Page 1
LASER INDUCED BREAKDOWN SPECTROSCOPY (LIBS):
CHARACTERIZATION OF GUNSHOT RESIDUE AND BULLET WIPE
by
KRISTINA TRUITT
ELIZABETH GARDNER, COMMITTEE CHAIR
JASON LINVILLE
MITCH RECTOR
A THESIS
Submitted to the graduate faculty of the University of Alabama at Birmingham,
in partial fulfillment of the requirements for the degree of
Master of Science
BIRMINGHAM, ALABAMA
2011
Page 2
ii
LASER INDUCED BREAKDOWN SPECTROSCOPY (LIBS):
CHARACTERIZATION OF GUNSHOT RESIDUE AND BULLET WIPE
Kristina Truitt
Master of Science in Forensic Science
Laser Induced Breakdown Spectroscopy (LIBS) can be used to quickly determine
the elemental composition of gas, liquid, and solid samples with minimal sample
preparation. A LIBS instrument commonly incorporates a ND:Yag Laser and a CCD or
Eschelle detector. The laser pulse ablates material from a sample, producing a high
temperature plasma. The plasma emits light at wavelengths that are characteristic of the
elements ablated from the sample. The emission of the plasma is collected and analyzed
by a detector within the LIBS system.
The advantages of LIBS are that the method is relatively non-destructive, very
little sample preparation is required, and the spectra can be obtained within just a few
minutes. The disadvantages are that the limit of detection is presently only 4-10 ppm and
that the percent composition of trace elements cannot be determined to the level of
accuracy required for forensic analyses. However, as shown by this project, LIBS can be
used as a quick test when the presence of specific elements are used to identify a sample,
such as lead for a bullet fragment or lead (Pb) and barium (Ba) for a suspected bullet
hole.
In cases that involve the use of firearms, traces of lead, barium, and antimony can
be detected on the victim, criminal, and other objects that have come in contact with the
firearm and/-or fired projectiles. However, when the point of impact is more than six feet
from the source of the gunshot, the only residue at the site of impact may be bullet wipe,
Page 3
iii
a dark ring of lead or barrel residue wiped from a bullet as it passes through the material.
This study focused on the effects that distance on the detection of gunshot residue and
bullet wipe on clothing, cement block, wood, and drywall shot at distance from 1” to 12’.
Peaks at 280.16, 368.49, and 405nm are characteristic of lead and at 455.4, 493.4, and
553.5 nm for barium.
The techniques developed in this project have the potential to establish an area of
bullet impact detection in the presence and absence of gunshot residue.
Keywords: Spectroscopy, Gunshot Residue, Bullet
Page 4
iv
DEDICATION
In Loving Memory of William Benson
Page 5
v
ACKNOWLEGDEMENTS
Dr. Elizabeth Gardner, Research Advisor
Dr. Jason Linville, UAB MSFS Director
Janey Deimling, UAB Graduate
Mitch Rector, Birmingham Police Department Forensic Scientist
Perry Gordon, Birmingham Police Department Firearms Examiner
Page 6
vi
TABLE OF CONTENTS
Page
ABSTRACT ............................................................................................................ ii
DEDICATION ....................................................................................................... iv
ACKNOWLDEGEMENTS .....................................................................................v
LIST OF TABLES ............................................................................................... viii
LIST OF FIGURES ............................................................................................... ix
INTRODUCTION ...................................................................................................1
MATERIALS AND METHODS ...........................................................................10
Preliminary Research Samples .........................................................................10
Preliminary Samples LIBS Procedure ..............................................................12
Experimental Samples ......................................................................................13
Experimental Controls ......................................................................................15
Experimental LIBS Procedure ..........................................................................16
RESULTS AND DISCUSSION ............................................................................18
Preliminary Research .......................................................................................18
Experimental Samples .....................................................................................24
T-shirt Samples ................................................................................................24
Cement Samples...............................................................................................29
Wood Samples .................................................................................................33
Drywall Samples ..............................................................................................36
CONCLUSION ......................................................................................................41
LIST OF REFERENCES .......................................................................................44
Page 7
viii
LIST OF TABLES
Table Page
1. Comparison of Wavelength and Intensity of the
Lead Standard and Shotshell 1 .....................................................................................19
2. Lead Spectral Lines of T-shirt Samples
and Controls .................................................................................................................25
3. Barium Spectral Lines of T-shirt Samples
and Controls .................................................................................................................26
4. Lead Spectral Lines of Cement Samples
and Controls .................................................................................................................30
5. Barium Spectral Lines of Cement Samples
and Controls .................................................................................................................31
6. Lead Spectral Lines of Wood Samples
and Controls .................................................................................................................34
7. Barium Spectral Lines of Wood Samples
and Controls .................................................................................................................35
8. Lead Spectral Lines of Drywall Samples
and Controls .................................................................................................................38
9. Barium Spectral Lines of Drywall Samples and Controls ...........................................39
Page 8
ix
LIST OF FIGURES
Figure Page
1. Image of pellet arrangement for Shotshell #3 (AR23L22) ..........................................11
2. Image of Cement Block shot at varied distances;
Label on left Sample #3, 6’; Label on right: Sample #4, 9’ .........................................14
3. Pb composition comparison of a PbNO3 Standard (#121)
and Shotshell 1 .............................................................................................................18
4. Shotshell Comparison: Pb composition comparison of four
different shotshells using spectral analysis overlay .....................................................20
5. Ba composition comparison of clean spot and
GSR spot on T-shirt .....................................................................................................21
6. Ba composition comparison of all preliminary samples
and Pb Standard ...........................................................................................................23
7. Pb composition comparison of all preliminary samples and
Pb Standard ..................................................................................................................22
8. Ba Composition of T-shirt Samples 1-5 (1” to 1’) ......................................................27
9. Pb and Ba Composition of T-shirt Samples 6-10 (1’to12’) .........................................28
10. Elemental Composition Comparison of Cement Sample and Control
Prominent Pb and Ba Peaks .........................................................................................32
11. Elemental Composition Comparison of Wood Sample and Control
Page 9
x
Prominent Pb and Ba Peaks .........................................................................................36
12. Elemental Composition Comparison of Drywall Sample and Control
Prominent Pb and Ba Peaks .........................................................................................40
Page 10
1
INTRODUCTION
Laser Induced Breakdown Spectroscopy (LIBS) uses a short pulse laser to
produce a plasma that can be analyzed for the elemental composition of gas, liquid, and
solid samples. The LIBS instrument uses a laser, commonly a ND: Yag Laser to produce
a plasma that evolves with time from the point of impact of the incident laser pulse. The
emission of the plasma is collected and analyzed by either a CCD or Eschelle detector
within the LIBS system. The intensities of the spectral emission lines vary with the
following conditions: the type of sample, the distance from the center of the plasma, and
the wavelength of the incident laser light.1 In most recent years, LIBS has been used to
detect trace elements in various types of samples in varied fields of study. Given LIBS’
ability to detect elements in a sample, LIBS could be used as a presumptive test for
gunshot residue in evidence samples. For example, if a fragment of metal, suspected to be
a fired bullet, was retrieved from a crime scene, LIBS could be used to detect elements
associated with gunshot residue on the metal fragment, further suggesting the fragment is
a bullet.
The advantages of applying LIBS as a presumptive test for gunshot residue is its
ability to identify the elemental composition of a sample in less than four minutes and
operate without the use of reagents and sample preparation. One limitation of LIBS as a
presumptive test is that if further analysis required the quantitation of the concentration of
the trace elements present in the sample, accomplished second analytical method would
Page 11
2
be required. In a case study done at Iowa State University, determination of trace
element concentrations lead to the class characterization of a bullet fragment as a
Cascade, Federal, Remington, or Winchester by identifying only three elements.2
Characterization of the bullet fragment involved detecting the trace elements bismuth,
arsenic, and copper, and calculating the concentration of these trace elements.
Unfortunately, LIBS lacks precision and the current technology does not allow for
quantitation. Most trace elements are present in low concentrations, making it difficult to
assign elements to individual peaks in the spectra and distinguish these trace elements
from background noise. The peak intensity for trace elements can be increased with an
increase in laser power, however if the main constituents are too concentrated the
detector becomes over saturated, making the wavelength measurements inaccurate.
Overall, LIBS precision in measuring the concentration of trace elements can range from
1-40% accuracy.
Researchers at the Fraunhofer- Institute for Laser Technology have taken
measures to gain insight into relevant parameters that have a significant effect on the
spectral data retrieved from ablating the surface of a sample.3 Wester and Noll developed
a heuristic model under the assumption that the plasma content and state of the plasma is
known in order to model or represent the emission spectra of the plasma created by LIBS.
The plasma in terms of the model is described as a spherical shell. In the model, the LIBS
plasma created by LIBS was recreated as two shells, the inner layer, the plasma core, and
the outer layer, ambient gas, both assumed to only consist of atoms, ions, and electrons.
With the formation of the two shell plasma model, the presence of an ambient gas, such
as Argon, was determined to have a positive influence the spectral data by lessening the
Page 12
3
variation in spectral conditions. They were able to determine that factors such as the
interaction of the laser with sample, the generation of the plasma, and the interaction of
the laser with the plasma all affect the conditions of the spectral data. In this model and as
assumed in many other studies, the spherical plasma created by LIBS is assumed to be in
local thermodynamic equilibrium (LTE). The LTE effect is used to depict the effects of
laser pulse interaction with the surface of the sample and how this correlates to plasma
emission.
Another method that incorporates the LTE effect and addresses the issue of
quantification of elements in a sample is the Calibration Free LIBS (CF-LIBS). This
particular method focuses on analyzing LIBS spectra generated from samples consisting
of multiple elemental components by measuring the relative intensity of peaks generated
and the property of the plasma generated from ablating the sample.4 Quantitative analysis
can be done based on assumptions regarding the Boltzmann population of levels of
electrons at their excited states. In order to ensure the LTE effect, this method employs
the McWhirter principle; which requires that at all electronic levels, the depopulation rate
of electron collision must be at least ten times greater than the depopulation rate of
electron radiation.
Based on a series of integration and log calculations of the transition between
particular electronic levels, the CF-LIBS method is able to accurately quantify the
concentration of elemental components of a sample. Their results showed that the CF-
LIBS method would be suitable for the detection of elements in metal alloys. The
disadvantage of CF-LIBS is that the calculation of concentrations in organic and mineral
Page 13
4
samples cannot be calculated with sufficient accuracy. This is due to organic and mineral
samples containing multiple elements at very small concentrations.
Abdel-Kareem et.al carried out a study on the analysis of historical metal threads
collected from different museums.5 Metal threads have been used since 3rd century BC
and are commonly used today in couture garments. The almost instant deterioration of
these threads causes scratches, significant changes in the surface morphology, and loss of
surface definition. The analysis of these samples using LIBS would help validate its
potential use for the detection of trace elements in degraded samples and its ability to
produce comparable, reproducible results. In order to test the efficiency of the spectral
data produced from LIBS analysis, the historical metal threads were also analyzed by
Scanning Electron Microscope (SEM) with energy-dispersive x-ray analyzer (EDX).
LIBS was compared to this particular technique since it had been proven to be the best
technique for the elemental analysis of samples such as the historical metal threads. From
analysis of the metal samples, it was determined that LIBS was a useful technique for the
detection of relatively small quantities, or trace amounts of copper (Cu), silver (Ag), and
gold (Au) in the historical metal elements.
The same elements were detected by LIBS even though the samples were in a
much degraded state. Their experimental procedures also indicated that the number of
times the samples were hit with the laser was based on the conditions of the sample being
analyzed. Spectral data could vary based on the amount of degradation of the samples
and the sensitivity of the surface area undergoing ablation. These results indicate that
LIBS could have the potential to be applied to other areas of interest such as the detection
of gunshot residue and bullet wipe.
Page 14
5
LIBS has also been used in research studies involving the analysis of different
types of cement. In an article published by Gondal et al. LIBS was used to detect the
elemental concentration of chloride in three different cement samples.6 The objective was
to develop a more cost effective and time efficient analytical technique to characterize the
durability and safety of concrete structures. The LIBS spectral data were compared to
spectral data results from Inductively Coupled Plasma Emission Spectroscopy (ICP-ES).
LIBS spectral data indicated that the primary elemental components of the cement
samples were calcium, aluminum, silicon, iron, chromium, magnesium, sodium, chlorine,
sulfur, phosphorous, and manganese. Calibration curves were calculated in order to
accurately measure the concentration of trace amounts of metal in each cement sample.
The concentration of each of the metal components was verified by comparing the results
to the data generated by the calibrated ICP spectrometer. The spectral data was compared
the standards from the NIST database and the data produced from ICP-ES analysis, and
were found to be within +/- 2%. From this comparison, it was determined that LIBS was
an efficient technique in the detection of elemental components. The reproducibility of
LIBS was calculated by determining the relative standard deviation (RSD), resulting in a
precision of 2-4%.
LIBS has also been used in forensic science laboratories for the analysis of trace
evidence. One research project in particular focused on the discrimination of glass
fragments for forensic applications.7 Rodriguez et al. made a comparison of the spectra of
glass fragments collected from various car windows, using linear and rank correlation
methods. The basis of this research relied on the sole fact that glass has a unique spectral
fingerprint that can be characterized. The potential of LIBS was evaluated based on its
Page 15
6
ability to discriminate glass samples of similar refractive index (RI) values. This was
accomplished by comparing the spectra generated from LIBS within the same day.
Analysis of the glass fragments involved comparing the spectra of unknown glass
fragments to a spectral database, direct comparison of the glass spectra to one another,
and validation of the instrument’s reproducibility. From this study, it was determined
that LIBS could provide an effective identification and discrimination of glass fragments
at a 95% confidence level. These results indicated that LIBS could become a useful
technique in the analysis of forensic glass samples.
Due to the ease in identifying trace elements, it may be possible to use LIBS to
quickly screen evidence samples for the presence of gunshot residue (GSR). Gunshot
residues are primarily composed of burnt and un-burnt particles from combustion and
residual components of the bullet, firearm, and primer.8 Currently, color tests are often
used to detect GSR in evidence samples. The main elements of detection in gunshot
residue are primer residue (lead, barium, and antimony), nitrites, and nitrates.9 Romolo
and Margot evaluated several color testing methods to determine the best inorganic
chemical method to use for GSR analysis. A portion of their surveys focused on the
comparison of various color tests used for presumptive testing. They compared the
paraffin test by Teodoro Gonsalez of the Mexico City Police to the chemical test of
Harrison and Gilroy produced in 1959.10
Both the paraffin test by Gonasalez and chemical test by Harrison and Gilroy
were able to detect GSR. The Paraffin Test consisted of using hot paraffin to make the
cast of a suspected shooter’s hand. Once the wax cooled, it was peeled off and the cast
was sprayed with N, N,-diphenylbenzidine in concentrated sulphuric acid. This chemical
Page 16
7
reaction results in a blue color change in the presence of nitrates. The color change to a
deep blue indicated the presence of partially burnt and un-burnt particles. The color tested
examined by Harrison and Gilroy, the sodium rhodizonate test, detected lead styphnate,
barium nitrate, and antimony sulphide from primer residue. Their GSR samples were
collected from the hands of the shooters using swabs moistened with hydrochloric acid.
The swabs were dried and then treated with two reagents: triphenylmethylarsonium
iodide and sodium rhodizonate. In their color testing, the first reagent produced an orange
color if antimony was present and the second reagent produced a red color if lead or
barium was present. If the swab turned red, dilute hydrochloric acid was added and the
sample turned a purple color if lead was present. The Harrison and Gilroy chemical test
was found to be more reliable in that it produced the least number of false positive
results.
Currently, the modified Griess test is used for the detection of nitrite residues.
Nitrite residues can be detected in unspent gunpowder residue that has not fully burned
when a bullet is projected out of the muzzle of a firearm. The residue occurs on surfaces
pierced by a bullet when nitrocellulose is not completely burned during flame ignition
after the firing pin is hit. The presence of GSR, in the form of nitrites, is indicated by an
orange color change. The modified Griess test consists of 1% sulfanilamide, 0.1% N-1-
naphthyl-ethylenediamine dihydrochloride, and 5% phosphoric acid. This chemical test is
most effective for the detection of nitrite gunshot residue on samples or materials shot
within 3’ or less. The level of GSR present is also dependant on the type of ammunition
and the type of firearm used.
Page 17
8
Although chemical testing is more efficient for detecting the presence of GSR
over a large area, it is limited to shots fired from less than 6’ from the target. The range
from which a firearm is shot determines the level of element detection. At close ranges,
the gaseous cloud of GSR particles is directly dispersed onto the surface of the target. In
farther ranges, the deposit of gunshot residue decreases and disperses before it reaches
the target. GSR can also be transferred to the target by having actual contact with the
bullet itself, this is known as bullet wipe. It has been determined that transfers of GSR, at
close range, from far range, or by bullet wipe, can be characterized and distinguished
from one another.11 If a dark blackish-grey area of GSR is visible on the sample, this is
an indication that the firearm was fired in close range. The presence of GSR in this dark
colored area, or in the bullet wipe area, could be characterized by the presence of the lead
peaks using LIBS.
According to Martiny, the characterization of gunshot residue is essential when
investigating firearm-related incidents. Martiny detected gunshot residue by using a
combination of chemical testing and Scanning Electron Microscopy/ Energy Dispersive
X-ray Spectroscopy (SEM/EDS) analysis.12 This study analyzed the elemental
composition of heavy-metal free environmental ammunition primers from Brazil. They
chose this ammunition type based on the fact that more lead free ammunition is being
produced in response to increased exposure to heavy metal-rich airborne particles. Their
analysis compared the morphology and characterization of samples produced by
Companhia Brasileira de Cartuchos (CBC) and Clean Range ammunition. They used
SEM/EDS to examine GSR from the shooter’s hands upon firing a firearm. Differences
were observed in the shape and composition for the Clean Range samples. The Clean
Page 18
9
Range samples consisted of first and second generation lead-free ammunition including 9
mm Luger, .380 AUTO, .38 Special, .40 Smith and Wesson, and .45 Automatic Colt
Pistol calibers. Their results indicated the detection of strontium with spherical shaped
particles in the first generation ammunition and the detection of potassium, aluminum,
silicon, and calcium with irregular shaped particles in the second generation ammunition.
From this study, it was concluded that the identification of gunshot residue from the CBC
ammunition was impossible without a distinct metallic marker within the composition of
the primer.
This project has been constructed to determine the validity and essential use of
LIBS in crime scene investigations that involve the use of firearms. The purpose of this
study is to determine whether LIBS has the potential to be used as a presumptive test for
the detection of gunshot residue and bullet wipe from suspected impact areas through the
analysis of bullet fragments, T-shirt, cement, wood, and drywall samples.
Page 19
10
MATERIALS AND METHODS
Preliminary Research Samples
The LIBS System (Crossfire Model, Photon Machines, Inc.) was used to analyze
the elemental composition of six types of samples, including a cotton T-shirt shot at close
range, three sets of known lead, unfired shotgun pellets, one unfired lead bullet, a
2”x4”wooden block and a cement block, each shot two years ago. The cement block and
2”x4” wooden block were shot from a distance of 10’. The known lead shot samples
analyzed were made by Remington and included:
Shotshell#1: Slugger 20 gauge Hollow Point rifled Slug, Lot # BP26B527
Shotshell#2: Game Load 20 gauge- 8 shot, Lot# BG16F520
Shotshell#3: Express Plastic 16 gauge-7 ½ birdshot, Lot # AR23L22
Shotshell #4: Premier Target Load 20 gauge- 9 shot, Lot # BF08U521
These samples were collected from a previous research project carried out by
University of Alabama at Birmingham Graduate, Janey Deimling between 2007 and
2008. The objective of her project was to use a microcrystalline test to detect the presence
of lead and prove it to be a much more suitable test than the currently used chemical test,
sodium rhodizonate. Research results showed that the microcrystalline test was
successful on samples that had at least 300µg/mL of lead. From the conclusions drawn
from this project, it was determined that collection of a sample’s elemental composition
would be easier with less sample preparation and faster data generation by using LIBS.
Page 20
11
Since the samples were already proven to contain a detectable amount of lead composite,
the application of LIBS presented a possible avenue to the detection of trace amounts of
elements such as lead.
A T-shirt that had been shot at close range, with visible gunshot on a wide area
residue surrounding the hole, was used to develop a gunshot residue detection method.
The T-shirt had been shot 2 years previous to the research and was analyzed primarily to
develop the experimental method. Approximately 1.2 cm samples were cut from four
spots on the T-shirt, at different distances (3/4” to 10”) from the bullet hole. The control
spectrum of the T-shirt was collected from an apparently clean section on the back, lower
portion of the T-shirt. The clean area was cut out using scissors and was mounted onto a
business card using double-sided tape. An approximate diameter of 1.2cm of the gunshot
residue sample was cut out near the hole using scissors as well.
The pellets from unfired cartridges were removed and separately mounted onto a
business card using double-sided tape. The pellets were placed in a close, compact
arrangement as shown in Figure 1.
Figure 1. Image of pellet arrangement for Shotshell #3 (AR23L22)
Page 21
12
The cement block and wood slab samples were gathered by chipping away pieces
of each material where discoloration from the bullet wipe was obvious. The pieces of
each material were mounted on a card using double-sided tape as well.
Preliminary Samples LIBS Procedure
The LIBS instrument is coupled with a Nd: Yag Laser with a CCD detector. LIBS
is used to identify the elemental composition of samples analyzed by laser pulsing the
samples, ablating the surface of the sample at 1500˚C, and using the detector to collect
the spectra of electrons after atoms relax from their excited state. Lead nitrate was used
for the lead standard (#121). The powder was pressed into pellets and analyzed by the
OOILIBS software (December 2008, Q-Switch delay -1.7us).
All samples were pulsed with the laser and the atomic spectral lines were
collected. The instrument was set at an integration of -0.42, a Q-switch delay time of -
0.4us, and the power level of 200mJ. The LIBS software, OOILIBS, was used to analyze
each sample and identify the elements of lead, barium, and antimony. The spectral data
collected from each sample was collected within the range of 200 to 600nm. Analysis of
the samples varied depending upon the intensities of the peaks detected; therefore the
OOILIBS settings of peak height (threshold) and search width were adjusted in order to
detect trace elements of lead in the samples. The threshold was adjusted been 300 and
800 counts based on the intensity of the peaks, and the search width varied based on the
location of distinct, sharp peaks in the spectra.
Page 22
13
Experimental Samples
In light of the preliminary work done to determine whether or not LIBS could
detect GSR, further research was done to determine if LIBS could be used to detect bullet
wipe without the presence of GSR patterns and tattooing. Samples analyzed included ten
cotton T-shirt samples, a cement block, dry wall samples, and a 2”x4”wooden block. The
T-shirt samples were shot at a distance of 1 inch to 12 feet in order to find a trend in
decreasing element detection as the distance increased.
The shooting range utilized to soot the samples analyzed was an open, training
field utilized by the Birmingham Police Department. The training field contained a
wooden fixture (setup) used for training; this particular wooden structure was used to
prop up the samples for shooting preparations. The shooting distance was measured by
using measuring tape to construct and assure the correct distance between the mounted
samples and the firearm. The cement block, wood sample, and dry wall were propped up
on a wooden stand before shooting. The cement block, wood sample, and dry wall
samples were divided into sections based on shooting distance by labeling the perspective
distance on each section of the sample, as shown in Figure 2. The samples collected for
experimental analysis included a total of 10 T-shirt samples (shot from 1 inch to 12 feet),
4 cement samples (shot from 3 feet to 12 feet), 5 wood samples (shot from 1 foot to 12
feet), and 5 dry wall samples (shot from 1 foot to 12 feet). Each sample was shot with a
lead semi-wad cutter bullet (Pb-SWC) using a Smith and Wesson, 357 Magnum Caliber
Revolver (Model 13).
Page 23
14
Figure 2. Image of Cement Block shot at varied distances; Label on left: Sample #3, 6’;
Label on right: Sample #4, 9’
The 10 T-shirt samples were separately mounted onto pieces of cardboard with
staples and properly labeled based on sample number and distance. The mounted T-shirt
samples were individually placed on a wooden stand and shot. The T-shirt samples were
shot from a distance of 1”, 3”, 6”, 9”, 12” (2), 3’, 6’, 9’ and 12’ at a 90˚ angle with
respect to the position of the firearm. Each bullet hole was analyzed by cutting a piece of
the fabric near the bullet hole entry using scissors. Each piece of fabric was mounted onto
a separate business card for LIBS analysis.
As shown in Figure 2, the cement block was divided into sections using a
permanent marker. Each section of the cement block was labeled with the respective
sample number and the distance. The block was shot once in each section from a distance
of 3’, 6’, 9’, and 12’ at an angle of 45˚ with respect to the position of the firearm. The
cement samples were then collected by chipping away pieces in the area of bullet impact
using a flat head screwdriver and a hammer. The pieces of cement from each section of
the cement block were mounted on separate, labeled business cards using double sided
tape.
Page 24
15
The wooden block was divided into sections like the cement sample in Figure 2.
The wood samples were also propped up on the wooden bench and shot from a distance
of 3’, 6’, 9’, and 12’ at a 90˚ with respect to the position of the firearm. The wood
samples were collected by cutting pieces from the area of bullet impact using a box-
cutter. Like the T-shirt and cement samples, the pieces from each section of the wood
block were mounted onto labeled business cards.
Five drywall samples were placed onto a wooden bench and shot from a distance
of 1’, 3’, 6’, 9’, and 12’ at a 90˚ with respect to the position of the firearm. The dry wall
samples were also collected by cutting away a piece from the area of bullet impact using
a box cutter. However, the dry wall sample had three layers, the chalky, white gypsum
material sandwiched between two layers of paper, so the top coating was sliced off for
easier, cleaner mounting and detection purposes.
Experimental Controls
The control samples consisted of the Pb-SWC bullet head and jacket (positive
controls), scissors, flat head screwdriver, box-cutter, and cardboard. Samples from the T-
shirt, cement, wood, and drywall controls before being shot were also collected as
negative controls in order to distinguish between bullet wipe and the elemental
composition of the samples themselves. The spectral data of the controls were collected
before they were used in the experimental procedures.
Page 25
16
Experimental LIBS Procedure
Like the preliminary samples, the experimental samples were analyzed using
OOILIBS software for the detection of lead and barium. AddLIBS software was also
used to detect trace amounts of elements in the sample that may not have been detected
using the OOILBS software. Collection of spectral data was collected in the 200 to
600nm range and the analysis of each sample varied based on the intensity of the peaks,
varying the threshold adjustment, but not accounting for elements detected below a
threshold of 300. All the samples were compared to each other to identify the similarities
or differences in their spectral data.. The elemental lines from each sample were
compared to The National Institute of Standards and Technology (NIST) Database of
common element wave numbers. The NIST Database was also used to validate the results
of the LIBS instrument.
All the samples were mounted onto a business card using double sided tape,
placed on the laser platform, and then ablated. For the analysis of the T-shirt, cement,
wood, and drywall samples, the instrument parameters were set at an integration of -
0.42., a Q-switch delay time of 0.4us, and the power level of 200mj. OOILBIS was used
to analyze each spectrum within the threshold range of 300 to 800 counts and 200 to 600
nm, with a +/- 3 search width of the spectrum. All samples were compared to each other,
to the spectra of the Pb-SWC bullet, and to the spectra of their perspective control
samples. T-shirt samples were also compared to T-shirt, cardboard, and scissor controls.
The spectra of the 4 cement samples were also compared to the flat head screwdriver
controls. The spectra of the wood sampled and drywall samples were also compared to
box-cutter controls.
Page 26
17
The controls were analyzed individually by mounting them on a business card,
using double-sided tape. The Pb-SWC bullet head was analyzed by mounting the bullet
onto the card and shooting it with a laser on various areas of the top portion of the bullet.
The spectral data of each of the controls were analyzed using the OOILIBS software at
the same settings. The Pb-SWC bullet jacket was also analyzed by mounting the bullet
onto a business card and shooting various areas of the lower portion of the bullet. The
metal portion of the flat head screwdriver and box-cutter were detachable and were also
placed on a business card and shot at various spots. The business card was simply placed
on the sample mount and shot at various spots as well. The T-shirt, cement, wood, and
drywall controls were collected from areas of the sample that had not come in contact
with the bullet itself. This was ensured by collecting the control samples before the
samples were shot at varied distances.
To prevent cross contamination between the collection of samples, the tools used
were cleaned with ethanol each time applied. The tools were washed with ethanol
between the collection of the same type of samples, however in some cases the same tool
was used for the collection of different sample types. A box-cutter was used for the
collection of both wood and drywall samples so to ensure that cross contamination did
not occur the blade was changed between collection of the wood samples and dry wall
samples.
Page 27
18
RESULTS AND DISCUSSION
Preliminary Research
The spectral data results from the LIBS instrument were consolidated into scatter
plots, as shown in Figures 3-4. The scatter plots generated for each sample were used to
compare the spectra by graph overlay and to distinguish each sample from one another
through element identification. Each sample was characterized by distinguishing sharp,
concise peaks of elements distinct from peaks generated by background noise.
Figure 3. Pb composition comparison of a PbNO3 Standard (#121) and Shotshell 1
In analyzing each sample, it was determined that there were significant
similarities between the Pb Standard and each shotshell sample. In a comparison of the
Page 28
19
Pb Standard and Shotshell 1, the major lead lines were detected at 280.19nm, 283.30nm,
363.95nm, 368.34nm, and 405.78 nm, as shown in Figure 3. There is a direct overlay of
the lead peaks detected in both the lead standard and Shotshell 1. These results validated
the operating system of LIBS and verified the prominent lead peaks that should be
detected in the bullet fragments.
There were also significant similarities between each shotshell sample. The
spectral lines in Table 1 indicate that the major lead lines in each shotshell sample were
again 280.19nm, 283.30nm, 363.95nm, 368.34nm, and 405.78nm. As shown in Figure 4,
the spectra from each of the four shotshells were very similar to each other. The primary
elemental compositions of the shotshells are shown.
Table 1
Comparison of Wavelength and Intensity of the Lead Standard and Shotshell 1
PbNO3 Standard #121 Shotshell 1 280.16 nm/ 1833cts. 280.19 nm/ 4095cts. 283.42 nm/1084cts. 283.30 nm/ 2381cts. 364.01 nm/ 1016cts. 363.95 nm/ 4095cts. 368.48 nm/ 2003cts. 368.34 nm/ 4095cts. 405.89 nm/ 2060cts. 405.78 nm/ 4095cts.
*Cts: The counts represent the relative intensity of a particular peak relative to other peaks detected in the spectra.
Page 29
20
Figure 4. Shotshell Comparison: Pb composition comparison of four different shotshells
using spectral analysis overlay
The composition shown in Figure 4 can be used as a control for the detection of
lead in bullet wipe. Lead was the major element detected in each bullet sample. The
spectra of the clean and GSR samples of the T-shirt are shown in Figure 5. There were
significant differences between the clean and gunshot residue spots, indicating that an
area that has come in contact with a bullet fired from a firearm can be distinguished from
an area that did not have contact with the fired bullet.
Intensity (counts) vs. Wavelength (nm)
Page 30
21
Figure 5. Ba composition comparison of clean spot and GSR spot on T-shirt
Along with lead, barium was the other prominent element detected in the GSR
spot, suggesting that these elements can characterize an area that has come in contact
with a fired bullet. The prominent barium lines were detected at 455.40nm, 493.41nm
and 553.54nm. Barium peaks were detected in the clean T-shirt sample; however, it was
determined that it was as a result of background noise.
Intensity (counts) vs. Wavelength (nm)
0
500
1000
1500
2000
2500
3000
0 100 200 300 400 500 600 700
Soiled frag
Clean frag
553.55 nm493.41
455..40 nm
Page 31
22
Figure 6. Ba composition comparison of all preliminary samples and Pb Standard
As shown in Figure 6, there were slight similarities between the cement block
sample, gunshot residue spot from the t-shirt, lead standard, and wood slab sample. In
Figure 6, the prominent barium lines were detected at 350.11nm and 553.55nm in the
cement block sample, gunshot residue spot, and wood slab sample. In Figure 7, the
prominent lead lines were detected at 280.2nm, 283.31nm, 363.96nm, and 368.35nm, and
405.78nm in the cement block sample, gunshot residue spot, and wood slab sample in
comparison to the lead standard. From the elemental composition analysis of each
sample, it was determined that LIBS could detect the prominent lead lines. These
Intensity (counts) vs. Wavelength (nm)
*This is a zoomed –in view of the spectral data, lead standard is not shown in this image.
Page 32
23
prominent lead and barium lines detected in the gunshot residue spot, wood slab, and
cement block indicated that bullet wipe had resulted from the bullet making contact with
each of the materials.
Figure 7. Pb composition comparison of all preliminary samples and Pb Standard
The T-shirt shot at close range was sampled at varied distances from the bullet
hole. Spectra results showed that the detection of elements becomes increasingly difficult
as the distance from the bullet hole increases. Thus far, the element detection was the
most effective nearest the hole and as far as 1 inch from the hole. The analysis of these
samples indicated that LIBS would be efficient for its use as a presumptive test for the
detection of gunshot residue and bullet wipe from suspected impact areas. In comparison
to chemical testing, LIBS would be more useful in the detection of bullet wipe if gunshot
residue is not visible. This reasoning is supported by the fact that there is an immediate
detection of gunshot residue in chemical testing a visible gunshot residue.
Page 33
24
Experimental Samples
From the preliminary results, it was determined that further work could be done
by determining limits of detection for bullet wipe in T-shirts and other samples shot at
varied distances and that the development of controls would be important for comparative
analysis of the materials analyzed and the tools that could be used to make holes in
construction materials.
T-shirt Samples
As previously mentioned, 10 T-shirt samples were analyzed. The first set, T-shirts
labeled 1 through 5, were shot from a distance of 1, 3, 6, 9, and 12 inches. Analysis of
these samples only showed the presence of prominent barium peaks at 455.40, 493.41,
and 553.55 nm, as shown in Figure 8 and Table 2 and 3. Of the first set of T-shirt
samples, Sample 5, shot from a distance of 12 inches, was the only sample to show Pb
Peaks. These peaks were identified at 368.346 and 405.781 nm, but at relatively low
intensities in comparison to the Pb peaks in the other samples analyzed. This finding
suggests that the soot and GSR produced from the firearm when fired did not have time
to disperse around the bullet entrance relative to the distance of the firearm from the T-
shirts. From the analysis of the first set of T-shirts, it would be difficult to detect evidence
of bullet wipe, therefore it would be more productive to apply the standard chemical test
for GSR at distances ranging from 1’ to 12’.
Page 34
25
Table 2
Lead Spectral Lines of T-shirt Samples and Controls
Pb Lines Wavelength 280.02nm 283.31nm 363.96 nm 368.35n
m 405.78nm
T-shirt Samples Sample 1 (1”) Sample 2 (3”) Sample 3 (6”) Sample 4 (9”) Sample 5 (1’) X X Sample 6 (1’) X X X Sample 7 (3’) X X X X Sample 8 (6’) X X X X X Sample 9 (9’) X X X X X Sample 10 (12’) X X X PbSWC Bullet Head
Sample: head-01 X X X X X PbSWC Bullet Jacket
Sample: jacket-01
X X X
T-shirt Control Sample: T-shirt-01
Cardboard Control Sample: Cardboard-01
*Yellow area: Spectral line may be due to background noise.
Page 35
26
Table 3
Barium Spectral Lines of T-shirt Samples and Controls
Ba Lines Wavelength 455.40nm 493.41nm 553.55nm T-shirt Samples Sample 1 (1”) X X X Sample 2 (3”) X X X Sample 3 (6”) X X X Sample 4 (9”) X X X Sample 5 (1’) X X X Sample 6 (1’) X X X Sample 7 (3’) X X X Sample 8 (6’) X X X Sample 9 (9’) X X X Sample 10 (12’) X X X PbSWC Bullet Head Sample: head-01 X X X PbSWC Bullet Jacket Sample: jacket-01 X X X T-shirt Control Sample: T-shirt-01 Cardboard Control Sample: Cardboard-01
Page 36
27
Figure 8. Ba Composition of T-shirt Samples 1-5 (1” to 1’)
The second set of T-Shirts, Samples 7 through10, were shot from a distance of 1,
3, 6, 9, and 12 feet, and the data generated from these spectra had slight variance in the
presence of lead and barium prominent peaks, as portrayed in Figure 9. Analysis of this
set of T-shirts showed both prominent Lead and Barium peaks. The spectra of T-shirt
Sample 6 contained Pb peaks at 36.96, 368.35, and 405.78 nm and Barium peaks at
455.40, 493.41, and 553.55 nm. T-shirt Sample 7 contained spectral peaks of Pb at 280.2,
283.31, 363.96, and 404.78 nm and peaks of Ba at 455.40, 43.4, and 553.55 nm. T-shirt
Samples 8 and 9 contained all of the prominent Pb and Ba peaks detected by LIBS. The
spectral data for T-shirt sample 10, shot from 12 feet, only showed 3 of the 5 prominent
Pb peaks which were 363.96, 368.35, and 405.78 nm.
Intensity (counts) vs. Wavelength (nm)
Page 37
28
Figure 9. Pb and Ba Composition Comparison of T-shirt Samples 6-10 (1’to 12’)
All of the T-shirt samples were compared to the Pb-SWC head and jacket of the
bullet, T-shirt, cardboard, and scissors. Based on the spectral data generated, it was
determined that the T-shirt control, cardboard, and scissors (negative controls) did not
make a contribution to the Pb and Ba lines detected in Sampled 6 through 10 since the
spectral data of these controls did not contain any of the prominent barium and lead peaks
detected by LIBS. The spectra of the head of the Pb-SWC bullet, which serves as a
positive control of the experiment, contained all the prominent Pb peaks at 280.2, 283.31,
363.31, 363.96, 368.35, and 405.78 nm and all of the prominent Ba peaks at 455.40,
493.41, and 553.55 nm. The spectral data of the jacket of the Pb-SWC bullet, which also
served as a positive control for analysis, contained all of the prominent lean and barium
peaks except for two of the Pb peaks at 283.31 and 363.96 nm. In comparison to the Pb-
Intensity (counts) vs. Wavelength (nm)
Page 38
29
SWC bullet control, it was determined that the lead and barium lines found in Samples 6
through 10 were an indication of bullet wipe.
Cement Samples
Cement samples 2 through 5, were shot from the distance of 3, 6, 9 and 12’, and
the data generated varied tremendously, making it difficult to track a trend or pattern of
detection based on distance. The cement block was not shot at a distance of 1’ due to
safety precautions taken to avoid injuries that could occur with shooting cement at such a
close range. Spectral data from cement sample 2, shot from 3’, had only one prominent
barium peak present at 553.55 nm. The data retrieved from the spectra of cement sample
3, shot from 6’, only indicated the presence of barium at 455.40 nm and 553.55 nm.
Cement Sample 4, shot from 9’, showed lead spectral lines at 280.2, 363.96, 368.35, and
405.78 nm and one barium spectral line at 553.55nm. Completely opposite to cement
sample 4, sample 5, shot from 12’, showed only one prominent lead peak at 280.2 nm and
two barium peaks at 455.40 and 553.55 nm.
The controls used for the analysis of the cement samples were the Pb-SWC bullet
controls, a cement control, and a flat head control. As previously stated the Pb-SWC
bullet controls contain the primary lead and barium peaks in spectral data to indicate the
presence of bullet wipe. The cement control spectral data contained 1 lead peak at 280.2
nm and 3 barium peaks at 455.40, 493.41, and 553.55 nm, as shown in Figure 10 and
Table 4 and 5.
Page 39
30
Table 4
Lead Spectral Lines of Cement Samples and Controls
Pb Lines Wavelength 280.02nm 283.31nm 363.96 nm 368.35nm 405.78nm
T-shirt Samples Sample 2 (3’) Sample 3 (6’) Sample 4 (9’) X X X X Sample 5 (12’) X
PbSWC Bullet Head
Sample: head-01 X X X X X PbSWC
Bullet Jacket
Sample: jacket-01
X X X
Cement Control Sample:
cement-01 X
Flat-head Control Sample: flat-011
Page 40
31
Table 5
Barium Spectral Lines of Cement Samples and Controls
Ba Lines Wavelength 455.40nm 493.41nm 553.55nm
Cement Samples Sample 2 (3’) X Sample 3 (6’) X X Sample 4 (9’) X Sample 5 (12’) X X
PbSWC Bullet Head Sample: head-01 X X X
PbSWC Bullet Jacket Sample: jacket-01 X X X Cement Control
Sample: cement-01 X X X Flat-head Control Sample: flat-011 X X X
Page 41
32
Figure 10. Elemental Composition Comparison of Cement and Cement Control, Prominent Pb and Ba Peaks
The flat head screwdriver spectral data did not have any of the lead or barium
lines; therefore use of the flat head to chip away pieces of the cement did not contribute
to the lead and barium peaks detected in the cement sample. However, since the cement
control contained the same prominent spectral lines lead peak at 280.19 and barium lines
at 455.40, 493.41, and 553.55 nm as some of the cement samples , the barium lines in
particular detected in sample 2, 3, and 5 cannot be used to indicate the occurrence of
bullet wipe. As shown in the graph overlay of a cement sample and the cement control in
Figure 10, it is difficult to differentiate between contribution of spectral barium lines
from the cement block itself and the contact of the bullet with the surface of the cement.
The spectral data looks identical in comparison aside from varied peak intensity.
Intensity (counts) vs. Wavelength (nm)
Page 42
33
Wood Samples
The spectral data from the wood samples showed a much more consistent trend in
lead and barium detection in comparison to the other samples analyzed. Wood Samples 1
through 5 were shot from a distance of 1, 3, 6, 9, and 12’. Indicated in Table 6 and 7,
sample 1 of the wood sampled contained all of the prominent lead peaks at 280.19,
283.31, 363.96, 368.35, and 405.78 nm and all of the prominent barium peaks at 455.40,
493.41, and 553.55 nm. Samples 2 and 4, shot from a distance of 3 and 9’, contained all
of the prominent lead and barium peaks except for one lead peak at 283.31 nm. Sample 3
of the wood samples, shot from a distance of 6’, contained all of the prominent lead and
barium peaks except for the lead lines at 280.2 and 283.31 nm. The only sample of the
wood samples that did not contain all of the primary barium spectral lines was sample 5,
which was shot from a distance of 12’. Sample 5 only contained the barium spectral peak
of 553.55 nm.
The wood samples were also compared to control samples in order to
determine whether barium and lead peaks were from bullet wipe or from the elemental
composition of the control material. All of the samples contained at least 3 of the 5
prominent lead lines that were detected in the Pb-SWC bullet control spectra. The wood
sample spectra were also compared to a box-cutter control and a wood control. In
comparison of the wood sample spectra to the box cutter, matching peaks were found at
280.2 nm for lead and 455.40, 493.41, and 553.55 nm for barium. From these results, it
was determined that the lead line shown at 280.2 nm in the wood samples could have
come from the box cutter used to cut pieces of the wood and not from bullet wipe. The
Page 43
34
spectral results from the wood control only showed a matching peak at the barium line of
553.55 nm, as depicted in Figure 11. This barium line was found in both the Pb-SWC
bullet control, wood control, and the box cutter so bullet wipe and elemental composition
of the sample and controls cannot be differentiated based on this barium spectral line
alone.
Table 6
Lead Spectral Lines of Wood Samples and Controls
Pb Lines Wavelength 280.02nm 283.31nm 363.96 nm 368.35nm 405.78nm
T-shirt Samples Sample 1 (1’) X X X X X Sample 2 (3’) X X X X Sample 3 (6’) X X X Sample 4 (9’) X X X X Sample 5 (12’) X X X X X
PbSWC Bullet Head
Sample: head-01 X X X X X PbSWC
Bullet Jacket
Sample: jacket-01
X X X
Wood Control Sample:
wood-011
Box-cutter Control
Sample: box-011
X
Page 44
35
Table 7
Barium Spectral Lines of Wood Samples and Controls
Ba Lines Wavelength 455.40nm 493.41nm 553.55nm
T-shirt Samples Sample 6 (1’) X X X Sample 7 (3’) X X X Sample 8 (6’) X X X Sample 9 (9’) X X X
Sample 10 (12’) X PbSWC Bullet Head
Sample: head-01 X X X PbSWC Bullet Jacket
Sample: jacket-01 X X X Wood Control
Sample: wood-011 X Box-cutter Control Sample: box-011 X X X
Page 45
36
Figure 11. Elemental Composition Comparison of Wood Sample and Wood Control,
Prominent Pb and Ba Peaks
Drywall Samples
The drywall samples were shot from a distance of 1’, 3’, 6’, 9’, and 12’, analyzed,
and then compared to their perspective control samples. The spectra from the drywall
sample, shot from a distance of 1’, indicated the presence of lead at peaks 363.96, 368.35,
and 405.78 nm and all 3 of the primary barium peaks at 455.40, 493.41, and 553.56 nm.
Drywall sample 2 and sample 4 were shot from a distance of 3 and 9’ and both of their
spectra indicated the same spectra lines as drywall sample 1 except it did not have the
lead spectral line at 363.957 nm. The spectral data retrieved for drywall sample three did
not give the expected results. Drywall sample 3, which was shot from 6’, only showed the
Intensity (counts) vs. Wavelength (nm)
Page 46
37
presence of 1 barium peak at 553.55 nm. The collection of sample 3 may have been
inadequate or there may not have been enough contact of the bullet with the surface area
of the drywall to detect bullet wipe. Sample 5 of the drywall samples only contained the
primary peaks of barium at 455.50, 493.41, and 553.55 nm. These spectral results did not
appear to portray a descending or ascending pattern of detection of bullet wipe.
The drywall samples were compared to both the paper and chalky, white gypsum
material drywall controls, as shown in Figure 12, and a box cutter control. The spectra of
the drywall paper control did not indicate the presence of any of the primary lead and
barium peaks. The lack of detection of lead and barium peaks in the drywall control
demonstrates that the elemental composition did not contribute to the presence of lead or
barium spectral lines in the experimental drywall samples. The other drywall control,
which consists of the white, chalky material, indicated the barium spectral line of 553.55
nm as shown in Table 8 and 9. The box cutter control showed similar peaks to the
drywall spectra at lines 455.40, 493.41, and 553.55 nm. The drywall samples, box cutter,
and drywall controls all indicated the presence of barium at 553.55 nm; therefore this
peak would not be used as a differentiating peak to identify bullet wipe on these drywall
samples. Also, since the box cutter contained the primary barium peaks of 455.40 and
493.41 nm, it was difficult to discern bullet wipe from the use of the box cutter as the
main contributor of these peaks.
Page 47
38
Table 8
Lead Spectral Lines of Drywall Samples and Controls
Pb LinesWavelength 2
80.02nm 2
83.31nm 36
3.96 nm 3
68.35nm 4
05.78nm T-shirt Samples
Sample 1 (1’) X X XSample 2 (3’) X XSample 3 (6’) Sample 4 (9’) X XSample 5 (12’)
PbSWC Bullet Head
Sample: head-01 X X X X XPbSWC
Bullet Jacket
Sample: jacket-01
X X X
T-shirt Control Sample:
T-shirt-01
Cardboard Control Sample:
Cardboard-01 X
Page 48
39
Table 9
Barium Spectral Lines of Drywall Samples and Controls
Ba Lines Wavelength 455.40nm 493.41nm 553.55nm
T-shirt Samples Sample 1 (1’) X X X Sample 2 (3’) X X X Sample 3 (6’) X Sample 4 (9’) X X X Sample 5 (12’) X X
PbSWC Bullet Head Sample: head-01 X X X
PbSWC Bullet Jacket Sample: jacket-01 X X X Drywall Control
Sample: paper-01 Drywall Control
Sample: drywall 2, w. X Cardboard Control
Sample: Cardboard-01 X X X *Yellow area: Spectral line may be due to background noise.
Page 49
40
Figure 12. Elemental Composition Comparison of Drywall Sample and Control, Prominent Ba and Pb Peaks
Intensity (counts) vs. Wavelength (nm)
Page 50
41
CONCLUSION
From the spectral data generated from the T-shirt, cement, wood, and drywall
samples, it was determined that LIBS could be applied as a presumptive test for the
detection of bullet wipe for certain materials. The data generated for the T-shirt samples
overall indicated that barium is detected heavily in close range and lead is detected more
heavily at a longer range. The spectral data demonstrated that bullet wipe could be
detected at distances ranging from 1” to 12’.
From analysis of the cement sample, it was determined that cement contains
barium as one of its main components. The presence of barium in spectra cannot be used
as a definitive indication of bullet wipe since it is an elemental component of cement. The
spectral data generated from the analysis of cement was the most complex of all the
spectral data produced among all the samples analyzed. Further analysis of the cement
samples could consist of recording other trace elements.
The wood samples produced spectral data that gave an overall indication that
bullet wipe could be detected on wood material at distances ranging from 1’ to 12’.
Although a distinct pattern was not seen for the five samples, the wood samples in
general showed indications of bullet wipe based on its comparison to the wood control,
box cutter control, and Pb-SWC controls. The only spectral data that was not applicable
for the detection of bullet wipe was the barium peak at 553.548 nm. This peak, as
Page 51
42
previously discussed, appeared in the spectral data of the wood samples, wood control,
Pb-SWC controls, and the box-cutter control, eliminating this peak as an integral part of
bullet wipe detection.
All of the drywall samples contained prominent barium lines; however the lead
lines were not prominent in the spectral data. The only lead lines that appeared in the
spectral data were at 363 and 368 nm. Since the box cutter was used for sample collection
and contained all of the prominent barium lines at 455, 493, and 553 nm, it was difficult
to determine whether the drywall or the box-cutter contributed to those peaks. Lead lines
would have to be present in a sample in order to indicate the presence of bullet wipe.
Overall, barium was the most prominent peak of existence in the spectral data of
the drywall samples. Barium lines were detected in all of the drywall samples at 455.403,
493.408, and 553.548 nm. Lead spectral lines were the least consistent in the drywall
sample spectra, however the presence of lead lines in the spectra would have to present
along with barium lines to indicate the presence of bullet wipe since the barium spectral
lines are not only present in the drywall sample spectra, but also in the drywall spectral
data.
After analysis of all the samples, controls, and tools used, it was determined that
the barium line at 553.548 nm was the most common amongst all of their spectral data.
Many of the tools used for collection of the samples contained barium, which made it
difficult to distinguish between bullet wipe and the tools as the contributor of the barium
lines. These factors would have to be further researched by comparing the relative
intensities of the peaks to each other to identify any distinct difference between them or
Page 52
43
determining the probability that the tools used would have a grave effect on the elements
detected in the sample spectra and what percentage of contribution that would be.
Page 53
44
REFERENCE MATERIAL
1. Rusak, D.A.; Castle, B.C.; Smith, B.W.; Winefordner, J.D. Fundamentals and
Applications of Laser-Induced Breakdown Spectroscopy, Analytical Chemistry 1997
27 257-290.
2. Li, Ming; Newman, James; Wang, Hao; Yang, Lingyuan; and Zhang, Li Do bullets
have fingerprints? http://mli.public.iastate.edu/bullets.pdf 200, 2007 last visited Feb.
2, 2010.
3. Wester, R.; Noll R. Heuristic modeling of spectral plasma emission for laser-induced
breakdown spectroscopy, Journal of Applied Physics 2009 106 123302-1 – 123302-
10.
4. Togoni, E. ; Cristoforetti, G.; Legnaioli, S. ; and Palleschi, V. Calibration-Free Laser
Induced Breakdown Spectroscoopy: State of the art Spectrochimica Acta Part B 2010
65 1-14.
5. Abdel-Kareem, O.; Kheder, A.; Abdelhamid, M.; Harith, M.A. Application of Laser
Induced Breakdown Spectroscopy (LIBS) Technique In Investigation of Historical
Metal Threads The 7th International Conference on Laser Applications – ICLA 2009
6. Gondal, M.A.; Yamani, Z.H.; Hussain, T.; and Al-Amoudi, O.S.B. Determination of
Chloride Content in Different Types of Cement Using Laser Induced Breakdown
Spectroscopy Spectroscopy Letters 2009 42 171-177.
Page 54
45
7. Rodriguez-Celis, E.M.; Gornushkin, I.B.; Hettmann, U.M.; Almirall, J.R.; Smith,
B.W.; Winefordner, J.D.; Omenetto, N. Laser Induced Breakdown Spectroscopy as a
tool for discrimination of glass for forensic application. s.l. : Springer-Verlag, 2008
391 1961-1968.
8. Dockery, Christopher R.; Goode , Scott R. Laser-Induced Breakdown Spectroscopy
for the detection of gunshot residues on the hands of a shooter, s.l. : Optical Society
of America, 2003.
9. Romolo, Francesco Saverio; Margot, Pierre Identification of gunshot residue: a critical
review, s.l. : Elsevier, 2001 119 195-211.
10 Harrison, H.C.; Gilroy, R. Firearms discharge residues, Journal of Forensic Science
1959 4 184-199.
11. Nag, N.K.; Sinha, P. A note on assessability of firing distance from gunshot residues,
Forensic Science International 1992 Vol. 56.
12. Martiny, Andrea; Campos, Andrea P.C.; Sader , Marcia S.; Pinto, Maj Andre´ L.
SEM/EDS analysis and characterization of gunshot residues from Brazilian lead-free
ammunition, Forensic Science International 2008 177, e9-e17.