- Email: [email protected]
Use of gold nanoparticles as molecular intermediates for the detection of fingermarks
Forensic Science International 168 (2007) 169–176 www.elsevier.com/locate/forsciint
Use of gold nanoparticles as molecular intermediates for the detection of fingermarks Andy Becue *, Christophe Champod, Pierre Margot Ecole des Sciences Criminelles, Institut de Police Scientifique, Universite´ de Lausanne - Baˆtiment Batochime, CH-1015 Lausanne-Dorigny, Switzerland Received 14 June 2006; received in revised form 12 July 2006; accepted 13 July 2006 Available online 21 August 2006
Abstract Among the numerous methods dedicated to the detection of latent fingermarks, the MultiMetal Deposition (MMD) offers, as a main advantage, the ability to be applied on a great number of porous and non-porous surfaces, e.g., paper, plastic, glass, latex, and polystyrene, even if wetted. While considered as a powerful and sensitive technique, MMD is often neglected, mainly because of operational limitations (siliconized vessels, restrictive pH domain, numerous immersion baths, . . .). In this contribution, we propose a modification of the standard MMD method so that the procedure is simplified with a number of baths reduced to a minimum. To reach this goal, it was necessary to obtain a fully operable solution which could detect fingermarks in a single step. We chose to take advantage of the molecular recognition mechanisms by functionalizing the gold nanoparticles with a molecular host able to bind itself to gold while keeping the ability to trap molecules in solution. Cyclodextrins were chosen as they can be easily chemically modified to offer gold-binding abilities. Moreover, they are widely used as hosts for various molecular guests (dyes, luminescent molecules, . . .). This new formulation has been tested on three different surfaces to attest the feasibility of this strategy. Successful results were obtained with detailed fingermarks offering a good contrast to allow their identification without the need to enhance the results (such as with a physical developer). If the new formulation behaves very similarly to the old one, in terms of experimental conditions, it offers the additional advantage to develop fingermarks after immersing them in only one bath. The goal is thus reached. # 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: MultiMetal Deposition; Latent fingermark; Gold nanoparticles; Modified cyclodextrins; Molecular recognition; Thiolated molecules
1. Introduction In the area of fingermark detection, various efficient techniques are currently available for a great variety of substrates, e.g., paper, paperboard, plastic, wood, human skin, . . . [1]. Scientific researches are continuously conducted to develop new efficient chemicals and procedures or to widen the application field of the existing ones. Efficiency criteria are focused on selectivity, sensitivity, cost, capacity to be integrated within detection sequences, health and safety issues, and finally versatility and easiness of use. It is certainly on that last operational criterion that the MultiMetal Deposition technique (MMD) suffers compared to other procedures. MMD was initially introduced in forensic science by Saunders, in 1989 [2], and greatly improved by Schnetz and Margot, in
* Corresponding author. Tel.: +41 21 692 46 39; fax: +41 21 692 46 05. E-mail address: [email protected] (A. Becue). 0379-0738/$ – see front matter # 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2006.07.014
2001 [3]. MMD presents the great advantage to be a very sensitive technique, efficient on a great number of different surfaces, being porous or non-porous, and more particularly on surfaces which may present some difficulties for other techniques, e.g., polystyrene packaging, thermal paper, wetted surfaces. But it suffers from restrictive experimental conditions with many immersion baths required. This often limits its application in actual investigation work whether in the laboratory or, obviously, at crime scenes. Therefore, we attempted to modify the standard MMD technique so that the experimental procedure is simplified and the detection modes widened, i.e., by using dyes or luminescent molecules. The chemical structure of the MMD main reactant, i.e., the gold nanoparticles, has been modified so that these could be considered as functionalized molecular intermediates. The chemical structure and properties of the molecular entities that participate in the MMD process, as well as those selected in the research, is described in this section. The experimental steps leading to a new operational formulation are
170
A. Becue et al. / Forensic Science International 168 (2007) 169–176
Fig. 1. Schematic representation of the MultiMetal Deposition technique. Gold nanoparticles are first deposited onto the fingermark residue and are then used as growing sites for the metallic silver deposition.
described in Section 2 as well as the established protocol for obtaining different types of latent fingermarks. The experimental results, in terms of gold modification and fingermark detection, appear in Section 3 and are followed by a discussion about the efficiency of this strategy and the possibility to further modify or optimize the MMD technique. 1.1. The MultiMetal Deposition technique MMD is based on the use of small particles of metallic gold suspended in water and called ‘‘colloidal gold’’. The optimal particle diameter, determined by Schnetz and Margot [3], is about 14 nm. It gives the aqueous colloidal gold an intense ruby-red color. When a sample bearing a latent fingermark is immersed in such a solution and under specific experimental conditions, the gold nanoparticles are attracted to some components of the fingermark residue. The widely accepted theory is that, at a low pH value, the negatively charged gold nanoparticles are electrostatically attracted to the positively charged molecules in the fingermark. Due to the small size of the colloidal gold, it is difficult to obtain a good contrast
in order to observe the detected fingermarks at this stage. Consequently, it is necessary to enhance the results with the deposition of a second metal onto the surface of the gold nanoparticles. For this purpose, silver Ag(I) is reduced into metallic silver Ag(0) through the action of a reducing agent, e.g., hydroquinone (as proposed by Schnetz and Margot [3]) or Fe(II)/Fe(III) redox couple as suggested by Saunders [2]. Each gold aggregate acts thus as a growing site for the silver particles. As a result, fingermarks distinctly appear as dark grey/black prints. This procedure is schematically depicted in Fig. 1, and examples of fingermarks detected by this method are shown in Fig. 2. Despite the advantages associated with the nanoparticles’ sensitivity and the detailed fingermarks that can be obtained using such a technique, MMD suffers from a major disadvantage which limits its application in actual investigation work: the number of immersion baths, i.e., water rinsing, colloidal gold deposition, second water rinsing, pre-silver treatment, silver deposition, and final wash, which sometimes discourages people from using this technique. We aimed to drastically reduce the number of immersion baths so that they
Fig. 2. (Top) Examples of fingermark detection by using the MMD; support: M-Office non-bleached paper. The following parameters were used: immersion time 10 min, pH 2.65, [Tween 20] = 1 ml/ml, silver bath: 15 min (see Ref. [3] for details). (Bottom) Fingermark details coming from the above treated samples.
A. Becue et al. / Forensic Science International 168 (2007) 169–176
171
are limited to the bare essential, that is, rinsing, detection, and final wash. That could also open ways to use MMD on crime scenes, especially given the non-toxicity of the reagents. To reach this goal, it was necessary to gather all the characteristics of MMD in a unique aqueous solution, that is, specific deposition of the gold nanoparticles onto the ridge skin secretions and visual detection requiring no enhancement step. Fully-functionalized gold nanoparticles with these characteristics could fulfil this role. 1.2. Functionalization of the gold nanoparticles Gold is a powerful template on which a great variety of molecular hosts can spontaneously bind. Indeed, if gold surfaces or colloids are chosen in various domains due to their relative chemical inertness and their resistance to oxidation, a main class of reactants strongly binds to metallic surfaces. These are the thiolated molecules (R–SH, where R stands for the rest of the molecule). When a gold surface is immersed (or mixed, for nanoparticles) in a solution containing thiolated molecules, highly-ordered self-assembled monolayers (SAMs) are rapidly formed [4]. As a consequence of the chemisorption of the thiolated molecules onto gold, it is possible to fully functionalize an initial bare gold surface or colloid with a molecular layer that is relatively resistant to washes and common experimental conditions (Fig. 3). Among the various applications of such modified colloidal gold, an important growing domain is related to the development of nanoscale biosensors, in which gold nanoparticles play the role of templates for the immobilization of biomolecules [5,6]. Nevertheless, it is widely used in immunocytochemistry [7] and cell biology [8]. Moreover, if the aliphatic chains are ended by another functional group, it is possible to obtain functionalized SAMs. Their applications are various and numerous, e.g., surface grafting of proteins [9] or thiolated DNA [9–11], molecular wires [12], nanoelectrodes [13], polyfunctionalized gold colloids [14,15]. In the context of the modification of MMD, it was necessary to find an efficient molecular host which would bind to gold and offer chelating abilities in solution, so that it could trap some dyes or photoactive molecules once gold is deposited on the fingermark. Modified cyclodextrins were chosen to play this role. Indeed, they are very powerful molecular hosts, as described
Fig. 3. Illustration of the chemisorption of alkanethiolates on a gold surface, leading to the formation of highly-ordered self-assembled monolayers (SAMs). R can be a methyl ending group (–CH3) or another chemical function, thus providing new properties to the gold surface.
Fig. 4. Structural representation of a b-cyclodextrin built on seven glucopyranose rings linked together (for native b-cyclodextrin, R0 = H). [image source: http://www.cyclodex.com/images/figure2.gif.]
below, and their chemical structures can easily be modified to offer new abilities, such as gold-binding. Cyclodextrins (CDs) are cyclic oligosaccharides composed of a well-defined number of ‘‘glucopyranose’’ rings linked together with glucosidic bonds (–C–O–C–) to form threedimensional (3D) truncated cones (Fig. 4). The commonly used CDs contain 6, 7 or 8 pyranose units and are named a-, b- and g-CDs, respectively. As a consequence of the 3D arrangement of the glucopyranose groups, the outer surface of the CDs is hydrophilic and allows thus their solubilization in aqueous solution. In opposition, the inner cavity is hydrophobic and is able to complex highly hydrophobic molecules in solution [16]. CDs are thus efficient chelating agents forming stable inclusion complexes with a wide range of compounds. The b-CD presents a major interest because it is the cheapest and the most useful since its cavity has the right size to bind a variety of aromatic and aliphatic compounds. Another main interest of CDs lies in the possibility to modify their native template with a wide variety of chemical functions or groups that can be linked in a regioselective manner, through chemical modifications of the outer –OH groups [17,18]. In the framework of this research, we were particularly interested in the synthesis of thiolated cyclodextrins, as the –SH groups located on their smaller face would give them the ability to bind to gold surfaces and, by the same way, to play the role of in situ hosts [19,20]. Three different thiolated CDs were retained, that is, 6-monodeoxy-6-monothio b-cyclodextrin, per-6-thio b-cyclodextrin, and 6-monodeoxy-6-monothiodecamethylenethio b-cyclodextrin, further referred in this article as CD–SH, CD–(SH)7, and CD–S–C10H20–SH, respectively (Fig. 5). The strategy that has been chosen is depicted in Fig. 6. First, the bare gold nanoparticles are functionalized by the thiolated CDs. This step is followed by the addition of a dye which will be trapped in the CD cavities. Finally, this solution is used to detect latent fingermarks in a single step.
172
A. Becue et al. / Forensic Science International 168 (2007) 169–176
Fig. 5. Simplified chemical structures of the three thiolated cyclodextrins that were synthesized during this study, that is, from left to right, per-6-thio b-cyclodextrin (CD–(SH)7), 6-monodeoxy-6-monothio b-cyclodextrin (CD–SH), and 6-monodeoxy-6-monothiodecamethylenethio b-cyclodextrin (CD–S–C10H20–SH). It has to be noted that the trapezoid shape in the middle is a simplification of the cyclodextrin template, which is depicted in Fig. 4.
Fig. 6. Schematization of the colloidal gold modification strategy. First, the gold nanoparticles are functionalized with thiolated cyclodextrins. A dye is then added in the solution and a sample is finally immersed. The detection of latent fingermarks is then performed in a unique step.
2. Materials and methods Except if specified otherwise, all chemicals were bought from Sigma– Aldrich, with high purity grade and used without further purification. All modified cyclodextrins were synthesized in our laboratory, starting from the native b-cyclodextrin. Tetrachloroauric acid (HAuCl4), which is necessary to prepare the colloidal gold solution, was bought from Merck (ref. 1.01582.001). Only the synthesis of the CD–SH is described in this section, since the two other CDs behave very similarly in terms of gold-binding and only CD–SH was retained for further research due to its simpler synthesis. The synthetic route to obtain CD–SH starting from b-CD is described by Nelles et al. [19] and depicted in Fig. 7. It requires the synthesis of mono(6-O-ptoluenesulfonyl) b-cyclodextrin (named CD-Tos), which is a molecular intermediate bearing a toluenesulfonyl group. This intermediate can be commercially bought from CycloLab (Hungary) or easily synthesized using the method described by Byun et al. [21]. Briefly, once synthesized, CD-Tos was reacted with thiourea in DMF, at 75 8C for 2 days. The crude product was recrystallized in acetone and then freeze dried. Finally, the isothiouronium salt was converted
to the thiol using sodium disulfite (Na2S2O5). At the end, a very light white powder is obtained (yield: 43%). The synthesis of gold nanoparticles in water is commonly performed via the reduction of tetrachloroauric acid (HAuCl4) by a reducing agent in solution. According to the relative quantity of reagents in solution, the nanoparticle diameters can vary from 1 to 100 nm. In this work, the synthesis described by Grabar et al. [22] was followed: 50 ml of sodium citrate (38.8 mM) were added to 500 ml of boiling HAuCl4 (1 mM), then kept boiling for 10 min before cooling to room temperature while stirring, for another 15 min. The resulting colloidal gold solution is very similar to the one used by Schnetz and Margot [3], since the nanoparticles have a diameter of 12.5 2.4 nm. The main differences are the following: (1) only sodium citrate is used as the reducing agent (i.e., no tannic acid), and (2) the obtained solution is about five times more concentrated in terms of Au nanoparticles. Indeed, it was possible to calculate an effective concentration of the two colloidal gold solutions by considering each Au nanoparticle as an actual molecular entity, instead of using the global concentration of gold. Molarities were then obtained by combining the quantity of gold ions converted during the syntheses with the gold density and the size of
Fig. 7. Synthesis of 6-monodeoxy-6-monothio b-cyclodextrin (right) according to Nelles et al. [19], starting from native b-cyclodextrin (left) and via mono(6-O-ptoluenesulfonyl) b-cyclodextrin (middle).
A. Becue et al. / Forensic Science International 168 (2007) 169–176
173
Fig. 8. Experimental procedure of the MMD method, as optimized by Schnetz and Margot [3].
the obtained nanospheres. The concentrations of the solutions are about 15.1 nM for the synthesis used by Grabar, compared to 3.0 nM for the Schnetz one. The functionalization of such colloidal gold solution by CD– SH (and by the other ones) is easily obtained by dissolving 13.29 mg of CD–SH (0.01 mmole) in 100 ml of colloidal gold for 2 days, at room temperature and in the dark. The modified gold nanoparticles are isolated by the addition of DMSO (mixed again for 48 h), then CH3CN to initiate their precipitation. The solution is centrifuged to obtain a red solid, which is rinsed by DMSO:CH3CN (1:1, v/v) and finally dissolved in 100 ml of pure water to give a ‘‘CD–SH’’modified colloidal gold solution characterized by a ruby-red color. This solution is stable if preserved at 4 8C until it has to be used for latent fingermark detection. Just before use, the solution is mixed for 2 h with a specific dye whose final concentration should be equal to 105 M, to allow its complexation in the CD cavities. In this case, we chose Acid Blue 25 (AB25), a blue organic dye built on an anthraquinone structure. AB25 has a very high affinity for the CD cavity, as shown by Crini [23]. Indeed, Crini showed that about 100% of the AB25 molecules in aqueous solution were adsorbed by an immersed cyclodextrin-based polymer, which is used as an efficient depolluting system. Different sets of latent fingermarks were collected for this study. To verify the efficiency of this modified colloidal gold solution, three kinds of surfaces were chosen for their different behaviour in terms of interactions with latent fingermark or with colloidal gold, that is, a porous one: ‘‘M-Office’’ nonbleached paper; and two non-porous ones: low density polyethylene (LDPE) bags and polypropylene (PP) sheets. In addition to this, six different donors were chosen, i.e., three males and three females, and a well-defined protocol has been set up for the deposition of fingermarks from both hands. After having periodically collected such samples, we were able to possess fresh (few days) and aged fingermarks (one to several months).
3. Results 3.1. Synthesis of the modified cyclodextrins and functionalization of gold The modified cyclodextrins were successfully synthesized following the procedures described by Nelles et al. [19]. Those syntheses require no specific glassware or hard experimental conditions. Each final product has been characterized by FT-IR and MALDI-TOF mass spectroscopy to confirm the presence of CD–SH, CD–(SH)7, and CD–S–C10H20–SH. The functionalization of the colloidal gold by thiolated cyclodextrins was easily performed as this chemisorption is a spontaneous mechanism. No specific experimental conditions were required since the reaction is performed at ambient temperature and pressure, without any catalyst. This is an important point because colloidal gold is very sensitive to environmental changes and may become unstable under certain circumstances. If so, it is easily observed as the gold colloids precipitate in the bottom of the flask, sometimes irreversibly, or the solution color changes from ruby-red to dark-blue. In our case, we observed no instability of the solutions during their
functionalization. The addition of DMSO and CH3CN allows the isolation of the modified gold colloids by precipitating them so that they can be purified. After rinsing, the red solid is solubilized again in pure water to obtain the intense ruby-red colored solution, characterizing the presence of 14 nm gold colloids in solution. Some samples of the cyclodextrinmodified gold nanoparticles (Au–S–CD) were precipitated and dried to be analyzed by FTIR. The analysis of the resulting spectra revealed the signature of cyclodextrins combined with gold. The modified gold solutions were kept in the fridge and were stable for months. The cyclodextrin-modified gold nanoparticles with the AB25 encapsulated dye will be referred to as ‘‘Au–S–CD(AB25)’’. 3.2. Detection of fingermarks using the new formulation As highlighted by Schnetz in his Ph.D. thesis work on MMD, strictly controlled experimental conditions are crucial to a successful interaction with the ridge skin secretions (Fig. 8) [24]. First, it is necessary to lower the pH to a value of 2.65 to allow the gold nanoparticles to be deposited on the fingermark components. Moreover, it is necessary to add a surfactant, which greatly helps lower the non-selective deposition of gold nanoparticles on the support background. Schnetz recommends Tween 20. Its quantity must be controlled; as too much Tween 20 leads to a washing effect of the fingermark. A concentration of 1 ml of Tween 20 per ml of gold solution is the optimal quantity. In the case of the Au–S–CD(AB25) formulation, a great attention has been paid to the stability of the resulting mixture. Several tests were performed to study the influence of the amount of Tween 20 as well as the pH value on the quality of the detection. The procedure which was finally obtained is the following: a sample is briefly rinsed into distilled water and immersed for 10–15 min in the modified colloidal gold solution with a pH adjusted to 2.65 and that contains 5 ml of Tween 20 per ml of gold solution. The sample is finally rinsed in water to remove the non-deposited gold nanoparticles and dried in the air (Fig. 9). As a result, the fingermarks may be observed under
Fig. 9. Experimental procedure of the new proposed MMD formulation.
174
A. Becue et al. / Forensic Science International 168 (2007) 169–176
Fig. 10. Illustration of the contrast obtained after having converted in grayscale the picture of a fingermark obtained by immersion of a sample in the new colloidal gold formulation; support: LDPE.
ambient light as dark-blue prints and present a good contrast. Thus, the resulting enhancement produces an image that can be directly used for identification (Fig. 10). We observed a good stability and efficiency of the solutions with all three CDs, observing no differences between them. This formulation was successful to detect fingermarks in one step for all surfaces: non-bleached paper, low-density polyethylene, and polypropylene (Fig. 11). Moreover, the behaviour of this new colloidal gold formulation is extremely similar to the one of Schnetz and Margot [3], meaning that: - The pH value of 2.65 remains the best value for a good deposition of the gold nanoparticles onto the fingermarks. - We increased the quantity of Tween 20 to five times the value of Schnetz, as our colloidal gold is more concentrated in gold nanoparticles. - It is recommended to fix the sample to be developed to the dish containing the colloidal gold solution, while shaking gently (50–100 rpm for horizontal shakers). - The same results were obtained regardless of the age of the latent fingermark (up to 6 months). - Little or no coloration of the paper background was observed. The background darkening is a phenomenon due to the nonselective deposition of gold nanoparticles on the support. This is a minor phenomenon and it does not interfere with the
Fig. 11. Examples of the results obtained on LDPE bags (top), PP sheets (middle), and paper (bottom) for latent marks which were 1 month old and detected after having followed the new procedure described in this article.
observation of the fingermark since a good contrast is still obtained. However, the silver bath used in the standard MMD method enhances this effect and may cause some loss of contrast.
A. Becue et al. / Forensic Science International 168 (2007) 169–176
4. Discussion This research has led to two main improvements of the MMD method: (1) the reduction of the number of immersion baths from six to three (two rinsing baths and one containing the active solution) and, by the same occasion, the obtaining of a solution that efficiently develops latent fingermarks and (2) the modification of the surface of the gold nanoparticles without interfering with their ability to deposit on the ridge skin components. The new formulation of the colloidal gold is obtained in three steps: (1) the thiolated cyclodextrins have to be synthesized, (2) a native solution of gold colloids has to be functionalized with those CDs, and (3) before use, a dye is added and the solution is mixed to let the dye be trapped in the CD cavities; then the pH is set to 2.65 and Tween 20 is added. The two first steps are the most time-consuming, because the syntheses require about 3–4 days and the functionalization of colloidal gold another 4 days. However, it does not constitute a drawback since these two steps can be performed well before the use of the new MMD formulation. Once synthesized, the thiolated cyclodextrins are under the form of powders which are dried and thus kept for future uses without any specific storage conditions. Moreover, the quantities that are synthesized in one batch are sufficient to prepare litres of colloidal gold. Then, once the colloidal gold has been functionalized by thiolated cyclodextrins, it is possible to keep the solution in the refrigerator for several months without observing any instability. Another time-consuming step is the solubilization of the dye, just before use. However, this time can be safely reduced (2 h were set by Crini [23] as a ‘‘security’’ time to complex 100% of the dye in solution). As a matter of fact, other authors reported inclusion times of 7 min for other organic molecules [25]. As explained in Section 1, the binding mode of the gold nanoparticles to the fingermark residues remains misunderstood, even if the electrostatic attraction is the most widely accepted theory. In the case of the modification of the gold nanoparticles, some citrate ions were logically moved or replaced by thiolated cyclodextrins. However, the nanoparticles still possess a global negative charge due to the remaining citrate ions at their surface. We can therefore postulate that the deposition is still driven by electrostatic attraction, and that the increased concentration in terms of gold nanoparticles (i.e., the density is about five higher than that in the Schnetz solution) may compensate the reduction of the negative charge. While studying the possibility to introduce a luminescent tag in the CD cavity, no luminescent fingermarks were observed using dansyl chloride or 6-p-toluidinylnaphtalene-2-sulfonate, two luminescent molecules known for their affinity for CD cavities. After fluorometric studies, it was concluded that the gold nanoparticles played a role in this phenomenon since gold in aqueous solution acts as an acceptor of the energy emitted by the luminescent tag through a non-radiative pathway (induced, e.g., by the heavy atom effect). This phenomenon caused the quenching of the fluorescence of the luminescent marker. Further researches will have to be performed to circumvent this
175
problem and to allow a luminescent detection of fingermarks, which should greatly enhance the contrast when working on patterned backgrounds. 5. Conclusions This research aimed at modifying and, more specifically, simplifying, the standard MultiMetal Deposition technique. Indeed, MMD constitutes a very powerful, efficient and sensitive technique, but is under-exploited due to the tediousness of its experimental procedure. The first step was to sharply reduce the number of immersion baths, so that the detection of fingermarks can be performed in a single step. To reach this goal, gold nanoparticles were functionalized with thiolated cyclodextrins that were synthesized in our laboratory. Once modified, the CDs offer the ability to be chemically bound to the gold surface and can moreover complex small organic molecules in solution, such as dyes, by trapping them in their cavity. In this study, it was shown that this strategy can successfully lead to the detection of fingermarks in only one immersion bath. The ridge skin appears in the form of dark-blue prints with sufficient contrast compared to the support (on which the gold nanoparticles do not deposit preferentially). Similarly to the traditional MMD, this technique is very efficient on various surfaces, i.e., porous and non-porous, and is not sensitive to the age of the fingermark, or if it has been previously wetted. This strategy should be followed by considering gold not as a final component, but rather as a virgin surface on which several molecules can be grafted. In this contribution, molecular hosts, such as cyclodextrins, were used to trap dyes or tentatively luminescent tags, but it is also possible to use molecular linkers which can completely modify the way gold is interacting with the fingermarks. Future research will try to answer the following questions: (1) ‘‘Is it possible to propose this formulation as a replacement of the standard MMD [3] or for use at crime scenes?’’ and (2) ‘‘At which level can this new formulation be put in sequence with the existing methods, e.g., physical developer, DFO, ninhydrin, and cyanoacrylate?’’. Moreover, on-going research is looking into the binding of luminescent biomolecules by following the same strategy. Acknowledgments The authors would like to thank the Swiss National Fund (SNF) for the grant allowed for this research (grant no. 105580). References [1] C. Champod, C. Lennard, P. Margot, M. Stoilovic, Fingerprints and Other Ridge Skin Impressions. International Forensic Science and Investigation Series, CRC Press LLC, Florida, 2004. [2] G. Saunders, MultiMetal Deposition technique for latent fingerprint development. international association for identification, in: 74th Annual Educational Conference, June, Pensacola, USA, 1989. [3] B. Schnetz, P. Margot, Technical note: latent fingermarks, colloidal gold and MultiMetal Deposition (MMD). Optimisation of the method, Forensic Sci. Int. 118 (2001) 21–28.
176
A. Becue et al. / Forensic Science International 168 (2007) 169–176
[4] L.H. Dubois, R.G. Nuzzo, Synthesis, structure, and properties of model organic surfaces, Ann. Rev. Phys. Chem. 43 (1992) 437–463. [5] S.G. Penn, L. He, M.J. Natan, Nanoparticles for bioanalysis, Curr. Opin. Chem. Biol. 7 (2003) 609–615. [6] M. Wang, L. Wang, G. Wang, X. Ji, Y. Bai, T. Li, S. Gong, J. Li, Application of impedance spectroscopy for monitoring colloid Auenhanced antibody immobilization and antibody–antigen reactions, Biosens. Bioelectron. 19 (2004) 575–582. [7] G. Mayer, R.D. Leone, J.F. Hainfeld, M. Bendayan, Introduction of a novel HRP substrate–nanogold probe for signal amplification in immunocytochemistry, J. Histochem. Cytochem. 48 (2000) 461–469. [8] M.A. Hayat, Colloidal Gold: Principles, Methods, and Applications, Academic Press, San Diego, 1989. [9] K. Castelino, B. Kannan, A. Majumdar, Characterization of grafting density and binding efficiency of DNA and proteins on gold surfaces, Langmuir 21 (2005) 1956–1961. [10] J.J. Storhoff, S.S. Marla, P. Bao, S. Hagenow, H. Mehta, A. Lucas, V. Garimella, T. Patno, W. Buckingham, W. Cork, U.R. Mu¨ller, Gold nanoparticle-based detection of genomic DNA targets on microarrays using a novel optical detection system, Biosens. Bioelectron. 19 (2004) 875–883. [11] M. Liu, G.-Y. Liu, Hybridization with nanostructures of single-stranded DNA, Langmuir 21 (2005) 1972–1978. [12] L. Patrone, S. Palacin, J.P. Bourgoin, Direct comparison of the electronic coupling efficiency of sulfur and selenium alligator clips for molecules adsorbed onto gold electrodes, Appl. Surf. Sci. 212 (2003) 446–451. [13] S.J. Green, J.J. Stokes, M.J. Hostetler, J. Pietron, R.W. Murray, Threedimensional monolayers: nanometer-sized electrodes of alkanethiolatestabilized gold cluster molecules, J. Phys. Chem. B 101 (1997) 2663–2668. [14] R.S. Ingram, M.J. Hostetler, R.W. Murray, Poly-hetero-v-functionalized alkanethiolate-stabilized gold cluster compounds, J. Am. Chem. Soc. 119 (1997) 9175–9178.
[15] A.C. Templeton, M.J. Hostetler, E.K. Warmoth, S. Chen, C.M. Hartshorn, V.M. Krishnamurthy, M.D.E. Forbes, R.W. Murray, Gateway reactions to diverse, polyfunctional monolayer-protected gold clusters, J. Am. Chem. Soc. 120 (1998) 4845–4849. [16] K.A. Connors, The stability of cyclodextrin complexes in solution, Chem. Rev. 97 (1997) 1325–1357. [17] A.R. Khan, P. Forgo, K.J. Stine, V.T. D’Souza, Methods for selective modifications of cyclodextrins, Chem. Rev. 98 (1998) 1977–1996. [18] C.J. Easton, S.F. Lincoln, Modified Cyclodextrins—Scaffolds and Templates for Supramolecular Chemistry, Imperial College Press, London, 1999. [19] G. Nelles, M. Weisser, R. Back, P. Wohlfart, G. Wenz, S. Mittler-Neher, Controlled orientation of cyclodextrin derivatives immobilized on gold surfaces, J. Am. Chem. Soc. 118 (1996) 5039–5046. [20] M. Weisser, G. Nelles, G. Wenz, S. Mittler-Neher, Guest–host interactions with immobilized cyclodextrins, Sens. Actuators B Chem. 38 (1997) 58– 67. [21] H.-S. Byun, N. Zhong, R. Bittman, 6A-O-p-Toluenesulfonyl-b-Cyclodextrin, Organic Syntheses 77 (1999) 225–230. [22] K.C. Grabar, R.G. Freeman, M.B. Hommer, M.J. Natan, Preparation and characterization of Au colloid monolayer, Anal. Chem. 67 (1995) 735– 743. [23] G. Crini, Studies on adsorption of dyes on beta-cyclodextrin polymer, Bioresour. Technol. 90 (2003) 193–198. [24] B. Schnetz, La re´ve´lation des empreintes digitales par l’or colloı¨dal: l’amplification du signal par des techniques biochimiques, Ph.D. Thesis, Institut de Police Scientifique et de Criminologie, UNIL, Lausanne, CH, 1999. [25] Y. Dotsikas, E. Kontopanou, C. Allagiannis, Y.L. Loukas, Interaction of 6p-toluidinylnaphthalene-2-sulphonate with b-cyclodextrin, J. Pharm. Biomed. Anal. 23 (2000) 997–1003.
Use of gold nanoparticles as molecular intermediates for the detection of fingermarks Andy Becue *, Christophe Champod, Pierre Margot Ecole des Sciences Criminelles, Institut de Police Scientifique, Universite´ de Lausanne - Baˆtiment Batochime, CH-1015 Lausanne-Dorigny, Switzerland Received 14 June 2006; received in revised form 12 July 2006; accepted 13 July 2006 Available online 21 August 2006
Abstract Among the numerous methods dedicated to the detection of latent fingermarks, the MultiMetal Deposition (MMD) offers, as a main advantage, the ability to be applied on a great number of porous and non-porous surfaces, e.g., paper, plastic, glass, latex, and polystyrene, even if wetted. While considered as a powerful and sensitive technique, MMD is often neglected, mainly because of operational limitations (siliconized vessels, restrictive pH domain, numerous immersion baths, . . .). In this contribution, we propose a modification of the standard MMD method so that the procedure is simplified with a number of baths reduced to a minimum. To reach this goal, it was necessary to obtain a fully operable solution which could detect fingermarks in a single step. We chose to take advantage of the molecular recognition mechanisms by functionalizing the gold nanoparticles with a molecular host able to bind itself to gold while keeping the ability to trap molecules in solution. Cyclodextrins were chosen as they can be easily chemically modified to offer gold-binding abilities. Moreover, they are widely used as hosts for various molecular guests (dyes, luminescent molecules, . . .). This new formulation has been tested on three different surfaces to attest the feasibility of this strategy. Successful results were obtained with detailed fingermarks offering a good contrast to allow their identification without the need to enhance the results (such as with a physical developer). If the new formulation behaves very similarly to the old one, in terms of experimental conditions, it offers the additional advantage to develop fingermarks after immersing them in only one bath. The goal is thus reached. # 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: MultiMetal Deposition; Latent fingermark; Gold nanoparticles; Modified cyclodextrins; Molecular recognition; Thiolated molecules
1. Introduction In the area of fingermark detection, various efficient techniques are currently available for a great variety of substrates, e.g., paper, paperboard, plastic, wood, human skin, . . . [1]. Scientific researches are continuously conducted to develop new efficient chemicals and procedures or to widen the application field of the existing ones. Efficiency criteria are focused on selectivity, sensitivity, cost, capacity to be integrated within detection sequences, health and safety issues, and finally versatility and easiness of use. It is certainly on that last operational criterion that the MultiMetal Deposition technique (MMD) suffers compared to other procedures. MMD was initially introduced in forensic science by Saunders, in 1989 [2], and greatly improved by Schnetz and Margot, in
* Corresponding author. Tel.: +41 21 692 46 39; fax: +41 21 692 46 05. E-mail address: [email protected] (A. Becue). 0379-0738/$ – see front matter # 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.forsciint.2006.07.014
2001 [3]. MMD presents the great advantage to be a very sensitive technique, efficient on a great number of different surfaces, being porous or non-porous, and more particularly on surfaces which may present some difficulties for other techniques, e.g., polystyrene packaging, thermal paper, wetted surfaces. But it suffers from restrictive experimental conditions with many immersion baths required. This often limits its application in actual investigation work whether in the laboratory or, obviously, at crime scenes. Therefore, we attempted to modify the standard MMD technique so that the experimental procedure is simplified and the detection modes widened, i.e., by using dyes or luminescent molecules. The chemical structure of the MMD main reactant, i.e., the gold nanoparticles, has been modified so that these could be considered as functionalized molecular intermediates. The chemical structure and properties of the molecular entities that participate in the MMD process, as well as those selected in the research, is described in this section. The experimental steps leading to a new operational formulation are
170
A. Becue et al. / Forensic Science International 168 (2007) 169–176
Fig. 1. Schematic representation of the MultiMetal Deposition technique. Gold nanoparticles are first deposited onto the fingermark residue and are then used as growing sites for the metallic silver deposition.
described in Section 2 as well as the established protocol for obtaining different types of latent fingermarks. The experimental results, in terms of gold modification and fingermark detection, appear in Section 3 and are followed by a discussion about the efficiency of this strategy and the possibility to further modify or optimize the MMD technique. 1.1. The MultiMetal Deposition technique MMD is based on the use of small particles of metallic gold suspended in water and called ‘‘colloidal gold’’. The optimal particle diameter, determined by Schnetz and Margot [3], is about 14 nm. It gives the aqueous colloidal gold an intense ruby-red color. When a sample bearing a latent fingermark is immersed in such a solution and under specific experimental conditions, the gold nanoparticles are attracted to some components of the fingermark residue. The widely accepted theory is that, at a low pH value, the negatively charged gold nanoparticles are electrostatically attracted to the positively charged molecules in the fingermark. Due to the small size of the colloidal gold, it is difficult to obtain a good contrast
in order to observe the detected fingermarks at this stage. Consequently, it is necessary to enhance the results with the deposition of a second metal onto the surface of the gold nanoparticles. For this purpose, silver Ag(I) is reduced into metallic silver Ag(0) through the action of a reducing agent, e.g., hydroquinone (as proposed by Schnetz and Margot [3]) or Fe(II)/Fe(III) redox couple as suggested by Saunders [2]. Each gold aggregate acts thus as a growing site for the silver particles. As a result, fingermarks distinctly appear as dark grey/black prints. This procedure is schematically depicted in Fig. 1, and examples of fingermarks detected by this method are shown in Fig. 2. Despite the advantages associated with the nanoparticles’ sensitivity and the detailed fingermarks that can be obtained using such a technique, MMD suffers from a major disadvantage which limits its application in actual investigation work: the number of immersion baths, i.e., water rinsing, colloidal gold deposition, second water rinsing, pre-silver treatment, silver deposition, and final wash, which sometimes discourages people from using this technique. We aimed to drastically reduce the number of immersion baths so that they
Fig. 2. (Top) Examples of fingermark detection by using the MMD; support: M-Office non-bleached paper. The following parameters were used: immersion time 10 min, pH 2.65, [Tween 20] = 1 ml/ml, silver bath: 15 min (see Ref. [3] for details). (Bottom) Fingermark details coming from the above treated samples.
A. Becue et al. / Forensic Science International 168 (2007) 169–176
171
are limited to the bare essential, that is, rinsing, detection, and final wash. That could also open ways to use MMD on crime scenes, especially given the non-toxicity of the reagents. To reach this goal, it was necessary to gather all the characteristics of MMD in a unique aqueous solution, that is, specific deposition of the gold nanoparticles onto the ridge skin secretions and visual detection requiring no enhancement step. Fully-functionalized gold nanoparticles with these characteristics could fulfil this role. 1.2. Functionalization of the gold nanoparticles Gold is a powerful template on which a great variety of molecular hosts can spontaneously bind. Indeed, if gold surfaces or colloids are chosen in various domains due to their relative chemical inertness and their resistance to oxidation, a main class of reactants strongly binds to metallic surfaces. These are the thiolated molecules (R–SH, where R stands for the rest of the molecule). When a gold surface is immersed (or mixed, for nanoparticles) in a solution containing thiolated molecules, highly-ordered self-assembled monolayers (SAMs) are rapidly formed [4]. As a consequence of the chemisorption of the thiolated molecules onto gold, it is possible to fully functionalize an initial bare gold surface or colloid with a molecular layer that is relatively resistant to washes and common experimental conditions (Fig. 3). Among the various applications of such modified colloidal gold, an important growing domain is related to the development of nanoscale biosensors, in which gold nanoparticles play the role of templates for the immobilization of biomolecules [5,6]. Nevertheless, it is widely used in immunocytochemistry [7] and cell biology [8]. Moreover, if the aliphatic chains are ended by another functional group, it is possible to obtain functionalized SAMs. Their applications are various and numerous, e.g., surface grafting of proteins [9] or thiolated DNA [9–11], molecular wires [12], nanoelectrodes [13], polyfunctionalized gold colloids [14,15]. In the context of the modification of MMD, it was necessary to find an efficient molecular host which would bind to gold and offer chelating abilities in solution, so that it could trap some dyes or photoactive molecules once gold is deposited on the fingermark. Modified cyclodextrins were chosen to play this role. Indeed, they are very powerful molecular hosts, as described
Fig. 3. Illustration of the chemisorption of alkanethiolates on a gold surface, leading to the formation of highly-ordered self-assembled monolayers (SAMs). R can be a methyl ending group (–CH3) or another chemical function, thus providing new properties to the gold surface.
Fig. 4. Structural representation of a b-cyclodextrin built on seven glucopyranose rings linked together (for native b-cyclodextrin, R0 = H). [image source: http://www.cyclodex.com/images/figure2.gif.]
below, and their chemical structures can easily be modified to offer new abilities, such as gold-binding. Cyclodextrins (CDs) are cyclic oligosaccharides composed of a well-defined number of ‘‘glucopyranose’’ rings linked together with glucosidic bonds (–C–O–C–) to form threedimensional (3D) truncated cones (Fig. 4). The commonly used CDs contain 6, 7 or 8 pyranose units and are named a-, b- and g-CDs, respectively. As a consequence of the 3D arrangement of the glucopyranose groups, the outer surface of the CDs is hydrophilic and allows thus their solubilization in aqueous solution. In opposition, the inner cavity is hydrophobic and is able to complex highly hydrophobic molecules in solution [16]. CDs are thus efficient chelating agents forming stable inclusion complexes with a wide range of compounds. The b-CD presents a major interest because it is the cheapest and the most useful since its cavity has the right size to bind a variety of aromatic and aliphatic compounds. Another main interest of CDs lies in the possibility to modify their native template with a wide variety of chemical functions or groups that can be linked in a regioselective manner, through chemical modifications of the outer –OH groups [17,18]. In the framework of this research, we were particularly interested in the synthesis of thiolated cyclodextrins, as the –SH groups located on their smaller face would give them the ability to bind to gold surfaces and, by the same way, to play the role of in situ hosts [19,20]. Three different thiolated CDs were retained, that is, 6-monodeoxy-6-monothio b-cyclodextrin, per-6-thio b-cyclodextrin, and 6-monodeoxy-6-monothiodecamethylenethio b-cyclodextrin, further referred in this article as CD–SH, CD–(SH)7, and CD–S–C10H20–SH, respectively (Fig. 5). The strategy that has been chosen is depicted in Fig. 6. First, the bare gold nanoparticles are functionalized by the thiolated CDs. This step is followed by the addition of a dye which will be trapped in the CD cavities. Finally, this solution is used to detect latent fingermarks in a single step.
172
A. Becue et al. / Forensic Science International 168 (2007) 169–176
Fig. 5. Simplified chemical structures of the three thiolated cyclodextrins that were synthesized during this study, that is, from left to right, per-6-thio b-cyclodextrin (CD–(SH)7), 6-monodeoxy-6-monothio b-cyclodextrin (CD–SH), and 6-monodeoxy-6-monothiodecamethylenethio b-cyclodextrin (CD–S–C10H20–SH). It has to be noted that the trapezoid shape in the middle is a simplification of the cyclodextrin template, which is depicted in Fig. 4.
Fig. 6. Schematization of the colloidal gold modification strategy. First, the gold nanoparticles are functionalized with thiolated cyclodextrins. A dye is then added in the solution and a sample is finally immersed. The detection of latent fingermarks is then performed in a unique step.
2. Materials and methods Except if specified otherwise, all chemicals were bought from Sigma– Aldrich, with high purity grade and used without further purification. All modified cyclodextrins were synthesized in our laboratory, starting from the native b-cyclodextrin. Tetrachloroauric acid (HAuCl4), which is necessary to prepare the colloidal gold solution, was bought from Merck (ref. 1.01582.001). Only the synthesis of the CD–SH is described in this section, since the two other CDs behave very similarly in terms of gold-binding and only CD–SH was retained for further research due to its simpler synthesis. The synthetic route to obtain CD–SH starting from b-CD is described by Nelles et al. [19] and depicted in Fig. 7. It requires the synthesis of mono(6-O-ptoluenesulfonyl) b-cyclodextrin (named CD-Tos), which is a molecular intermediate bearing a toluenesulfonyl group. This intermediate can be commercially bought from CycloLab (Hungary) or easily synthesized using the method described by Byun et al. [21]. Briefly, once synthesized, CD-Tos was reacted with thiourea in DMF, at 75 8C for 2 days. The crude product was recrystallized in acetone and then freeze dried. Finally, the isothiouronium salt was converted
to the thiol using sodium disulfite (Na2S2O5). At the end, a very light white powder is obtained (yield: 43%). The synthesis of gold nanoparticles in water is commonly performed via the reduction of tetrachloroauric acid (HAuCl4) by a reducing agent in solution. According to the relative quantity of reagents in solution, the nanoparticle diameters can vary from 1 to 100 nm. In this work, the synthesis described by Grabar et al. [22] was followed: 50 ml of sodium citrate (38.8 mM) were added to 500 ml of boiling HAuCl4 (1 mM), then kept boiling for 10 min before cooling to room temperature while stirring, for another 15 min. The resulting colloidal gold solution is very similar to the one used by Schnetz and Margot [3], since the nanoparticles have a diameter of 12.5 2.4 nm. The main differences are the following: (1) only sodium citrate is used as the reducing agent (i.e., no tannic acid), and (2) the obtained solution is about five times more concentrated in terms of Au nanoparticles. Indeed, it was possible to calculate an effective concentration of the two colloidal gold solutions by considering each Au nanoparticle as an actual molecular entity, instead of using the global concentration of gold. Molarities were then obtained by combining the quantity of gold ions converted during the syntheses with the gold density and the size of
Fig. 7. Synthesis of 6-monodeoxy-6-monothio b-cyclodextrin (right) according to Nelles et al. [19], starting from native b-cyclodextrin (left) and via mono(6-O-ptoluenesulfonyl) b-cyclodextrin (middle).
A. Becue et al. / Forensic Science International 168 (2007) 169–176
173
Fig. 8. Experimental procedure of the MMD method, as optimized by Schnetz and Margot [3].
the obtained nanospheres. The concentrations of the solutions are about 15.1 nM for the synthesis used by Grabar, compared to 3.0 nM for the Schnetz one. The functionalization of such colloidal gold solution by CD– SH (and by the other ones) is easily obtained by dissolving 13.29 mg of CD–SH (0.01 mmole) in 100 ml of colloidal gold for 2 days, at room temperature and in the dark. The modified gold nanoparticles are isolated by the addition of DMSO (mixed again for 48 h), then CH3CN to initiate their precipitation. The solution is centrifuged to obtain a red solid, which is rinsed by DMSO:CH3CN (1:1, v/v) and finally dissolved in 100 ml of pure water to give a ‘‘CD–SH’’modified colloidal gold solution characterized by a ruby-red color. This solution is stable if preserved at 4 8C until it has to be used for latent fingermark detection. Just before use, the solution is mixed for 2 h with a specific dye whose final concentration should be equal to 105 M, to allow its complexation in the CD cavities. In this case, we chose Acid Blue 25 (AB25), a blue organic dye built on an anthraquinone structure. AB25 has a very high affinity for the CD cavity, as shown by Crini [23]. Indeed, Crini showed that about 100% of the AB25 molecules in aqueous solution were adsorbed by an immersed cyclodextrin-based polymer, which is used as an efficient depolluting system. Different sets of latent fingermarks were collected for this study. To verify the efficiency of this modified colloidal gold solution, three kinds of surfaces were chosen for their different behaviour in terms of interactions with latent fingermark or with colloidal gold, that is, a porous one: ‘‘M-Office’’ nonbleached paper; and two non-porous ones: low density polyethylene (LDPE) bags and polypropylene (PP) sheets. In addition to this, six different donors were chosen, i.e., three males and three females, and a well-defined protocol has been set up for the deposition of fingermarks from both hands. After having periodically collected such samples, we were able to possess fresh (few days) and aged fingermarks (one to several months).
3. Results 3.1. Synthesis of the modified cyclodextrins and functionalization of gold The modified cyclodextrins were successfully synthesized following the procedures described by Nelles et al. [19]. Those syntheses require no specific glassware or hard experimental conditions. Each final product has been characterized by FT-IR and MALDI-TOF mass spectroscopy to confirm the presence of CD–SH, CD–(SH)7, and CD–S–C10H20–SH. The functionalization of the colloidal gold by thiolated cyclodextrins was easily performed as this chemisorption is a spontaneous mechanism. No specific experimental conditions were required since the reaction is performed at ambient temperature and pressure, without any catalyst. This is an important point because colloidal gold is very sensitive to environmental changes and may become unstable under certain circumstances. If so, it is easily observed as the gold colloids precipitate in the bottom of the flask, sometimes irreversibly, or the solution color changes from ruby-red to dark-blue. In our case, we observed no instability of the solutions during their
functionalization. The addition of DMSO and CH3CN allows the isolation of the modified gold colloids by precipitating them so that they can be purified. After rinsing, the red solid is solubilized again in pure water to obtain the intense ruby-red colored solution, characterizing the presence of 14 nm gold colloids in solution. Some samples of the cyclodextrinmodified gold nanoparticles (Au–S–CD) were precipitated and dried to be analyzed by FTIR. The analysis of the resulting spectra revealed the signature of cyclodextrins combined with gold. The modified gold solutions were kept in the fridge and were stable for months. The cyclodextrin-modified gold nanoparticles with the AB25 encapsulated dye will be referred to as ‘‘Au–S–CD(AB25)’’. 3.2. Detection of fingermarks using the new formulation As highlighted by Schnetz in his Ph.D. thesis work on MMD, strictly controlled experimental conditions are crucial to a successful interaction with the ridge skin secretions (Fig. 8) [24]. First, it is necessary to lower the pH to a value of 2.65 to allow the gold nanoparticles to be deposited on the fingermark components. Moreover, it is necessary to add a surfactant, which greatly helps lower the non-selective deposition of gold nanoparticles on the support background. Schnetz recommends Tween 20. Its quantity must be controlled; as too much Tween 20 leads to a washing effect of the fingermark. A concentration of 1 ml of Tween 20 per ml of gold solution is the optimal quantity. In the case of the Au–S–CD(AB25) formulation, a great attention has been paid to the stability of the resulting mixture. Several tests were performed to study the influence of the amount of Tween 20 as well as the pH value on the quality of the detection. The procedure which was finally obtained is the following: a sample is briefly rinsed into distilled water and immersed for 10–15 min in the modified colloidal gold solution with a pH adjusted to 2.65 and that contains 5 ml of Tween 20 per ml of gold solution. The sample is finally rinsed in water to remove the non-deposited gold nanoparticles and dried in the air (Fig. 9). As a result, the fingermarks may be observed under
Fig. 9. Experimental procedure of the new proposed MMD formulation.
174
A. Becue et al. / Forensic Science International 168 (2007) 169–176
Fig. 10. Illustration of the contrast obtained after having converted in grayscale the picture of a fingermark obtained by immersion of a sample in the new colloidal gold formulation; support: LDPE.
ambient light as dark-blue prints and present a good contrast. Thus, the resulting enhancement produces an image that can be directly used for identification (Fig. 10). We observed a good stability and efficiency of the solutions with all three CDs, observing no differences between them. This formulation was successful to detect fingermarks in one step for all surfaces: non-bleached paper, low-density polyethylene, and polypropylene (Fig. 11). Moreover, the behaviour of this new colloidal gold formulation is extremely similar to the one of Schnetz and Margot [3], meaning that: - The pH value of 2.65 remains the best value for a good deposition of the gold nanoparticles onto the fingermarks. - We increased the quantity of Tween 20 to five times the value of Schnetz, as our colloidal gold is more concentrated in gold nanoparticles. - It is recommended to fix the sample to be developed to the dish containing the colloidal gold solution, while shaking gently (50–100 rpm for horizontal shakers). - The same results were obtained regardless of the age of the latent fingermark (up to 6 months). - Little or no coloration of the paper background was observed. The background darkening is a phenomenon due to the nonselective deposition of gold nanoparticles on the support. This is a minor phenomenon and it does not interfere with the
Fig. 11. Examples of the results obtained on LDPE bags (top), PP sheets (middle), and paper (bottom) for latent marks which were 1 month old and detected after having followed the new procedure described in this article.
observation of the fingermark since a good contrast is still obtained. However, the silver bath used in the standard MMD method enhances this effect and may cause some loss of contrast.
A. Becue et al. / Forensic Science International 168 (2007) 169–176
4. Discussion This research has led to two main improvements of the MMD method: (1) the reduction of the number of immersion baths from six to three (two rinsing baths and one containing the active solution) and, by the same occasion, the obtaining of a solution that efficiently develops latent fingermarks and (2) the modification of the surface of the gold nanoparticles without interfering with their ability to deposit on the ridge skin components. The new formulation of the colloidal gold is obtained in three steps: (1) the thiolated cyclodextrins have to be synthesized, (2) a native solution of gold colloids has to be functionalized with those CDs, and (3) before use, a dye is added and the solution is mixed to let the dye be trapped in the CD cavities; then the pH is set to 2.65 and Tween 20 is added. The two first steps are the most time-consuming, because the syntheses require about 3–4 days and the functionalization of colloidal gold another 4 days. However, it does not constitute a drawback since these two steps can be performed well before the use of the new MMD formulation. Once synthesized, the thiolated cyclodextrins are under the form of powders which are dried and thus kept for future uses without any specific storage conditions. Moreover, the quantities that are synthesized in one batch are sufficient to prepare litres of colloidal gold. Then, once the colloidal gold has been functionalized by thiolated cyclodextrins, it is possible to keep the solution in the refrigerator for several months without observing any instability. Another time-consuming step is the solubilization of the dye, just before use. However, this time can be safely reduced (2 h were set by Crini [23] as a ‘‘security’’ time to complex 100% of the dye in solution). As a matter of fact, other authors reported inclusion times of 7 min for other organic molecules [25]. As explained in Section 1, the binding mode of the gold nanoparticles to the fingermark residues remains misunderstood, even if the electrostatic attraction is the most widely accepted theory. In the case of the modification of the gold nanoparticles, some citrate ions were logically moved or replaced by thiolated cyclodextrins. However, the nanoparticles still possess a global negative charge due to the remaining citrate ions at their surface. We can therefore postulate that the deposition is still driven by electrostatic attraction, and that the increased concentration in terms of gold nanoparticles (i.e., the density is about five higher than that in the Schnetz solution) may compensate the reduction of the negative charge. While studying the possibility to introduce a luminescent tag in the CD cavity, no luminescent fingermarks were observed using dansyl chloride or 6-p-toluidinylnaphtalene-2-sulfonate, two luminescent molecules known for their affinity for CD cavities. After fluorometric studies, it was concluded that the gold nanoparticles played a role in this phenomenon since gold in aqueous solution acts as an acceptor of the energy emitted by the luminescent tag through a non-radiative pathway (induced, e.g., by the heavy atom effect). This phenomenon caused the quenching of the fluorescence of the luminescent marker. Further researches will have to be performed to circumvent this
175
problem and to allow a luminescent detection of fingermarks, which should greatly enhance the contrast when working on patterned backgrounds. 5. Conclusions This research aimed at modifying and, more specifically, simplifying, the standard MultiMetal Deposition technique. Indeed, MMD constitutes a very powerful, efficient and sensitive technique, but is under-exploited due to the tediousness of its experimental procedure. The first step was to sharply reduce the number of immersion baths, so that the detection of fingermarks can be performed in a single step. To reach this goal, gold nanoparticles were functionalized with thiolated cyclodextrins that were synthesized in our laboratory. Once modified, the CDs offer the ability to be chemically bound to the gold surface and can moreover complex small organic molecules in solution, such as dyes, by trapping them in their cavity. In this study, it was shown that this strategy can successfully lead to the detection of fingermarks in only one immersion bath. The ridge skin appears in the form of dark-blue prints with sufficient contrast compared to the support (on which the gold nanoparticles do not deposit preferentially). Similarly to the traditional MMD, this technique is very efficient on various surfaces, i.e., porous and non-porous, and is not sensitive to the age of the fingermark, or if it has been previously wetted. This strategy should be followed by considering gold not as a final component, but rather as a virgin surface on which several molecules can be grafted. In this contribution, molecular hosts, such as cyclodextrins, were used to trap dyes or tentatively luminescent tags, but it is also possible to use molecular linkers which can completely modify the way gold is interacting with the fingermarks. Future research will try to answer the following questions: (1) ‘‘Is it possible to propose this formulation as a replacement of the standard MMD [3] or for use at crime scenes?’’ and (2) ‘‘At which level can this new formulation be put in sequence with the existing methods, e.g., physical developer, DFO, ninhydrin, and cyanoacrylate?’’. Moreover, on-going research is looking into the binding of luminescent biomolecules by following the same strategy. Acknowledgments The authors would like to thank the Swiss National Fund (SNF) for the grant allowed for this research (grant no. 105580). References [1] C. Champod, C. Lennard, P. Margot, M. Stoilovic, Fingerprints and Other Ridge Skin Impressions. International Forensic Science and Investigation Series, CRC Press LLC, Florida, 2004. [2] G. Saunders, MultiMetal Deposition technique for latent fingerprint development. international association for identification, in: 74th Annual Educational Conference, June, Pensacola, USA, 1989. [3] B. Schnetz, P. Margot, Technical note: latent fingermarks, colloidal gold and MultiMetal Deposition (MMD). Optimisation of the method, Forensic Sci. Int. 118 (2001) 21–28.
176
A. Becue et al. / Forensic Science International 168 (2007) 169–176
[4] L.H. Dubois, R.G. Nuzzo, Synthesis, structure, and properties of model organic surfaces, Ann. Rev. Phys. Chem. 43 (1992) 437–463. [5] S.G. Penn, L. He, M.J. Natan, Nanoparticles for bioanalysis, Curr. Opin. Chem. Biol. 7 (2003) 609–615. [6] M. Wang, L. Wang, G. Wang, X. Ji, Y. Bai, T. Li, S. Gong, J. Li, Application of impedance spectroscopy for monitoring colloid Auenhanced antibody immobilization and antibody–antigen reactions, Biosens. Bioelectron. 19 (2004) 575–582. [7] G. Mayer, R.D. Leone, J.F. Hainfeld, M. Bendayan, Introduction of a novel HRP substrate–nanogold probe for signal amplification in immunocytochemistry, J. Histochem. Cytochem. 48 (2000) 461–469. [8] M.A. Hayat, Colloidal Gold: Principles, Methods, and Applications, Academic Press, San Diego, 1989. [9] K. Castelino, B. Kannan, A. Majumdar, Characterization of grafting density and binding efficiency of DNA and proteins on gold surfaces, Langmuir 21 (2005) 1956–1961. [10] J.J. Storhoff, S.S. Marla, P. Bao, S. Hagenow, H. Mehta, A. Lucas, V. Garimella, T. Patno, W. Buckingham, W. Cork, U.R. Mu¨ller, Gold nanoparticle-based detection of genomic DNA targets on microarrays using a novel optical detection system, Biosens. Bioelectron. 19 (2004) 875–883. [11] M. Liu, G.-Y. Liu, Hybridization with nanostructures of single-stranded DNA, Langmuir 21 (2005) 1972–1978. [12] L. Patrone, S. Palacin, J.P. Bourgoin, Direct comparison of the electronic coupling efficiency of sulfur and selenium alligator clips for molecules adsorbed onto gold electrodes, Appl. Surf. Sci. 212 (2003) 446–451. [13] S.J. Green, J.J. Stokes, M.J. Hostetler, J. Pietron, R.W. Murray, Threedimensional monolayers: nanometer-sized electrodes of alkanethiolatestabilized gold cluster molecules, J. Phys. Chem. B 101 (1997) 2663–2668. [14] R.S. Ingram, M.J. Hostetler, R.W. Murray, Poly-hetero-v-functionalized alkanethiolate-stabilized gold cluster compounds, J. Am. Chem. Soc. 119 (1997) 9175–9178.
[15] A.C. Templeton, M.J. Hostetler, E.K. Warmoth, S. Chen, C.M. Hartshorn, V.M. Krishnamurthy, M.D.E. Forbes, R.W. Murray, Gateway reactions to diverse, polyfunctional monolayer-protected gold clusters, J. Am. Chem. Soc. 120 (1998) 4845–4849. [16] K.A. Connors, The stability of cyclodextrin complexes in solution, Chem. Rev. 97 (1997) 1325–1357. [17] A.R. Khan, P. Forgo, K.J. Stine, V.T. D’Souza, Methods for selective modifications of cyclodextrins, Chem. Rev. 98 (1998) 1977–1996. [18] C.J. Easton, S.F. Lincoln, Modified Cyclodextrins—Scaffolds and Templates for Supramolecular Chemistry, Imperial College Press, London, 1999. [19] G. Nelles, M. Weisser, R. Back, P. Wohlfart, G. Wenz, S. Mittler-Neher, Controlled orientation of cyclodextrin derivatives immobilized on gold surfaces, J. Am. Chem. Soc. 118 (1996) 5039–5046. [20] M. Weisser, G. Nelles, G. Wenz, S. Mittler-Neher, Guest–host interactions with immobilized cyclodextrins, Sens. Actuators B Chem. 38 (1997) 58– 67. [21] H.-S. Byun, N. Zhong, R. Bittman, 6A-O-p-Toluenesulfonyl-b-Cyclodextrin, Organic Syntheses 77 (1999) 225–230. [22] K.C. Grabar, R.G. Freeman, M.B. Hommer, M.J. Natan, Preparation and characterization of Au colloid monolayer, Anal. Chem. 67 (1995) 735– 743. [23] G. Crini, Studies on adsorption of dyes on beta-cyclodextrin polymer, Bioresour. Technol. 90 (2003) 193–198. [24] B. Schnetz, La re´ve´lation des empreintes digitales par l’or colloı¨dal: l’amplification du signal par des techniques biochimiques, Ph.D. Thesis, Institut de Police Scientifique et de Criminologie, UNIL, Lausanne, CH, 1999. [25] Y. Dotsikas, E. Kontopanou, C. Allagiannis, Y.L. Loukas, Interaction of 6p-toluidinylnaphthalene-2-sulphonate with b-cyclodextrin, J. Pharm. Biomed. Anal. 23 (2000) 997–1003.