1
`
`Sci, July 2002, Vol. 47, No. 4 J Forensic
`
`
`Paper ID JFS200 I 356_ 474
`
`
`Available online at: www.astm.org
`
`
`
`TECHNICAL NOTE
`
`1
`2
`3
`Keiji G. Asano,
`
`M.S.; Charles K. Bayne,
`
`
`Ph.D.; Katie M. Horsman,M.S.;
`1
`
`and Michelle V. Buchanan,
`Ph.D.
`
`Chemical Composition of Fingerprints for
`
`
`
`Gender Determination*
`
`This work investigates the chemical nature of fingerprints to ascertain whether differences in chemical composition or the existence
`
`
`
`
`
`
`
`
`
`ABSTRACT:
`
`
`
`
`
`
`
`
`
`
`of chemical markers can be used to detennine personal traits, such as age, gender, and personal habits. This type of information could be useful for
`
`
`
`
`
`
`
`
`
`
`
`reducing the pool of potential suspects in criminal investigations when latent fingerprints are unsuitable for comparison by traditional methods. Fin­
`
`
`
`
`
`
`
`
`gertip residue that has been deposited onto a bead was extracted with a solvent such as chloroform. Samples were analyzed by gas chromatogra­
`
`
`
`
`
`
`
`
`
`
`phy/mass spectrometry (GC/MS). The chemical components identified include fatty acids, long chain fatty acid esters, cholesterol and squalene.
`
`
`
`
`
`
`
`
`
`
`The area ratios of ten selected components relative to squalene were calculated for a small preliminary experiment that showed a slight gender dif­
`
`
`
`
`
`
`
`
`
`ference for three of these components. However, when the experiment was repeated with a larger, statistically designed experiment no significant
`
`
`
`
`
`
`
`
`
`
`differences between genders were detected for any of the component ratios. The multivariate Hotelling' s T2 test that tested all ten-component ratios
`
`
`
`
`simultaneously also showed no gender differences at the 5% significance level.
`
`
`
`
`
`
`
`
`
`gas chromatography/mass spectrometry characterization, chemical KEYWORDS: forensic science, fingerprints,
`
`
`
`Fingerprints at crime scenes are crucial in identifying suspects.
`
`
`
`
`glands produce inorganic components (chlorides, metal, phos­
`
`
`
`
`
`
`
`
`
`
`However, if the fingerprints are smudged or are only partial prints,
`
`
`
`
`
`phate) and amino acids. Sebaceous glands produce components
`
`
`
`they may not be suitable for Automated Fingerprint Identification
`
`
`
`
`such as squalene, fatty acids, and alcohols. It is believed that slight
`
`
`System (AFlS) processing. If there were a way to characterize the
`
`
`
`variations in the composition of the sebaceous fatty acid mixtures
`
`
`individual who left the fingerprint, such as age, gender, ethnic
`
`
`
`give individuals a unique scent (2). It is this compositional varia­
`
`
`
`
`background or personal habit, the list of suspects might be reduced.
`
`tion that is believed to provide trained dogs the ability to track an
`
`
`
`
`
`
`Whether such information, specifically gender determination, can
`
`
`
`individual. We hypothesized that if this were true, then there is a
`
`
`
`
`be extracted from fingerprint residue is the basis of this report. It
`
`
`
`
`
`possibility that general differences in chemical residue from fin­
`
`
`
`should be noted that these experiments were performed on unpro­
`
`
`
`gertips could be used to distinguish between males and females and
`
`
`cessed fingerprint residue to determine whether any differences
`various age groups.
`
`
`
`
`could be detected before studying what effects visualization pro­
`Initial experiments in this area were studying the chemical com­
`
`
`
`
`
`cessing might have on these results.
`
`
`
`
`position of children and adults' fingerprints. This study showed
`
`
`
`
`Lipids on the skin surface originate from the sebaceous gland
`
`
`
`
`
`children's fingerprints (pre-pubescent) had less non-volatile com­
`
`
`
`and the epidermis. The palm and fingertips primarily have eccrine
`
`
`ponents, such as long chain fatty acid esters, than adult fingerprints
`glands, although
`
`
`the material found on these areas is usually con­
`
`
`
`
`
`due primarily to inactive sebaceous glands. The chemical compo­
`
`
`
`
`taminated with sebaceous gland secretions due to frequent contact
`
`
`
`
`sition of adults' prints sampled in this study showed some compo­
`
`
`with regions rich in this gland, such as the face ( 1 ). The eccrine
`
`
`
`sitional differences that raised the possibility of characterizing
`
`
`
`
`
`individuals based on their fingerprint residue. A preliminary ex­
`
`
`
`
`periment was conducted to study the sources of variations of the
`
`1 Chemical Sciences Division. Oak Ridge National Laboratory, Oak Ridge,
`
`TN
`
`
`
`
`chemical components in fingertip residue that was sampled on four
`
`2 Computer Science and Mathematics Division, Oak Ridge National Labora­
`
`
`
`
`
`different days, for three individual subjects at two different times
`
`
`tory, Oak Ridge, TN.
`
`
`
`of the day (morning and afternoon). The results from this study
`
`3 Present Address: Department of Chemistry, University of Virginia, Char­
`
`
`
`
`
`
`
`showed little variation due to the different sampling days and repli­
`
`lottesville, VA.
`*This research was sponsored by the Office of Research and Development,
`
`
`
`
`
`
`
`cate measurements. The majority of the variation for the chemicals
`
`
`
`U.S. Department of Energy under contract DE-AC05-00OR22725 with Oak
`
`
`
`measured arose due to the three different subjects. These data
`
`
`
`
`
`Ridge National Laboratory, managed and operated by UT-Battelle, LLC for the
`
`
`
`
`showed that variations in three components, specifically palmitic
`
`
`
`
`
`U.S. Department of Energy. KMH acknowledges support from the Professional
`
`
`
`acid (Cl6 acid), palmitoleic acid (C16:1 acid), and oleic acid
`
`
`
`
`
`Internship Program (PIP) administered jointly by Oak Ridge National Labora­
`
`
`
`(Cl 8: 1 acid), could be gender dependent. This led to new experi­
`
`tory and Oak Ridge Associated Universities.
`
`
`
`Received 12 Oct. 2001; and in revised form 10 Jan. 2002; accepted 10 Jan.
`
`
`
`
`ments with a larger sampling group to determine whether this ob­
`
`2002; published 14 June 2002.
`
`
`servation still would be statistically valid.
`
`Reactive Surfaces Ltd. LLP
`
`Ex. 1029 (Rozzell Attachment J)
`
`
`PA 19428-2959.,<eiw�.oot <W �1W1s'ri.lt1n�t!"rJRlf�fil1:i o8� llJf?J5irav�'?Pal!Zx C700, West Conshohocken,
`rinted'by
`'-oown'l'oacied/p
`
`
`Reactive Surfaces Ltd. LLP v. Toyota Motor
`
`
`to License Agreement. No further
`
`
`
`
`James D Rozzell (Sustainable Chemistry Solutions) pursuant
`
`reproductions authorized.
`Corp.
`IPR2016-01914
`
`
`
`
`
`
`
`

`

`2 JOURNAL OF FORENSIC SCIENCES
`
`Experimental
`
`Subjects
`
`All approvals required by the Department of Energy to perform
`research with human volunteers were acquired before actual work
`proceeded. All necessary steps to protect the volunteers’ privacy
`and confidentiality were taken and the entire sample collection and
`analysis processes were conducted in accordance with the proto-
`cols approved by the Oak Ridge Associated Universities/Oak
`Ridge National Laboratory (ORAU/ORNL) Human Studies Insti-
`tutional Review Board (IRB). For these preliminary experiments,
`volunteers were not asked to provide a fingerprint containing dis—
`tinct ridgelines or any other information that could serve to breach
`the confidentiality of this study. The volunteers were required to
`read and sign an informed consent form before providing a sample.
`
`into the jar and screwed on the Teflon—lined cap. Each sample was
`labeled with a code number to insure privacy.
`
`Sample Analysis
`
`The residue left on the glass bead was extracted with chloroform
`(Burdick and Jackson) and evaporated to dryness. The sample was
`reconstituted with chloroform to a final volume of 150 ML. One mi-
`croliter was injected onto a Thermo Finnigan Voyager GC/MS
`(San Jose, CA). The GC parameters were as follows: 30 111 DE—
`SMS capillary column (J&W Scientific). The initial oven tempera—
`ture of 50°C was held for l min and then ramped at 10°C/min to a
`final temperature of 310°C and held for 20 min. All samples were
`processed within 90 min of collection to prevent any sample loss or
`degradation due to aging of the print.
`
`Sample Collection
`
`Experimental Design
`
`Sample collection consisted of using a glass bead (purchased at
`local hobby store) and jars with Teflon-lined caps that were both
`washed prior to collection. The use of the glass bead was adopted
`from Bernier et al. who demonstrated that material from skin can
`
`be collected by handling glass beads (3). Samples were obtained
`from volunteers in a manner that did not expose the volunteers or
`sampling personnel to chemicals. The volunteer simply rubbed
`his/her fingertips across his/her forehead, removed the glass bead
`from the sampling jar and rubbed the bead between his/her finger-
`tips for about 15 seconds. The volunteer then placed the bead back
`
`This fingerprint study was designed as a nested experiment (4)
`with four sources of uncertainty or variation. The four sources of
`variation are designated as day-to-day variation, periods within a
`day (am and pm), gender within a period, and experimental error
`due to replicate measurements.
`Ten females and ten males volunteered to examine the differ-
`
`ences between female and male fingerprint chemical components
`that were sampled on five different days. Two replicate measure-
`ments on each subject were made (one in the morning and one in
`the afternoon) for a total of eight measurements per day. Two dif—
`
`RT: 500 - 48.00
`
`1 0 I
`
`90
`
`80
`
`7O
`
`60
`
`50
`
`40
`
`3O
`
`20
`
`10
`
`0
`
`
`
`RelativeAbundance
`
`-
`
`e
`
`f
`
`1
`
`b
`
`a\\
`
`L...._____.-_
`
`5
`
`10
`
`15
`
`20
`
`25
`Time (min)
`
`3O
`
`35
`
`40
`
`45
`
`FIG. l—Typical gas chromatogram obtainedfrom chloroform extraction offingertip residue ofi‘glass bead with selected components identzfied——(a) do-
`decanoic acid, (b) tetradecenoic acid, (c) myristic acid (CI4), (d) palmitioleic acid (C16:1), (e) palmitic acid (C16), (f) oleic acid (C18:1), (g) stearic acid
`(C18), (11) squalene, (i) cholesterol, (j) 9-hexadecenoic acid tetradecyl ester, (k) 9-hexadecenoic acid hexadecyl ester, (1) 9-hexadecenoic acid octadecyl
`em ("eggtggtflgycmfii‘i anagrataraafite cttmggaogercidflwm em
`Downloaded/printed by
`James D Rozzell (Sustainable Chemistry Solutions) pursuant to License Agreement. No further reproductions authorized.
`
`

`

`ASANO ET AL. ~ GENDER DETERMINATION 3
`
`1 00
`
`90
`
`80
`
`7O
`
`60
`
`50
`
`4O
`
`30
`
`20
`
`1 0
`
`-10
`
`
`
`%ofSqualene'sArea
`
`- r3- - male
`
`—6— female
`
`
`
`FIG. 2—Compan‘son offemale and male average percentages ofsqualene ’s area plotted with +/- one standard deviation oferror.
`
`Chemical Component
`
`0
`sf
`0
`
`6
`ON
`
`0
`sf
`
`(if
`0
`
`ferent females and two different males were sampled each day for
`the five-day sample period. The subjects were randomly allocated
`over the five days and the measurements were taken in random or—
`der within each morning and afternoon time period.
`
`Results And Discussion
`
`Figure 1 shows a typical chromatogram obtained from a finger-
`print sample. The base peak in most fingerprint samples was squa-
`lene (peak h; t,=27 .00 min). The other components identified
`include unsaturated fatty acids of various carbon length, monoun—
`saturated fatty acids, and cholesterol. long chain fatty acid esters
`were also observed, but usually at lower abundance than the previ-
`ously mentioned components. Using the results from our prelimi-
`nary experiments, ten components were chosen as possible indica-
`tors for gender differentiation. These components were myristic acid
`(C14), pentadecanoic acid (C15), palmitic acid (C16), palmitoleic
`acid (C1621), methyl palmitate (C16me), methyl palrnitoleate
`(C16: lme), methyl stearate (C18me), oleic acid (C18: 1), stearic acid
`(C18), and cholesterol. For each subject, the peak area for these
`components was divided by the squalene peak area, the largest peak
`in each fingertip residue chromatogram. As previously noted, sam—
`ples were processed within 90 min of collection to avoid sample loss
`or degradation due to aging of the prints. Earlier studies (not re—
`ported) showed that peak area ratios were constant for up to six
`hours at room temperature. Figure 2 shows the average value for
`each component—to-squalene ratio for males and females and their
`standard deviations. Although the averages for palmitic acid, palrni—
`toleic acid, and oleic acid, are slightly higher in males than females,
`incorporating standard deviations negates any statistically signifi—
`cant difference.
`
`For each chemical component, the sources of variation were sta-
`Uéssyal taaanqnararmsasaeafiuearreséraasaiear F-
`Downloaded/printed by
`James D Rozzell (Sustainable Chemistry Solutions) pursuant to License Agreement. No further reproductions authorized.
`
`test was used to test whether each variation source made a significant
`contribution to the total uncertainty. No source of variation due to
`day-to-day, period—within—day, and gender—widrin-period gave a 5%
`significant test for any of the chemical measurements. This means
`that all of the variation can be attributed to experimental random
`variation. No statistically significant gender effects were detected.
`An additional statistical test was performed considering all ten com—
`ponents simultaneously using the Hotelling’s T2 test (5). This also
`showed no gender differences at the 5% significance level.
`The chemical composition of fingerprints holds much informa-
`tion that could be useful in criminal investigations. Although this
`work did not find compositional variations that could be used to
`distinguish between genders, it is possible that there are other dis-
`tinguishing components in fingerprints.
`
`References
`
`1. Nicolaides N. Skin lipids: Their biochemical uniqueness. Science 1974',
`18:19—26.
`2. Knowles AM. Aspects of physicochemical methods for the detection of
`latent fingerprints. J Phys E: Sci Instrum 1978;11:713—21.
`3. Bernier UR, Booth MM, Yost RA. Analysis of human skin emanations by
`gas chromatography/mass spectrometry 1. Thermal desorption of attrac-
`tants for the yellow fever mosquito (aedes aegypti) from handled glass
`beads. Anal Chem 1999;71:1—7.
`Searle SR. Linear Models, New York: Wiley & Sons, Inc. 1971;155—63.
`Morrison DF. Multivariate Statistical Methods, New York: McGraw—Hill
`1967;127—9.
`
`E"?
`
`Additional information and reprint requests:
`Keiji Asano, M.S.
`Research Associate
`Oak Ridge National Laboratory
`1 Bethel Valley Road
`Building 5510, MS—6365
`Oak Ridge, TN 37831-6365
`E-mail: asanokg@ornl.gov
`
`