LIBRARY

Michigan State
Unlverslty

 

 

 

PLACE IN RETURN BOX
to remove this checkout from your record.
To AVOID FINES return on or before date due.

 

MTE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1!” COMM“

THE DIFFERENTIATION OF LASER PRINTER TONERS
USING DIFFUSE REFLECTANCE INFRARED FOURIER TRANSFORM
SPECTROPHOTOMETRY
By

Troy Jonathan Ernst

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

MASTER OF SCIENCE

School of Criminal Justice

1 997

ABSTRACT

THE DIFFERENTIATION OF LASER PRINTER TONERS
USING DIFFUSE REFLECTANCE INFRARED FOURIER TRANSFORM
SPECTROPHOTOMETRY

By

Troy J. Ernst

In this project, laser printer toners were analyzed using diffuse reflectance
infrared Fourier transform spectroscopy (DRIFTS). The purpose of this research
was to evaluate DRIFT S as a method for differentiating laser printer toners. The
spectra were collected, differentiated, and classified based on distinguishing
peaks of absorption. A library of spectra of laser printer toners was produced,
which was tested by re-running several samples and comparing their spectra to

the library.

ACKNOWLEDGMENTS

I would like to extend my appreciation to Dr. Jay A. Siegel, who provided a
plan, knowledge of the subject and of the instrumentation, and helped things to
run as smoothly as a research project can. His guidance and assistance enabled
the research to overcome several obstacles and ultimately reach a conclusion.

Also, thanks go out to Jerry Bockes, who was my primary source for

toners.

TABLE OF CONTENTS

LIST OF FIGURES ..................................................................................... v
LIST OF ABBREVIATIONS ........................................................................ vi
INTRODUCTION ........................................................................................ 1
Laser Printers .......................................................................................... 2
DRIFTS ................................................................................................... 4
Past Research Focusing on Differentiation of Toners ............................. 5
PROCEDURE ............................................................................................. 8
RESULTS ................................................................................................... 1 1
DISCUSSION ............................................................................................. 23
APPENDIX A — NAMES OF TONERS ........................................................ 28
APPENDIX B - TONER GROUPS ............................................................. 29
APPENDIX C — INFRARED SPECTRA OF TONERS ................................ 31

BIBLIOGRAPHY ......................................................................................... 57

LIST OF FIGURES

Figure 1 - Path of Laser in DRIFTS Unit .................................................... 4

Figure 2 - Groups I through VII .................................................................. 13
Figure 3 - Groups VIII through XIII ............................................................. 14
Figure 4 - Spectra of Top Matches from Library ......................................... 16
Figure 5 - Spectra of Top Matches from File .............................................. 17
Figure 6 - Overlays of Unknown with T008 and T004 ................................. 19
Figure 7 — Overlays of Unknown with T003 and T002 ................................. 20
Figure 8 - Overlays of Unknown with T015 and T001 ................................ 21

Figure 9 — Overlays of Unknown with T009 and T007 ................................. 22

LIST OF ABBREVIATIONS
cm'1 - wavenumbers .
DRIFTS - Diffuse Reflectance Infrared Fourier Transform Spectrophotometry
FTIR - Fourier Transform Infrared
GCIMS - Gas Chromatography I Mass Spectrometry
HPLC - High Performance Liquid Chromatography
IR #- Infrared

KBr -- Potassium Bromide

INTRODUCTION

In the field of forensic science, there is much room for empirical research.
This research focuses on questioned document analysis, which is a discipline
within the forensic sciences. The term "questioned document" refers to any
document of which the source is unknown (Saferstein, 1990). Examples of
questioned documents include forgeries, notes of extortion, and letters of threat.

Questioned documents can be handwritten, typewritten, copied, or printed
by an ink-jet printer, a laser printer, etc. Another form is the document that is
made by cutting letters out of magazines and newspapers and affixing them to
paper to form the desired words.

Each form of questioned document requires a different method of
examination. A handwritten document is compared to exemplars taken from
suspects, in order to determine the similarities and differences in the printing or
writing styles between the documents. An analyst of a typewritten document
compares the typefaces used and the irregularities in certain letters due to the
normal wear and damage to the typewriter’s moving parts in the questioned
document with a specimen collected on the suspect’s typewriter. Questioned
documents prepared on a copier or computer printer may be analyzed by
studying the makeup of the toners or inks present on the paper. Cut-and-paste
questioned documents allow for the comparison of adhesives found on the paper
with adhesives obtained from the suspect.

The research presented in this thesis focuses solely on the analysis of

questioned documents prepared by a laser printer. Because laser printers and

electrostatic copiers operate on the same principles, the conclusions of this
research apply to copier toner analysis as well.

With the rise in technology and the lowering of prices for advancements in
printing and copying procedures, a growing number of people have greater
access to equipment that can produce printed documents of seeming anonymity.
A result of this situation is the need for improved measures for the differentiation

of these documents.

madam:

’ Laser printing requires several steps to produce the final printout. First, a
cylindrical photosensitive drum receives a cleaning from the last image it made
by a cleaning blade and an erase light. The blade removes excess toner, and
the light erases any prior images. A negative 600-volt charge is given to the
drum by a thin wire called the primary corona. It is then ready for a new image,
which is drawn on the drum with a laser beam reflected by a movable mirror.
The new image is made wherever the laser strikes the drum by changing the
voltage to negative 100 volts. Because the toner has magnetic properties, the
toner is deposited on the drum where there is a more positive surface. This is
done when the drum rotates past the toner particles. The next step is the
transfer of the toner to the paper, which has been given a positive 600-volt
charge by another thin wire, the transfer corona. Because of the paper's positive
charge, the toner is transferred to the paper. The paper then undergoes a
reduction in charge. The final step is fusion of toner onto the paper. The paper
is rolled through a metal roller at a temperature of 180 degrees Celsius, which
melts the toner onto the paper (Minasa, 1992).

Laser printer toners are comprised of varying amounts of pigments,

binding agents, and various additives that depend on the brand and function of

the toner. The pigments give the toner its color. The most common pigment is
carbon black, although additional pigments are sometimes added. The binding
agent affixes the pigment to the paper and is a mixture of synthetic organic
resins. Additives ensure the toner particles have the correct polarity to remain on
the charged paper before the fusion of the toner and the paper takes place
(Mazzella, Lennard, & Margot, May 1991).

Laser printers are sufficiently common that attempting to trace a printed
paper to a particular source is not normally feasible. Only when a particular
printer leaves unique marks on every page of paper it prints could
individualization of a document to a specific printer take place (Brunelle, 1982).
Since individualization is the exception, methods need to be developed that can
associate a printed paper with a class of laser printers. The class could include a
single brand and model of printer, or a variety of brands and models, each
producing the same distinguishing features.

Analysis of the toner on the printed paper represents one method to
classify documents. Laser printer models specify which type of toner is to be
used in printing. For example, one model of laser printers may specify Oasis
Eclipse laser printer toner to be used in it, while another model may call for
Laser1 Aspen Toner. If a method could be found that could differentiate between
the toners, then a document examiner could compare the test results of an
unknown source to the results of known toners. The result would be that some
laser printers would be eliminated as the possible source of the questioned
document, narrowing the scope of the criminal investigation. If it could be shown
that the suspect had access to a laser printer with toner that matched the
questioned document, then it is a piece of corroborative evidence, linking the

suspect to the crime.

QBIELS

Diffuse reflectance infrared Fourier transform spectrophotometry
(DRIFTS) uses properties of molecules when they are irradiated with infrared
radiation. Radiation in this region includes the wavenumbers from 4000 to 650
cm4. A laser beam provides this energy. The beam is directed at an oblique
angle onto a powdered sample using a series of mirrors (Figure 1). The
powdered sample is mixed with a background such as potassium bromide (KBr),

which does not absorb radiation in the infrared region (Skoog & Leary, 1992).

ya m

// ..;i::.i:§3}23%:3.;2:2% 1.; 5::';%.52.'3:.;.:-

To (-
Detector

 

 

 

Z/////////////////////////////////////////%///////////{////////c

Figure 1 - Path of Laser in DRIFTS Unit

Molecules absorb energy at wavenumbers that correspond to the required
energy needed to excite vibrational modes. These wavenumbers are highly
specific. Energy at wavenumbers that is not absorbed is scattered in all
directions. The scattered radiation is reflected through another series of mirrors
to a detector, which measures the wavenumbers of energy that reaches it. The
computer records this information for every scan that is made of the reflected
energy. The sample is replaced by a microcup containing only the background
and the same number of scans are run on it. The Fourier transform function of
the machine subtracts the average reflectance of the scans of the sample with
background from the average reflectance of the scans of the background to
obtain the infrared spectrum of the sample (reflectance v. wavenumber). The

reflectance spectrum is converted to absorbance through a mathematical

function and a highly specific absorbance v. wavenumber spectrum for that

sample is obtained (Skoog & Leary, 1992).

 

A few studies of methods attempting to classify and differentiate toners
have been attempted. These have analyzed primarily copier toners, but their
conclusions are relevant to laser printer toners as well. Past studies have
attempted normal infrared spectrophotometry, pyrolysis gas chromatography /
mass spectrometry (GCIMS), and DRIFTS to classify toners.

Normal infrared spectrophotometry of toners was studied by Kemp and
Totly (1981), with some discrimination between different toners. However, their
research indicated differences in the spectrum of the raw toner versus that of the
toner extracted from a printed paper. Therefore, it could not conclusively link the
two types (raw and extracted) to be from the same source, even when, in fact,

they were. The rationale given for this finding was that the process of binding the

toner to the paper caused a chemical change to take place, thereby causing a
different spectrum to be produced. This theory does not prove to be the case,
however. Later studies have shown that spectra of a toner in its raw and
extracted forms are indistinguishable.

Pyrolysis GCIMS experiments on toners were published by Levy and
Wampler in 1986. Only nine samples from four manufacturers were tested. The
results showed that toners from different manufacturers were distinguishable,
while toners made by the same manufacturer had similar spectra. Pyrolysis
GCIMS, therefore, may be capable of classifying a toner by the manufacturer. A
larger sample size is needed to verify this conclusion, however.

A combination of IR spectrophotometry and pyrolysis gas chromatography
was used by Zimmerman, Mooney, and Kimmett (1986) in an effort to
characterize copier toners. They placed 35 toners into 18 categories based on
their IR spectra. Their next step used pyrolysis gas chromatography to further
break down the categories. In their largest category based on IR absorbance,
which included eight toners, differences in the pyrolysis gas chromatography
spectra distinguished between each of them.

DRIFTS on toners was first studied in 1990 (Mazzella et al., March 1991).
149 toners were classified into 36 distinct groups, based on differences in the
infrared spectrum. In this study, 119 black copier toners were analyzed, along
with several color toners and laser printer toners. Characteristic peaks were
found in the wavenumber region between 2100 cm.1 to 700 cmd.

Raw and extracted samples of the same toner were analyzed and
compared, giving identical spectra. Tests on specific toners used in different
copiers yielded matching spectra also (Mazzella, et al., May 1991). These two
findings are significant. The first allows for comparison tests to be done between

raw toners as the known source and extracted toners as the unknown. The

second indicates that different copy machines do not cause the same type of
toner to produce different spectra. The data for laser printer toners was
suggestive of similar results.

Clearly, more research is needed with regards to the analysis of toners.
Relying on a small set of studies has inherent dangers. Without studies of
confirmation of results and conclusions, assumptions by future researchers
based on past studies may be faulty. These assumptions may then lead to
flawed or irrelevant studies conducted and incorrect conclusions reached.

The following research attempts to extend the analysis of toners by

DRIFTS to include a greater number of laser printer toners.

PROCEDURE

Laser printer toners are collected in both the raw and printed form. The
raw toners need only to be mixed with KBr to be prepared for analysis. The
toners in the printed form first need to be extracted from the paper. This is done
by heating a microscope slide on a hot plate, placing the slide on the printed ‘
paper, and removing the slide. The toner is melted by the hot slide and lifted off
the paper. The toner on the microscope slide is then scraped off using a razor
blade and mixed with KBr.

A small amount of KBr is ground using a WIg-L-Bug® for approximately 15
seconds. This grinds the KBr into a fine powder, which is desirable for DRIFTS.
A portion of the ground KBr is placed into a microcup, which is approximately
three mm in diameter and two mm deep. The toner sample is then added to the
unused portion of KBr in an estimated 10:1 ratio of KBrztoner. This mixture is
ground with the WIg-L-Bug® for an additional 15 seconds to produce a fine,
intimate mixture. This sample is placed in another microcup. Both microcups'
are taken to the DRIFTS unit for analysis.

For the first nine toner samples, a Nicolet FTIR fitted with a DRIFTS unit
was used for the testing. Wrth this machine, the main bench and the DRIFTS
unit use the same fitting apparatus, which means that the main bench needs to
be replaced by the DRIFTS unit every time testing is done. After placing the
DRIFTS unit in the machine, the laser beam needs aligning in order to be aimed
at the sample in such a way as to permit the most reflectance of infrared

radiation. This is accomplished by raising or lowering the apparatus on which the

microcups are seated.

When the bench is aligned, the sample is placed on the microcup seat and
the door to the DRIFTS unit is closed. Because water and carbon dioxide distort
spectra significantly, a wait of approximately 15 minutes is involved to allow the
gases in the chamber to purge. After the fifteen minutes elapse, 200 scans of
the toner sample are taken. The toner sample is replaced by the background
sample and, after 15 minutes of purging, 200 scans of the background are taken.
The computer Fourier transforms the spectra of the samples, and the infrared
spectrum of the toner is seen.

This spectrum needs to be converted to absorbance using a function of
the computer. The spectrum's baseline then needs correcting, as it is prone to
drifting during the run. A straight baseline is obtained in order to compare two or
more spectra. The baseline scale is then set so that the minimum absorbance is
equal to zero, since the computer’s software did not contain this function.

After these nine spectra were collected, the computer system supporting
the FTIR crashed, preventing any additional spectra from being taken. A newer
Nicolet model, with more advanced computer programming was ordered, which
allowed the spectra on the old model to be transferred without the loss of data.
Five of the samples which were analyzed on the old model were again analyzed
on the new model in order to test the consistency between the models. In each
case, the original spectrum and the new spectrum were indistinguishable. Wrth
these results, testing continued.

The newer model has the DRIFTS unit in a compartment by itself, which
means that it is permanently mounted and requires aligning only once. The new
DRIFTS unit has a detector that requires extremely low temperatures to operate
at a satisfactory level. Liquid nitrogen needs to be added to the detector at the

beginning of every day of research, and after every two or three hours of use.

10

The remaining toners were analyzed on the newer model, using the same
procedures as before, except there is no wait for the gases to purge. The new
model has a function that eliminates the water and carbon dioxide peaks without
having to purge for fifteen minutes, and this function is set up according to the
manual.

The toners' spectra are saved on the hard drive, and a library of laser
printer toners is formed of the 26 toners. A report form is made for the toners to
print out the whole spectrum (4000 - 650 wavenumbers), with the "fingerprint"
region (2000 - 650 wavenumbers) directly beneath it. The 26 spectra are printed
using this report form.

Another function of the computer displays the wavenumber of all of the
peaks above a certain threshold. These results are noted for comparative
purposes, and the toners are placed into groups according to distinguishing
peaks.

Raw toners were labeled T001 through T016, and toners extracted from
paper were labeled T101 through T109.

When the collection of toner spectra ended and the spectra were entered
into a searchable library, several toners were retested and checked against the
library. Promising, yet cautionary, results were obtained, which will be discussed
later. In addition, tests were performed to check the accuracy of assigning toners

to groups.

RESULTS

The 26 toners displayed several different spectra. Based on
distinguishing peaks, 13 groups were formed. The largest group contained four
toners, and six of the toners were placed into an individual category. Twenty-
three of the toners shared eleven prominent peaks, and were broken down by
the presence or lack of additional peaks. The remaining three toners had little in
common with the others, but shared some similarities with each other.

It should be noted that there are error margins for the peaks. Sharp peaks
may vary by up to two wavenumbers, while broad peaks may vary by as many as
ten wavenumbers. The reason for the variations is that the computer’s resolution
is two wavenumbers. This means that peaks differing by two wavenumbers or
less are not able to be distinguished by the computer. The broad peaks’ error
margin results from the absorbance of radiation throughout a broad region of the
infrared spectrum. When the computer labels a broad peak at a certain
wavenumber, it chooses the highest point of the peak as the spectrum is
displayed. A slight adjustment of the baseline may alter the point at which the
computer labels the peak.

Group I consists of a single toner that contains only the eleven basic
peaks (Figure 2). Toners in Groups II through X consist of the eleven basic
peaks plus the additional peaks that set them apart from the rest. Toners in
Group II contain a peak at 1780 wavenumbers (Figure 2). Toners in Group III
absorb at 1780 and 1376 wavenumbers (Figure 2). Group IV toners have a
small peak at 1269 and have no significant absorbance at 1376 (Figure 2).

II

Q-Q

12

Toners in Group V have a small peak at 1376, no significant peak at 1269, and a
significant peak at 1114 (Figure 2). Group VI toners have a large peak at 1376
and no peak at 1269. A small peak is present at 1114, but it is much less
pronounced than that in Group V (Figure 2). Group VII toners have significant
peaks at 1376, 1269,‘and 1114 (Figure 3). Toners in Group VIII contain Group
Vll’s peaks, but the peak at 1269 is much smaller than the peak at 1160 (Figure
3). Also, a small peak at 1780 may be present. Group IX toners have the same
peaks as Group VII, and also a significant peak at 1347 (Figure 3). Toners in
Group X contain similar peaks to those in Group VII, but two of the peaks are
shifted more than the two wavenumbers allowed for narrow peaks and ten
wavenumbers for broad peaks. There is absorption at 1724, and not 1729; there
is absorption at 1180, and not 1160. In addition, there is a peak at 970 (Figure
3).

Groups Xl through XIII contain toners that differ markedly from those in
Groups I through X. Further, Group XI and Xll toners have very few peaks in
common with each other (Figure 3). The toner that makes up Group XIII shares
several peaks with the Group XI toner, and several others with the Group XII
toner (Figure 3).

The groups, with the toner labels and distinguishing peaks are printed in
Appendix B.

A limited blind test was conducted to test the library’s ability to match the
new spectrum with the spectra in the library and to test the grouping of the toners
discussed previously. This blind test was not a true blind test. The toner was
known to be one that was collected and placed in the library. Therefore, the
chance that the toner would not match any within the library was not present, as
it should be in a true blind test.

When a spectrum is compared to the library of toners, the computer lists

I3

 

 

""ABsoi’cance Absorbance Absorbance Absorbance Absomance Absorbance Afisaibance

FE-..

 

 

 

 

 

 

 

 

 

 

- roup l ,,.
. Group I
005* Fri Oct 24 13:48:35 1997
‘ ‘- i l J:
‘ ,I ', I I' I‘ l. 4 F J, wI/ Jr 4'
0.00- we,» \I I'MM (I \yvV/I W\\'\, \A. ~z, \ \ s.
0.2: roup II
0.1 -
r/‘WWLWA ft - x
' ’\.J I ' \N/flb-A‘ ”V
0.10_ roup lll .
0.0541 A . A I
A I U /" I
. l A ,‘ . ;
0°00 ‘ w/‘J/U MI W LVN/W VNJ \v k
0.2: roup IV :1 .
0.1g
0 05..Group V
. I
0.104 IOUP VI I .
.. ’2 t
0.05- , Ir \ ?
- [0‘ j I I“ I
0.10: WW" I
I i g
0.051 . , i. ll E
0.0mm; I: [MW] U \
2000 1500 1000

Wavenumbers (cm-1)

Figure 2 - Groups I through VII

14

 

 

 

 

 

 

 

 

 

 

2000

I 0.10 roup VIII f
I§ - Group VIII
F3 005: Tue Jan 28 14. 50: 52 1397
$8 I I -
I O-OO___A—M"I INA/VI I‘m; I/I L/JIW/ IV’\ \' I"\~'\’\ ‘~I_."\.- \ 'JJ
I8 0.101 '°“'°'X
I -
I: q
ICU
IE, 0. 053 , , .
2 III
000va MIIIIII WWW/\m III \,
0.10— roupX Ir ,
g I
_ W I I .
g 0.054 II III/I III II .
m .
g I III‘JxIIV‘IIv IL/I
: 0.00: I' MWvIW/IIINWVL IL
18 W
. III
IE 0.05- I III
“ I I
g 4 I III [\II I
- I ;‘ I' I I I
1 0.004 WWW/\Wm ’ V II“
_ « roup XII II 1
:8 0.14 II ‘
I: .
a I
. - III/WI
” MII ‘
I“ 0.0._._,\.,/I \IIM IWw/I/vafl/WI I
: roup xm I ,
I8 0 10: 'II III II
g I; I III I I I
Lg 0'05“ II I MI I I
to I 1 I I
m

1 500 1 000

Wavenumbers (cm-1)

 

Figure 3 - Groups VIII through XIII

15

the top ten matches based on a 100-point scale. A match rating of 100 indicates
a perfect match of the spectra. This scale is a quantitative matching based on
the location and intensity of peaks. Out of these ten matches, the top four are
displayed below the new spectrum.

When the blind test of a toner was performed in order to test the library,
the library displayed the names and ratings of the top ten matches. Because
T008 was accidentally added to the library in triplicate, three of these matches
were for T008. Exclusion of two of these from further evaluation led to a total of
eight toners which were compared to the unknown spectrum. Instead of
comparing only the library’s top four matches, the top eight were compared
(Figure 4).

Because the spectra in the library have much poorer resolution than those
saved on the hard drive, the according files were opened for a more telling
comparison (Figure 5). By comparing the unknown toner with each of the
library’s top matches, some of the toners could be eliminated as the probable
source. The toners are eliminated when the spectrum of the unknown is
distinguishable from the spectrum of the known toner. The presence of a peak in
one spectrum that is not present in the other spectrum is grounds for ruling that
the two are from different toners.

A toner is not eliminated when its spectrum is indistinguishable from that
of the unknown toner. Two spectra are indistinguishable when all of their peaks
are found within the error margin of two wavenumbers. Further, the relative
intensities of the peaks should be similar. In this research, relative intensities
were considered in major peaks, that is, those that are at least ten percent of the
height of the largest peak. Differences in relative intensities were considered

distinguishable when one peak was at least twice as high as the other peak.

l6

 

Abs Abs Abs Alfs Abs Abs Abs Abs

Abs

 

08 Canon OEM SX

0.53 . .
3 WW. AMA/A“ '

 

l #L

004 (Canon?) CX
0.5-1

 

: 003 Graphic Black SX
0.5—

 

: lC on ro omponen p Ilp Graphics
0.5-

M Ma

 

. fiknown
005; Unknown ’
- Thu Nov20 1548391

0.00—

 

jTOlS Apple Easei’Wfiter 300, 310, 360

0.5;
+

 

~ 001 Lexmark 4U49
0.54

 

. 009 Mirage SX
0.5—

A- AALJAA- A ---

A
. H rap rcs ec noogles'e eone" se e ‘
0.5—

“W‘ AAA

2000 1500 1000
Wavenumbers (cm-1)

 

 

Figure 4 - Spectra of Top Matches from Library

 

 

”V’Abs ‘

Abs ‘Abs Abs ~Abs Abs Ab?" TAA 3‘

Abs

 

0.1 DOB Canon OEMSX

r,

04 (Canon?) CX

I
ll l
‘l

ongwjvw Mvd Kim/“W “\l K
Eligjwltwwmlll

 

0.1_ l 003 Graphic Black SX

 

 

_ nknown

 

ppe ase

exma

Irage SX

 

 

00: T009 Mirag SX
0.2: 007 Graphics Technologies Velvetone HP LaserJet 4

 

2000 1500 1000
Wavenumbers (cm-1)

Figure 5 — Spectra of Top Matches from File

l
l

4 Kmf 4-4 .4.

J

18

The overlays of the top matches with the unknown toner are found in
Figures 6 through 9. T008 and T001 were eliminated as the source of the
unknown toner due to their peaks at 1780 wavenumbers, which was nonexistent
in the unknown toner's spectrum. T004 and T003 were eliminated due to their
peaks at 1269, which was not present in the unknown. T007 was eliminated as
the possible source by the lack of significant absorbance at 1376 and the
presence of a peak at 1269. The three remaining toners, T002, T009, and T015,
displayed peaks similar to the unknown at all points in their spectra. These
three, therefore, could not be eliminated as the source of the unknown toner.

Next, the unknown toner was placed into a group according to its peaks.
In addition to the eleven basic peaks, it possessed a peak at 1376 wavenumbers.
This eliminated Groups I, II, IV, and XI - XIII. The lack of a peak at 1269
wavenumbers eliminated the remaining groups except Group VI and possibly V.
By comparing the unknown’s absorbance at 1114 with that of toners in each of
these groups, Group V was eliminated. The logical choice would be to assign
the unknown toner to Group VI.

In fact, the unknown toner was Ieamed to be T009, which was one of the
three finalists after searching the library, and which is a member of group VI.
The other finalists in the library search, T002 and T015, are also in Group VI,
which further validates the grouping of laser printer toners discussed in this

thesis.

l9

 

"—Afisfil‘fia—ncew "

- T008 Cano :OEM ,
Sat Nov 02 {:5321ll6 19 a";
1.

 

 

53
I
:2
l l:

 

I .

a
I
it

llSX a llUnknown

Wavenumbers (cm-1L

 

 

 

aria” ' U

- E sort

0.020

0.010

11l11|1|11111111|

0.0005

  

048:2

 

 

)CXrland ll

1 1996i

 

l

 

 

l

 

 

 

 

I'll
Ill W

 

 

m

 

 

 

2000

1500
Wavenumbers (cm-1)

1obo

Figure 6 - Overlays of Unknown with T008 and T004

__-____._§._E-V

20

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

W c ‘.
f T003 Graphic Black sx and Unknown 5
0.10__ Fri Oct 04162226 1996 g
0.06:
8 :
5 0 06 ‘
e - 1
o _
m a
'0 '1
< 0.04: |
002—: ‘ U
I» II
2000 1500 1000
Wavenumbers (cm-1)
T002 Static Control Components LX
0.109 llpllllp Graphics and Unknown
; Fri Oct 0417;11:191996
0.06.}
8 :
5 0 06 ‘
e : I {I
O a
3 -
< 0.04.; I
0.02.: I
coo—“W VI w
2000 1500 1000
Wavenumbers (cm-1)

 

 

Figure 7 — Overlays of Unknown with T003 and T002

21

 

Absorbance

 

  
      

 

0.043 T015 Applel ,310.360 - .;
: Unknown

0023': 061240 I

 

 

 

 

 

   

 

 

I ‘ ll

2000 1530 1000
Wavenumbers (cm-1)

 

 

 

 

 

Absorbance

 

‘ T001 Lexmark 4049 and Unknown
‘ Thu Oct 03 08:50:16 1996

 

 

 

 

 

 

 

 

 

 

 

 

 

2000 1500 1000
Wavenumbers (cm-1)

 

 

Figure 8 — Overlays of Unknown with T015 and T001

22

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

» 3,...mgm

'l

l

00

  

 

. 012%
l :
i 010;
8 1108;
g 1
"E (106;
(0
2 :
0.043
ET009 Mirag Xa Uu- .....
002% Sun Nov 03 0.51. 19 I w
000%
2000 1500 1000
Wavenumbers (cm-1)
l
l I
‘ l
l I
‘ l
I8 I
.C
Q 1
1'5 0 ‘
Li I ‘ I
l
l
l
l

 

0W WM U

 

 

 

1000
Wavenumbers (cm-1)

Figure 9 — Overlays of Unknown with T009 and T007

DISCUSSION

The goal of this project was to evaluate DRIFTS as a method to
differentiate laser printer toners. This research has shown that DRIFTS is a
valuable technique in distinguishing between laser printer toners. Laser printer
toners contain differences in chemical makeup, which can be used to
differentiate between them using DRIFTS. The toners did not all display unique
spectra, but were broken down into classes containing between one and four
toners.

It is interesting to note that the toners which were known to be of the same
brand did not necessarily fall into the same groups. For example, the seven
Hewlett Packard laser printer toners were placed into four groups. This finding is
evidence that the companies that produce laser printer toner use different mixes
in their types of toners. Knowing this, it is possible for the researcher to
differentiate not by company, but by individual toner. In other words, instead of
identifying the company who made the toner, the results indicate the type of
toner. Since the type of toner is specific to a single or small set of printer models,
the spectrum links the sample to this set of laser printer models.

In forensic science, almost all tests are comparison tests. In a questioned
documents case in which the document was prepared on a laser printer, the
examination would be to compare the toner of the questioned document with that
of a document of a known source to determine if they could have come from the
same source. This research provides evidence that laser printer toners can be

distinguished using DRIFTS, and would prove useful for this examination. If the

23

24

spectra of the two documents’ toners were distinguishable from each other, the
known source could be eliminated as the possible source of the questioned
document. If, on the other hand, the spectra were indistinguishable, the source
of the known document could not be ruled out as the source of the unknown.
With class evidence, which is what this evidence is, the value comes in excluding
possible sources of the questioned document. This research has shown that
DRIFTS can exclude toners on the basis of their infrared spectra.

Another notable finding is that toners manufactured by different
companies for use in the same printer model produced distinguishable infrared
absorbance spectra. For example, in addition to the toner made by Hewlett
Packard for their LaserJet 4 printers, Graphics Technologies makes a toner
designed for this printer. The spectra of these two toners are distinguishable.
This means that the same manufacturer is not used among toner companies for
toners made for the same model. The significance of this finding is that if a
certain printer can use two different toners, and an unknown toner sample’s
spectrum is indistinguishable from one of these, then the printers using the other
toner can be eliminated as the source of the unknown. In the example from
above, if a spectrum indistinguishable from that of Graphics Technologies’ toner
for the LaserJet 4 was obtained from a questioned document, laser printers that
contained toner from Hewlett Packard could be eliminated as the possible source
of the questioned document. However, this only holds for printers that have used
only one of the two types of toners during the timespan in question.

Of the eight toners that were retested and compared to the toner library,
the library search results indicated that the correct toner was the first or second
best match five of these times. On two occasions, however, the correct toner
was not in the top five choices, as given by the computer's rating of the degree of

match between the tested toner and the toners in the library. The reason for this

25

is that the resolution of the spectra that are saved in the toner library is much
lower than that of the toners when they are tested. When a spectrum is saved to
a library, the resolution is decreased in order to accommodate more spectra in
the same amount of memory. The result of this is that the spectra in the library
may be missing small peaks or may contain peaks shifted by up to eight
wavenumbers.

When the computer searches the library to compare a recently acquired
spectrum, it is not infallible. Discretion must be exercised by the operator in
determining whether there is a match or not. When the computer provides the
spectra that are most like the new spectrum, the operator should open the files
that correspond to these spectra to compare the original spectra. This practice
reduces the shifting of peaks to a maximum of two wavenumbers. When a
narrow peak from one spectrum is three wavenumbers different than a peak on a
second spectrum, there is a difference between the spectra.

Although a database of 26 laser printer toners is by no means exhaustive,
this provides a solid base for future research and for practical use. Researchers
in the future may add spectra from additional toners as the toners become
available. The new toners may be placed into the groups defined in this research
or, if other distinguishing peaks are present, new groups may be formed.

In theory, spectra from DRIFT S collected from one instrument should be
interchangeable with DRIFT S spectra from other instruments. In other words, a
spectrum of a toner from one DRIFT S unit should be similar to a spectrum of the
same toner produced on another DRIFTS unit. The presence of searchable
libraries of spectra obtained from infrared analysis indicates that FTIR
instruments produce interchangeable spectra. Also, the spectra obtained in this
research came from two different units, which produced indistinguishable spectra

when testing the same toner.

26

The reason interchangeable spectra are produced is because DRIFTS
does not depend on the settings of the machines such as temperature, flow rate,
or length or type of column. Instead, DRIFTS measures a physical characteristic
of matter, infrared absorption, that is present at any given time. Any differences
between spectra from DRIFTS units would possibly be the result of different
resolutions of the detector, which would affect the sharpness of the peaks, but
not the wavenumbers at which the infrared radiation is absorbed.

The practical use of this research comes in questioned document
analysis. When a document that has been printed with a laser printer is
analyzed, the examiner may perform DRIFTS of the toner and compare its
spectrum to that of known sources. If the toners display distinguishable spectra,
then the sources of the two documents are different.

In cases in which there is no known source with which to compare the
questioned document, this research provides a base from which to begin. The
laboratory conducting the examination could obtain a disk with the 26 toners’
spectra, and the spectrum from the toner of the questioned document could be
compared to those on file. If the spectrum was indistinguishable from any of the
toners, then the laser printers that use these toners are possible sources. All of
the printers that use toners whose spectra are distinguished from the questioned
document’s toner’s spectrum would be eliminated as the source. Although this
research is most useful in providing evidence of the characterizability of laser
printer toners using DRIFTS in order to distinguish a questioned document from
possible known sources, it can provide a starting point in a case that contains no
suspected sources.

This research would be more useful in this regard if the entire population
of laser printer toners were analyzed. Unfortunately, laser printer toners are not

easily obtained from the companies that manufacture or distribute them. This

27

research contains 26 of the estimated 50 toners on the market in the United
States.

An examiner would yield the optimum results if DRIFTS analysis of toners
was used in conjunction with another differentiating method, such as pyrolysis
GCIMS. The toners could be analyzed first by DRIFTS and placed in a group.
The next step would involve subjecting the toners with spectra that match the
questioned document to pyrolysis GCIMS, which possibly could differentiate
between the toners within the group.

Future research in this area should focus on extending the database of
DRIFTS spectra and attempting other tests to differentiate between toners within
the groups defined in this study. Pyrolysis GCIMS has provided promising
results to meet this need. High-perfomiance liquid chromatography (HPLC) could
possibly be used to differentiate toners based on the synthetic organic resins
present in the toners.

This research showed that it is possible to differentiate laser printer toners
based on infrared absorption using DRIFTS. The results did not show that the
toners were individualizable, but many different spectra were seen, and the
toners were grouped accordingly. The field of questioned document analysis
benefits from this research. Documents printed by a laser printer are able to be
differentiated by the type of toner, and therefore by the brand and model of
printer. This is an additional piece of evidence that may be used to determine
the outcome of the case, which is the ultimate goal of all forensic science

examinations.

APPENDICES

APPENDIX A

T001
T002
T003
T004
T005
T006
T007
T008
T009
T010
T011
T012
T013
T014
T015
T016
T017
T101
T102
T103
T104
T105
T106
T107
T108
T109

Lexmark 4049

Static Control Components LX — llp/lllp Graphics
Graphic Black SX

(Canon?) CX

Laser1 Aspen Toner

Graphics Technologies Legend HP LX Graphics
Graphics Technologies Velvetone HP LaserJet 4
Canon OEM SX

Mirage SX

No Brand - SX only

Oasis Eclipse EX

Older LX - Unknown Brand

Static Control Components SX - Ultra Graphics
Ricoh FT Type 410

Apple LaserWriter 300,310,360

IBM 4019

Oasis Mirage 5PNX

Hewlett Packard LaserJet 03903A

Hewlett Packard LaserJet 92298A

Hewlett Packard LaserJet 92291A

Hewlett Packard LaserJet 92274A

Hewlett Packard LaserJet 92295A

Hewlett Packard LaserJet 03900A

Hewlett Packard LaserJet C3909A

Brother TN-200HL

Lexmark 1382150

28

APPENDIX B

.—.—

Group I: T104
peaks: 1729 1601 1493 1452 1160 1070 1028 906 842 758 699

Group II: T001 T101 T105 T107
peaks: peaks from Group I; 1780

Group III: T008 T017
peaks: peaks from Group I; 1780 1376

Group IV: T007 T016 T109
peaks: peaks from Group I; small 1269; not significant 1376

Group V: T102 T103
peaks: peaks from Group I; small 1376; 1114; not significant 1269

Group Vl: T002 T009 T015
peaks: peaks from Group I; large 1376; no 1269

Group VII: T003 T004 T013
peaks: peaks from Group I; 1376 1269 1114

Group VIII: T011 T012 T106

peaks: peaks from Group Vll, but 1269 much smaller than 1160; small
1780 may be present

Group IX: T006
peaks: peaks from Group VII; 1347

Group X: T010
peaks: peaks from Group VII; 1724, not 1729; 1180, not 1160; 970

Group XI: T014
peaks: 1729 ~1500 ~1450127212421158111711031017 829 761
730 700

Group XII: T005

peaks: 1722 1607 1510 1461 1299 1231 1183 1075 1042 997 939 829
729 676

29

30

Group XIII: T108
peaks: 172216071510 ~145614081271 1245118311181041 1017
996 931 830 731 700

APPENDIX C

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I -
02°“ T001 Lexmark 4049
: Thu Oct 03 08:50:16 1996
l -
0.15—
8
C
3 ..
o 0.10—
3
< -
0.05-
o'oo—MWAM a
4000 3000 2000 1000
Wavenumbers (cm-1)
020‘ 1001 Lexmark 4049
. Thu Oct 03 08:50:16 1996
0.15—
8
C
5 .7
o 0.10—
W _
2 _
0.05—
0.004 ,
2000 1500 1000

 

Wavenumbers (cm-1)

 

T001 Lexmark 4049

31

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T002 Static Control Components LX - I .
0.10-— llp/lllp Graphics .
2 Fri Oct 04 17:11:19 1996 I
I 0.08 : II I
I8 : II
I9 0.06: I I .
I3 1 I I
g I I: I I
0'04 j I I II II I
I 7 l' I I II
I I II I I II ,
I 0022 IIWI l WM
III I I I
I III/I III I III IIIM/II I IIIIIIIIIIII II III IIIIII
I 0 00“WMWIIIII WWI II III/II IIIIIIIIII"
I 4000 3000 2000 1000
I Wavenumbers (cm-1) I
T002 Static Control Components LX -
0.10— llp/lllp Graphics
3 Fri Oct 0417:11z191996
008—:
8 :
5 0 06 ‘
E ' i
8 I
.0 -I
“ 0.041
0.02:
0.00—1
2000 1500 1000
Wavenumbers (cm-1)

 

 

 

T002 Static Control Components LX - llp/Illp Graphics

T003 Graphic Black SX

 

 

 

 

 

 

 

 

 

 

 

0105 Fri Oct 04 16:22:26 1996
I 0.08:
IQ) :
I0 _I
Is -
e 0.06:
a
< 0.041
I 0.02%
I I
0.00.:wa
40 00 3000 2000 1000
Wavenumbers (cm-1)
3 T003 Graphic Black sx
0101 Fri Oct 0416: 22- 26 1996
0.08:
8 :
E
g 0.06:
O a
0’ I
‘3 I
< 0.04:
0.02 _
0.00:
20001500 1000
Wavenumbers (cm-1)

 

T003 Graphic Black SX

34

 

-fi —7
.3
c:
on
C
11

E T004 (Canon?) cx
0-070; Fri Oct 04 16:48:25 1996 ;

Absorbanoe
o
2
O

 

 

 

 

 

 

 

 

 

 

 

 

4000 3000 2000 1000
Wavenumbers (cm-1)

 

 

 

; T004 (Canon?) CX
0.070: Fri Oct 04 16:48:25 1996

0.040%

Absorbanoe

 

 

 

0.020; I

 

0.010

11111
<

 

0.0003

2000 1500 1000
Wavenumbers (cm-1)

 

 

 

T004 (Canon?) CX

3S

 

Absorbance

40

0.12.: T005 Laser1 Aspen Toner
2 Mon Oct 07 15:44:45 1996

010—: II I

0.08; II I
0.06% I I
0.04:
0.02—E I

0.00%»4411 III II I

 

 

00 3000 2000 1000 I
Wavenumbers (cm-1)

 

Absorbance

0.02

0.00
20

 

0.12: T005 Laser1 Aspen Toner
2 Mon Oct 07 15:44:45 1996
010—:
0.08 —

0.06;

IIIIIIII

I?

111111111

 

 

 

 

 

00 1500 1000
Wavenumbers (cm-1)

T005 Laser1 Aspen Toner

36

 

L

L

l

T006 Graphics Technologies Legend HP
LX Graphics
Mon Oct 07 16:07:53 1996

0.10

0.08

L1111111111

I l

Afisorbanoe
o
O
?’

 

 

(102;

 

 

 

(100} I
4000 3000 2000 1000
Wavenumbers (cm-1)

 

 

 

.1

l l

: T006 Graphics Technologies Legend HP
010— LX Graphics
~ Mon Oct 07 16:07:53 1996

-r

.J

(1082

q

1106{

q

Absorbance

]
004-1

.4

 

 

d

(1021

 

 

a:

(1003

2000 1500 1000
Wavenumbers (cm-1)

 

 

 

 

 

T006 Graphics Technologies Legend HP LX Graphics

 

Absofiiance

 

0205
018;
016i
014%
012i
010%
008%
006i
0045
002—
000

T007 Graphics Technologies Velvetone
HP LaserJet 4
Sat Nov 02 15:56:16 1996

 

 

40

2000
Wavenumbers (cm-1)

00 3000

1000

 

 

 

ancé 4 4 4 4

I4 A556?!)

0203
018%
016;
014%
012%
010i
006%
006i
004i
0023

I T007 Graphics Technologies Velvetone

HP LaserJet 4

; Sat Nov 02 15:56:16 1996

I I11
II III I I

 

0.00_ a

3W I WIN/WI III I I

I
IVVA ALI IIv/VIJUJ

 

2000

f\/
1 500

Wavenumbers (cm-1)

T007 Graphics Technologies Velvetone HP LaserJet 4

1000

38

 

0.10

3 T008 Canon OEM sx
0 08 Sat Nov 02 17:53:16 1996

111111111

0.06

(1041

Abso?oanoe

 

 

 

 

 

 

 

0.02% d

 

(1001

 

 

4000 3060 2060 1060
Wavenumbers (cm-1)

 

 

 

 

 

p
3%

T008 Canon OEM SX
Sat Nov 02 17:53:16 1996

0.08

0.06

1141111111111

)9

 

Absorbance

 

 

 

 

 

2000 15b0 1obo
Wavenumbers (cm-1)

 

 

 

T008 Canon OEM SX

39

 

.9

A

N
11111

Absorbance
.0
o
o:
1 1 1

i T009 Mirage SX
Sun Nov 03 00:51:16 1996

WW I I

4000 sobo 2060 1060
Wavenumbers (cm-1)

 

 

 

 

 

 

 

 

 

 

 

 

 

Absorbance
.9
O
9’

 
    

 

 

3T009 Mirag SX
2 Sun Nov 03 0:51:

  

 

 

 

2000 15b0 1obo
Wavenumbers (cm-1)

 

 

 

T009 Mirage SX

4o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

010—:
008—:
8 E
5 0.06:
I; :
3 I
< 0.04: I
3 T010 No Brand - SX only
002; Tue Jan 28 15:20:08 1997
000‘:me ' II III
4000 3000 2000 1000
Wavenumbers (cm-1)
(1105
0.08:
8 _
5 (106—
'9 :
O
_8 I
< 0.04—
:T010 No Br d - SX onl
0.02; Tue Jan 28 :20: 8 19 \L
(1003
2000 1500 1050
Wavenumbers (cm-1)

 

 

 

T010 No Brand - SX only

41

 

(1101 I

(1083

-<

1106:

—1

d

(1041

~—

Absorbance

j T011 Oasis Eclipse EX
002— Tue Jan 28 14:50:52 1997

(100ésmuumrc~«~,~,«}III1A-vaaaac_~e,/VMJ I 1

4000 3000 2000 1000
Wavenumbers (cm-1)

 

 

 

 

 

 

 

 

 

 

O.10~

AbsoT‘banoe

 

 

 

 

 

 

 

2000 1500 1000
Wavenumbers (cm-1)

 

 

T011 Oasis Eclipse EX

42

 

Afisorbanoe

.0
—I

0.12

11111111111

I
_|.
C

1

(108%

1

‘III‘

1

1L111111

I

F3 s: F’
E3 53 83

1 1

l

(1003

 

J

T012 Older LX - Unknown Brand
Thu Mar 27 13:02:00 1997

 

 

 

4000 3000 2000

Wavenumbers (cm-1)

1000

 

 

Absalbance

 

1114.
(1125

(110;

008—:

‘1
cl

.1

0.06-

0.04

111111111

0.02 —~

u

j
1100;

 

I 1
T012 Older 1 - Unkno I Brand
Thu Mar 27 3:02:00 19

 

 

 

2000 1500

Wavenumbers (cm—1)

1000

 

T012 Older LX - Unknown Brand

 

 

 

43

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I
I 0.12% I
I 3 :
'~ 0.10{
0.083
§ ;
“5 :
g 0.06j I
§ :
0.04%
; T013 Static Control Components SX Laser
0 02; Toner - Ultra Graphics
' : Thu Mar 27 13:36:2 1997 R
0.00—: I I
4000 3000 2000 1000
Wavenumbers (cm-1)
0.125
0.105
0.08j
§ :
m _
E 0.06:
3 3
< 2
0.04: I
3 T013 Static Iontrol Co . .. ents SX Las
1 Toner- Ult : Graphics |
0-02: Thu Mar 27 :36: 7 19
0.004:
2000 1500 1000

 

Wavenumbers (cm-1)

 

T013 Static Control Components SX Laser Toner - Ultra Graphics

 

 

Absorbance
o
12
f’

 

 

0.0203: T014 Ricoh FT Type 410
: Wed Oct 08 12:50:02 1997

0.0105 . .
: II I .
000an l ' II

4000 3000 2000 1000
Wavenumbers (cm-1)

 

 

 

 

 

 

 

Absofliance
o
12
1’

0.020; T014 Ricoh ' ype4 1'
3 Wed Oct 08 1: 0:02 '1 .3

 

 

 

 

 

 

 

2000 1500 1000
Wavenumbers (cm-1)

 

 

 

T014 Ricoh FT Type 410

45

 

0.161: 1015 Apple LaserWriter 300,310,360

: Fri Oct 24 08:47:41 1997
0.145

0.12;
0.10.?
008—?
0.06%
0.04%
0.02.5
0.00—3

Absorbance

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

4000 3000 2000
Wavenumbers (cm-1)

1000

 

 

0.165 T015 Apple LaserWriter 300,310,360
3 Fri Oct 24 08:47:41 1997
11145

0.125

I

0.10.;
0.06%

-1

AbsorBanoe

0.065

d

0.04%

 

-

0.025

 

 

0.00—i

 

 

 

2000 1500
Wavenumbers (cm-1)

 

1000

 

 

T015 Apple LaserWriter 300,310,360

46

 

AbsorBanoe

 

(1145
(112%
(110%

(106i

 

- T016 IBM 4019
Fri Oct 24 09:20:

51 1997

 

 

 

 

 

 

3000 2000
Wavenumbers (cm-1)

1000

 

 

 

Absorbance

 

0.145

(112:

(110;

(106i

—(

(106%

(1045

q

(102%

.1

(100%

 

3 T016 IBM 4019
3 Fri Oct 24 09:20:51 1997

 

 

 

 

2000

1500
Wavenumbers (cm-1)

1000

 

 

T016 IBM 4019

47

 

I 0.12 :
I T017 Oasis Mirage 5PNX
I 010—: Fri Oct 24 09:58:18 1997

I 008—:
I3 0.065
E 0 04 E
E E
(< 0.02g

E II
0.00 :MMWI’ 1W

-0.02—f III

 

 

 

 

 

-0.04 3
40 00 3000 2000
Wavenumbers (cm-1)
0.12 i
; T017 Oasis Mirage 5PNX
— Fri Oct 24 09:58:18 1997
0.10 —
0.08

11111111

AbsofBance
o
o
a)
1 1 1 1 1

.3

A

0.025

 

0.00;

 

 

20(00 1500 1000
Wavenumbers (cm-1)

 

 

 

T017 Oasis Mirage 5PNX

48

 

Absorbance

 

' 0000—:

0060—:

q

0050—:

0.040—~

—(

0.030:

0.020.:

0.0105

.1

 

T101 Hewlett Packard LaserJet 03903A
Wed Oct 15 10:27:53 1997

 

 

 

 

 

4000 3000 2000

Wavenumbers (cm-1)

1000

 

 

Absorfianoe

 

0.060:
0.0503

0.0401

‘

0.0305

d

0.0203

d

0.010-1

.I

0.0003

 

T101 Hewlett Packard LaserJet C3903A

_ Wed Oct 15 10:27:53 1997

 

 

I

 

 

 

2000 1500
Wavenumbers (cm-1)

1000

 

 

T101 Hewlett Packard LaserJet C3903A

49

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 050; T102 Hewlett Packard LaserJet 92298A I

' - Thu Oct 16 13:44:01 1997 I
00401
§ 0.030:
a 1
g :
9 0.0201
1
0.010—

4000 3000 2000 1000
Wavenumbers (cm-1)
0 050; T102 Hewlett Packard LaserJet 92298A
' _ Thu Oct 16 13:44:01 1997

0.0401
2:3 0.0303
W -
'9 _
‘9’ 0 0201
2 ' _
0.0105
1

0.0001 III

2000 1500 1000

Wavenumbers (cm-1)

 

 

T102 Hewlett Packard LaserJet 92298A

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 0503 T103 Hewlett Packard LaserJet 92291A
- . Fri Oct 24 12:55:29 1997
0040—;
§ 0.030{
(B
e :
g 0.020: I
< . I
0010—: N I
°~°°°9WIWWW I
00105
4000 3000 2000 1000
Wavenumbers (cm-1)
0: T103 Hewlett Packard LaserJet 92291A
0-05 j Fri Oct 24 12:55:29 1997
0.0403
2:3 0.030:
(g a
19 q
3 00203
2 ' :
0.0101
0.0001
2000 1500 1000

 

 

Wavenumbers (cm-1)

T103 Hewlett Packard LaserJet 92291A

51

 

0080—? T104 Hewlett Packard LaserJet 92274A
Fri Oct 24 13:46:35 1997

Absorbance
.0
§
1 1 1 LL

 

 

 

 

 

 

V 0.000%WI’WI III/«MI

4000 3000 2000 1000
Wavenumbers (cm-1)

 

 

 

 

0080-? T104 Hewlett Packard LaserJet 92274A
3 Fri Oct 24 13:48:35 1997

Absorbance
O
E
O
1 1

 

 

 

 

 

 

 

0.030:
0.0205 I .
: I
00105 I i
1
0.000; 11
2000 1500 1000

 

 

Wavenumbers (cm-1)

 

T104 Hewlett Packard LaserJe192274A

52

 

Absorbance

 

 

0.14—

—(

(1125

.4

l

—(
—1

0.10
0.08

0.06

52

111111111111111111111111

0.02

0.003

:Hw11NI15axT““JIIIIT““”“TOVr-~490~I I

T105 Hewlett Packard LaserJ6192295A f
Fri Oct 24 15:03:21 1997 ‘

 

 

 

 

 

 

4000 3000 2000 1000

 

 

Wavenumbers (cm-1)

 

 

Absorbance

 

0.14—

0.12

0.10 4

0.08

0.02

p p
11111111111111111111111111111111111

.0
o
o

l l

T105 Hewlett Packard LaserJet 92295A
Fri Oct 24 15:03:21 1997

 

 

 

 

 

2000 1500 1000

 

Wavenumbers (cm-1)

 

T105 Hewlett Packard LaserJet 92295A

53

 

Absorbance

. 0.000

.1

0.070%
0.060%
0.050%
0.040%
0.030%
0.020%

0.010%

111114

.1

‘ T106 Hewlett Packard LaserJet C3900A

Fri Oct 31 15:11:31 1997

 

 

 

 

 

 

 

 

 

4000 3000 2000

Wavenumbers (cm-1)

1000

 

 

Absorbance

 

0.070%
0.060%
0.050%
0.040%
0.030%

0.020%

T106 Hewle Packard L; rJet C3900A
Fri Oct 311 : 1:31199 *

 

 

 

 

 

1500
Wavenumbers (cm-1)

T106 Hewlett Packard LaserJet C3900A

1000

 

54

 

Absorfianoe

 

1
T107 Hewlett Packard LaserJet C3909A

Tue Oct 28 14:41 :47 1997

___..—_—._._._,

 

 

 

 

 

4000 3000 2000

Wavenumbers (cm-1)

 

 

Absorbance

 

1111111

.4

0.020.}

-1

0.0104:

0.000;

 

T107 Hewl i= Packard I: - rJet C3909A
Tue Oct .: :41:4719

 

 

 

2000 1500

Wavenumbers (cm-1)

1000

 

 

T107 Hewlett Packard LaserJet C3909A

55

 

.0
o
00
J

morbance

.0
o
N

T108 Brother TN-2

1111111111111111111111

.0
o
o

 

WW

¢>
O
N

1

Fri Oct 31 09:31:151 7

4000 3000 2000 1000

L I I I

 

Wavenumbers (cm-1)

 

 

_o

_|

N
1

.0
_l
O
1 1

.o

o

a)
L111L1111111

006:

.0

o

A
|

   

Afiscfiaance

T108 Brot er N-

002 FnOd31 9

0.00

11111111111

1

0
9

 

 

2000

 

1500 1000
Wavenumbers (cm-1)

 

 

T108 Brother TN-200HL

56

 

__.___4___ fl

0.251
1

 

 

 

 

 

 

.- -. _..‘ __._......

 

 

 

 

 

 

 

 

 

 

 

0201
g I
1. 0.15—
.e _
O
(I) _ I I
2 0.109 I
:T109 Lexmark 1382150
005— Fri Oct 31 09:55:25 1997
4000 3000 2000 1000
Wavenumbers (cm-1)
0.25.?
1
1
0.204
I
8 -
5 0.15—
Fe -
O -
m ‘1
2 0.10—
- T109 Lexm 1362150
: Fri Oct 31 0 55:25199
0.05-
0.00%
2000 1500 1000
Wavenumbers Lcm—1)

 

T109 Lexmark 1382150

 

 

 

BIBLIOGRAPHY

BIBLIOGRAPHY

Brunelle, R.L. MW” Chapter 14, edited by
Saferstein, R. Prentice-Hall: Englewood Cliffs, NJ, 1982.

Kemp, GS. and Totty, R.N. “The Differentiation of Toners Used in Photocopy
Processes by Infrared Spectroscopy.” Forensic Science lntemational, Vol. 22.
1983, pp. 75-83.

Levy, E.J. and Wampler, T.P. “Application Of Pyrolysis Gas Chromatography/
Mass Spectrometry to Toner Materials from Photocopiers.” Joumal of Forensic
Sciences, JFSCA, Vol. 31, NO. 1. Jan 1986, pp. 258-271.

Mazzella, W.D., Lennard, C.J., and Margot, P.A. “Classification and Identification
Of PhotOOOpying Toners by Diffuse Reflectance Infrared Fourier Transform
Spectroscopy (DRIFTS): I. Preliminary Results.” Joumal Of Forensic Sciences,
JFSCA, Vol. 36, No. 2. March 1991, pp. 449-465.

Mazzella, W.D., Lennard, C.J., and Margot, P.A. “Classification and Identification
of Photocopying Toners by Diffuse Reflectance Infrared Fourier Transform
Spectroscopy (DRIFTS): II. Final Report.” Journal Of Forensic Sciences,
JFSCA, Vol. 36, No. 3. May 1991, pp. 820-837.

Minasi, M. “Hardware Clinic: You Look Marvelous.” Compute. Jan 1992, pp.
68+.

Skoog. DA. and Leary. J.J. WW5 4th ed
Saunders College Publishing: Fort Worth, TX, at al. 1992.

Totty, R.N. “Analysis and Differentiation of Photocopy Toners." Forensic
Science Review, Vol. 2, No. 1. June 1990.

Saferstein,R.Q[1m1[laII$1I12$._An_InILmuQIIQn.10_EQLen§lQ_SQle.nQe.4th ed.
Prentice-Hall: Englewood Cliffs, NJ, 1990.

Zimmerman J., Mooney D., and Kimmett, M.J. “Preliminary Examination of
Machine Copier Toners by Infrared Spectrophotometry and Pyrolysis Gas
Chromatography.” Journal of Forensic Sciences, JFSCA, Vol. 31, NO. 2. April
1986, pp. 489-493.

57

"11111111111111