Editors' Highlight pubs.acs.org/ac
Strategies for Potential Age Dating of Fingerprints through the Diffusion of Sebum Molecules on a Nonporous Surface Analyzed Using Time-of-Flight Secondary Ion Mass Spectrometry Shin Muramoto* and Edward Sisco National Institute of Standards and Technology (NIST), Gaithersburg, Maryland 20899, United States S Supporting Information *
ABSTRACT: Age dating of fingerprints could have a significant impact in forensic science, as it has the potential to facilitate the judicial process by assessing the relevance of a fingerprint found at a crime scene. However, no method currently exists that can reliably predict the age of a latent fingerprint. In this manuscript, time-of-flight secondary ion imaging mass spectrometry (TOF-SIMS) was used to measure the diffusivity of saturated fatty acid molecules from a fingerprint on a silicon wafer. It was found that their diffusion from relatively fresh fingerprints (t ≤ 96 h) could be modeled using an error function, with diffusivities (mm2/h) that followed a power function when plotted against molecular weight. The equation x = 0.02t0.5 was obtained for palmitic acid that could be used to find its position in millimeters (where the concentration is 50% of its initial value or c0/2) as a function of time in hours. The results show that on a clean silicon substrate, the age of a fingerprint (t ≤ 96 h) could reliably be obtained through the extent of diffusion of palmitic acid. ass spectrometric (MS) imaging of latent fingerprints is an area of increasing interest due to the enormous amount of chemical information that can be extracted regarding the donor. Fingerprints contain both endogenous compounds such as fatty acids, sterols, squalene, and wax esters that can be used to identify the biology of the donor such as age1,2 and gender3,4 as well as exogenous compounds such as cosmetics, drugs, and explosives that can be used to connect the donor to a crime scene.5,6 Attributing spatial information to such compounds provides new perspectives to latent fingerprint analysis that is relevant to forensics, such as the activities of the donor prior to or during deposition of the fingerprint, which may potentially aid in the reconstruction of a crime scene or may assist in its tactical investigation.7,8 Of particular interest is to estimate the time of deposition of a fingerprint, which could be used to identify fingerprints that are related to the crime, ideally being able to assess the relevance of the evidence in a crime scene. Despite the significant change in chemistry a latent fingerprint may experience over a short period of time,9,10 no forensic method exists that can reliably estimate the age of a fingerprint found at a crime scene. The highly variable chemical composition of a fingerprint from donor to donor precludes the use of particular endogenous biomarkers for determining age, and the complex composition also affects the aging process.11 Recently, it has been found that the extent of oxidation of proteins and lipids can be used to determine the age of a fingerprint,12 just as the age of a blood stain can be monitored through the oxidation products of hemoglobin.13 However, the oxidation process is sensitive to environmental factors such as
M
This article not subject to U.S. Copyright. Published XXXX by the American Chemical Society
temperature and light, which renders the method quite challenging in situations where these factors cannot be controlled. Another idea that has been proposed but has yet to be investigated is the determination of the fingerprint age through the extent of surface diffusion of biomolecules with time.8 To test the feasibility of this method, time-of-flight secondary ion mass spectrometry (TOF-SIMS) is used to observe the diffusion of saturated fatty acid molecules ranging in molecular weight from 143 g/mol to 395 g/mol on the surface. The instrument has the ability to detect and identify multiple chemical species simultaneously, which completely eliminates the need for markers such as tagged-antibodies,5,14−16 and its demonstrated submicrometer spatial resolution of fingerprints6,17 combined with its picogram to even femtogram sensitivity of organic molecules on surfaces18,19 makes it perfectly suited to study the diffusion of sebaceous molecules on complex surfaces.
■
EXPERIMENTAL SECTION 10 cm Si(100) wafers purchased from Virginia Semiconductors (Fredericksburg, VA) (Certain commercial equipment, instruments, or materials are identified in this paper to adequately specify the experimental procedure. Such identification does not imply recommendation or endorsement by the National Received: May 29, 2015 Accepted: July 17, 2015
A
DOI: 10.1021/acs.analchem.5b02018 Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Editors' Highlight
Figure 1. (a) TOF-SIMS ion images of the fingerprint, showing the distribution of the C16H31O2− ion (palmitic acid) on top of a bare silicon wafer at t = (1, 24, 48, 72, and 96) h after deposition. Images are 12 mm × 12 mm. The red rectangle shows the region of interest (150 pixels wide) that was used to obtain a portion of the linescans shown in part b. (b) Linescans showing the intensity of palmitic acid from the edge of the fingerprint as a function of time. Open markers show the experimental values, and the solid lines show the mathematical fits using eq 4.
valley regions and out onto the bare substrate (Figure 1b). As soon as 48 h after initial deposition, the ridges become chemically indistinguishable and more molecules are seen to diffuse out onto the substrate with time. Interestingly, not all molecules in the sebum were found to have the same mobility. Ions such as CNO− and SO4H− that are thought to be fragments of large molecules, generated upon ionization, were not seen to diffuse significantly within the time span of the experiment (Figures S-1, Supporting Information), suggesting a molecular weight dependence on surface diffusion.20 Diffusion refers to the unresolved random thermal motions of a fluid, and is characterized by the spreading of molecules on the surface.21 The behavior of such “random walkers” can be modeled mathematically, with solutions to hundreds of Fick’s diffusion problems tabulated in the monographs of Crank and of Carslaw and Jaeger.22,23 In this study, Fick’s second law is used to model the migration of molecules from the outer edge of the fingerprint, in this case solving for the concentration, c(x,t), at a given position, x, and time, t. The problem is modeled as a one-dimensional diffusion problem rather than two-dimensions (the spreading in this case is radial), because latent fingerprints rarely exist as a circular blot on the surface. Thus, it is simpler to determine the linear extent of diffusion of molecules from the edge of the fingerprint and to apply that to real world samples. The differential continuity equation for the migration of a molecule in one-dimension is
Institute of Standards and Technology nor does it imply that the materials or equipment identified are necessarily the best available for the purpose.) were diced into 15 mm × 15 mm square pieces using the Disco DAD341 dicing saw equipped with a 15 μm diamond impregnated metal blade (Tokyo, Japan). The individual pieces were soaked overnight in an 18 MΩ/cm deionized water obtained from a Milli-Q ultrapure water system (EMD Millipore, Billerica MA) to remove salts and sonicated sequentially for 10 min each in methylene chloride, acetone, and methanol (Sigma-Aldrich, Co., St. Louis, MO) to remove organic contaminants. The substrates were then dried under a stream of dry nitrogen gas. Fingerprints from an anonymous donor (All of the handling and residue experiments involving human subjects were approved and conducted in accordance to our Institutional Review Board.) were deposited onto the substrates through a circular mask (a 7 mm diameter hole made in aluminum foil by an office hole puncher) and kept underneath an aluminum foil with a controlled climate of 21.8 °C and 22% relative humidity. At t = (1, 24, 48, 72, and 96) h after deposition, the samples were imaged using an IONTOF IV (Münster, Germany) instrument equipped with a 25 kV Bi3+ analysis source at an incidence angle of 45°, operated at a current of 0.12 pA. The beam was rastered within a 500 μm × 500 μm area with a pixel density of 128 × 128 pixels, with 5 pulses at each pixel. These scans were stitched together using the IONTOF Surface Explorer software to create 12 mm × 12 mm images. The ion dose density used was 3.27 × 109 ions/cm2, below the static limit of 1 × 1012 ions/cm2, which refers to the ion dose density at which 1% of the surface is sampled. All data points reflect the average of four linescan measurements, obtained from the top and bottom, and left and right sides of the circular fingerprint. The error bars represent their standard deviations.
∂c ∂ ⎛⎜ ∂c ⎞⎟ ∂ 2c = =D 2 D ∂t ∂x ⎝ ∂x ⎠ ∂x
(1)
where c is the concentration of the molecule in normalized counts, and D is its diffusivity in mm2/h. The second order derivative calls for two boundary conditions, one of which is an initial condition at t = 0 h when the sebum is first deposited since the problem at hand is a nonsteady state flow that varies with time. For the boundary condition, the concentration of a given molecule at the edge of the fingerprint is assumed to be a constant since the amount of molecules present inside a ridge is not expected to change significantly during the time frame of the experiment. The initial and boundary conditions are then
■
RESULTS AND DISCUSSION Figure 1a shows the secondary ion images of the fingerprint deposited on silicon, where a sequence of images are used to illustrate the extent of diffusion of the palmitic acid molecules at t = (1, 24, 48, 72, and 96) h after deposition. At t = 1 h, the friction ridge patterns of the fingerprint are clearly visible, with paltimic acid molecules residing along the friction ridges while the valley regions remain unoccupied. After 24 h, the ridge patterns become less obvious as the molecules diffuse into the
c=0 B
at
t=0
(2) DOI: 10.1021/acs.analchem.5b02018 Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Editors' Highlight
Figure 2. (a) Diffusivities of saturated fatty acid molecules plotted as a function of molecular weight. The carbon chain lengths of even-numbered straight chain fatty acid molecules ranged from C8 to C26, with their diffusivities tabulated in Table S-1, Supporting Information. (b) The position at which the concentration of palmitic acid is equal to c0/2, plotted as a function of time, for values that were measured (diamond) and values that are predicted (circle) by eq 4.
c = c0
at
x=0
the migration of the molecules to continue indefinitely without reaching a maximum extent of diffusion; the equation x = 0.02t0.5 shown in Figure 2b does not reach an asymptote. This cannot be correct, since as t approaches infinity, the molecules present in the fingerprint will be exhausted. Moreover, molecules can degrade or become oxidized over time and upon exposure to various environmental factors such as light and ultraviolet rays.9,10 To simulate these cases, the continuity equation was solved where the concentration decreases at x = 0 with increasing t, and also for a case where the concentration decays with time to model molecular breakdown, but neither models were able to appropriately fit the line scans in Figure 2a. This suggested that the initial assumption regarding a constant concentration at x = 0 was correct for the time interval studied (t ≤ 96 h) but also suggested that a different set of assumptions and boundary conditions may be required for large t values (t ≥ 96 h).
(3)
where at t = 0 h, the fingerprint is deposited onto the substrate, from when the diffusion of the molecules start. The solution to the problem is given by ⎡ ⎛ x c(x , t ) = c0⎢1 − erf⎜ ⎝ 4Dt ⎣
⎞⎤ ⎟⎥ ⎠⎦
(4)
Figure 1b shows the line scans of the palmitic acid molecules at t = (1, 24, 48, 72, and 96) h after deposition, normalized to the highest intensity at x = 0. The lines that go through the data points are the mathematical models based on eq 4. For each molecule, the equation is used to solve for the diffusivity that will satisfy the line scans at all t. For palmitic acid, this diffusivity was determined to be (4.5 ± 0.3) × 10−4 mm2/h. When the diffusivities of saturated fatty acid molecules (CnH2n−1O2+) were plotted as a function of molecular weight (Figure 2a), the result was a trend that followed a power function, very similar to what was observed for the diffusivities of polymers on a surface with different molecular weights.24−26 The molecular weight dependence on diffusion suggests that the migration of large molecules can be monitored to determine the age of a fingerprint that has been around for a long time. However, large fatty acid molecules were generally associated with large errors since their intensities were rather low compared to smaller molecules (Figure S-2, Supporting Information), and noise in their line scans greatly affected the curve fitting. Saturated fatty acids such as myristic acid (C14), palmitic acid (C16), and stearic acid (C18) displayed high intensities and thus are considered to be good candidates for determining the diffusion using this technique. As an example, Figure 2b shows the diffusion of palmitic acid with time that was calculated using eq 4, which could be used to determine the age of a fingerprint. The plot predicts the time where 50% of the initial concentration of palmitic acid (c = c0/2) will reach a position x. One limitation of this approach is that the diffusivity of a molecule may differ significantly depending on the type of the substrate, its roughness, and contaminant coverage on the surface. Therefore, diffusivities may need to be evaluated individually for the substrate on which the fingerprint is present. Another limitation of the model is that eq 4 predicts
■
CONCLUSIONS The use of TOF-SIMS imaging for the age dating of fingerprints was demonstrated on a well-controlled model system. The rapid acquisition of large, spatially resolved images with monolayer sensitivities were useful for the visualization of the diffusion of sebaceous molecules originating from the fingerprint. It was found that the diffusion of the molecules from relatively fresh fingerprints (t ≤ 96 h) could be modeled by an error function, with higher molecular weight species showing lower diffusivities that followed a power function. For the type of molecules studied, the diffusivities seemed to be dependent only on molecular weight. The age dating of fingerprints using TOF-SIMS is still a work in progress, but initial indications are promising. Future work will focus on increasing the time interval for diffusion to over 240 h as well as determining the diffusivities of the same molecules on a wide variety of nonporous substrates such as metal and paint.
■
ASSOCIATED CONTENT
S Supporting Information *
Additional TOF-SIMS ion images of the fingerprint and table of tabulated diffusivity values for the saturated, even-numbered straight chain fatty acid molecules. The Supporting Information C
DOI: 10.1021/acs.analchem.5b02018 Anal. Chem. XXXX, XXX, XXX−XXX
Analytical Chemistry
Editors' Highlight
is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b02018.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: 1-301-975-5997. Fax: 1-301-417-1321. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■ ■
ACKNOWLEDGMENTS Research was performed in part at the NIST Center for Nanoscale Science and Technology (CNST). REFERENCES
(1) Antoine, K. M.; Mortazavi, S.; Miller, A. D.; Miller, L. M. J. Forensic Sci. 2010, 55, 513−518. (2) Williams, D. K.; Brown, C. J.; Bruker, J. Forensic Sci. Int. 2011, 206, 161−165. (3) Ferguson, L. S.; Wulfert, F.; Wolstenholme, R.; Fonville, J. M.; Clench, M. R.; Carolan, V. A.; Francese, S. Analyst 2012, 137, 4686− 4692. (4) Asano, K. G.; Bayne, C. K.; Horsman, K. M.; Buchanan, M. V. J. Forensic Sci. 2002, 47, 805−807. (5) Leggett, R.; Lee-Smith, E. E.; Jickells, S. M.; Russell, D. A. Angew. Chem. 2007, 119, 4178−4181. (6) Szynkowska, M. I.; Czerski, K.; Rogowski, J.; Paryjczak, T.; Parczewski, A. Surf. Interface Anal. 2010, 42, 393−397. (7) Weyermann, C.; Ribaux, O. Sci. Justice 2012, 52, 68−75. (8) van Asten, A. C. Sci. Justice 2014, 54, 170−179. (9) Weyermann, C.; Roux, C.; Champod, C. J. Forensic Sci. 2011, 56, 102−108. (10) Archer, N. E.; Charles, Y.; Elliott, J. A.; Jickells, S. Forensic Sci. Int. 2005, 154, 224−239. (11) Girod, A.; Ramotowski, R.; Weyermann, C. Forensic Sci. Int. 2012, 223, 10−24. (12) van Dam, A.; Schwarz, J. C. V.; de Vos, J.; Siebes, M.; Sijen, T.; van Leeuwen, T. G.; Aalders, M. C. G.; Lambrechts, S. A. G. Angew. Chem., Int. Ed. 2014, 53, 6272−6275. (13) Li, B.; Beveridge, P.; O’Hare, W. T.; Islam, M. Sci. Justice 2013, 53, 270−277. (14) McDonnell, L. A.; Heeren, R. M. A. Mass Spectrom. Rev. 2007, 26, 606−643. (15) Hazarika, P.; Jickells, S. M.; Wolff, K.; Russell, D. A. Angew. Chem., Int. Ed. 2008, 47, 10167−10170. (16) Hazarika, P.; Jickells, S. M.; Wolff, K.; Russell, D. A. Anal. Chem. 2010, 82, 9150−9154. (17) Bailey, M. J.; Ismail, M.; Bleay, S.; Bright, N.; Elad, M. L.; Cohen, Y.; Geller, B.; Everson, D.; Costa, C.; Webb, R. P.; Watts, J. F.; de Puit, M. Analyst 2013, 138, 6246−6250. (18) Odom, R. W. Appl. Spectrosc. Rev. 1994, 29, 67−116. (19) Benninghoven, A. Angew. Chem., Int. Ed. Engl. 1994, 33, 1023− 1043. (20) Lodge, T. P. Phys. Rev. Lett. 1999, 83, 3218−3221. (21) Bird, R. B.; Stewart, W. E.; Lightfoot, E. N. Transport Phenomena, rev. 2nd ed.; John Wiley & Sons: New York, 2007. (22) Carslaw, H. S.; Jaeger, J. Conduction of Heat in Solids, 2nd ed.; Oxford University Press: Oxford, U.K., 1959. (23) Crank, J. The Mathematics of Diffusion; Clarendon Press: Oxford, U.K., 1975. (24) Bartels, C. R.; Crist, B.; Graessley, W. W. Macromolecules 1984, 17, 2702−2708. (25) Von Seggern, J.; Klotz, S.; Cantow, H. J. Macromolecules 1991, 24, 3300−3303. (26) Pearson, D. S.; Fetters, L. J.; Graessley, W. W.; Ver Strate, G.; von Meerwall, E. Macromolecules 1994, 27, 711−719.
D
DOI: 10.1021/acs.analchem.5b02018 Anal. Chem. XXXX, XXX, XXX−XXX
