Forensic Science International 253 (2015) 28–32
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Application of imaging ellipsometry to the detection of latent fingermarks Ilsin An * Department of Applied Physics, Hanyang University, Ansan, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Article history: Received 27 February 2015 Received in revised form 1 May 2015 Accepted 7 May 2015 Available online 18 May 2015
Imaging ellipsometry (IE) is applied to visualize latent fingermarks on specular surfaces. Instead of a real image, IE provides images related to the polarization states, which are changed by the imprinted layer on a surface. Fingermarks formed on the surfaces of various materials are investigated, including a shiny metal and a black-colored plastic. Relatively clear IE images are obtained from most surfaces on which the optical properties are distinguishable from those of the fingermarks. Also, it is shown that discernible IE images can be obtained even after a fingermark is vigorously rubbed with lab tissues. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Fingermark Imaging Ellipsometry Non-destructive Polarization
1. Introduction Latent fingermarks may provide critical evidence in criminal investigations. However, the visualization of fingermarks is sometimes challenging when the marks are faint or the surface conditions are not suitable for imaging. The surfaces on which the fingermarks are formed may have various physical constraints for detection. Therefore, a variety of detection methods have been developed, such as powder dusting [1], vacuum vapor deposition [2–4], fuming [5,6], luminescence [7,8], and other techniques [9,10]. In most techniques, various foreign materials are introduced onto the marks in order to enhance the visibility. However, chemical or physical processes involved in the use of such materials can potentially degrade or damage the marks. Furthermore, the use of such materials may prevent further forensic testing of the same sample. The success of deposition techniques depends upon the proper choice of materials, as well as the deposition conditions. Without a priori knowledge of the optimum conditions, the use of this technique can irreversibly degrade the quality of the evidence. Therefore, non-destructive fingermark visualization techniques are preferable if applicable. Optical techniques are possible noninvasive candidates. The laser-induced luminescence technique can be used, but the natural fluorescence signal from a fingermark
* Tel.: +82 314005478; fax: +82 314003817. E-mail address: [email protected] http://dx.doi.org/10.1016/j.forsciint.2015.05.009 0379-0738/ß 2015 Elsevier Ireland Ltd. All rights reserved.
is generally very weak unless prior treatment is performed with strongly fluorescent chemicals or powders [7]. Also, the background noise can be problematic when luminescence is produced by the underlying surface. Recently, specular reflection of polarized light was used to image fingermarks [11]. This optical method utilizes the fact that the reflected light is partially polarized with the plane of polarization perpendicular to the plane of incidence. The difference in polarization-dependent reflection between the fingermark and the background allows the visualization of the image of the latent fingermark. This method works well for dielectric surfaces but is not suitable for metallic surfaces, because the underlying metal reflects more light than the upper fingermark. In the present work, we introduce a novel optical method called imaging ellipsometry (IE). This technique utilizes fully polarized light for the visualization of latent fingermarks. Instead of a real reflection image based on a difference in reflected intensity, IE provides images of the change in polarization state induced by the fingermark layer on the surface. In this paper, many example applications of this technique are demonstrated. 2. Imaging ellipsometry 2.1. Ellipsometry Ellipsometry is a powerful technique for characterizing the optical properties and microstructures of thin films through analysis of changes in the polarization states upon reflection. In
I. An / Forensic Science International 253 (2015) 28–32
general, the values measured by ellipsometry are expressed as D and C. These values are related to the complex reflection coefficients for p- and s-polarized light (rp and rs). Here, p(s)polarization indicates the component of an electric field of light that is parallel (perpendicular) to the plane of incidence. r p ¼ r p eid p ; r s ¼ jr s jeids ;
(1)
where dp(s) is the change in phase of p(s)-polarized light after reflection. The two ellipsometry parameters, D and C, are defined as follows: jr j D ¼ dP dS ; C ¼ arctan P (2) jr S j That is, D measures the phase difference between p- and spolarized light upon reflection. Meanwhile, C measures the magnitude ratio of reflection between these two waves. Physical information about the sample, such as the thickness of the film or the refractive indexes of the materials, can be deduced through optical analysis of the D and C values. When D and C are obtained as a function of wavelength, this represents spectroscopic ellipsometry (SE). Meanwhile, IE is based on single-wavelength ellipsometry. Ellipsometry shows monolayer sensitivity in the measurement of ultrathin films. However, its lateral resolution is limited by its beam spot size (typically a few mm or less). As most deposition techniques produce uniform films over a wide area, the lateral resolution of ellipsometry is not a concern in most thin film studies. 2.2. Imaging ellipsometry IE has been developed and drawn much attention because it shows the merits of both ellipsometry and optical microscopy [12,13]. In IE, a microscope and an image sensor such as a chargecoupled device (CCD) or a complementary metal oxide semiconductor array detector are integrated into an ellipsometer to acquire ellipsometric images over a microscopic area of a surface. In other words, IE provides D and C images rather than an optical image like that obtained with a conventional microscope. Fig. 1 shows the IE system used in this study. Its operating principle is based on rotating compensator ellipsometry [14]. The system consists of: (1) a light-emitting diode, (2) a diffuser, (3) a focusing lens, (4) a band-pass filter (520 nm), (5) a polarizer, (6) a rotating waveplate, (7) a sample stage, (8) an analyzer, (9) zoom optics, and (10) a CCD camera. The zoom lens provides variable primary magnifications of 0.7–4.5. With the current optical elements, only partial images of a fingermark can be observed. If
29
one is interested in viewing the full fingermark, macroimaging optics can be adopted. In this study, the angle of incidence is fixed at 708, and the wavelength is fixed at 520 nm. One can change these parameters for different samples in order to enhance the sensitivity of the system. The intensity images collected at different angular positions of the waveplate are processed in order to acquire D and C images. 2.3. Optical properties of a fingermark A fingermark usually consists of ridges of skin residue composed of inorganic (salt and water) and organic (amino acids, lipids, etc.) materials [15]. The mixture of these substances is rather transparent and can be assumed as a dielectric material in optical analysis. When a finger touches an object, a dielectric-like thin film bearing a fingermark pattern is imprinted on the surface. Thus, IE can detect the image of a thin film as long as the optical properties of the thin film are different from those of the underlying surface. For investigation of its optical properties, a fingermark was imprinted on a silicon wafer. This mark was rubbed to form a uniformly thin film, and SE measurement was performed. Fig. 2 shows SE spectra collected over this sample (circles). Lines are fit to the data assuming dielectric-like optical properties of the fingermark (see inset). The overall fit quality is acceptable considering the poor quality of the fingermark film, which supports the assumption. This also suggests that the image of a fingermark formed on dielectric material such as glass will be less distinct compared to that on semiconductor or metal due to the similar optical properties of the fingermark and the underlying surface.
3. Results and discussion 3.1. Fingermarks on various surfaces For IE studies, fingermarks are imprinted on various surfaces in order to investigate the sensitivity of the technique and the guidelines are followed which are provided by the International Fingerprint Research Group (IFRG) [16]. Four donors are employed and six different substrates are used overall. All marks are natural and most are aged at least a day. First, fingermarks on a slide glass, a silicon wafer, and a chromium-coated surface are studied. These are representatives of the three branches of optical materials, i.e., dielectrics, semiconductors, and conductors. Fig. 3 shows the D and C images of the fingermarks on each surface. Both D and C provide discernible
Fig. 1. Schematic and photograph of the imaging ellipsometer used in this study.
30
I. An / Forensic Science International 253 (2015) 28–32
colors of the surfaces, reasonably fine images are obtained from both surfaces (Fig. 4(a) and (b)). Also fingermark obtained from the backside of a single-side polished silicon wafer is shown in Fig. 4(c). The backside of this wafer is not specular and its surface looks like that of finely-ground glass. Despite the diffused reflection from this substrate, IE is able to detect the image by increasing the mount of light entering the detector. 3.2. Effect of wiping on the detection of fingermarks
Fig. 2. SE spectrum of a fingermark film on a silicon surface (circles). Lines are fit to the data. The refractive index used in the fitting calculation is shown in the inset, and the extinction coefficient of the fingermark is assumed to be zero.
images of the fingermarks from all surfaces. The images on a slide glass (Fig. 3(c)) are somewhat faint due to the dielectric-like optical properties of the fingermark. In other words, the variations in D and C values in the images are small compared to those of other surfaces. In the cases of the silicon wafer and the chromium sample, strong specular reflection is expected from the underlying surface. However, the images of fingermarks obtained from neither surface are affected in this way (Fig. 3(a) and (b)). This is another merit of IE, which relies only on the changes in polarization state rather than the intensity of reflected light. The nonuniform lines of the ridges indicate either variation in thickness or nonuniformity in optical properties. Additionally, fingermarks on a dark plastic surface and a black cell phone screen are measured with IE. The fingermark on the plastic is imprinted with moderate pressure, and the one on the cell phone is formed by the soft touch of a finger. Despite the dark
As another demonstration of the performance of IE, measurements are conducted after wiping the fingermark on the silicon wafer using lab tissues (Kimwipes1, Kimberly-Clark Co. U.S.A.). Lab tissues are folded in half several times to produce a thick square brush of sorts, and the surface of the fingermark is brushed with moderate pressure. Fig. 5 shows images taken before and after multiple brushings. Although some damage appears on the right side of the image after 100 brushes, the patterns of the ridges are still identifiable, as shown in Fig. 5(c). Moreover, Fig. 5(d) shows that the fingermark survives 300 brushes followed by a vigorous rubbing action. This indicates that some elements in the fingermark sufficiently adhere to the wafer surface to endure vigorous mechanical rubbing. Although the thickness of the ridges is expected to be reduced greatly by rubbing, the D images are still visible thanks to the phase sensitivity of IE. For comparison, images obtained by diffused reflection and reflection with coaxial illumination are presented in Fig. 6 along with D images. Each technique yields almost comparable visibility for untouched fingermark (upper ones). Meanwhile, the images of rubbed fingermark are somewhat different from each other (lower ones). In diffused reflection, the image is deteriorated by the debris from rubbing process. And in the image of reflection with coaxial illumination, contrast reduction is more vivid. From the ellipsometric analysis of rubbed fingermark, the height differences between ridges and valleys can be estimated as 1–3 nm. This corresponds to 0.05–0.15% difference in reflection between ridges and valleys, which explains the low contrast in the image of reflection with coaxial illumination.
Fig. 3. Ellipsometric images of fingermarks on different surfaces (upper: D, lower: C). (a) Silicon wafer, (b) chromium, and (c) slide glass. Scales along the sides of each image display the pixel numbers of the CCD sensor. Graphs on the top and right sides of each image show the D and C values of cross-sections in the horizontal and vertical directions, respectively. The dark straight lines in the images of the slide glass are defects in the glass. All marks are aged a week.
I. An / Forensic Science International 253 (2015) 28–32
31
Fig. 4. Ellipsometric images (upper: D, lower: C) of fingermarks on (a) dark-colored plastic, (b) black-colored cell phone screen, and (c) backside of a single-side polished silicon wafer. Each mark is from different donor.
Fig. 5. Ellipsometric images (upper: D, lower: C) of a fingermark on a silicon wafer taken (a) as-imprinted, (b) after 10 brushes, (c) after 100 brushes, and (d) after 300 brushes and vigorous rubbing.
Fig. 6. Images (upper: before rubbing, lower: after rubbing) of fingermarks obtained by (a) imaging ellipsometry, (b) diffused reflection, and (c) reflection with coaxial illumination. Lower images are taken a week after rubbing.
32
I. An / Forensic Science International 253 (2015) 28–32
In order to obtain similar results as shown in Fig. 5 in destructive techniques, hundreds of fingermarks are required. In IE, however, no multiple samples are required, as it is a nondestructive technique. Although these results are preliminary, they show that IE can be used to study the survivability of fingermarks in various conditions [17–20]. 4. Conclusions It is demonstrated that IE can be used to visualize latent fingermarks on various specular surfaces. Discernible ellipsometric images are obtained from not only a shiny metal surface, but also an absorbing black-colored object. Also, it is shown that IE works well even when the ridges of fingermark become very thin due to rubbing. These are possible because IE detects the changes in polarization states rather than the intensity of reflected light. However, its applications are limited to small objects which have smooth surfaces. Since it does not interfere with other methods, it might be worth trying this method if any metal or glass piece brought from crime scenes is expected to have very thin fingermarks on it due to environmental conditions or aging. It is also demonstrated that IE can collect sequential images during rubbing process. This preliminary result shows that IE can be useful for real-time studies. Fingermarks in crime scene will be damaged and fade over time and this process will be expedited by weather and environmental conditions. Thus, IE can be applied to aging and survivability studies of fingermarks in various conditions. As IE is a non-destructive technique, a single fingermark can be measured repeatedly as a function of time, humidity, or temperature. Further work with IE is currently being undertaken to understand the degradation process of fingermark in water. In this case, the images of fingermark will be taken in real-time with the substrate positioned in water. Acknowledgments This work was supported by the Future Semiconductor Device Technology Development Program (10049185) funded by the MOTIE (Ministry of Trade, Industry & Energy) and KSRC (Korea Semiconductor Research Consortium).
References [1] G.S. Sodhi, J. Kaur, Powder method for detecting latent fingerprints: a review, Forensic Sci. Int. 120 (2001) 172–176. [2] R.C. Shaler, A. Lakhtakia, J.W. Rogers, D.P. Pulsifer, R.J. Martin-Palma, Columnarthin-film acquisition of fingerprint topology, J. Nanophoton. 5 (2011) 51509– 51511. [3] J. Dutta, S.A. Ramakrishna, I.M. Alaoui, Fingerprint visualization enhancement by deposition of columnar thin films and fluorescent dye treatment, Forensic Sci. Int. 228 (2013) 32–37. [4] G.W. Batey, J. Copeland, D.L. Donnelly, C.L. Hill, P.L. Laturnus, C.H. McDiarmid, K.J. Miller, A.H. Misner, A. Tario, A.B. Yamashita, Metal deposition for latent print development, J. Forensic Identif. 48 (1998) 165–175. [5] J.E. Watkin, D. Wilkinson, A.H. Misner, A.B. Yamashita, Cyanoacrylate fuming of latent prints: vacuum versus heat/humidity, J. Forensic Identif. 44 (1994) 545–557. [6] D.T. Burns, J.K. Brown, A. Dinsmore, K.K. Harvey, Base-activated latent fingerprints fumed with a cyanoacrylate monomer. A quantitative study using Fourier-transform infra-red spectroscopy, Anal. Chim. Acta 362 (1998) 171–176. [7] E.R. Menzel, Detection of latent fingerprints by laser-excited luminescence, Anal. Chem. 61 (1989) 557A–561A. [8] E.R. Menzel, M. Takatsu, R.H. Murdock, K. Bouldin, K.H. Cheng, Photoluminescent CdS/dendrimer nanocomposites for fingerprint detection, J. Forensic Sci. 45 (2000) 770–773. [9] G. Williams, H.N. McMurray, D.A. Worsley, Latent fingerprint detection using a scanning Kelvin microprobe, J. Forensic Sci. 46 (2001) 1085–1092. [10] C.G. Worley, S.S. Wiltshire, T.C. Miller, G.J. Havrilla, V. Majidi, Detection of visible and latent fingerprints using micro-X-ray fluorescence elemental imaging, J. Forensic Sci. 51 (2005) 57–63. [11] S. Lin, K.M. Yemelyanov, E.N. Pugh Jr., N. Engheta, Polarization-based and specular-reflection-based noncontact latent fingerprint imaging and lifting, J. Opt. Soc. Am. A 23 (2006) 2137–2153. [12] Z.H. Wang, G. Jin, A label-free multisensing immunosensor based on imaging ellipsometry, Anal. Chem. 75 (2003) 6119–6123. [13] A.J. Choi, T.H. Ghong, Y.D. Kim, J.H. Oh, J. Jang, Imaging ellipsometry study on the Ni-mediated crystallization of a-Si, J. Appl. Phys. 100 (2006) 113529–113533. [14] J. Lee, P.I. Rovira, I. An, R.W. Collins, Rotating-compensator multichannel ellipsometry: applications for real time Stokes vector spectroscopy of thin film growth, Rev. Sci. Instrum. 69 (1998) 1800–1810. [15] C. Weyermann, C. Roux, C. Champod, Initial results on the composition of fingerprints and its evolution as a function of time by GC/MS analysis, J. Forensic Sci. 56 (2011) 102–108. [16] International Fingerprint Research Group (IFRG), Guidelines for the assessment of fingermark detection techniques, J. Forensic Identif. 64 (2014) 174–200. [17] Y. Cohen, E. Rozen, M. Azoury, D. Attias, B. Gavrielli, M. Elad, Survivability of latent fingerprints Part I: Adhesion of latent fingerprints to smooth surfaces, J. Forensic Identif. 62 (2012) 47–53. [18] Y. Cohen, M. Azoury, M. Elad, Survivability of latent fingerprints Part II: The effect of cleaning agents on the survivability of latent fingerprints, J. Forensic Identif. 62 (2012) 54–61. [19] R. Sutton, C. Grenci, L. Hrubesova, A comparison on the longevity of submerged marks in field and laboratory conditions, J. Forensic Identif. 64 (2014) 143–156. [20] K. Wertheim, Fingerprint age determination: is there any hope? J. Forensic Identif. 53 (2003) 42–49.
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
Application of imaging ellipsometry to the detection of latent fingermarks Ilsin An * Department of Applied Physics, Hanyang University, Ansan, Republic of Korea
A R T I C L E I N F O
A B S T R A C T
Article history: Received 27 February 2015 Received in revised form 1 May 2015 Accepted 7 May 2015 Available online 18 May 2015
Imaging ellipsometry (IE) is applied to visualize latent fingermarks on specular surfaces. Instead of a real image, IE provides images related to the polarization states, which are changed by the imprinted layer on a surface. Fingermarks formed on the surfaces of various materials are investigated, including a shiny metal and a black-colored plastic. Relatively clear IE images are obtained from most surfaces on which the optical properties are distinguishable from those of the fingermarks. Also, it is shown that discernible IE images can be obtained even after a fingermark is vigorously rubbed with lab tissues. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Fingermark Imaging Ellipsometry Non-destructive Polarization
1. Introduction Latent fingermarks may provide critical evidence in criminal investigations. However, the visualization of fingermarks is sometimes challenging when the marks are faint or the surface conditions are not suitable for imaging. The surfaces on which the fingermarks are formed may have various physical constraints for detection. Therefore, a variety of detection methods have been developed, such as powder dusting [1], vacuum vapor deposition [2–4], fuming [5,6], luminescence [7,8], and other techniques [9,10]. In most techniques, various foreign materials are introduced onto the marks in order to enhance the visibility. However, chemical or physical processes involved in the use of such materials can potentially degrade or damage the marks. Furthermore, the use of such materials may prevent further forensic testing of the same sample. The success of deposition techniques depends upon the proper choice of materials, as well as the deposition conditions. Without a priori knowledge of the optimum conditions, the use of this technique can irreversibly degrade the quality of the evidence. Therefore, non-destructive fingermark visualization techniques are preferable if applicable. Optical techniques are possible noninvasive candidates. The laser-induced luminescence technique can be used, but the natural fluorescence signal from a fingermark
* Tel.: +82 314005478; fax: +82 314003817. E-mail address: [email protected] http://dx.doi.org/10.1016/j.forsciint.2015.05.009 0379-0738/ß 2015 Elsevier Ireland Ltd. All rights reserved.
is generally very weak unless prior treatment is performed with strongly fluorescent chemicals or powders [7]. Also, the background noise can be problematic when luminescence is produced by the underlying surface. Recently, specular reflection of polarized light was used to image fingermarks [11]. This optical method utilizes the fact that the reflected light is partially polarized with the plane of polarization perpendicular to the plane of incidence. The difference in polarization-dependent reflection between the fingermark and the background allows the visualization of the image of the latent fingermark. This method works well for dielectric surfaces but is not suitable for metallic surfaces, because the underlying metal reflects more light than the upper fingermark. In the present work, we introduce a novel optical method called imaging ellipsometry (IE). This technique utilizes fully polarized light for the visualization of latent fingermarks. Instead of a real reflection image based on a difference in reflected intensity, IE provides images of the change in polarization state induced by the fingermark layer on the surface. In this paper, many example applications of this technique are demonstrated. 2. Imaging ellipsometry 2.1. Ellipsometry Ellipsometry is a powerful technique for characterizing the optical properties and microstructures of thin films through analysis of changes in the polarization states upon reflection. In
I. An / Forensic Science International 253 (2015) 28–32
general, the values measured by ellipsometry are expressed as D and C. These values are related to the complex reflection coefficients for p- and s-polarized light (rp and rs). Here, p(s)polarization indicates the component of an electric field of light that is parallel (perpendicular) to the plane of incidence. r p ¼ r p eid p ; r s ¼ jr s jeids ;
(1)
where dp(s) is the change in phase of p(s)-polarized light after reflection. The two ellipsometry parameters, D and C, are defined as follows: jr j D ¼ dP dS ; C ¼ arctan P (2) jr S j That is, D measures the phase difference between p- and spolarized light upon reflection. Meanwhile, C measures the magnitude ratio of reflection between these two waves. Physical information about the sample, such as the thickness of the film or the refractive indexes of the materials, can be deduced through optical analysis of the D and C values. When D and C are obtained as a function of wavelength, this represents spectroscopic ellipsometry (SE). Meanwhile, IE is based on single-wavelength ellipsometry. Ellipsometry shows monolayer sensitivity in the measurement of ultrathin films. However, its lateral resolution is limited by its beam spot size (typically a few mm or less). As most deposition techniques produce uniform films over a wide area, the lateral resolution of ellipsometry is not a concern in most thin film studies. 2.2. Imaging ellipsometry IE has been developed and drawn much attention because it shows the merits of both ellipsometry and optical microscopy [12,13]. In IE, a microscope and an image sensor such as a chargecoupled device (CCD) or a complementary metal oxide semiconductor array detector are integrated into an ellipsometer to acquire ellipsometric images over a microscopic area of a surface. In other words, IE provides D and C images rather than an optical image like that obtained with a conventional microscope. Fig. 1 shows the IE system used in this study. Its operating principle is based on rotating compensator ellipsometry [14]. The system consists of: (1) a light-emitting diode, (2) a diffuser, (3) a focusing lens, (4) a band-pass filter (520 nm), (5) a polarizer, (6) a rotating waveplate, (7) a sample stage, (8) an analyzer, (9) zoom optics, and (10) a CCD camera. The zoom lens provides variable primary magnifications of 0.7–4.5. With the current optical elements, only partial images of a fingermark can be observed. If
29
one is interested in viewing the full fingermark, macroimaging optics can be adopted. In this study, the angle of incidence is fixed at 708, and the wavelength is fixed at 520 nm. One can change these parameters for different samples in order to enhance the sensitivity of the system. The intensity images collected at different angular positions of the waveplate are processed in order to acquire D and C images. 2.3. Optical properties of a fingermark A fingermark usually consists of ridges of skin residue composed of inorganic (salt and water) and organic (amino acids, lipids, etc.) materials [15]. The mixture of these substances is rather transparent and can be assumed as a dielectric material in optical analysis. When a finger touches an object, a dielectric-like thin film bearing a fingermark pattern is imprinted on the surface. Thus, IE can detect the image of a thin film as long as the optical properties of the thin film are different from those of the underlying surface. For investigation of its optical properties, a fingermark was imprinted on a silicon wafer. This mark was rubbed to form a uniformly thin film, and SE measurement was performed. Fig. 2 shows SE spectra collected over this sample (circles). Lines are fit to the data assuming dielectric-like optical properties of the fingermark (see inset). The overall fit quality is acceptable considering the poor quality of the fingermark film, which supports the assumption. This also suggests that the image of a fingermark formed on dielectric material such as glass will be less distinct compared to that on semiconductor or metal due to the similar optical properties of the fingermark and the underlying surface.
3. Results and discussion 3.1. Fingermarks on various surfaces For IE studies, fingermarks are imprinted on various surfaces in order to investigate the sensitivity of the technique and the guidelines are followed which are provided by the International Fingerprint Research Group (IFRG) [16]. Four donors are employed and six different substrates are used overall. All marks are natural and most are aged at least a day. First, fingermarks on a slide glass, a silicon wafer, and a chromium-coated surface are studied. These are representatives of the three branches of optical materials, i.e., dielectrics, semiconductors, and conductors. Fig. 3 shows the D and C images of the fingermarks on each surface. Both D and C provide discernible
Fig. 1. Schematic and photograph of the imaging ellipsometer used in this study.
30
I. An / Forensic Science International 253 (2015) 28–32
colors of the surfaces, reasonably fine images are obtained from both surfaces (Fig. 4(a) and (b)). Also fingermark obtained from the backside of a single-side polished silicon wafer is shown in Fig. 4(c). The backside of this wafer is not specular and its surface looks like that of finely-ground glass. Despite the diffused reflection from this substrate, IE is able to detect the image by increasing the mount of light entering the detector. 3.2. Effect of wiping on the detection of fingermarks
Fig. 2. SE spectrum of a fingermark film on a silicon surface (circles). Lines are fit to the data. The refractive index used in the fitting calculation is shown in the inset, and the extinction coefficient of the fingermark is assumed to be zero.
images of the fingermarks from all surfaces. The images on a slide glass (Fig. 3(c)) are somewhat faint due to the dielectric-like optical properties of the fingermark. In other words, the variations in D and C values in the images are small compared to those of other surfaces. In the cases of the silicon wafer and the chromium sample, strong specular reflection is expected from the underlying surface. However, the images of fingermarks obtained from neither surface are affected in this way (Fig. 3(a) and (b)). This is another merit of IE, which relies only on the changes in polarization state rather than the intensity of reflected light. The nonuniform lines of the ridges indicate either variation in thickness or nonuniformity in optical properties. Additionally, fingermarks on a dark plastic surface and a black cell phone screen are measured with IE. The fingermark on the plastic is imprinted with moderate pressure, and the one on the cell phone is formed by the soft touch of a finger. Despite the dark
As another demonstration of the performance of IE, measurements are conducted after wiping the fingermark on the silicon wafer using lab tissues (Kimwipes1, Kimberly-Clark Co. U.S.A.). Lab tissues are folded in half several times to produce a thick square brush of sorts, and the surface of the fingermark is brushed with moderate pressure. Fig. 5 shows images taken before and after multiple brushings. Although some damage appears on the right side of the image after 100 brushes, the patterns of the ridges are still identifiable, as shown in Fig. 5(c). Moreover, Fig. 5(d) shows that the fingermark survives 300 brushes followed by a vigorous rubbing action. This indicates that some elements in the fingermark sufficiently adhere to the wafer surface to endure vigorous mechanical rubbing. Although the thickness of the ridges is expected to be reduced greatly by rubbing, the D images are still visible thanks to the phase sensitivity of IE. For comparison, images obtained by diffused reflection and reflection with coaxial illumination are presented in Fig. 6 along with D images. Each technique yields almost comparable visibility for untouched fingermark (upper ones). Meanwhile, the images of rubbed fingermark are somewhat different from each other (lower ones). In diffused reflection, the image is deteriorated by the debris from rubbing process. And in the image of reflection with coaxial illumination, contrast reduction is more vivid. From the ellipsometric analysis of rubbed fingermark, the height differences between ridges and valleys can be estimated as 1–3 nm. This corresponds to 0.05–0.15% difference in reflection between ridges and valleys, which explains the low contrast in the image of reflection with coaxial illumination.
Fig. 3. Ellipsometric images of fingermarks on different surfaces (upper: D, lower: C). (a) Silicon wafer, (b) chromium, and (c) slide glass. Scales along the sides of each image display the pixel numbers of the CCD sensor. Graphs on the top and right sides of each image show the D and C values of cross-sections in the horizontal and vertical directions, respectively. The dark straight lines in the images of the slide glass are defects in the glass. All marks are aged a week.
I. An / Forensic Science International 253 (2015) 28–32
31
Fig. 4. Ellipsometric images (upper: D, lower: C) of fingermarks on (a) dark-colored plastic, (b) black-colored cell phone screen, and (c) backside of a single-side polished silicon wafer. Each mark is from different donor.
Fig. 5. Ellipsometric images (upper: D, lower: C) of a fingermark on a silicon wafer taken (a) as-imprinted, (b) after 10 brushes, (c) after 100 brushes, and (d) after 300 brushes and vigorous rubbing.
Fig. 6. Images (upper: before rubbing, lower: after rubbing) of fingermarks obtained by (a) imaging ellipsometry, (b) diffused reflection, and (c) reflection with coaxial illumination. Lower images are taken a week after rubbing.
32
I. An / Forensic Science International 253 (2015) 28–32
In order to obtain similar results as shown in Fig. 5 in destructive techniques, hundreds of fingermarks are required. In IE, however, no multiple samples are required, as it is a nondestructive technique. Although these results are preliminary, they show that IE can be used to study the survivability of fingermarks in various conditions [17–20]. 4. Conclusions It is demonstrated that IE can be used to visualize latent fingermarks on various specular surfaces. Discernible ellipsometric images are obtained from not only a shiny metal surface, but also an absorbing black-colored object. Also, it is shown that IE works well even when the ridges of fingermark become very thin due to rubbing. These are possible because IE detects the changes in polarization states rather than the intensity of reflected light. However, its applications are limited to small objects which have smooth surfaces. Since it does not interfere with other methods, it might be worth trying this method if any metal or glass piece brought from crime scenes is expected to have very thin fingermarks on it due to environmental conditions or aging. It is also demonstrated that IE can collect sequential images during rubbing process. This preliminary result shows that IE can be useful for real-time studies. Fingermarks in crime scene will be damaged and fade over time and this process will be expedited by weather and environmental conditions. Thus, IE can be applied to aging and survivability studies of fingermarks in various conditions. As IE is a non-destructive technique, a single fingermark can be measured repeatedly as a function of time, humidity, or temperature. Further work with IE is currently being undertaken to understand the degradation process of fingermark in water. In this case, the images of fingermark will be taken in real-time with the substrate positioned in water. Acknowledgments This work was supported by the Future Semiconductor Device Technology Development Program (10049185) funded by the MOTIE (Ministry of Trade, Industry & Energy) and KSRC (Korea Semiconductor Research Consortium).
References [1] G.S. Sodhi, J. Kaur, Powder method for detecting latent fingerprints: a review, Forensic Sci. Int. 120 (2001) 172–176. [2] R.C. Shaler, A. Lakhtakia, J.W. Rogers, D.P. Pulsifer, R.J. Martin-Palma, Columnarthin-film acquisition of fingerprint topology, J. Nanophoton. 5 (2011) 51509– 51511. [3] J. Dutta, S.A. Ramakrishna, I.M. Alaoui, Fingerprint visualization enhancement by deposition of columnar thin films and fluorescent dye treatment, Forensic Sci. Int. 228 (2013) 32–37. [4] G.W. Batey, J. Copeland, D.L. Donnelly, C.L. Hill, P.L. Laturnus, C.H. McDiarmid, K.J. Miller, A.H. Misner, A. Tario, A.B. Yamashita, Metal deposition for latent print development, J. Forensic Identif. 48 (1998) 165–175. [5] J.E. Watkin, D. Wilkinson, A.H. Misner, A.B. Yamashita, Cyanoacrylate fuming of latent prints: vacuum versus heat/humidity, J. Forensic Identif. 44 (1994) 545–557. [6] D.T. Burns, J.K. Brown, A. Dinsmore, K.K. Harvey, Base-activated latent fingerprints fumed with a cyanoacrylate monomer. A quantitative study using Fourier-transform infra-red spectroscopy, Anal. Chim. Acta 362 (1998) 171–176. [7] E.R. Menzel, Detection of latent fingerprints by laser-excited luminescence, Anal. Chem. 61 (1989) 557A–561A. [8] E.R. Menzel, M. Takatsu, R.H. Murdock, K. Bouldin, K.H. Cheng, Photoluminescent CdS/dendrimer nanocomposites for fingerprint detection, J. Forensic Sci. 45 (2000) 770–773. [9] G. Williams, H.N. McMurray, D.A. Worsley, Latent fingerprint detection using a scanning Kelvin microprobe, J. Forensic Sci. 46 (2001) 1085–1092. [10] C.G. Worley, S.S. Wiltshire, T.C. Miller, G.J. Havrilla, V. Majidi, Detection of visible and latent fingerprints using micro-X-ray fluorescence elemental imaging, J. Forensic Sci. 51 (2005) 57–63. [11] S. Lin, K.M. Yemelyanov, E.N. Pugh Jr., N. Engheta, Polarization-based and specular-reflection-based noncontact latent fingerprint imaging and lifting, J. Opt. Soc. Am. A 23 (2006) 2137–2153. [12] Z.H. Wang, G. Jin, A label-free multisensing immunosensor based on imaging ellipsometry, Anal. Chem. 75 (2003) 6119–6123. [13] A.J. Choi, T.H. Ghong, Y.D. Kim, J.H. Oh, J. Jang, Imaging ellipsometry study on the Ni-mediated crystallization of a-Si, J. Appl. Phys. 100 (2006) 113529–113533. [14] J. Lee, P.I. Rovira, I. An, R.W. Collins, Rotating-compensator multichannel ellipsometry: applications for real time Stokes vector spectroscopy of thin film growth, Rev. Sci. Instrum. 69 (1998) 1800–1810. [15] C. Weyermann, C. Roux, C. Champod, Initial results on the composition of fingerprints and its evolution as a function of time by GC/MS analysis, J. Forensic Sci. 56 (2011) 102–108. [16] International Fingerprint Research Group (IFRG), Guidelines for the assessment of fingermark detection techniques, J. Forensic Identif. 64 (2014) 174–200. [17] Y. Cohen, E. Rozen, M. Azoury, D. Attias, B. Gavrielli, M. Elad, Survivability of latent fingerprints Part I: Adhesion of latent fingerprints to smooth surfaces, J. Forensic Identif. 62 (2012) 47–53. [18] Y. Cohen, M. Azoury, M. Elad, Survivability of latent fingerprints Part II: The effect of cleaning agents on the survivability of latent fingerprints, J. Forensic Identif. 62 (2012) 54–61. [19] R. Sutton, C. Grenci, L. Hrubesova, A comparison on the longevity of submerged marks in field and laboratory conditions, J. Forensic Identif. 64 (2014) 143–156. [20] K. Wertheim, Fingerprint age determination: is there any hope? J. Forensic Identif. 53 (2003) 42–49.
