88.2 ~83 82 25
~83 --~ 283 ~_~.~ ~22
-~2 82.2 --.~ 82 :82
~22 2-.~ --.~ -~.~ 222
2 35 33

~38 >

~82 222-2

~22 22-2-2

~22 22-0-2

.2822 .228: ~22—Sam

 

74

95202
38288
38202
25288
95288
838288

<95m

3628:
36288
36282
2,8288
36288
95288

< 25 m

839202

2,6288
38288
95288
8,5288
2,6288

m 85 <

8528:

85288

$7.28:

38288
36288
38288

m 95 <

 

 

.889: 8288.23“ 3 322%:sz 82 838% MO :85an0 682558 fie 832—. 58:33..

 

 

 

 

 

 

 

 

 

 

 

 

 

awed 386 3.86 v36

28.? 833 SEN 8mg

33..— oflmg $.36 32

855 Vde 886 mmo.m

83m Sam fimwd man

83.: owned mum-us 33w
2

53%

832

2:2

SR;

r3

mmmod $.85 aged 39°

33.3 28.2 amt-N 313-

8mm; 3.32 ‘98.: £3

manna matum :36 wvod

3mm;- vawvd New; ogd

mmmnfi $3.2 336 25.2
2

23%

$32

$3.—

Em.—

fihcou< «nu—3m ma

 

 

 

 

 

 

 

885 $85 $86 «mod «2%
882 £23 £86 83 E5
88.. at: 286 a: «25
”32 3.88 236 23 msfiz
Rm? 8%: 888 on? 3»:
3m?“ 28a £86 Ea Eaz
2 8.5 33

22.3 >

822 82-8-2

”22 053-2

~22 25-0-2

6.500(- aux—cm u—N—flam

885 83° 286 «85 25
882 886 8.8.: 88 25
~82 3:: :86 £2 «25
”82 $23 286 23 3.6.2
Rm? can: 886 $2 3»:
was 23m £88 £2 2R2
2 8.5 33

$3“ >

ENE 25-2-2

$23 053.2

~22 25-0-2

Fauu< «£66m can—am

 

75

.082: 328.282 >2 62822285 82 88—92% .20 28.82an0 68:22:00 26 2an 8.205292

838202 38202
2,8288 2,8288
28202 28202
o>8288 28202
28288 28288
o>8288 28288

< 22 m m 22 <

28202 38288
28288 2,8288
2,8202 o>8288
28202 28202
38288 ”$8202
2,8288 28288

< 82 m m 85 <

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6366 $86 N666 .6 636
man 2 Rm 293.: 63.“- 6 mmvfim
5mm; 82.— :m66 vow.—
Nme .m hgmd 3.666 omwd
NBS-v gmmd m: 6N .6 Nm~.m
wax-.2 Nam: 63». 6 m _ 6m
2 2
39%
man.—
af m2
mg m2
mun—«=0 820 ha
336 memo-c N N 66 6 $66
N _ 3.2 monas- Navv6 3nd
m2 m2 mum _ ._ 6686 6mm.—
Omoad wmwmd N686 find
39: Suva; X4266 Km.—
2 ~86 Kcmd 936 6 emu-m
E 0—
Evin
can;
5 2 m2
52 m2

 

 

 

 

 

 

 

$23 $82 282 ~82 «222
283 28.“ 22.2 8% 220
~82 an: :22 22 2.20
mg: 2.~v.~ 28.2 22~ 8:22
~82 2%: 222 82 282
2.82 282 22.2.2 22” 222
2 045 302

22.2 >

222 22-2-2

2:2 22-2-2

~22 22-0-2

283‘ 2:3: via—am

$23 88.0 382 ~23 «228
232 223 222 83 220
~82 2E: :22 2: «220
282 $~2~ 282 S2~ 8222
an? 222 ~32 22 282
222 222 222 23 352
2 0<> 302

22.8 A >

282 22-2-2

2:? 22-2-2

~22 22-0-2

2.884 «ES: 225m

 

76

822222
822222
822288
822222
822288
32288

<82 m

922202
32202
82388
32222
822288
32288

<82 m

32222
95288
022288
822202
8,2288
2,2202

m32 <

2,2222
32222
822288
85202
822222
82228

m 32 <

 

 

.2825 8288qu 3 3222-2225 22 Z 22223 .22 282-82220 6822228 26 2229. 2282.5

 

 

 

 

 

 

 

 

 

 

 

 

 

$32 $82 282 323
83.8 23.: 22%. 22.2
82.2 82.2 2.32 22
22.2. 82.2 882 23
322 22.2 222 :2
22.2 23% 22.2 22.2
2 2

22.2

82.2

2:2

G23

'3

823 $82 $82 236
22.3 22.: 32.2 22.2
22.2 822 :22 33
~83 $22 2222 22.2
~83. 22.2 22.2 ~22
22.2 22% 233 2;

2 2

232

22.2

222
~22 ,

Inn-lasso also 2

 

 

 

 

 

82:. 282 222.2 22:. 225
E22 228. 2.8.2 3.2 25
52 82.2 :82 SE 226
882 22.2 232 22.2 £22
£22 322 :82 28.2 222
22.2- 228 :22 Si 222
2 2 8.5 322
22.2 >

82 22-2 - 2

2:2 22-2-2

822 22-2-2

I3 222.5

8222 $82 2832 «.32 «222
E22 22.2. 28.2 3.2 25
52 82.2 :82 SI «25
882 38.2 28.2 23 2222
322 22.2 :82 28.2 222
2.23 82.2 :22 a: 352
2 2 8.5 322
222 >

82 22-2-2

2:2 22-2-2

823 22-2-2

 

 

o—Dam HEP—02 #6

222%

77

8,2202
8,2288
95288
95288
2,2288
85288

< 25 m

2,2202
2,2202
32208
32288
32288
25288

< 25 m

2,2282
2,2202
2,2288
8,2288
2,2288
2,2288

m 82 <

32202
2,2202
85288
32288
85288
85288

m 82 <

 

 

.8825 gauged 3 3222255 82 832% Mo :85an0 62.228 _.w 2an 228%?

 

 

 

 

 

 

 

 

 

 

 

 

wowed memod ages mmod
3%.th 39963- chm-m .3 mm. 5
2.32 mm:._ QNNQG aw_._
mmwvd ngd N366 3N6
med 232 mm: 6 acmd
mmmvd >3: .0 :vad 35.5
E 0—
SQNm
ENQ—
Noam;
mug;
53.50 than ca
836 38.0 macs wmod
882K Vmfimd aqé .N N gvdv
R3”.— Pfig :36 cm;
Cmmd wmvwd $256 wmod
owgé NCSN :36 32m
$2.3 acmmfi nmmad mmm.:
E 2
Ngém
amNmA
03 m2
v2 m2

 

32:25 «MER— a

 

 

 

 

 

 

 

822 :85 2.8.5 835 25.5
8%.” 386 83.5 8% 25
:51 28.2 235 9.2 25
825 382 £85 :22 2222
as: $82 28.5 82 222
22:. «.83 £25 83 222
5 2 8; 33
83 >

£2 25-2-2

~82 25-2-2

:22 25-0-2

«0:00 3: ch. aw u—Naam

823 285 Re: 2.3 «225
R32 EMS. 38.2 222 26
52 822 :86 83 2,56
85.... 382 285 22 £22
«22 Emma :85 23 322
2.83 225 53 a: 222
5 2 SS 33
52% >

82 25-2 - 2

2:2 25-2-2

83.2 25-0-2

22.5 3.82 3 via—am

 

78

8,2202
2,2282
32288
832202
32202
32288

<92 m

2,2202
8,2288
2,2288
25202
32288
32288

<92 m

32202
25202
03.2288
32288
95202
85288

m 82 <

022202
32288
32202
2,2202
32288
32288

m32<

 

 

.8805 8282.82 .3 32222235 «oz 838$ .20 282.2an0 ”3:228 H6 2an 228992

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

885 285 225 ES
22.2 :22 82.: www.2-
222 :22 882 22
~83. 2R~ 32:0 :3
~22. $-~ 222 22
823 23: 222 22:
2 2

22.2

E~2

S~2

E2

m 222.0 220 a

22:0 823 2.82 2.20
83.0 ~22. ~22 SE
2.22 82: 2.32 m-._
203 :S.~ 322 ~$~
ES 2:2 ~83 22
~23. 22: ~22 22

2 2

22m

:~2

$2

222

23.. S~ 2:25 2

 

 

 

 

 

 

 

~23 885 38.5 ”85 25cm
32.: 38.n- §~§ 82mm 220
:22 ~82 282 22 2220
”ES -2~ 3.22 at: 2222
823 52 ~82 $2. 322
R22 82.: ~22 22.2 222
5 2 8.5 302
83.2 >

:~2 25-2-2

S~2 25-2-2

E2 25-0-2

.2355 0225.— am 2.52m

223 285 :22 $3. 25%
3.5% 28.8 ~23. 88.: 220
232 2-2 232 ~52 dw22.0
§~.~ 223 882 2.2 2222
28.~ «~22 2:2 ~m~.~ 322
22.: 2.22 22.3 83 222
5 2 0<> Bo.—
~om.mm >

-~2 25-2-2

~22 25-2-2

222 22-0-2

15ou a..._=lo.<_ 3. 2 Sam

   

 

79

3.2202
37.202
2,2288
32202
95202
32202

< 32 m

022202
32202
32288
2,2202
2,2202
32288

< 25 2

32202
32202
32288
2,2202
32202
32202

m 82 <

032202
32202
32288
26202
32202
222202

m82 <

 

 

.562 gnombom 3 3222225 82 332$ 20 28222200 62228 #6 2an 22893

 

 

 

 

 

 

 

 

 

 

 

 

 

mmmod meod .6866 v.36

mwooéo mfiodm 6mm?“ 823»

comm; cmmog Ema-6.6 m—NA

aNNmfi wNRN 2666 $66

360*. vmwvd N626 ovmd

N202 352.3 666.66 mac-N—
0—

23%

can;

3;;

3m;

Ego< 223mm mw

nmmod 5356 $666 $56

28.9. Nomm.wm 236 wowdm

33”; 332 R666 mum.—

Nonod aunt-N 66666 Sum-N

samba Sim-N 6566 meM

33.: owned mum-m6 3&2
2

534%

can;

53m.—

22m;

lg

 

 

 

 

 

 

 

 

acvcd N986 «<66 6 N36 «265
gov-co NVE .2 663- .n Sag-m M20
38; _ $66 336 m8; 2220
33 .m mehfi 6666 6 mea.m 3.422
6 ~ 33‘ NON-6N MMVN 6 oofim ZR:
5 _ 8.2 83.x Nm6u 6 63.3 222
2 U<> >901—

ofl 36w >

van.— o2_-m - a

win; o2_-Q - a

33.— 2.2-0-2

EU 3 flan-am

6636 $86 266 6 ~36 mZEm
goo-cc Nag .2 663. N 565m M20
Smog _ Sad m6 3 6 m8; «ZED
32 .m vwmbN 6666 6 Mead 3222
m 2 31a Nomad MWVN 6 confi- Jim—2
23.2 332 N63 6 03.3 222
o_ U<> >90:—

3 madm >

mem; 027m - a

can; o2_-Q - a

$32 0270- a

20 Va 295%

 

80

03335
03385
36388
03335
039.88
039588

< 85 m

36288
023320
95288
33288
039588
03335

< 95 m

.59: 038.30% 3 352%.“:me 82 $3.5m no 833800 60:55.8 fie 2an £259?

25283
033.2:
03335
02335
36288
033—88

m 35 <

2522:
2522:
2,6336
2,6288
023—38
023—05

mSE <

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Jana-sch 53. {2.3. Inn

1

 

336 mama-c aged wmod
wamwdw— Sam-9“- 35». .mm. 8a.:-
hvwmé 80¢.— E 3 6 Nb g A
Go: wag-N NNNQG mwfifi
nmmfiv 38% 3.26 an.m
Kemd ~ cmwvd 3‘66 at:
E 2
agém
comm;
>3 m;
:3.—
[a
good 38.0- 3.36 Vmod
wade goo-Nb.- msm- .2» 03.3
Nmmmg one; 3.36 mag
wag .m eth eQaa 6 NNQN
mug-v v2 _.m 3K Nd awed
33.2 van-w «mm-NR 6 50.2
E 0—
303
mean.—
8%.—
:3.—

 

 

 

 

 

 

 

285 2.85 583» «.35 «225
8:3 5%:- 332 2.3.2 25
22: 835 $85 33 «25
39.5 «~82 :85 32 3,32
3.35 55.: ~55 83 3m:
23% 2;.“ 3.2% R3 32
5 SS 33

88.2 >

252 25-2-2

52 25-5-22

ES 25-0-2

22.5 ma 253.5

235 $85 35.5 «35 «225
3:3 5%: saw 58.: 226
:2: $2.3 $3.5 33 «25
$35 ~23 £85 83 3552
$85 52.: ~55 8.2 3»:
232 2:: 52% E: 222
2 BS 33

88.2 >

252 25-2-2

52 25-5-3

ES 25-0-2

555 a 2.5.5

 

81

32305
2,2305
2,2388
2,2305
9,2205
92205

< 25 m

2,2205
2,2288
2,2286
32288
32283
32288

< 85 m

32205
2,2305
2,2388
32288
32205
033.02

m35<

032:2:
32205
32288
32288
32288
020288

m 35 <

 

 

.089: 352.20% 3 Baflawauma 82 moEEmm 60 5820an0 6025500 66 033. 26:25.2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

weeod 386 3666 636
mmomdh 33.: QRN6~ omvdw
mama; oomfi— 3236 m2;
mmaad «mafim 2‘66 EQN
ommwé omhhd 23.6 unfim
wmocd— anew-- 3666 35.3
E 0—

man-mm

hmmné

cam.—

33."

6.8,.— ha

336 38.0 3666 wmod
$3.3 win-V 666‘: £65.
mug; 336 6.366 aNmA
hmvwd flaw-N 6366 NS.N
mcwvé _mca.N awn-N6 wmhd
366N— mmnhfi 63.56 Gad

E 0—

35mm

comm.—

@634

2.34

239520

 

 

 

 

 

 

 

 

«~85 385- 23.5 3:. «225
$2.28 83.3- 22.:- 222. 230
~83 885 33.5 23 «220
53“ 892 88.5 23 3222
52m.“ 322 252 83 222
2.2.2 ".22 23: E5: 222
5 2 0<> 33
532% >

222 05-2-2

822 25-0-2

3:2 25-0-2

3 2225

28.5 $85 28.5 my; «225
8:3 5%: 322 .032 220
22: 83.5 285 33 «220
032m 2:: 530 83 £32
$82 0%: 22¢ 82 3»:
233 35.“ 2:2 E: 3.22
5 5 0<> 33
88.2 >

252 25-2-2

:22 25-0-2

52 25-0-2

22.5 ma 225m

 

82

922305
32205
32288
9,2205
2,2205
8,230.:

< 95 m

3228:
32205
32288
32205
$228:
32288

< 35 m

3228:
32288
32288
32303
2,2205
2,2388

m 95 <

32208
32288
32288
32288
32288
32388

m 95 <

 

 

2825 250.38% an 322.2555 202 mead-2m mo 50228580 80:23:00 fie 033. 528%?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

222 $22 222 222

22.2 22.: 222 22.2

222 2:2 222 22

N82” 222 222 822

N82. 222 222 N22

28:: N2: 222 22
2 2

22.2

222

2:2

822

222.0 2.5 2

882 $22 222 222

E22 222 322 2.22

52 822 :22 53

882 28.2 222 222

22.2 222 :22 22

28.2 822 :22 S:
2 2

22.2

82

2:2

2:2

13

 

 

 

 

 

 

 

222 822- 222 32 2222
22.22 82.2- 22.2- 22.22 220
22: 232 222 oz: «220
222 822 222 222 $222
«22 222 222 222 222
22.2 22.2 232 22.: 222
5 2 0<> 33
89.2 >

222 25-2-2

222 25-5-2

3:2 25-0-2

2.290 >2.0 mm 222.2

222 8.8.2. 222 S2 2222
2228 82.2- 22.:- 222 220
82: 232 2:2 2:: 2220
22% 822 222 252 2222
22.2 222 222 222 222
22.2 22.2 2%: 22.: 222
2 2 0<> >52
22.2 >

222 25-2-2

822 25-5-2

222 25-0-2

2.26 >26 ma 22:32

 

83

03335
03385
36288
032:2:
039:2:
25205

<85 m

039:2:
038—88
033—86
«>332:
023—88
033—88

< 35 m

36205
02333
36288
2522:
3632:
2522:

m 35 <

2532:
aims—88
25288
25305
aims—88
023—83

m 95 <

 

 

#qu o>=o~bom .3 vegan—magma “oz moan—am mo cemtquoU 62.5280 fic 2an 5339‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

836 38.0 539% wmod
ocmodh $36 $36.: vaNv
hag.— 2‘34 :36 cmmg
Emma wmvwfi @366 wood
owfivé mead mums weed
$2.3 onnd «mad nmmA—
E o—
31%
awn.—
can;
$34
1%
owned 2 8.9 369% 95.0
33.: 3.3.3 Q36.» Sada
8:; mace; ngd 5o;
wwhmd Nm_o.m @356 cafim
mafia PEN a: N d emvd
awmwd :Nad mwmmd 35$
E 0—
mhhém
wcmmg
a2 m4
m2 m4
vnkN pom—«.30 aw

 

 

 

 

 

 

 

 

23¢ 88.? $86 83 «25
«$38 88.3- 22.3. at? 25
~83 83... £86 23 25
53% ENE 885 Sea wsfiz
53 9A? 233 Sam 3m:
$8.2 3%.» 22: mg: :fiz
E 2 39 BS
8%.? >

mm? 052-2

222 053-3

«22 25-0-2

«Ea—am

33¢ 38d $86 ~36 «25
23.8 £3: :36 3.: who
8%; 3.2 an; SE «25
03A 33 $26 23 wsfiz
can: as? £86 £3 3»:
£86 33 R86 83. 23A
3 2 BS 33
88% >

SN? 25472

8%; 053-2

EB 25-93

I235 Eam- 3 295m

 

84

32288
3222:
32288
032:2:
2,2202
022:2:

< 22 m

2522:
2,2305
26388
32288
3222:
03222:

< 22 2

32205
039205
25388
32305
32305
2,2305

22::-

32205
32205
322020
32288
32305
2,222:

m 85 <

 

 

.52.: 0282.202 3 3222222me 82 moi—cam mo comm-89:00 602228 _.e 223- 262392

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

good 33.? cde vmod

5363 floods- max-.2- Ema-mo

mmmmg own: $.36 33

wmofim gnu-N 836 mad

mums. 22:” F226 m3...”

cwmwd~ 3.32 3.de 52.3
o—

awmwm

mean;

can;

:3;

wrung. o_ .H 9...... u. .15.. m. an

336 38d $.36 wmod

vwmoém win-v “532.: ewfidv

mum-w.— 332 3.226 am:

SSW-N wavd 336 mum-N

now: Read and wan-m

mac—.2 mmmhs 33.: Sad
0—

SSH

comm;

can;

05m.—

332-220

 

 

 

 

 

 

:23 282 28.2 :22 «222
22.5 28.2- 2322 23m 25
$32 822 28.2 M2: «25
:2:- 322 282 :3 2222
882 2x2 «23 R:- 322
E2: 322 22.2 25.2 232
2 2 u<> >52
32.3 >

322 222-2

222 22-2-2

222 22-0-2

«38.5

:32 282 282 :22 22.2
22.5 28.2- 2322 23m 25
3.: 822 28.2 83 226
:22 :22 88.2 :3 $22
882. 232 22.2 2:- 3w:
5.2: 222 32.2 892 222
2 2 SS >52
22.: >

32.2 22-2-2

222 22-2-2

222 22-0-2

2: 8.5”.— >25 mm can—am

 

 

85

3622:
023—05
033.38
2,6288
2622:
93335

< 25 m

023.2:
039:2:
023—88
2522:
aims—88
03335

< 85 m

033.05
033.2:
aims—ox»
“56388
2522:
2522:

m 95 <

033.05
033—88
033—83
3622:
023—88
aims—88

mSE <

 

 

48?: gauge-m 3 Rafi—$565 .02 8383 MO :85an0 62.588 fie 2an 598%?-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

memo-o wcmc-o Read 356

mag-3 33.5 MMQQN awe-S.

m3"; 336 336 N34

emom-m ovum-N @366 m:.m

awn-v mug-N .3.de bee-m

Sac-2 awmhd aha-6 0N1:
E o—

way-mm

anmmé

33.—

~93.—

EUBEO ma

Ewe-c Name-o téad mead

RAE-m 3%.: Mam-m. NSdN

2.3”; vmmfi— bum-ad mmmg

mmmwd ambim Qua-N6 mmm-N

cmmvd cmwwé c236 $~.N

33.x «Eh-v QQVm-d ham-o
E 2

Nun-mm

Sum;

mam.“

meg;

lg

 

 

 

 

 

    

 

 

   

$8.: 385 Va: 35 «2%
83.3 38.: 22.2 8...? E6
”.32 SN: $85 a: «25
«82 ~82 52 :3 wzhz
8m:- afi: 52 :2 3»:
38.: 3ch ~83 3:: 3.2
a 2 Q<> 3.:
83m >

52 052-2

922 053-2

«22 25-0-2

2:. B 2955

:85 $86 386 :55 «23
”:32 38.2. 3%.2 23“ MB
$3; 835 386 $3 «25
82:“ ES 3S6 :3 $32
886 $35 8%.: 2:- 3»:
RE: 33m 336 $5 2%
a 2 SS 33
33% >

89: as; -2

32 25-92

as? 25-0-2

t 8.5% >25 mm

   

 

86

029:2:
aims—oi
2,6288
2,6288
aims—0E
032:2:

< 35 m

2622:
032:2:
23388
03335
2,622:
2,622:

<02: m

36205
26205
033—88
3632:
2522:
023—05

m 35 <

029:2:
033—05
032:2:
033.88
2522:
033—2:

m8£<

 

 

.565 gag-aux an gammawcuma 82 3.95m mo scar—3800 603558 fie Bash 565%?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mmaod memo-c- 66 N 6 6 god
vam.w6N 83.5m- 6253. 63.:
866.— Namad 3. N 6 6 6::
hmmfm hams-N 6666 6 ch.m
Vnwmd Sag-N n66w6 coma
Nina-2 mamas 6va< www.:
0—
ommvfim
menA
62 m.—
VE mg
«.5200 >25 mm
mama-o 2 No.6 V6666 eve-o
She-wk. Eva: Rafi-N N acm-Su
36 fl ._ beam-o V366 vac.—
wwamd Sofia 6666 .6 mac-m
vwché Qua-N «mm-6N6 Ema
603.3 55.0 Vmwm 6 ~94 _
o—
flee-mm
ammng
3_ m4
m3 m4

 

 

 

 

 

 

 

 

 

 

38.0 386 3666 ave-o «22m
mmcmdh ammo-E 33.3 omvdv MRO
wacmg com: $36 3: «ZED
mmmafl «mohm 2‘66 _nw.~ 3,32
own: 352 ES ES 3»:
mace-2 mmows $36 3:: 2R2
E 2 U<> BOA

man-mm >

3mm; Bid - a

022 25d - E

33; 25-0 - a

98...— 5 0385

$86 name-o “6666 956 «ZEm
Sou-2. $8.: Emmi awe-me MRO
wag; com: $36 3: «ZED
9mg...” SEN >266 Swd wvéaz
cmmhé amt-N RN56 <36 36»:
$8.2 Rows N636 vow-S 232
E 2 U<> >901—

mandm >

53m.— BE-m - a

62m; 25-n— - a

334 25-0 - E

6.8,..— 5 nan—am

 

87

032:2:
02.2205
033—020
033.05
023—020
030202

< 85 m

030.205
020.205
023—05
02.239:
03.2305
023—05

< 25 0

033—05
033—05
0302020
030305
030305
033—05

m 85 <

020205
02.23020
039:2:
0322—05
032:2:
030-205

0 25 <

 

 

.2065 02002.63. 3 60.3sz“me 32 8383 .20 cam-26800 60:50:00 2.6 030.6. 56:06an

 

 

 

 

 

 

 

 

 

 

 

 

 

:36 626.6 6N666 N36
«Rm-3 mam-6N 6626 86.9»
32.6 wove-6 3666 666;
cam-m Sew-N M6666 $6.2m
wmmcé :mmd EKN .6 666.?
806.2 comm-Z V‘6V6 632$
E 0—

Non-mm

moan.—

Nam;

63m.—

._0=u>aU 5:0 ma

366.6 owned 66666 N36
wmwmém moon-T 2326 SN:
63: 6666.6 66M66 N66.—
¢vwc.m amowN m6: .6 mbmfi
m~mmé mum—.m R6VN6 mmwfi
Mme-3 cum—.2 Gav-6 68.9

E o—

36.3

mom-m;

can;

62m.—

lg

 

 

 

 

 

 

 

 

885 285 28.5 :55 226.5
0868 22.2 33. «Ron 220
v82 8:; 2236 $2 2220
~23 222 ~33 52 £22
22:. ES 22.5 23 7602
38.2 822 22.5 22.: 222
5 2 0<> 30.2
5.2 >

822 25-2-2

52 25-5-2

022 25-0-2

«Seam 3.: I ma 0.685

:35 3.85 285 28¢ 225
2222 $8.2. $2.2 :32 220
$23 82: :82 «:5 2250
29.2 282 28.5 23 2222
m2:- 882 223 22 302
:32 8.22 35.5 at: 222
5 2 0<> 305
22.: >

825 25-2-2

52 25-0-2

EB 25-0-2

X‘— tcomm 6.8% 0.685

 

88

32202
32202
32202
32288
32202
32202

< 82 m

32202
32208
32202
32202
32202
32202

< 22 m

32288
32202
32202
32288
32202
32202

m 82 <

32288
32288
32288
32202
32288
32288

m 82 <

 

 

.0865 3689.635 .3 62226225 82 226526 .8 5823800 ”605228 5.6 833. 228692

 

 

 

 

 

 

 

 

 

 

 

 

 

736.6 6666.6 6 ~ 66 .6 666.6
6666.665 5666.66- 6666 .66 $6.:
$32 ~666._ R6 6 .6 m2;
$66.6 wax-6d 6666.6 6626
666:. 6666.6 66626 N666
K666— 6N6¢.6 6:6 .6 69%:
E 2
666.66
66N6A
h6_ 6 2
_S 6.2
NA «.5026 6.5..—
2666 6666.6 66 66 .6 N666
2 66.66 6666.3 66.2 .6 6666*
36 2 A 663; 6m66 .6 6662
6NNN.6 6666.N 6666.6 326.6
66666 366.6 V62 ~ .6 666.6
636.2 6628.: 366.6 636$
E 6—
N66.66
66N6A
N6 _ 6 ._
665 6. _
8:960 >30 66

 

 

 

 

 

 

 

 

82:0 285- 3.35 «85 22.5
2222 28.2- 22.2. 58.8 20
~22 522 23.5 82 2220
22:2 222 822 ~22 2222
222. 2:2 255 2.22 222
32.2 28.2 22% 222 222
5 2 0<> 33
222 >

82.5 05-2-2

822 25-2-2

EB 25-0 - 2

.20ch 28:3 66 83:26

:22 285- 6.35 23. «222
38.2. 28.2 22.5 ~22-N 020
25: $32 530 $2 2220
88.2 52.2 222 83 2222
an? :2: :55 :22 222
22.2 52.2. 822 52.2 222
5 2 0<> 35.—
222 >

822 25-2-2

222 25-2-2

822 25-0-2

lg 2222

 

89

030205
030205
030283
020288
039205
033205

< 25 0

020205
030322
039288
039288
030305
039205

< 95 0

030205
030288
039288
030205
039205
030288

m 35<

03020.2
020205
020388
03028.20
03030.2
029205

mBfi <

 

 

£025 0>20N20M .3 20202095me 82 838$ .20 80203800 60:50:00 2.2 035. 0220:2522.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ameod owned «63.2 N35
wawmém 825.2- 23:2 3N:
emf; 336 386 N2:
avwcd 33:0. 2226 mad
mane 3&3 3‘3 035.
$2.: 82.: 23.6 80.2
E 2

35%

SN?

can;

023

lasagna?

38.0 38.0 $8.0 $0.0
$8.8 0:: 8%.: 333
mg: 32.2 33.0 am?
523 $30 33.0 $3
Sm: $02 $3.0 mm;
82.2 mm: $5.0 Gad

E 02

26mm

comm;

023

022

0.32505

 

 

 

 

 

 

886 $8.0. 33.2 305 azim
2%de good»- 222. .2. cmmdc MED
Nmmmg 2mm: $3.2 ma: «ZED
waofim ownhd 33.2 mmmN wERZ
g: E 2” 53 $3 302
83.3 Vmomd on? .2 30.3 234
E 2 U<> BOA

awmdm >

SN: 0:22 - 2

80m; 0:2..Q - a

2:2 052.0.52

03:35. 500:: mm 0.955

Seed $.56- $86 $56 aZEm
$3.32 vmooNh- «5:. .2. 336 MED
NmmNA 0mm: 3.36 no: «ZED
mmo_.m cvpd 88.2 NNQN 3602
0224. v: 2m 03 ~ .2 mod $32
33.2 Vmomd 03m .2 $2.2 ZRZ
E 0— U<> BOA

$03 >

won: 002-2 - E

802 0:25 - a

22m; 0:26-52

#3 235

 

9O

032222
032222
032228
032222
032228
032228

< 35 0

032222
032222
032228
032228
032222
032228

< 022 m

032222
032222
032228
032222
032228
032228

m 32 <

032222
032222
032228
032222
032222
032228

m 022 <

 

 

.2022 3200.202 .3 2022:2529 82 8:28am .20 28223800 ”2022280 2.2 038. 222823

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mmcod ommod «68.2 N32
223 22:- 22.» new:
22: Some 228.2 83
$22 223 2.2.8 22
22.2. 222 32.0 2223
22.: 82.: 33% 92.2
E 2

85m

$22

023

at?

3.202 3

2.35 88.0 28.2 23
:23: «comm- ‘82: 2.8.2
:22 23: 222.0 23
2.5m «~23 £86 23
£26 :22 Reg 23.
E22 82.: 253 22.2

E 2

83.2

:2:

82.:

ER:

Essa 82.8.— 2

 

 

 

 

 

 

 

:23 23° 2.82 #3 «2%
5:: 22.: 32 ~52 25
2%.: EN: 286 22 «25
222 ~22 an; 22 $32
22.2 322 286 $3 3»:
28.x 325. 3:6 33 222
E 2 29 33
202 >

82.: 25.2.:

23 25-0-2

232 25-0-2

“Locum tho-m va o—maum

2:3 285 386 $3. «2%
22.2 :82 .82.: 222. 28
m2? :2: $36 $2 «25
23m 22 a»: :3 $32
233. $22 as? 22 .302
28.2 :3: 32S 22: 222
E 2 SS >52
22.2 >,

22.: 25-2-2

82.: 25-2-2

2:3 25-0-2

m 22:6 85 2 0.25am

 

 

91

032222
032222
032228
03220.2
032202
032222

< 8.2 m

032222
032202
032228
032222
032222
032222

< 022 m

032222
032222
032228
032222
032222
032222

m 8.2 <

032202
032222
032228
032222
032202
0322822

m 0:2 <

 

 

2822 03208-202 >2 2022:2225 82 moan—am mo 52.29200 60.22280 26 032.2. 22282?-

 

 

 

 

 

 

 

 

 

 

 

 

 

336 286 ESQ 6 03.0
53.5. h 23%.: ~Qmm.~ N mow-EV
$3 .2 836 V366 3c.—
wwmmd neawN QQQQ 6 macaw
595v «mamfi memd 236
83.3 schmd 3.3 6 2.8.2 2
22 2
286m
Amman.—
32m;
336 $86
83.3 33”.: mag N Ammo. 3»
m2: .— haaad we 3 6 N3.—
vMomxn ovgd c356 a fin
mgmé 882 mwmmd boo-m
239m— awmhd awn-n 6 £2:
22 2
32mm
mmmm2
am 2 m2
N02 m2

 

 

, III-ll

EUR—20 ma

 

 

 

 

835 285 282 22. 225
88.2 23”.» 232 222. 23
52 R22. :82 82 «25
222 222 2.2.2 23 wzaz
2:3. 2:2 222 32m 3»:
32.2, $2.» 222 “2...: 222
2 2 35 BS
22% >
22.: 22-2-2
822 223-2
, .22 22-0-2
1%; 225
$35 285 282 2% 2225
88.2 23.6 23.2 222. 25
52 R22 :32 22 «25
222 222 $22 22 2222
2:1. 2:2 222 $3 3»:
$2.: 82.” 22.2 22: 322
2 2 2.5 33
22% >
2%.: 22-2-2
222 22-2-2

3:2

 

22-0 - 2
lulsssm 0 25 3 2.25m

 

92

033—05
3622:
2532:
033—88
aims—88
033.88

< 95 m

2522:
033—88
033—88
039:2:
03335
36335

<85 m

9532:
33238
058205
36288
aims—88
033.88

m3§<

03335
03335
25288
etc-205
033—88
aims—axe

m8§<

 

 

48?: ”Swag-«om 3 @053»:me 82 mpg—95m no newt-88°C ”325:8 fie 2an £23.?-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

vavod memcd Ngcd mgd
~306— momafi N3€6 30¢
«NZ-fl mug; “586 on;
omoad wmva Nabad VNBN
$3: owe-NA VNVQG Km;
:Nm.m fiend $.36 v2.6
3 0—
fimvém
emmmg
hm _ m;
33;
.093H. .3: ch- mm
:36 wmmod QNQQS N36
224% 2.35m 3.2 6 mcodv
2% fig wove-M fleas o2:
ONNN.m 3ch m6de 25.n-
wmmoé :35 SK ~ 6 vac-V
egg? 88.: ngd 3‘12
E 0—
memfim
mean.—
Ne m4
of mg
3:950 >20 ma

 

 

 

 

 

 

 

£35 885 ~82 an: «25
225 8:6- Skfi $2:- $6
«:2 8:; $86 $2 «25
Emma ".32 886 $3 msfiz
SN? ”83 £56 58 3»:
38.2 $3.” $3 $3: Zaz
2 SS 33

83m >

ca? 85.2

”$2 25-92

~22 25-0-2

.37. .6 a macaw

god 3.86 286 3:. «.25
53;. man: 53 5.8 M5
av? an: 286 :2 ~25
ea: flag Mam; 5:“ wsfiz
52 88; 3&3 :3 .3»:
5; 8o:- ~m§ 33 352
2 93 BS

mama >

SN? 052-2

922 25-9-2

8:; 25-0-2

AU Eam- Ech min—am

 

93

032205
932288
25288
022:2:
032288
032288

< 85 m

032:2:
Dims—oi
32286
2622:
32288
32288

< 22 2

022205
02222:
95388
032205
032288
25388

m35<

032:2:
03222:
032288
032288
032288
032285

m85<

 

 

2865 gauged 3 322.2585 82 mafia-mm no 28229200 6022528 26 033. 262332

 

 

 

 

 

 

 

 

 

 

 

 

 

 

336 $86 MN666 36.0

30566 33.5 «665$ $9;

@232 336 R266 ~34

wwemd ownwd 62.66 6:5

2:6 mmamd Max-NW6 $66

$3.2 $2.6 awn-n6 6%.:
E 0—

wavdm

mam.—

$3.2

N034

20220 ma

mowed $86 $666 956

RVm.Nm Emmé awn-66 3.in

5N3; Sam; K666 8v;

No_o.m vmmod 6366 mmw.~

vm_a.m Emma” R666 56d

vmmafi 386 R26 5:
E 2

52mm

can;

3;.—

$34

[fling

 

 

 

 

 

 

823 $82 28.2 252 2222
22.2 2.8.: 22:- 222 25
N22 332 2222 RS 226
283 222 222 ~22 2222
222 2:2 ~23 ~22 222
$322 23.2 283 at.” 222
2 2 35 BS
22% >

$22 22-2-2

2:2 22-2-2

~22 220-2

3 295m

232 882 282 282 222
22.2 2.28- 23: 222- 28
2:2 2.22 28.2 :22 «25
Emma 822 88.2 282 2222
$22- 283 25.2 33” 322
28.2 22.2 25.2 23: 222
2 2 35 >62
53m >

222 22-2-2

2:2 22-2-2

~22 22-0-2

.57. :0 an 29.2w

 

 

94

2,2322
2,2305
2,2288
32288
32288
32288

<35 m

022:2:
32286
32233
02232:
95238
32288

< 22 2

32222
32205
2,2288
2,2388
033—88
2,2288

m 35 <

32283
2,2288
32288
32288
032286
25288

m 85 <

 

 

48?: gauged 3 32222525 82 moafiam mo 2828820 ”322528 fie 033- 22220222..

 

 

 

 

 

 

 

 

 

 

 

 

mono-o memo-o 3626 Sod

mag-we 33”.: «Hag-u 23.?

avg; 336 V236 N3;

Vmcmfi avg-N 2356 3:”

23nd mam-N mvmmd bee-m

38.n— mwmhd awn-m6 emf:
E 0—

wavdm

awn;

$3.2

NEW-2

[Illa-v.3”

memo-o memo-o mags $5.0

mach-v0 33”.: MMQQN 23.?

avg; 3226 $36 N34

imam-m ovN-wfi 2366 3:.-

QVmé mag-N .3de Sod

$22.2 amp-a amaze 09.:
E 0—

wag-mm

mam.—

$3.2

N03;

SUB—:6 ma

 

 

 

 

 

 

82:. $82 2332 $3 «222
52m 222. 32.2 3.2 25
52 82.2 :82 SI «25
8:2 382 282 83 2222
$22 222 28.2 2.82 3m:
«83 233 :22 a: 222
2 2 SS 322
23.2 >

33 22-2 - 2

2:3 22-2-2

323 22-2-2

1% 3&an

823 28.2. 23.2 232 «222
22.2 E; 33 $2: 25
as? 822 28.2 :3 «25
«83 222 2:2 2: 2222
332 882 :22 a? 322
82.» 283 2.23 23 222
2 2 SS 322
88.2 >

52 22-2-2

2:2 222.2

$2? 22-0-2

LI

 

(Hts—6&0 Uflmufléam @@

228$

95

2,828:
2,828:
aims—88
833—88
3828:
833—88

<35 m

95305
38388
38388
2,6205
£5288
£8388

< 82 m

9528:
3828:
85388
25288
83.0.22:
aims—88

m 95 <

9528:
3838:
833—88
023—05
833—88
833—88

m 33 <

 

 

.889: 838.38% 3 32818585 “oz main—Sm mo comm-89:00 683538 fic 835- fivfixfi<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mcmod memo-o Mmoed 35d
moon-3 haemS— MMQQN mmoév
ave; Sam-o 336 «no;
gem-m AVE-Wm 0356 ¢:.m
933V mmmmd vaNN-e boo-m
#392 amp-a 3.3.6 31.:
2
31mm
¢mNmA
$34
SEA
203:6 ma
335 $86 mged Eyed
mack-4% 3342 MMQQN $93
35“; Road «636 N8.—
VMcmd ovgd “:de a:.m
m—Vmé mmaad mVNNG boo-m
~39? amp-a awn-ad 8?:
0—
mow-mm
mmmmé
$34
NEW—
EUEEO ma

 

 

 

 

 

 

 

$86 $85 $86 Sod «2%
82.8 $3.: $2.» moi-m MB
an? 83 286 $2 «23
~82 SEN :36 8mm msfiz
~83- fimg Egg Nam :82
88.2 as; $63 2; :82
E 2 8; 33
”8.: >

”82 258-2

2:2 035-2

~22 25-0-2

3226 2.5 8 2958

$23 $85 as: Sod seam
83.8 28.: 2:8 2&2 8:6
8%; 88: £86 82 «25
22a $88 $86 82 wgz
5.3 5:: 8:6 :2 3»:
Ex: 83% mag ”$6 2%
E 2 SS 33
82.8 >

SN? oar-TE

8:? 25-92

§2 25-0-2

'3 285m

 

96

03335
25238
023—88
3622:
022:2:
aims—05

< 35 m

38205
33288
25286
25288
36388
023.88

< 25 m

.565 grow-aux 3 605333me 82 moan—am .«o aomtdeoU 6255.8 66 Baa-w xxx—25¢.

023:2:
032:2:
033—83
2622:
039:2:
3622:

m 35 <

033—05
3632:
25288
25205
«>633».
033—88

m 23 <

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$36 mowed $66 .6 636
Chem-mam 3.3..ch Emu-6m ohm-mm
She.— wgad ~NN66 :64
«Neg-m vaN «‘2 6 wam
3566 ewmvd 666N6 «ma-v
unmod— cmwm6_ RE 6 Q 03
E o—
- AVG-mm
van;
ow _ m;
5 _ m4
lg
Nvmod memo-o 3.66 6 636
aha-mam $3.63- E wvfin ohm-mm
:2: wgad ~NN66 :6.“
$93” met-N RR: 6 wwo.m
mwhcé Saw-m 666m 6 N36
mewég omwmdfi RE 6 S 02
E 0—
636m
3mm;
ow _ w.—
$~ m._

In] a? .30.: «a

 

 

 

 

 

 

 

286 $85 $86 3.3 «2%
38.8 28.9. at.“ 813- M5
2:2 32: R36 22 «25
as? “WEN 2&6 ”3m £32
3%.,- vawea S26 9.? 3w:
«3.2 5:: $83 25.2 262
o. 9.5 33

23% >

can; Ban-TE

2:2 22-9-2

22 3-0-2

F83< «23% mm 29:3

885 $85 $86 Sod «2%
:8? 83% SEN 8mg M5
:3: e22 286 $2 «25
88a 52 886 R3 mzaz
can; oz? and £3 3»:
$8.: 88¢ $3 $3 262
2 35 BS

53% >

$22 Ban-TE

2:2 25-92

822 82-0-2

:80 8: oh ea «Ea—am

 

97

3222:
032505
25288
32288
32205
32288

< 85 m

2,2205
25288
033.88
2,2305
32205
32305

< 25 m

033—05
2,2388
25388
25288
033—05
023—88

mgfi <

2,2205
2,2205
32288
033.05
32305
32388

m 25 <

 

 

.0865 2.608.63— 3 63222525 82 moi—Sum no 5033800 6035.80 66 035. x6523

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mace-6 cmmcd 66666 N36
wmwmdm mock-.7 .2326 mom-S
cnwfig 3666 6366 Nae;
659m amend m6: 6 mhmd
Ewmé mnmfim 86:6 636
awe—.3 82.: 336 omv.-
0.
356m
man;
can;
62m;
385m :6
owned N— 86 6m. 66 6 $56
Gem-5» ahemdm 6N2.» 666.?“
82.— wmoog 66m 6 6 $6.—
owhmfi N266 66666 636
Nan-N fined 62 ~ 6 Exam
awnofi 336 MV6N6 02k.
2
mun-3
mean;
663;
mtmg
VNuN 2:930 an

 

 

 

 

 

 

 

33¢ :85 280 285 55.5
28.8 5:? 8%.: £33 080
m5: m8: 822 82 «230
832 28$ 8.2.5 23 £32
«£3. 88.~ 825 82 .3»:
82.2 a: 825 8.8 :82
5 2 0<> 33
8:2 >

083 25-2-2

8:; 25-0-2

as? 25-0-2

0.32.265 oEEam

$85 285 $85 Sod «25
8.8.2 3.8.2 28.2 3.8 2R0
88.. 082 8:2 2: «230
$82 882 $25 $3 “0232
882 2%; $2.5 83 3m:
82.” 38+ $3 23 32
5 2 0<> 33
«2.2 >

882 25-2-2

882 25.0-2

ES 25-0-2

can—am

 

 

98

.89: gnaw-aux B con—2.2525 82 838% no 58528.80 683528 2.0 838.2. 298an

822.85 022.85
32288 32205
32388 823288
32388 823—88
823—88 82388
32338 ”52288

< 35 m m 35 <

3232: 823.2:
822.85 32205
823—88 823—88
3222: 822.82
32:85 822288
822.82 22388

< 22 m m 22 <

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

22.85 >26 8

2.28 88.5 882 83

88.28 88.2:- 512 822

:82 885 286 :3

88.2. ER 233 23

28.2- 282». 286 N2:-

882 88.2 233 .222
2

228

882

2:3

8:2

[3

:23 882. 88.8 $3

28.8 88.8 82.» 828

322 832 E86 83

882 322 38.8 :52

88+ :82 8:2 83.

88.2 88.: 3.82 83.2
2

8.2

M82

832

. 8:2

 

 

 

 

 

 

 

336 886 N356 mvod «28m
23.8 33.5 Non-V6 Smd MED
mam; m5: 386 can; 2220
@382 memN N956 van-N 3222
$3.2 94m; wfiéd EA 222
28$ Rena 2.36 3K2” 222
2 U<> >901—

Evém >

can; 25-2 - a

$3.2 25-2 - a

83.2 25-0 - E

898,—- 3: on. ma «Ea—am

vaod ommod 336 Quad «ZEm
mom-n.3— mommd .83 .2 wmwdc MED
amt-:2 man: :36 22 «220
mummd 8de Quad ~86 3222
8.2- 882 E 8.8 282 3»:
8:2 83.x 383 one? 222
2 U<> BOA

v3.9 >

mean; 2.2.2 - a

323 255 - E

822 25-0 - E

58,—- 3.5 va oEEam

 

99

aims—05
039:2:
3822:
2532:
032:2:
33205

< 85 m

03335
25288
03,-3.on
aims—88
25288
26288

< 35 m

2522:
033—88
033—2:
033—2:
033—2:
3632:

m 25"

25205
023—88
023—88
033—88
aims—88
023.83

m3fi<

 

 

 

 

 

 

 

$36 $86 28 6 $66
38.8 Nva .2 36m. N 25.5”
38.— :wmd «656 mm:
32 .m mehd babes mead
231g Nona-N mwvms 02.6
530.3 83.x Nam .Q 03.2
E o—
3 mmdm

«www.—

ow _ m;

5 ~ m4

EU ea
33d Nomad 28 6 N36
$8.8 avg .2 36%. N Sag-m
«ewe; :wad Mb 36 m8.—
nae _ .m mehd eggs mcafi
936 mogfi mmvmd @056
23.3 wocww when 6 93.3
E 2
3 86m

memA

cm _ m4

5 _ m.—

 

 

 

20 ea

48?: gauged E 3&3?sz “oz 838mm mo cam-"38°C 628880 fie 03¢.“- 530mg

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

$85 385 $86 Sod 22m
33%; «33:- EEK 2me M5
ES ”3% :86 :3 «25
39% SEN :53 £3 mzzz
3%.... 08$ 35% $3 3»:
$8.: 832 R36 «3.2 :fiz
E 2 35 BS
292 >

«22 Bad-E

2:2 aid-E

3:3 25-0-2

.8 a3 .30.! am 295m

323 885 35s m2; «25
232 83s a»; 32 M5
ES m5: 88¢ 82 «25
893 mafia 836 $2 wzhz
$9: 8:; R23 :2 3w:
:qu 5.3 3.36 $2 3.2
E 2 SS BS
51% >

can; Baum-2

2:2 Quad-E

§2 25-0-2

I3 «flea

 

100

32288
832288
83.0.32:
832:2:
832:8:
83.0.28:

< 22 m

3228:
832.85
83228:
95288
32:85
2,232:

< 22 m

83.0.28:
822:8:
2538:
2,238:
3238:
2,228:

m 22 <

3828:
3.2288
8,238:
3238:
832:8:
2,228:

m 85 <-

 

 

235 gauge-m >2 32222::me 8 Z 8.28% we 28288.80 683538 fie oSaH “£88292

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mmmod $86. .336 mmod

memdom ogmfin- 33:: 31mm

mace.“ Nomad 3.36 oz:

hmmvfi SNEN gees Nwo.m

Emma chm.N 33.6 cow.”-

NVmNS mmmwfi QNQQ mum-Z
0—

cmmvdm

mam.—

cam;

$202

1%

amwod ommod 386 N36

wawmdm meg-.7 23.6 3N:

cmwfifi 336 386 Nae.—

mvwcd amwwfi 3:6 mhmd

2wmé mum—.0.“ ngvmd mmwfi

$3.3 ommfi: Baud enqmg
0—

35.3

mommé

cam.—

ELF

«to-Sm— ca

 

 

 

 

 

 

 

823 885 222 33 «2%
23.8 2%.: «23 23.:- 25
$22 835 $22 N2: «25
«$2 252 22.2 2; W222
232. 282 2&3 22m 322
38.2 222 222 2%.: 222
2 2 8; 33
222 >

882 22-2-2

222 22-2-2

~22 22-0-2

2885 2 228.2

:2; 286 322 :52 ~25
2:22 2.82- E%.2 28.8 26
2:2 826 22.2 m2: 228
:8.»- EZ 38.2 :2 mzhz
882 222 882 23 3»:
E2: 322 32.2 822 3.2
2 2 u<> 33
33% >

882 22-2-2

222 22-2-2

, 832 22-0-2

to 8.5m >25 mm 8.285

 

 

101

2,6303
2,8303
3.6303
03883
25388
25388

< 93 m

eta-303
2,8303
03,8303
838303
03.8303
38303

< 22 m

38303
25303
38388
2,8303
2,8388
2,8388

m 22 <

2538:
38303
38388
85303
25303
833.03

m33<

 

 

.889: grow-aux .3 333%:sz 82 8.95% «o acmtaEcU 62.3280 fie 838. 838593

 

 

 

 

 

 

 

 

 

 

 

 

 

m _ mod wwmod mgad mad
meg-g wan 3.x. Nam-9N 03d 3
m3?— mcvad mass #83
3‘91” mas-m 3.de wcm.m
owned chi-.3 N34 6 and
awamd mm 36 «mm-:6 53.5
3 3
88.3
monA
33 m3
:3 m3
8.8.;— mm
38.0 memo-o- mas-Q mmod
vam.wom 83.5m- bNNm-«g‘ 38mm
Soc.— Nmmad 3.36 oz:
hmmvd hams-N eagaé awed
www.md Sum-N 2.5.0.6 coma
NVmNc— mmmms 9N3; mum-Z
E 3
ommv.mm
mem._

can.—

 

 

 

 

 

 

 

 

823 885 886 ~38 «25
88.8 832. 23.» mom: who
ca: 322. 886 83 «25
$82 $23 253 Ea 8:82
2%.... m5."- 886 an: 382
$2.: 82.: 32% 822 32
E 2 8; 33
35% >

882 22-2-2

832 22-2-2

at? 228-2

3.28 8 mafia

33¢ 28... £85 $3 225
$8.? :32: 38.: 823 $6
$22 835 2.86 :82 25
88.". 823 886 83. wzhz
355. «83 3.3 Rm.»- 3»:
88.: 835 5a.: Rm: _<\~z
E 2 89 BS
$23 >

322 22-2-2

332 22-2-2

2:2 22-8-2

2:3 :95 £30 ma «En—am

 

102

022:2:
022:2:
0202020
022:2:
022:88
0202020

< 22 m

022:2:
02020.:
022:2:
022:2:
022:2:
022:2:

< 0:: m

022:2:
023—020
023—88
022:2:
023—020
02335

m 22 <

022:2:
022:2:
022:2:
022:2:
022:2:
022:2:

m o::<

 

 

 

.202: 02208-202 a: 3:232:35 82 838mm .20 28:29:00 60:52:80 :6 03.»:- 0:309?

 

 

 

 

 

 

 

 

 

 

 

 

 

wwmod Vwmod mgcé mgd
Nam-mm Rea-m: m~mm.m- www.mm
meg: chm—.2 @036 5:.—
avoflm oamo.m cc: 6 356
cmch 93%.— 3.2 d wmfim
amvwfi moomé Nu mu 6 N36
2: 0—
www.mm
3%.:
8%.:
K: w:
200.28 .5202 Na
mhmod wmmod $de :56
ONE}; m «Long-2- MNVM 6m. mow-mk-
NmeA gem: 3.36 am:
bmomd ommvd mused god
:56 NEON SNNd in...”
ace: .: amgé “.366 and
2: 2
www.mm
can:
can:
mzm:

 

 

.525 5.. a. ma

 

 

 

 

 

 

 

885 285 28.5 285 «225
«82.22 2:32- %32 80.2 25
822 32.2 236 22 «25
3‘2 882 225 E: £22
3%... 352 25.0 $3 3m:
32.: 3%.” 2:2 82: 222
5 2 9.9 23
22.2 >

E? 25-2-2

222 25-92

EB 25-0-2

3.52225 2 222%

88.0 28... $8.0 285 «225
£222 2:82- %3.& 822 26
822 «32 23.0 22 «25
2.3.,” 083 32.0 23 msfiz
as? 35? 825 32m 302
32.2 5.3 2.3.5 8a.: 222
5 2 9; 33
28.2 >

END 222-2

8:: 25-9-2

2:: 25-0-2

3522.5 2 0.22am

 

103

022:2:
022:2:
0202020
0202020
022:020
0202020

< 02: m:

022:2:
0202020
0202020
022:020
022:020
022:020

m 0:: <

 

 

20:00: 02202-23: b: 02:22:23.5 :07: $588 :0 52.29800 6025200 :6 050:. 2053.3

 

 

 

owned «:86 369% $66
:mcmfiv anomdm Q36.» 8a.?”
02:: wmoo: 3036 $9:
cwbmd mm:o.m mcbad 3:6
max-N KEN 92 ~ 6 emvd
mwmed :mmad mvmmd 0:8
2: 0:
mtém
wcmm:
3:0:
mtw:
v~uN 202930 am

 

 

 

 

 

 

 

 

88.0 280 280 25.0 «225
£222 N222- 208.2 80.2 020
82.: 22.: 230 22 ‘0720
35m 882 22.0 23 0222
283 $52 202.0 «22 202
$8.2 3%.” 23.0 20.: 32
5 2 0<> 23
22.2 >

can: 85-2-2

2:: 25-0-2

2:: 25-0-2

02:23:20 2 22:50

 

 

104

Appendix VII

Summary Data for Refractive Index and Elemental Analysis Comparisons

l. Refractive Index Only

 

N = 2450
Included = 186 (7.6%)
Excluded = 2264 (92.4%)

 

 

 

Appendix Table 7.1: Summary of Refractive Index Data.

2. Elemental Comparisons

a) Different Refractive Index

 

N = 906
Inclusive = 63
Exclusive = 696

Inconclusive = 147

 

 

 

Appendix Table 7.2: Summary of Elemental Comparisons for

Samples with Different Refractive Indices.

b) Same Refractive Index

 

N = 186
Inclusive = 20

Exclusive = 131

 

 

Inconclusive = 35

 

Appendix Table 7.3: Summary of Elemental Comparisons for

Samples with Indistinguishable Refractive Indices.

105

c) Refractive Index Ignored

 

N = 1092
Inclusive = 83
Exclusive = 827

Inconclusive = 182

 

 

 

Appendix Table 7.4: Summary of Elemental and Refractive Index Comparisons.

3. For all 2450 comparisons,

 

Inclusive (RI & SEM-EDS) = 20 (0.8%)
Exclusive (R1, or RI & SEM-EDS) = 2395 (97.8%)

 

 

Inconclusive (RI & SEM-EDS) = 35 (1.4%)

 

Appendix Table 7.5: Summary of All Comparisons.

106

Notes

Chapter 1: Introduction

 

' Dabbs, M.G.D and BF. Pearson. 1972. Some physical properties of a large number of
window glass specimens. Journal of Forensic Sciences 17(1): 70-78.

" Reeve, V., J. Mathiesen and W. Fong. 1976. Elemental analysis by energy dispersive x-
ray: a significant factor in the forensic analysis of glass. Journal of Forensic Sciences
21(2): 291-306.

'" Miller, ET. 1982. Forensic glass comparisons. In Forensic Science Handbook, ed. R.
Saferstein. Englewood Cliffs, NJ: Prentice-Hall Inc.

" Ibid.

Chapter 2: Scanning Electron Microscopy m the Analysis of Glass

V PFG. (1999). Window glass. About glass http://www.pfg.co.za/
about_glass_window.asp (29 September 2000).

"‘ Saint Gobain Glass. ( ). Float glass. Glass teach-ins http://www.dockrellglass.ie
/prod/teachin/float.htm (29 September 2000).

V" PFG. (1999). Window glass. About glass http://www.pfg.co.za/

about _glass_window.asp (29 September 2000).

V‘“ Saint Gobain Glass. ( ). Float glass. Glass teach-ins http://www.dockrellglass.ie
/prod/teachin/float.htm (29 September 2000).

"‘ Reeve, V., J. Mathiesen and W. Fong. 1976. Elemental analysis by energy dispersive x-
ray: a significant factor in the forensic analysis of glass. Journal of Forensic Sciences
21(2): 291-306.

Chapter 3: Review of Literature

" Brown, G.A. 1985. Factors affecting the refractive index distribution of window glass.
Journal of Forensic Sciences 30(3): 806-813.

"‘ Ibid.

"“ Hickman, D. 1981. A classification scheme for glass. Forensic Science International,
no. 17: 265-281.

"‘“ Koons RD. and J. Buscaglia. 1999. The forensic significance of glass composition and
refractive index measurements. Journal of Forensic Sciences 44(3): 496-503.

’“V Ibid.

"V Andrasko J. and AC Maehly. 1978. The discrimination between samples of window
glass by combining physical and chemical techniques. Journal of Forensic Sciences
23(2):250-262.

“i Ibid.

xvii Ibid.

Reeve, V., J. Mathiesen and W. Fong. 1976. Elemental analysis by energy dispersive x-
ray: a significant factor in the forensic analysis of glass. Journal of Forensic Sciences
21(2): 291-306.

107

 

""‘ Brown, G.A. 1985. Factors affecting the refractive index distribution of window glass.
Journal of Forensic Sciences 30(3): 806-813.

"" Andrasko J. and A.C. Maehly. 1978. The discrimination between samples of window
glass by combining physical and chemical techniques. Journal of Forensic Sciences
23(2):250-262.

""‘ Reeve, V., J. Mathiesen and W. Fong. 1976. Elemental analysis by energy dispersive
x-ray: a significant factor in the forensic analysis of glass. Journal of Forensic Sciences
21(2): 291-306.

Cypter 4: Development of Suitable SEM-EDS Operating Parameters
“ii Ary D., Jacobs LC. 1976. Introduction to Statistics: purposes and procedures.
Winston NYzHolt Rinehart.

C_h2_tDter 5: Conclusions_ and Discussion

“i“ Saint Gobain Vitrage. ( ). Glass Manufacture. Fabrication du verre http://
www.saint-gobain-vitrage.com/english/manufacture/fr_fab.htrn (3 April 2001).

”“V Ibid.

x“ Reeve, V., J. Mathiesen and W. Fong. 1976. Elemental analysis by energy dispersive
x-ray: a significant factor in the forensic analysis of glass. Journal of Forensic Sciences
21(2): 291-306.

“Vi Advanstar Communications. (2001). A guide to inductively coupled plasma mass
spectrometry. Spectroscopy. http://www.spectroscopymag.com/articles/Ol03_articles
/0103 _poster/0103 _poster.html (3 April 2001).

“V“ Parouchais T., et al. 1996. The analysis of small glass fragments using inductively
coupled plasma mass spectrometry. Journal of Forensic Science 41(3): 351-360.

108

Bibliography

Advanstar Communications. (2001). A guide to inductively coupled plasma mass
spectrometry. Spectroscopy. http://www.spectroscopymag.com/articles
/0103_articles/0103 _poster/0103 _poster.html (3 April 2001).

Andrasko J. and A.C. Maehly. 1978. The discrimination between samples of window
glass by combining physical and chemical techniques. Journal of Forensic
Sciences 23(2):250-262.

Ary D., Jacobs LC. 1976. Introduction to Statistics: Purposes and Procedures. Winston
NY: Holt Rinehart.

Brown, G.A. 1985. Factors affecting the refractive index distribution of window glass.
Journal of Forensic Sciences 30(3): 806-813.

Cooper, BE. 1969. Statistics for Experimentalists. New York: Pergarnon Press.

Dabbs, M.G.D and BF. Pearson. 1972. Some physical properties of a large number of
window glass specimens. Journal of Forensic Sciences 17(1): 70-78.

Dudley, R.J., et al. 1980. The discrimination and classification of small fragments of
window and non-window glasses using energy dispersive x-ray fluorescence
spectrometry. X-ray Spectrometry 9(3): 119- 122.

F legler, S.L., J .W. Heckrnan, and KL. Klomparens. 1993. Scanning and Transmission
Electron Microscopy: An Introduction. New York: Oxford University Press.

Goldstein, J .I. and H. Yakowitz, eds. 1975. Practical Scanning Electron Microscopy,
Electron And Ion Microprobe Analysis. New York: Plenum Press.

Hickman, D. 1981. A classification scheme for glass. Forensic Science International, no.
1 7: 265 -28 1 .

Hickman, D.A., G. Harbottle, and E.V. Sayne. 1983. The selection of the best elemental
variables for the classification of glass samples. Forensic Science International,
no. 23:189-212.

Koons RD. and J. Buscaglia. 1999. The forensic significance of glass composition and
refractive index measurements. Journal of Forensic Sciences 44(3): 496-503.

Lloyd, J .B.F . 1981. Fluorescence spectrometry in the identification and discrimination of

float and other surfaces on window glasses. Journal of Forensic Sciences 26(2):
325-342.

109

Parouchais T., et a1. 1996. The analysis of small glass fragments using inductively
coupled plasma mass spectrometry. Journal of Forensic Science 41(3): 351-360.

PFG. (1999). Window glass. About glass http://www.pfg.co.za/about_glass_window.asp
(29 September 2000).

Reeve, V., J. Mathiesen and W. Fong. 1976. Elemental analysis by energy dispersive x-
ray: a significant factor in the forensic analysis of glass. Journal of Forensic
Sciences 21(2): 291-306.

Saint Gobain Glass. ( ). Float glass. Glass teach-ins http://www.dockrellglass.ie
/prod/teachin/float.htm (29 September 2000).

Saint Gobain Vitrage. ( ). Glass Manufacture. Fabrication du verre http://
www.saint-gobain-vitrage.com/english/manufacture/fr_fab.htrn (3 April 2001).

110

   

IIIIIIIIIIIIIIIIIIIII

ll'l'lllllllllllIt'll