Anal. Chem. 2009, 81, 5267–5272
Sensitive Determination of Hexamethylene Triperoxide Diamine Explosives, Using Electrogenerated Chemiluminescence Enhanced by Silver Nitrate Suman Parajuli and Wujian Miao* Department of Chemistry and Biochemistry, The University of Southern Mississippi, Hattiesburg, Mississippi 39406 Sensitive detection and quantification of hexamethylene triperoxide diamine (HMTD), which is one of commonly used explosives by terrorists, was presented on the basis of electrogenerated chemiluminescence (ECL) technology coupled with silver nitrate (AgNO3) enhancement in acetonitrile at a platinum electrode. Upon anodic potential scanning, HMTD irreversibly oxidized at ∼1.70 V vs Ag/Ag+ (10 mM) at a scan rate of 50 mV/ s, and the ECL profile was coincident with the oxidation potential of HMTD in the presence of ruthenium(II) tris(bipyridine) (Ru(bpy)32+) luminophore species, which showed a half-wave potential of 0.96 V vs Ag/Ag+. The addition of small amounts of AgNO3 (0.50-7.0 mM) into the HMTD/Ru(bpy)32+ system resulted in significant enhancement in HMTD ECL production (up to 27 times). This enhancement was determined to be largely associated with NO3- and was linearly proportional to the concentrations of NO3 and Ag+ in solution. Homogeneous chemical oxidations of HMTD by electrogenerated NO3• and Ag(II) species proximity to the electrode were proposed to be responsible for the ECL enhancement. On the basis of cyclic voltammetry (CV) and CV digital simulations, standard potential values of 1.79 V vs Ag/Ag+ (or 1.98 V vs NHE) and 1.82 V vs Ag/Ag+ (or 2.01 V vs NHE) were estimated for Ag(II)/Ag(I) and NO3•/NO3- couples, respectively. A detection limit of 50 µM of HMTD was achieved with the current technique, which was 10 times more sensitive than that reported previously, which was based on a high-performance liquid chromatography/Fourier transform infrared (HPLC/FT-IR) detection method. Legal authorities have witnessed an increased number of threats of illegal use of peroxide-based explosive materials by terrorists. These compounds are very popular among terrorists because they can be synthesized readily from commercially available chemicals.1 These explosives, which are also known as “unconventional explosives” (for the reason that they have no use for military purposes, because of their high instability and powerful * To whom correspondence should be addressed. Tel.: +1 601-266 4716. Fax: +1 601-266 6075. E-mail:
[email protected]. (1) Schulte-Ladbeck, R.; Kolla, P.; Karst, U. Analyst 2002, 127, 1152–1154. 10.1021/ac900489a CCC: $40.75 2009 American Chemical Society Published on Web 06/10/2009
initiating explosive capability), are basically linked to terrorists.2 There is an urgent need for detection of these “home-made” explosives, especially in the checkpoints of the mass-transit facilities and other government and public facilities.3-6 Hexamethylene triperoxide diamine (HMTD) is a representative of the aforementioned peroxide-based explosives. It is a white solid with cyclic structure7 (see Scheme 1), which was first synthesized by Legler in 1885,8 that is sensitive to friction, impact, and electrical discharge.9 The friction sensitivity of HMTD is comparable to other well-known explosives, such as triacetone triperoxide (TATP) and trinitrotoluene (TNT), although its impact sensitivity is approximately half of TATP.10-12 Several analytical techniques have been used to detect peroxide explosives, including HMTD. These techniques include separationbased gas chromatography coupled with mass spectrometry (GC/MS),13-15 liquid chromatography (LC)/MS,9 time-of-flight (TOF)/MS,16 desorption electrospray ionization (DESI)/MS,6,17 liquid chromatography/Fourier transform infrared (LC/FT-IR),10 and ultraviolet-visible (UV-vis) spectrometric-based methods.1 Mass spectroscopy (MS)-based methods are very sensitive and can detect peroxide explosives at nanogram to picogram levels; however, the instruments are generally very expensive and not portable for field tests. Chromatographic, UV-vis, and infrared (IR) detection techniques are often time-consuming in sample (2) Pumera, M. Electrophoresis 2008, 29, 269–273. (3) Munoz, R. A. A.; Lu, D.; Cagan, A.; Wang, J. Analyst 2007, 132, 560–565. (4) Schulte-Ladbeck, R.; Vogel, M.; Karst, U. Anal. Bioanal. Chem. 2006, 386, 559–565. (5) Pumera, M. Electrophoresis 2006, 27, 244–256. (6) Cotte-Rodriguez, I.; Chen, H.; Cooks, R. G. Chem. Commun. 2006, 953– 955. (7) Baeyer, A.; Villiger, V. Ber. Dtsch. Chem. Ges. 1900, 33, 2479–2487. (8) Legler, L. Ber. Dtsch. Chem. Ges. 1885, 18, 3343–3351. (9) Crowson, A.; Beardah, M. S. Analyst 2001, 126, 1689–1693. (10) Schulte-Ladbeck, R.; Edelmann, A.; Quintas, G.; Lendl, B.; Karst, U. Anal. Chem. 2006, 78, 8150–8155. (11) Yinon, J. Forensic and Environmental Detection of Explosives; John Wiley & Sons: New York, 1999. (12) Modern Methods and Applications in Analysis of Explosives; Yinon, J., Zitrin, S., Eds.; John Wiley & Sons: New York, 1996. (13) Gielsdorf, W. Fresenius’ J. Anal. Chem. 1981, 308, 123–128. (14) Sigman, M. E.; Clark, C. D.; Fidler, R.; Geiger, C. L.; Clausen, C. A. Rapid Commun. Mass Spectrom. 2006, 20, 2851–2857. (15) Muller, D.; Levy, A.; Shelef, R.; Abramovich-Bar, S.; Sonenfeld, D.; Tamiri, T. J. Forensic Sci. 2004, 49, 935–938. (16) Mullen, C.; Irwin, A.; Pond, B. V.; Huestis, D. L.; Coggiola, M. J.; Oser, H. Anal. Chem. 2006, 78, 3807–3814. (17) Cotte-Rodriguez, I.; Hernandez-Soto, H.; Chen, H.; Cooks, R. G. Anal. Chem. 2008, 80, 1512–1519.
Analytical Chemistry, Vol. 81, No. 13, July 1, 2009
5267
Scheme 1. Synthesis of Hexamethylene Triperoxide Diamine (HMTD)
preparation and insufficiently sensitive for trace amounts of explosive quantification. Electrogenerated chemiluminescence (ECL), which is a process of light production, as a result of electrochemical reactions at an electrode, has been proven to be a powerful analytical technique, because of its inherent features (such as high sensitivity, good selectivity, low background, integrated versatility, and fast sample analysis).18-21 ECL has been widely used for many types of target detection and quantification under a broad variety of areas, which include DNA probe, immunoassay, pharmaceutical study, food and water testing, and biowarfare agent detection.18 Very recently, we reported an ultrasensitive detection of TNT, which was accomplished based on the sandwich-type TNT immunoassay combined with ECL technology.22 The limit of detection (e0.10 ± 0.01 ppt) is ∼600 times more sensitive, compared to the most sensitive TNT detection method, based on surface plasmon resonance23 in the literature, and the absolute detection limit in mass (∼0.1 pg) is only ∼0.5% of that from mass spectroscopy.24 In the present study, an ECL detection and quantification method for HMTD, using AgNO3 as an ECL enhancing agent, will be reported. This method is based on the fact that HMTD contains tertiary amine moieties with R-C hydrogens (see Scheme 1), which could act as an ECL co-reactant such as tri-npropylamine (TPrA) in the presence of an ECL luminophore such as Ru(bpy)32+ (ruthenium(II) tris(bipyridine)) up on anodic potential scanning.25 Remarkable enhancement by AgNO3 for HMTD ECL generation and relevant electrochemical and ECL mechanisms will be described. The ECL enhancement strategy used in this paper could be extended to other systems, where the electrochemical oxidations of the analyte (co-reactant) or the luminophore at an electrode are suppressed, because of the nature of the working electrode,26 or “delayed”, because of the slow heterogenous electron-transfer rate of the compound. EXPERIMENTAL SECTION Chemicals. The following materials were used in the study: hexamethylenetetramine (99+%, Alfa Aesar, Ward Hill, MA); silver nitrate (99.5%) and tetra-n-butylammonium perchlorate (TBAP, (18) Bard, A. J., Ed. Electrogenerated Chemiluminescence; Marcel Dekker: New York, 2004. (19) Miao, W. Chem. Rev. 2008, 108, 2506–2553. (20) Richter, M. M. Chem. Rev. 2004, 104, 3003–3036. (21) Yin, X.-B.; Dong, S.; Wang, E. Trends Anal. Chem. 2004, 23, 432–441. (22) Pittman, T. L.; Thomson, B.; Miao, W. Anal. Chim. Acta 2009, 632, 197– 202. (23) Shankaran, D. R.; Gobi, K. V.; Sakai, T.; Matsumoto, K.; Imato, T.; Toko, K.; Miura, N. IEEE Sens. J. 2005, 5, 616–621. (24) Lee, M. R.; Chang, S. C.; Kao, T. S.; Tang, C. P. J. Res. Natl. Bur. Stand. 1988, 93, 428–430. (25) Miao, W.; Choi, J.-P. In Electrogenerated Chemiluminescence; Bard, A. J., Ed.; Marcel Dekker: New York, 2004; Chapter 5, pp 213-272. (26) Rubinstein, I.; Bard, A. J. J. Am. Chem. Soc. 1981, 103, 512–516.
5268
Analytical Chemistry, Vol. 81, No. 13, July 1, 2009
99+%, electrochemical grade) (each obtained from Fluka, Milwaukee, WI); hydrogen peroxide (30%), anhydrous citric acid (>99.5%, ACS reagent), sulfuric acid (95%-98%, ACS reagent), tris(2,2′-bipyridyl)dichlororuthenium(II) hexahydrate (99.95%), silver benzoate (99%), silver tetrafluoroborate (98%), and acetonitrile (MeCN, 99.93+%, HPLC grade) (all from Sigma-Aldrich, Milwaukee, WI) were used as received. Synthesis and Characterization of HMTD. As outlined in Scheme 1, HMTD was synthesized by dissolving an appropriate amount of hexamethylenetetramine in an ice bath-cooled 30% H2O2 solution, followed by the addition of anhydrous citric acid into the solution with constant stirring.27 Dried HMTD was characterized with FT-IR spectroscopy and nuclear magnetic resonance (NMR) spectroscopy, where a characteristic C-O stretch10 at 1225 cm-1 and a singlet 1H NMR peak at 4.82 ppm were observed, respectively (see the Supporting Information for more details). [CAUTION: HMTD is very sensitive to explosion when present as a dry solid. It should be handled carefully with appropriate precautions and should not be kept in large quantities. For long-term storage, HMTD should be dissolved or kept wet.] Electrochemical and ECL Studies. Cyclic voltammetry (CV) was performed with a Model 660A electrochemical workstation (CH Instruments, Austin, TX). The ECL signals, along with the CV responses, were measured simultaneously with a homemade ECL instrument, as described previously.19,28 This instrument, combined the 660A electrochemical workstation with a photomultiplier tube (Hamamatsu R928, Japan, biased at -700 V DC) installed under a conventional three-electrode cell, in which a platinum wire was used as the counter electrode, a Ag/Ag+ (10 mM AgNO3 and 0.10 M TBAP in MeCN) was used as the reference electrode (∼0.186 V vs NHE), and a disk composed of either glassy carbon (GC, 3 mm in diameter), platinum (Pt, 2 mm in diameter), or gold (Au, 2 mm in diameter) was used as the working electrode. The working electrode was polished with 0.3-0.05-µm alumina slurry, thoroughly rinsed with water, and dried using Kim wipes facial tissue and then an air blower before each experiment. The internal electrolyte solution and the Vycor tip of the reference electrode were changed periodically, to eliminate possible contamination of other species within the cell. No degassing was needed, because all electrochemical scans were conducted in the positive potential region. RESULTS AND DISCUSSION Cyclic Voltammetric (CV) and ECL Studies of HMTD. The CV and ECL responses of a MeCN solution that consisted of 1.0 mM HMTD, 0.70 mM Ru(bpy)3Cl2, and 0.10 M TBAP electrolyte (27) Schaefer, W. P.; Fourkas, J. T.; Tiemann, B. G. J. Am. Chem. Soc. 1985, 107, 2461–2463. (28) Rosado, D. J., Jr.; Miao, W.; Sun, Q.; Deng, Y. J. Phys. Chem. B 2006, 110, 15719–15723.
Figure 1. CV (black line) and ECL (blue line) responses obtained from 1.0 mM HMTD-0.70 mM Ru(bpy)3Cl2-0.10 M TBAP in MeCN at a 2-mm-diameter platinum electrode with a scan rate of 50 mV/s.
Figure 2. Effect of Ru(bpy)3Cl2 concentration and working electrode material on the ECL intensity of the Ru(bpy)32+-HMTD system in 0.10 M TBAP MeCN. Scan rate ) 50 mV/s, [HMTD] ) 1.0 mM.
Scheme 2. Proposed ECL Mechanism of the Ru(bpy)32+-HMTD System in MeCN upon Anodic Potential Scanninga
a HMTD ) hexamethylene triperoxide diamine, RuII ) Ru(bpy)32+, RuIII ) Ru(bpy)33+, and RuII* ) excited state Ru(bpy)32+*. P1 and P2 are the oxidation products of HMTD free radicals.
are shown in Figure 1. In the anodic potential scanning range of 0-2.0 V vs Ag/Ag+, Ru(bpy)32+ oxidizes reversibly to form Ru(bpy)33+ at a platinum electrode with a half-wave potential of 0.96 V vs Ag/Ag+, whereas HMTD displays an irreversible oxidation at ∼1.7 V (See Figure S2 in the Supporting Information for more details). Because two identical tertiary amine moieties exist in HMTD (recall Scheme 1), an overall two-electron transfer oxidation process is expected, which is consistent with the oxidation current ratio of HMTD to Ru(bpy)32+. The ECL response of the Ru(bpy)32+/HMTD system, which appears at ∼1.45 V and reaches a maximum at ∼1.76 V, is coincident with the oxidation potential of HMTD. Scheme 2 summarizes the proposed ECL mechanism of this system, in which the ECL contribution that is associated with the HMTD dication species, [•HMTD•]2+ (eq 2), relative to the overall ECL emissions (eq 3), could be relatively small, because the oxidation of HMTD•+ to [•HMTD•]2+ is most likely the rate-determining step after the one-electron electro-oxidation of HMTD (eq 1). The ECL intensity of the Ru(bpy)32+/HMTD system is dependent on the added Ru(bpy)32+ concentration, as well as the nature of the working electrode. As revealed in Figure 2, for 1.0 mM HMTD, the maximum ECL appears at 0.70 mM Ru(bpy)3Cl2 for all three working electrodes studied, in which the Pt electrode shows slightly stronger ECL responses than the Au electrode does, and the GC electrode gives relatively poor ECL signals. Similar behavior has been reported previously for the Ru(bpy)32+-TPrA system in MeCN.29 (29) Miao, W.; Bard, A. J. Anal. Chem. 2004, 76, 5379–5386.
Figure 3. (A) CV (black line) and ECL (blue line) responses obtained from 1.0 mM HMTD-0.70 mM Ru(bpy)3Cl2-0.10 M TBAP in MeCN in the presence of 7.0 mM AgNO3, and (B) CV of 7.0 mM AgNO3 in MeCN that contains 0.10 M TBAP at a 2-mm-diameter Pt electrode with a scan rate of 50 mV/s.
ECL Enhancement with AgNO3. As shown in Figures 3A and 4, after the addition of AgNO3 into a MeCN solution that contains HMTD and Ru(bpy)3Cl2, a significant enhancement in the ECL signal is observed. For example, in 1.0 mM HMTD-0.70 mM Ru(bpy)3Cl2-0.10 M TBAP MeCN, ∼9× and 27× increases in the ECL intensity were obtained in the presence of 2.0 mM and 7.0 mM AgNO3, respectively, with respect to that in the absence of added AgNO3. The ECL peak intensity is linearly proportional to the concentration of AgNO3 added over a concentration range of 0-7.0 mM (see Figure 4a). In the absence of HMTD, no ECL background (when [AgNO3] e 2.0 mM) or values equal to 1.4 mM. These data suggest that the ECL enhancement is primarily due to NO3ions. In fact, the contribution from Ag+ ions is generally
