Forensic Science International 233 (2013) 257–264

Contents lists available at ScienceDirect

Forensic Science International journal homepage: www.elsevier.com/locate/forsciint

Determining the effects of routine fingermark detection techniques on the subsequent recovery and analysis of explosive residues on various substrates Sam King a, Sarah Benson b, Tamsin Kelly a, Chris Lennard a,* a b

National Centre for Forensic Studies, University of Canberra, Canberra, ACT 2601, Australia Forensics, Australian Federal Police, GPO Box 401, Canberra, ACT 2601, Australia

A R T I C L E I N F O

A B S T R A C T

Article history: Received 10 July 2013 Received in revised form 13 September 2013 Accepted 17 September 2013 Available online 25 September 2013

An offender who has recently handled bulk explosives would be expected to deposit latent fingermarks that are contaminated with explosive residues. However, fingermark detection techniques need to be applied in order for these fingermarks to be detected and recorded. Little information is available in terms of how routine fingermark detection methods impact on the subsequent recovery and analysis of any explosive residues that may be present. If an identifiable fingermark is obtained and that fingermark is found to be contaminated with a particular explosive then that may be crucial evidence in a criminal investigation (including acts of terrorism involving improvised explosive devices). The principal aims of this project were to investigate: (i) the typical quantities of explosive material deposited in fingermarks by someone who has recently handled bulk explosives; and (ii) the effects of routine fingermark detection methods on the subsequent recovery and analysis of explosive residues in such fingermarks. Four common substrates were studied: paper, glass, plastic (polyethylene plastic bags), and metal (aluminium foil). The target explosive compounds were 2,4,6-trinitrotoluene (TNT), pentaerythritol tetranitrate (PETN), and hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), as well as chlorate and nitrate ions. Recommendations are provided in terms of the application of fingermark detection methods on surfaces that may contain explosive residues. ß 2013 Elsevier Ireland Ltd. All rights reserved.

Keywords: Explosive residues Fingermark detection Fingerprints Organic explosives Chlorate Nitrate

1. Introduction The use of explosives for criminal purposes, including terrorism, is of ongoing concern as it can lead to scenarios involving significant loss of life and injury to persons and property. These events usually involve the use of improvised explosive devices (IEDs), which can be based on simple ‘‘homemade’’ explosives made from materials readily available to the public or more sophisticated devices assembled with high-grade military or commercial explosives. High explosives such as 2,4,6-trinitrotoluene (TNT), hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) and pentaerythritol tetranitrate (PETN) are heavily regulated and generally difficult to obtain, but can still fall into the wrong hands through misappropriation. However, improvised explosives, including propellants and various explosive mixtures, are of significant concern due to how easily the component materials can be obtained. As a result, they have been used with increasing frequency in many high profile incidents such as the Unabomber

* Corresponding author. Tel.: +61 2 6201 2160; fax: +61 2 6201 2461. E-mail address: [email protected] (C. Lennard). 0379-0738/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.forsciint.2013.09.018

(USA, 1978–1995), the World Trade Centre bombing (USA, 1993), the Oklahoma City bombing (USA, 1995), attacks on public transport systems (Madrid, 2004, and London, 2005), several terrorist attacks in Indonesia (2002–2005), and the more recent Boston Marathon bombings (2013). Counter-terrorism initiatives rely heavily on the detection and identification of explosive material found at scenes related to incidents under investigation. The analysis of explosive substances can involve: (i) detection and identification of explosives in ‘‘pre-blast’’ scenarios (i.e., unexploded bulk material, location where a device was constructed, etc.); and (ii) the identification of explosive residues in ‘‘post-blast’’ environments (i.e., after detonation). The location where an explosive device may have been constructed could hold many other forms of trace evidence to assist an examiner in determining possible suspects for such incidents. Latent fingermarks are an example of such evidence. Not only could these fingermarks link an individual to a particular scene, they may also contain exogenous material such as explosive residues that is relevant to the investigation. If the explosive residues found in these fingermarks are identified as being the same as those associated with a planned or actual bombing, then this may become crucial evidence. However, detecting these latent

258

S. King et al. / Forensic Science International 233 (2013) 257–264

fingermarks is a task in itself. As the name suggests, latent fingermarks are not visible to the naked eye and their detection and recording generally requires the application of physical and/or chemical treatments [1]. So, before any potential explosive material present in the fingermarks can be collected and analysed, the fingermarks themselves must first be located. The difficulty is that the application of a fingermark detection technique to the evidence risks the removal of any explosive residues that may be present. This must be considered before determining how to proceed. It was the purpose of this research to provide information that could assist with this decision-making process. Various studies have been published on the direct analysis of exogenous materials (including explosives) in fingermarks. Examples include the following:  The in situ detection and identification of trace explosives in fingermarks by Raman spectroscopy [2].  The detection and identification of drugs of abuse and adulterants, present as contaminants in both latent fingermarks and fingermarks developed by cyanoacrylate fuming, using Raman spectroscopy [3,4].  The simultaneous chemical analysis and imaging of fingermarks using Raman chemical imaging [5–7].  The use of mass spectrometric methods, including surfaceassisted laser desorption/ionisation time of flight mass spectrometry (SALDI-TOF-MS) and direct analysis in real time mass spectrometry (DART-MS), to detect common explosives in latent fingermarks and in fingermarks developed using black powder [8,9].  The detection and identification of explosive residues in fingermarks using attenuated total reflection Fourier transform infrared (ATR-FTIR) microspectroscopy [10].  The use of infrared spectroscopic imaging techniques and multivariate analysis for fingermarks that reflect a specific chemical history, such as exposure to explosives [11].  The detection of microscopic particles, including high explosives, present as contaminants in latent fingermarks by means of synchrotron radiation-based Fourier transform infrared (SRFTIR) micro-imaging [12].  The analysis of explosive residues in fingermarks using optical catapulting laser-induced breakdown spectroscopy (OC-LIBS) [13].

However, these studies have generally ignored either the way latent fingermarks are routinely detected and enhanced by fingerprint examiners or how a forensic chemist typically processes items submitted for explosive residue analysis. These two examinations are generally undertaken separately and it is largely unknown what effects the fingerprint processing may have on the subsequent isolation, detection and identification of any explosive residues that may be present. The primary objective of this research was to determine the effects that common fingermark detection techniques could have on the detection of any explosive residues that may be present in latent fingermarks. The completion of this task was achieved through two separate phases. The first phase was determining the likely mass of explosive residues that would be present in latent fingermarks deposited by an individual who had recently handled a bulk amount of explosive material. This was intended to simulate a scenario in which an individual responsible for assembling an explosive device handled bulk explosive material and subsequently touched other surfaces, depositing fingermarks contaminated with explosive residues. Once specific values were determined via handling experiments, these were used in the second phase to determine realistic starting amounts of each target compound that

would be subjected to selected fingermark detection techniques. In the second phase, quantitative methods were applied to determine any loss of target material (from the chosen starting amounts) that resulted from the action of each fingermark detection method on each test substrate. The test substrates consisted of one porous surface (paper) and three non-porous surfaces (glass, clear polyethylene, and aluminium foil). Routine fingermark detection methods were applied individually and in sequence, as indicated in Fig. 1. 2,4,6-Trinitrotoluene (TNT) was chosen as a representative nitroaromatic, pentaerythritol tetranitrate (PETN) as a representative nitrate ester, and hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) as a representative nitramine. Potassium chlorate and ammonium nitrate were selected as representative inorganic compounds relevant to explosive residue analysis. These compounds were chosen as they are commonly encountered in either military, commercial or improvised explosives. They also possess low volatility (low vapour pressure) and would therefore result in residues remaining for significant periods in contaminated fingermarks. In addition, analytical methods are readily available for the quantitation of these compounds. Organic peroxides (such as triacetone triperoxide; TATP) were not considered as target compounds as their high volatility would lead to a rapid loss when deposited in a fingermark. 2. Materials and methods 2.1. Analytical methods The analytical methods used for the quantitation of each target compound were adapted from previous research work reported by Song-im et al. [14]. The target organic explosives (TNT, PETN and RDX) were quantified by highperformance liquid chromatography (HPLC) using o-nitrotoluene as an internal standard. The target inorganic species (nitrate and chlorate ions) were quantified by capillary electrophoresis (CE). Sulfate and thiocyanate ions were chosen as internal standards for the chlorate and nitrate ion analyses, respectively. The organic analyses were performed on an Agilent 1120 high performance liquid chromatography system comprising a quaternary pump, vacuum degasser, standard autosampler, thermostatted column compartment and a diode array detector. Instrumental control, data acquisition and analysis were accomplished using EZChrom EliteTM Chromatography Data System software (Version 3.3.2). The chlorate and nitrate ion analyses were performed on an Agilent 7100 capillary electrophoresis system equipped with a diode array detector. Instrumental control, data acquisition and analysis were accomplished using 3D-CE ChemStation software (Version B.04.02). Using standard solutions across a range of concentration for each of the target compounds, analytical regression curves were prepared on each day that sample analyses were performed. The quantitative analyses were performed over the following concentration ranges: 0.04–4 ppm for TNT, PETN and RDX; 1–29 ppm for chlorate; and 1–20 ppm for nitrate. 2.2. Chemicals High-purity water was obtained from a Satorius arium 611 water purification system. HPLC grade methanol was obtained from Chem-Supply (Gillman, Australia)

Porous Substrates

Indanedione-Zinc

Ninhydrin

Non-Porous Substrates

Powder

Cyanoacrylate

Rhodamine 6G

Physical Developer Fig. 1. Fingermark detection sequences for porous and non-porous substrates applied in this study. These techniques were applied individually and in sequence to determine their effects on the subsequent recovery and analysis of explosive residues.

S. King et al. / Forensic Science International 233 (2013) 257–264

259

and LiChroSolv1 gradient grade acetonitrile was supplied by Merck Pty. Ltd. (Kilsyth, Australia). Both methanol and water were filtered under vacuum through a 47 mm nylon filter membrane with a pore size 0.45 mm (supplied by Grace Davison Discovery Sciences, Rowville, Australia) prior to use. Sodium chlorate (99.8% purity), ammonium nitrate (99.5% purity, ACS reagent grade), anhydrous sodium sulfate (99.7% purity, ACS reagent grade), sodium thiocyanate (99% purity, reagent grade), chromium(VI) oxide (99.9% purity), sodium chromate tetrahydrate (99% purity) and Trizma1 base (99.9% purity) were purchased from Sigma–Aldrich Pty. Ltd. (Castle Hill, Australia). Hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), 2,4,6-trinitrotoluene (TNT), and pentaerythritol tetranitrate (PETN), at a certified concentration of 1000 mg/mL in acetonitrile, and o-nitrotoluene (99.5% certified purity) were obtained from ChemService, Inc. (West Chester, USA). The internal standard o-nitrotoluene was prepared as a 1000 mg/mL stock solution in acetonitrile and then a pre-determined volume was added to each working solution, to give a final concentration of 2 ppm, for the quantitative analysis of the three organic explosives (PETN, RDX and TNT) by HPLC. Sodium chlorate, ammonium nitrate, anhydrous sodium sulfate and sodium thiocyanate were used to prepare single-anion stock solutions at a concentration of 1000 mg/L. A 0.1 N sodium hydroxide solution was prepared for preconditioning the capillary by diluting 1.0 N NaOH solution obtained from Agilent Technologies Australia (Forest Hill, Australia). Hexadimethrine bromide (96% purity) was used to prepare a 1% aqueous solution to coat the internal wall of the capillary. The background electrolyte (BGE) was composed of 10 mM chromium(VI) oxide (CrO3), 10 mM sodium chromate tetrahydrate (Na2CrO4) and 40 mM Trizma1 base (TRIS). For the chlorate analyses, a pre-determined volume of the sulfate stock solution (internal standard) was added to each working solution to give a final concentration of 25 ppm. For the nitrate analyses, a pre-determined volume of the thiocyanate stock solution (internal standard) was added to each working solution to give a final concentration of 15 ppm. For the Phase 1 handling experiments, the bulk explosives/compounds listed in Table 1 were employed. For the Phase 2 experiments, methanolic stock solutions of sodium chlorate (2000 mg/mL) and ammonium nitrate (3000 mg/mL) were prepared to generate deposits of the representative inorganic compounds on the test fingermarks. Deposits for the organic explosives (RDX, TNT, and PETN) were generated from the commercially-sourced certified 1000 mg/mL standard solutions indicated above.

‘‘low’’ concentration sets). Sets of ‘‘blank’’ fingermarks were also prepared (with no explosives contamination) on both the glass and paper substrates. For the glass slides, one fingermark was deposited per slide so that there was a total of 5 slides/samples per set. For the paper, the A4 sheets were bisected and a set of 5 contaminated fingermarks deposited per half sheet with the impressions outlined with pencil. The glass slides were subsequently stored in plastic slide mailers and the paper samples stored in A5 paper envelopes prior to laboratory analysis. For the analysis of the organic target compounds, the samples were prepared as follows (based on previously published methods [15]):

2.3. Substrates

The starting amounts indicated in Table 2 were used in the spiked fingermarks (with the target compounds deposited as individual analytes in separate fingermarks). These amounts were achieved using aliquots of the standard solutions described previously (i.e., 1000 mg/mL in acetonitrile for TNT, PETN and RDX; 2000 mg/mL in methanol for sodium chlorate; 3000 mg/mL in methanol for ammonium nitrate). These starting amounts, based on the Phase 1 results and chosen as being pragmatic from an analytical perspective (to ensure that quantitation was still possible even if relatively large losses were encountered), are also consistent with those used in a previous study assessing target compound stability in solution, on alcohol wipes and on a glass surface [15]. For each of the four substrates, each fingermark treatment (as described below) and each target compound, three latent fingermarks (from the thumb, index and middle fingers) were deposited from one male donor. An aliquot of standard solution was added to each fingermark, as specified in Table 2. The solution was allowed to dry on the fingermark; the fingermark was then stored in the dark under normal laboratory conditions prior to treatment. After the application of the relevant fingermark treatments, the target compounds were quantified in the same manner as for the Phase 1 experiments (with the plastic and aluminium substrates processed as for the glass substrate).

The selected substrates were as follows: glass microscope slides (representative glass surface; Livingstone International, Pty. Ltd., Rosebery, Australia); white A4 printer paper (representative paper surface; Australian Paper, Mount Waverley, Australia); clear polyethylene film (representative plastic surface; taken from snapseal plastic bags supplied by Dowlings Canberra Pty. Ltd., Fyshwick, Australia); and heavy duty catering aluminium foil (representative metallic surface; Confoil Pty. Ltd., Bayswater, Australia). These substrates were used as received (with no pre-cleaning). Blank analyses on substrate washings confirmed that there were no detectable levels of any of the target compounds. 2.4. Phase 1 experiments For each of the bulk explosives/compounds listed in Table 1, the exposed material (after the removal of any packaging) was held in the left hand of a volunteer for 20 s. A set of fingermarks was then immediately deposited on the test substrate and these constituted the ‘‘high concentration’’ impressions. A set of ‘‘low concentration’’ impressions was collected by repeating the whole process with the exception that all loose material adhering to the skin was removed by brushing the hands together (as you would if you were removing dust from your hands). ‘‘High’’ and ‘‘low’’ concentration samples were collected for each of two substrates: paper and glass. The hands of the volunteer were thoroughly washed with soap and hot water before and after each handling experiment (including in between the ‘‘high’’ and

 Glass  Each sample (single fingermark on glass slide) was washed once into a 2 mL volumetric flask using a continuous stream of HPLC grade methanol until the correct final volume was achieved.  Aliquots of this solution were taken for quantitative analysis.  Paper  Each sample (single fingermark) was cut from the paper sheet and then further cut into small pieces.  These pieces were placed in a glass vial (sealed with a screw cap and Parafilm) and the sample extracted with 2 mL of HPLC grade methanol using sonication for 10 min.  After cooling to room temperature, the extract was removed with a pipette and transferred to a 2 mL volumetric flask and the volume made up with additional methanol.  Aliquots of this solution were taken for quantitative analysis.

For the analysis of the inorganic target compounds, the same process was applied for each sample type (glass and paper) but high-purity water was employed as the wash/extraction solvent instead of methanol. In each case, the volume of the aliquot taken for analysis (to which the internal standard was then added) was adjusted as required to ensure that the final concentration of target compound fell within the relevant quantitation range. Each analysis was repeated 3 times and an average value calculated. 2.5. Phase 2 experiments

2.6. Fingermark treatments The effects of the fingermark treatments listed in Table 3, on the recovery and analysis of the target compounds, were assessed. For these fingermark treatments, the reagent formulations and development procedures currently in use by the

Table 1 Bulk explosives/substances used for the Phase 1 handling experiments. (Note that the concentrations indicated for each major component are as provided by the manufacturer or are from literature sources. These values are relevant to the explosive material only – after the removal of any packaging or casing – and were not independently verified.) Material

Major component(s)

Source

Description

TNT flakes Plastic explosive No. 4 (PE4) Detasheet Detonation cord Anzomex G cast booster BST cast booster Ammonium nitrate prill Potassium chlorate

TNT (>99%, w/w) RDX (90%, w/w) PETN (60%, w/w) PETN (>99%, w/w) TNT (30–60%, w/v) + PETN (30–60% w/v) TNT (40–60%, w/v) + RDX (40–60%, w/v) NH4NO3 (>98%, w/w) KClO3 (>99%, w/w)

AFP supplies ADI Ltd. (Australia) ADI Ltd. (Australia) AFP supplies ICI Australia Ltd. Beston Australia Pty. Ltd. AFP supplies May & Baker Ltd.

Pale yellow flakes White, greasy plasticised solid Greasy, olive green plasticised sheet Fine white powder in plastic casing Mixed booster – solid beige cylinder in cardboard casing Mixed booster – solid beige cylinder in cardboard casing Small, white beads Fine, white powder

260

S. King et al. / Forensic Science International 233 (2013) 257–264

Table 2 Starting amounts for each target compound used to spike fingermarks in the Phase 2 experiments. Target compound

Starting quantity in spiked fingermarks

Aliquot required to achieve this amount

TNT PETN RDX Sodium chlorate Ammonium nitrate

15 mg 15 mg 15 mg 30 mg 60 mg

15 mL 15 mL 15 mL 15 mL 20 mL

Table 3 Fingermark treatments (single methods and sequences) applied to the spiked fingermarks in the Phase 2 experiments. Substrate

Fingermark treatment

Porous (paper)

No treatment (control) Indanedione-zinc (IND-Zn) Ninhydrin (NIN) Physical developer (PD) Sequence IND-Zn + NIN Sequence IND-Zn + NIN + PD

Non-porous (glass, plastic, metal)

No treatment (control) Black magnetic fingerprint powder Cyanoacrylate fuming (CA) Sequence CA + rhodamine 6G stain (R6G)

Australian Federal Police were employed [16]. These procedures are summarised in Table 4.

3. Results 3.1. Phase 1 None of the target compounds were detected in the extracts from the ‘‘blank’’ fingermarks (no explosives contamination). In addition, there were no interfering signals observed in the chromatograms/electropherograms, originating from either the substrates or endogenous components of the fingermarks, that may have compromised the quantitative analyses. Target compounds were detected and quantified for the majority of the contaminated fingermarks collected during the handling experiments. One notable exception was with Detasheet, where no PETN was detected in any of the fingermarks collected after handling this material. For the paper substrate, no PETN was detected in the fingermark samples following the handling of the Anzomex booster. In addition, no chlorate ion was detected in the ‘‘low’’ concentration chlorate samples on both

(1000 mg/mL (1000 mg/mL (1000 mg/mL (2000 mg/mL (3000 mg/mL

in in in in in

acetonitrile) acetonitrile) acetonitrile) methanol) methanol)

glass and paper, and no nitrate ion was detected in the ‘‘low’’ concentration nitrate samples on glass. For the remaining fingermark sets where the target compounds were detected, approximate mass ranges and average amounts per fingermark were determined and are summarised in Table 5. Some representative examples of results obtained from the Phase 1 handling experiments are displayed in Fig. 2. The quantities of target compounds detected in the fingermarks were highly variable across the experiments and across the individual fingermarks for each experiment. For each set of five fingermarks, there was no obvious correlation between quantity of deposited target compound and the particular finger of the hand. As expected, the ‘‘high’’ concentration sets gave, on average, higher quantities of each target compound than the corresponding ‘‘low’’ concentration sets, indicating that the brushing of the hands to remove loose material was effective in reducing the level of contamination on the fingers. Based on these results and in the interests of working with realistic quantities of each target compound (that could be readily quantified even after some loss), the Phase 2 experiments involved the spiking of latent fingermarks with the target compound amounts specified in Table 2. 3.2. Phase 2 The average percent recoveries for each target compound after each fingermark treatment are summarised in Fig. 3, for the three non-porous surfaces, and in Fig. 4, for the paper substrate. 3.2.1. Black magnetic fingerprint powder For the fingermarks on glass, the application of powder had no significant effect on the recovery of the target compounds (Fig. 3). On plastic, some loss was observed for RDX and nitrate. For the aluminium foil, small losses were observed for PETN and RDX; however, these can be considered insignificant given the relatively large standard deviations for these measurements.

Table 4 Summary of the general procedure employed for the application of each fingermark detection method applied in the Phase 2 experiments. Reagents were prepared as per current Australian Federal Police procedures [16]. Fingermark treatment

General procedure

Indanedione-zinc (IND-Zn)

Samples were dipped briefly in the working solution, allowed to dry, and then heated in a dry heat press at 160 8C (5 8C) for 10 s between sheets of clean absorbent paper. Samples were dipped briefly in the working solution and then allowed to dry. Items were treated as follows (in clean glass trays): 1. Water wash (15 min), repeated 3 times. 2. Immersion in maleic acid solution (5 min). 3. Water wash (5 min). 4. Immersion in PD working solution (15 min). 5. Water wash (15 min), repeated 3 times. 6. Allowed to dry. Black magnetic fingerprint powder was applied with a magnetic wand and the wand used to remove excess powder. Samples were fumed for 30 min in a commercial fuming chamber (Forensic Cyanoacrylate Cabinet; Carter-Scott Design, Australia) using 20 drops of Loctite1 401 cyanoacrylate adhesive. The working solution was applied with a pipette to each cyanoacrylate fumed fingermark. Just enough solution was added to cover the fingermark area. This area was then rinsed with a minimal amount of water, to remove excess stain, and then allowed to dry.

Ninhydrin (NIN) Physical developer (PD)

Black magnetic fingerprint powder Cyanoacrylate fuming (CA) Rhodamine 6G (R6G) stain

S. King et al. / Forensic Science International 233 (2013) 257–264

261

Table 5 Approximate ranges and average amounts detected for each target compound, per fingermark, for the contaminated fingermarks analysed in the Phase 1 experiments. Paper

Glass

High Low High Low High Low High Low High Low

TNT PETN RDX Chlorate Nitrate

Range (mg)

Average (mg)

Range (mg)

Average (mg)

6–67 0–16 3–33 2–18 3–217 1–47 13–28 Not detected 0–76 Not detected

24 4 15 5 58 16 21 Not detected 43 Not detected

1–6 0.8–2 4–13 2–5 1–6 0–2 192–209 Not detected 125–186 54–129

3.1 1.4 7 3.6 3.0 0.9 202 Not detected 167 94

PETN (detonation cord) – Paper

35

16

30

14

Calculated Mass (µg)

Calculated Mass (µg)

PETN (detonation cord) – Glass

25 20 15 10 5

12 10 8 6 4 2

0

0

Thumb

Index

Middle

Ring

Little

Thumb

Middle

Ring

Little

RDX (PE4) – Paper

240

6

200

5

Calculated Mass (µg)

Calculated Mass (µg)

RDX (PE4) – Glass

Index

160 120 80 40 0

4 3 2 1 0

Thumb

Index

Middle

Ring

Little

Thumb

TNT (BST booster) – Glass

Index

Middle

Ring

Little

TNT (BST booster) – Paper 7

25

Calculated Mass (µg)

Calculated Mass (µg)

6 20 15

10 5 0

5 4 3 2 1 0

Thumb

Index

Middle

Ring

Little

Thumb

Index

Middle

Ring

Little

Fig. 2. Examples of the results obtained from the Phase 1 handling experiments. Each column represents the average calculated mass of target compound (in mg) for that fingermark, from three repeat analyses, with the error bar representing one standard deviation. (The black columns are for the ‘‘high’’ concentration samples and the grey columns are for the ‘‘low’’ concentration samples.).

S. King et al. / Forensic Science International 233 (2013) 257–264

262

Glass 100

Average Percent Recovery (%)

90 80 70 60

Control

50

Powder

40

CA

30

CA + R6G

20 10 0

TNT

PETN

RDX

Chlorate

Nitrate

Plastic 100

Average Percent Recovery (%)

90 80 70 60

Control

50

Powder

40

CA

30

CA + R6G

20 10 0

TNT

PETN

RDX

Chlorate

Nitrate

Aluminium Foil 100

Average Percent Recovery (%)

90 80 70 60

Control

50

Powder

40

CA

30

CA + R6G

20 10 0

TNT

PETN

RDX

Chlorate

Nitrate

Fig. 3. Effects of black magnetic fingerprint powder, cyanoacrylate fuming (CA) and CA followed by rhodamine 6G staining (CA + R6G) on the subsequent recovery of each target compound from the three non-porous surfaces tested. Average percent recoveries are indicated for control samples (no treatment; black columns) and for samples after the application of the specified treatment. The error bars represent one standard deviation for nine measurements (three spiked fingermarks and three repeat analyses per fingermark extract).

3.2.2. Cyanoacrylate fuming (CA) On glass, the CA fuming appeared to have no significant effect on the recovery except for nitrate, where a reduced percent recovery was observed (Fig. 3). On plastic, reduced recoveries were observed for both RDX and nitrate. On the aluminium foil, a small loss was observed for TNT and more significant losses for PETN and nitrate. 3.2.3. Sequence CA + rhodamine 6G staining The rhodamine 6G formulation used in this research is water based and treatment with the stain solution is followed by a brief water rinse. Not surprisingly, the inorganic target compounds

(chlorate and nitrate) did not survive this treatment. For the organic target compounds, losses were observed for PETN on glass, RDX on plastic, and both TNT and PETN on aluminium foil (Fig. 3). 3.2.4. Ninhydrin (NIN) Ninhydrin treatment resulted in observed reductions in percent recoveries for TNT, RDX and nitrate (Fig. 4). There was a marginal reduction in PETN recovery and no effect on chlorate. 3.2.5. Indanedione-zinc (IND-Zn) With the IND-Zn treatment, significant losses were observed for all of the target compounds except chlorate (Fig. 4).

S. King et al. / Forensic Science International 233 (2013) 257–264

263

Paper 100

Average Percent Recovery (%)

90 80 70

Control

60

NIN

50

IND-Zn

40

IND-Zn + NIN 30

PD 20 10 0

TNT

PETN

RDX

Chlorate

Nitrate

Fig. 4. Effects of indanedione-zinc (IND-Zn), ninhydrin (NIN), IND-Zn followed by ninhydrin (IND-Zn + NIN), and physical developer (PD) on the subsequent recovery of each target compound from the paper substrate tested. Average percent recoveries are indicated for control samples (no fingermark treatment; black columns) and for samples after the application of the specified treatment. The error bars represent one standard deviation for nine measurements (three spiked fingermarks and three repeat analyses per extract).

3.2.6. Sequence IND-Zn + NIN The results obtained from the combined IND-Zn + NIN sequence were nearly identical to those obtained for IND-Zn alone (Fig. 4). This suggests that the observed losses were largely due to the effects of the IND-Zn treatment and that the residue that survived the IND-Zn treatment tended to also survive the subsequent ninhydrin treatment. 3.2.7. Physical developer (PD) The PD technique, as indicated in Table 4, involves a significant number of water washes in addition to treatments with an aqueous solution of maleic acid and the aqueous PD reagent itself. It was assumed from the outset that the two inorganic salts (chlorate and nitrate) would not survive this treatment so this was not tested. For the organic target compounds, only PETN could be detected after the PD treatment as indicated in Fig. 4. 3.2.8. Sequence IND-Zn + NIN + PD After the application of this sequence on the paper samples, none of the target compounds could be detected. This is understandable given the cumulative effect of the losses observed for each technique in isolation, particularly the significant losses attributed to the IND-Zn and PD treatments.

4. Discussion The analytical methods chosen for the detection and quantitation of the target compounds performed as expected, with good reproducibility for the repeat analyses. No interfering signals were observed, from extracts taken from treated or untreated fingermarks, that may have otherwise compromised the quantitative measurements. For the non-porous substrates tested, the application of a fingerprint powder had, in general, minimal effect on the subsequent recovery of the target compounds. Where some loss was observed, this may have been due to physical removal of some of the target compound when the excess fingerprint powder was removed from the surface. Black magnetic powder was employed in this study. While not tested, it would be expected that a greater loss due to physical removal would result if a conventional fingerprint brush was employed (due to the abrasive effect of the bristles). To minimise the loss of potential explosive residues, it is recommended that magnetic powder is used and that the powder is applied sparingly (i.e., light powdering only).

Cyanoacrylate fuming had minimal effect on the percent recoveries from the glass substrate, but some losses were observed from plastic and aluminium foil. It is likely that these losses were due to the target compound being trapped in the deposited cyanoacrylate polymer. This would reduce the percent recovery achieved using the methanol or water washes as these treatments will not dissolve the polymer. While not tested in this research, losses (such as those observed with RDX and nitrate ion on plastic) might be reduced via the use of a solvent, such as acetone or acetonitrile, that would dissolve the CA polymer and capture any trapped material. The inorganic target compounds (chlorate and nitrate) did not survive the water-based rhodamine 6G treatment applied in this study. To minimise such losses, the use of a non-aqueous stain solution and the avoidance of a final water rinse are recommended. For example, a non-aqueous rhodamine 6G stain that does not require a subsequent water wash was reported by Margot and Lennard [17]. For the paper substrate, some losses were observed as a result of ninhydrin treatment. Such losses may be the result of physical remove of the target compounds due to the action of dipping the paper in the ninhydrin solution (resulting in the washing off of loose particles and/or via dissolution in the solvent mixture that makes up the ninhydrin formulation). While not tested in this research, it may be possible to reduce these losses by using a fine spray to apply a minimal quantity of reagent. Care would need to be taken to only apply enough solution to wet the surface. With the indanedione-zinc treatment, percent recoveries were significantly lower for nitrate and all three organic explosives tested compared to the control samples. While some of these losses may have been due to the action of dipping the paper samples in the solution (as for ninhydrin), losses may also be due to transfer onto the absorbent paper when the samples were placed in the heat press and loss due to the effect of the heat applied to develop the fingermarks (160 8C for 10 s). Further research would be required to determine which of these factors contributed to the greatest loss of target compound. To minimise such losses, it is recommended to apply the reagent as a fine spray and to either conduct the development at room-temperature over several days or use an oven rather than a heat press (e.g., 100 8C for 30 min) to avoid physical contact with the surfaces that may carry residues of interest. The PD technique as applied in this study involves a significant number of water washes and treatments with an

264

S. King et al. / Forensic Science International 233 (2013) 257–264

aqueous solution of maleic acid and the aqueous PD reagent itself. It was assumed from the outset that the two inorganic salts (chlorate and nitrate) would not survive this treatment so this was not specifically tested. For the organic target compounds, only PETN could be detected, at less than 40% recovery, after PD application. This persistence of PETN is in agreement with the observations made by Kamyshny and colleagues who found that the detection of PETN on high-density polyethylene, linoleum, glass and aluminium remained possible even after a month of soaking in seawater [18]. They reported that traces of PETN were consistently more persistent in water than the other organic explosives that were tested. TNT and RDX were not detected after the PD treatment. Given this observed impact and the loss of inorganic residues, the use of PD should be avoided if it is intended that any explosive residues that may be present will be recovered and analysed. With respect to the inorganic target compounds, our results are consistent in general terms with those reported in a recent article by Love and co-workers who looked at the detection, using ion chromatography, of a range of anionic species in fingermarks developed on porous and non-porous surfaces [19]. They found that powdering did not significantly reduce the recoveries of these species from a glass substrate. It was also determined that trace anions could be detected and quantified in ninhydrin-developed fingermarks on filter paper, albeit with some variance in percent recoveries. In instances where both fingerprint and explosive residue evidence is sought, the challenge is to modify the application of the fingermark detection methods to minimise the loss of potential explosive residues while not compromising their ability to detect latent fingermarks. This is clearly an area that would benefit from further research.

5. Conclusions The results suggest that, with most of the fingermark detection methods tested, it will still be possible to recover and analyse any explosives residues that may have been originally present in the fingermarks when they were deposited. However, each detection method applied in a sequence is likely to reduce the amount of residue remaining. To minimise such losses, it is recommended that only a limited number of techniques be applied in the detection sequence, that certain methods are avoided (e.g., physical developer on porous surfaces and water-based cyanoacrylate stains on non-porous surfaces), and that methods are applied in a certain way (e.g., application of amino acid reagents in the form of a fine spray rather than by immersion). The adoption of such recommendations will provide the greatest chance of success with respect to the detection and identification of explosive residues in fingermarks from individuals who have recently handled bulk explosives.

Acknowledgements The authors would like to acknowledge financial support for this project from the Combating Terrorism Technical Support Office, USA, and the National Security Science & Technology Centre, Defence Science and Technology Organisation, Australia. Dr Nopporn Song-im provided advice and assistance in the early stages of the project and this is gratefully acknowledged. Special thanks to Dr Tony Cantu for the original discussions that inspired the development, funding and completion of this work. References [1] C. Champod, C. Lennard, P. Margot, M. Stoilovic, Fingerprints and Other Skin Ridge Impressions, CRC Press LLC, Boca Raton, 2004. [2] C. Cheng, T.E. Kirkbride, D.N. Batchelder, R.J. Lacey, T.G. Sheldon, In situ detection and identification of trace explosives by Raman microscopy, J. Forensic Sci. 40 (1995) 31–37. [3] J.S. Day, H.G.M. Edwards, S.A. Dobrowski, A.M. Voice, The detection of drugs of abuse in fingerprints using Raman spectroscopy I: latent fingerprints, Spectrochim. Acta A 60 (2004) 563–568. [4] J.S. Day, H.G.M. Edwards, S.A. Dobrowski, A.M. Voice, The detection of drugs of abuse in fingerprints using Raman spectroscopy II: cyanoacrylate-fumed fingerprints, Spectrochim. Acta A 60 (2004) 1725–1730. [5] E.D. Emmons, A. Tripathi, J.A. Guicheteau, S.D. Christesen, A.W. Fountain III, Raman chemical imaging of explosive-contaminated fingerprints, Appl. Spectrosc. 63 (2009) 1197–1203. [6] A. Tripathi, E.D. Emmons, P.G. Wilcox, J.A. Guicheteau, D.K. Emge, S.D. Christesen, A.W. Fountain III, Semi-automated detection of trace explosives in fingerprints on strongly interfering surfaces with Raman chemical imaging, Appl. Spectrosc. 65 (2011) 611–619. [7] J.A. Guicheteau, H. Swofford, A. Tripathi, P.G. Wilcox, E.D. Emmons, S.D. Christesen, J. Wood, A.W. Fountain III, Sequential Raman chemical imaging and biometric analysis on fingerprints for rapid identification of threat materials and individuals, J. Forensic Ident. 63 (2013) 90–101. [8] D.R. Ifa, N.E. Manicke, A.L. Dill, R.G. Cooks, Latent fingerprint chemical imaging by mass spectrometry, Science 321 (2008) 805. [9] F. Rowell, J. Seviour, A.Y. Lim, C.G. Elumbaring-Salazar, J. Loke, J. Ma, Detection of nitro-organic and peroxide explosives in latent fingermarks by DART- and SALDITOF-mass spectrometry, Forensic Sci. Int. 221 (2012) 84–91. [10] Y. Mou, J.W. Rabalais, Detection and identification of explosive particles in fingerprints using attenuated total reflection-Fourier transform infrared spectromicroscopy, J. Forensic Sci. 54 (2009) 846–850. [11] T. Chen, Z.D. Schultz, I.W. Levin, Infrared spectroscopic imaging of latent fingerprints and associated forensic evidence, Analyst 134 (2009) 1902–1904. [12] A. Banas, K. Banas, M.B.H. Breese, J. Loke, B. Heng Teo, S.K. Lim, Detection of microscopic particles present as contaminants in latent fingerprints by means of synchrotron radiation-based Fourier transform infra-red micro-imaging, Analyst 137 (2012) 3459–3465. [13] M. Abdelhamid, F.J. Fortes, M.A. Harith, J.J. Laserna, Analysis of explosive residues in human fingerprints using optical catapulting-laser-induced breakdown spectroscopy, J. Anal. At. Spectrom. 26 (2011) 1445–1450. [14] N. Song-im, S. Benson, C. Lennard, Evaluation of different sampling media for their potential use as a combined swab for the collection of both organic and inorganic explosive residues, Forensic Sci. Int. 222 (2012) 102–110. [15] N. Song-im, S. Benson, C. Lennard, Stability of explosive residues in methanol/ water extracts, on alcohol wipes and on a glass surface, Forensic Sci. Int. 226 (2013) 244–253. [16] M. Stoilovic, C. Lennard, Fingermark Detection & Enhancement, 6th ed., National Centre for Forensic Studies, Canberra, 2012. [17] P. Margot, C. Lennard, Fingerprint Detection Techniques, 6th ed., University of Lausanne, Lausanne, 1994. [18] A. Kamyshny, S. Magdassi, Y. Avissar, J. Almog, Water-soaked evidence: detectability of explosive traces after immersion in water, J. Forensic Sci. 48 (2003) 312–317. [19] C. Love, E. Gilchrist, N. Smith, L. Barron, Detection of anionic energetic material residues in enhanced fingermarks on porous and non-porous surfaces using ion chromatography, Forensic Sci. Int. 231 (2013) 150–156.