Sonochemically induced decomposition of energetic materials in aqueous media

Chemosphere 50 (2003) 1107–1114 www.elsevier.com/locate/chemosphere

Sonochemically induced decomposition of energetic materials in aqueous media Lala R. Qadir, Elizabeth J. Osburn-Atkinson 1, Karen E. Swider-Lyons, Veronica M. Cepak 2, Debra R. Rolison * Surface Chemistry Branch (Code 6170), Naval Research Laboratory, 4555 Overlook Avenue, SW, Washington, DC 20375, USA Received 14 September 2001; received in revised form 5 September 2002; accepted 4 November 2002

Abstract This study demonstrates that ultrasound rapidly degrades the energetic compounds RDX (cyclo-1,3,5-trinitramine2,4,6-trimethylene) and ADN (ammonium dinitramide) in aqueous microheterogeneous media. The conditions for effective degradation of these nitramines, as monitored by UV absorption spectroscopy, were determined by varying sonication time, the heterogeneous phase and its suspension density, and the concentration of NaOH. In the presence of 5 mg/ml of aluminum powder and at pH
12 (10 mM NaOH), 74% of the RDX and 86% of the ammonium dinitramide (ADN) in near-saturated solutions decompose within the first 20 min of sonication (20 kHz; 50 W; 6 5 °C). Sonication without Al powder and base yields minimal degradation of either RDX and ADN (
5–10%) or the nitrite/ nitrate ions that are expected byproducts during RDX and ADN degradation. Sonication at high pH in the presence of dispersed aluminosilicate zeolite, alumina, or titanium dioxide also yields minimal degradation. Preliminary electrochemical studies and product analyses indicate that in situ ultrasonic generation of metallic aluminum and/or aluminum hydride drives reductive denitration of the nitramines. Sonochemical treatment in the presence of a reductant offers an effective and rapid waste remediation option for energetic waste compounds. Published by Elsevier Science Ltd. Keywords: High explosives; Ultrasound; Reductive denitration; Environmental remediation; RDX; Ammonium dinitramide; Nitramines

1. Introduction Recent global demilitarization has created a need for effective destruction or remediation of the high explosive materials used in nuclear weapons delivery systems and rocket propellants. An estimated 3:13  108 kg of munitions waste exists in military stockpiles alone (Heil-

*

Corresponding author. Tel.: +1-202-767-3617; fax: +1-202767-3321. E-mail address: [email protected] (D.R. Rolison). 1 Present address: Department of Chemistry, Linfield College, McMinnville, OR 97128, USA. 2 Present address: Eltron Research, Inc., 4600 Nautilus Court South, Boulder, CO 80301, USA. 0045-6535/03/$ - see front matter Published by Elsevier Science Ltd. PII: S 0 0 4 5 - 6 5 3 5 ( 0 2 ) 0 0 7 7 0 - 1

mann et al., 1996). These hazardous waste stockpiles present both environmental and health problems, especially since some explosives, such as RDX, cyclo-1,3,5 trinitramine-2,4,6-trimethylene, are known carcinogens (Heilmann et al., 1996; Hundal et al., 1997). An ancillary hazard is the wastewater that results after energetic manufacturing and washdown, explosive melting, and steam-cleaning operations of reject warheads. Current treatment technologies include open-air incineration, alkaline hydrolysis, adsorption on carbon beds, and advanced oxidation, but these methods are often capital and energy intensive, and can produce concentrated toxic byproducts that need to be treated further (Heilmann et al., 1996; Hundal et al., 1997). This study focuses on the ultrasonic treatment of high explosives in aqueous media with emphasis on the

1108

L.R. Qadir et al. / Chemosphere 50 (2003) 1107–1114

Fig. 1. Molecular formulae for the energetic nitramines, RDX and ADN.

nitramines RDX and ammonium dinitramide (ADN) (Fig. 1). RDX, a white crystalline heterocyclic solid first synthesized in 1899 for medicinal use, is a stable energetic material with nearly 130% of the explosive power of trinitrotoluene, TNT (McLellan et al., 1988). Because of the inherent stability of RDX and its relative insensitivity to shock, it does not readily degrade by conventional treatment methods. Bioremediation methods take days and sometimes weeks at elevated temperatures to achieve significant degradation of RDX (Freedman and Sutherland, 1998). ADN, another shock-resistant high explosive and effective oxidizer, is a stable, white ionic salt used in explosives and fuels (L€ obbecke et al., 1997). Previous studies have shown that the localized and extreme conditions generated by ultrasound, as well as the resulting formation in aqueous media of hydrogen and hydroxyl radicals, create chemical species that can degrade chemical contaminants in wastewater (Bhatnagar and Cheung, 1994; Cheung and Krup, 1994; Hoffmann et al., 1996; Kr€ uger et al., 1999). Sonochemical effects are largely attributed to phenomena that result from cavitation. During sonication, an oscillating pressure field drives periodic cycles of expansion and contraction of gas-filled microbubbles in the liquid. The implosive collapse of these microbubbles (cavitation) generates high local temperatures and pressures (5000 °C and 1000 atm, respectively) that drive sonochemical reactions directly (Suslick, 1990, 1997) or indirectly through reactants generated via these extreme conditions (Hoffmann et al., 1996). Ultrasonic degradation of organic species in aqueous solutions may follow multiple reaction pathways, including: (1) a high concentration of oxidizing species such as hydroxyl radicals; (2) pyrolytic decomposition by the high localized temperatures and pressures; or (3) supercritical water oxidation (Hoffmann et al., 1996). Substances that have been remediated sonochemically include pesticides, phenols, esters, TNT, and chlorinated organic compounds (Sierka, 1984; Hoffmann et al., 1996). Ultrasonic degradation of chlorinated volatile organic compounds (VOCs) and chlorofluorocarbons (CFC-11 and CFC-13) proceeds via cleavage of the carbon–halogen bonds (Cheung and Krup, 1994; Kr€ uger

et al., 1999). It is therefore feasible that ultrasound could also remediate RDX and ADN, because both molecules are rendered non-hazardous by denitration, i.e., cleavage of the nitramine moiety (N–NO2 ) to form nitrites (Owens and Sharma, 1980; Heilmann et al., 1996; Hundal et al., 1997; Thompson and Doraiswamy, 1999). Reduction of the nitro group is one means of achieving cleavage of the N–NO2 bond (Hundal et al., 1997). Previous studies indicate that Fe metal promotes remediation of RDX in soil via reductive denitration (Hundal et al., 1997). We choose to test Al powder because its reducing power (with an EMF of )1.67 V) is far greater than that of Fe metal ()0.44 V) (Bard et al., 1985), and ultrasound can be used to generate, in situ, a fresh surface of metallic aluminum, which can then function as a reducing agent. The ultrasonic degradation of RDX and ADN is probed by studying the UV absorbance of filtrates derived from aqueous microheterogeneous media as a function of sonication time, pH, and the nature and suspension density of the dispersed particulate. The effect of aluminosilicate zeolite, alumina, or titanium dioxide on the degradation effectiveness was also studied as experimental controls for aluminum powder. Plausible mechanisms for the denitration of RDX and ADN by ultrasound in the presence of Al powder are proposed on the basis of the results obtained from absorbance measurements (to monitor the loss of RDX or ADN), X-ray photoelectron spectroscopy, XPS (to determine nitrogen speciation in the post-reaction filtrates), and electrochemical studies (as a second means to generate Al(0) and aluminum hydride in aqueous solution).

2. Experimental procedures 2.1. Chemicals Analytical reagent grade cyclo-1,3,5-trinitro-1,3,5triamine (Naval Research Laboratory, Washington, DC and Naval Surface Warfare Center, Indian Head, MD) and ADN (NRL-DC) were used as received. Stock solutions of RDX and ADN (127 and 151 lM, respectively) were made with 18 MX cm water (Barnstead Nanopure) and stirred overnight to equilibrate. The solid ADN and all ADN solutions were stored and studied in a dark environment to avoid photolytic decomposition. Standard solutions of sodium nitrate (Fisher Scientific; 0.77 mM) and sodium nitrite (Fisher Scientific; 0.94 mM) were prepared in order to test the stability of the nitrate and nitrite ions when exposed to ultrasound and Al metal. A stock solution of NaOH (1.00 M) was prepared by diluting 5.15 ml of a freshly opened solution of 19.4 M NaOH (Fisher Scientific) to 100 ml.

L.R. Qadir et al. / Chemosphere 50 (2003) 1107–1114

2.2. Equipment and instrumentation The sonochemical reaction vessel consisted of a 25 ml pear-shaped, three-necked glass flask, into which the ultrasonic horn (titanium alloy, 3.5-mm-diameter tip) and temperature probe (Cole Palmer, Inc. model 850216 thermistor) were inserted. The reactor was placed into a circulatory bath (Neslab model RTE-110) with a set temperature of 5 °C and irradiated with 50 W, 20 kHz ultrasound (Sonics and Materials Inc. model VC50T ultrasonicator). The temperature was thermostatted below room temperature because sonochemical reactions are typically not favored at high temperatures, due to cushioning of the imploding bubble by the greater kinetic energy of the vapors within the bubble (McLellan et al., 1988; Hoffmann et al., 1996). All aqueous standards and reaction samples were analyzed by UV– vis spectroscopy (Hewlett-Packard photodiode array model 8452A). A limited set of standards and samples was analyzed by XPS (Surface Science Instrument model SSX-100-03; AlKa X-rays).

sampling procedures produced the lowest minimum replicate error (8%) in the absorbance at the monitoring wavelength. We define a change that is <8% as indicating that no RDX or ADN degradation occurs. The samples were monitored at secondary UV absorption maxima (236 nm (RDX) and 286 nm (ADN)) to avoid interference from hydroxide, nitrate, and nitrate ions and particulate matter that also absorb between 190 and 220 nm. Calibration curves for RDX, ADN, NO 2, and NO species were obtained using standard solutions. 3 Aliquots (50 ll) of nitrate/nitrate-derived standards and selected post-sonication samples were evaporated onto 1 cm  1 cm  0:1-mm indium foil (Alfa, 99.99%) for XPS analysis. 3. Results and discussion Aqueous solutions of RDX and ADN exhibit two UV absorption maxima (RDX: kmax ¼ 198 and 236 nm; ADN: kmax ¼ 216 and 286 nm), as seen in Fig. 2. 3.1. Effect of non-Al(0) solids on RDX degradation Microheterogeneous media containing TiO2 , alumina, or NaY were studied as control solids because 1.2

RDX λmax = 236 nm

1.0

Absorbance

The following powders were used as received in the heterogeneous ultrasound studies: aluminum powder (historical supply), c-alumina (Alfa-Aesar), sodium-ioncompensated synthetic type Y aluminosilicate zeolite (NaY; a gift from UOP, LLC), and titanium dioxide (P25; a gift from Degussa). The mean diameter of the Al powder, as determined by scanning electron microscopy (Leo model 1550 electron microscope), is <10 lm with a wide size distribution of agglomerated primary particles. Aluminum hydride was prepared by electrochemically charging a 0.133-mm thick strip (5:2 cm  1:2 cm) of Al foil at )4.00 V vs Pt mesh in a 10 mM NaOH solution.

1109

0.8 0.6 0.4

2.3. Experimental protocol

0.2

The reaction samples consisted of 9.9 ml of the energetics-contaminated wastewater plus 0.10 ml of 1.00 M NaOH. Controls for neutral solutions of RDX, ADN, NaNO2 , and NaNO3 were run with 0.10 ml of water substituted for the NaOH stock solution to maintain the same concentration of contaminant in the sample. Powders of Al, Al2 O3 , NaY, or TiO2 were added in 5-, 50-, or 500-mg portions to 10-ml samples of the energetics wastewater. The reaction samples were sonicated for 60 min with 300-ll aliquots taken every 5–10 min. After each run, aliquots from RDX samples were diluted to 750 ll, bringing the concentration to 50.8 lM (in the absence of any molecular degradation). This dilution was based on the molar absorptivity of the RDX chromophore. Aliquots taken from the sonochemical reaction of ADN were not diluted. All controls and post-reaction samples were centrifuged for 30 min and pressure filtered through a 0.02-lm disk (Whatman Anotop). These

0.0 200

225

250

275

300

325

350

375

400

Wavelength / nm 1. 2

ADN λmax = 286 nm

Absorbance

1. 0 0. 8 0. 6 0. 4 0. 2 0. 0

200

250

300

350

400

450

500

Wavelength (nm ) Fig. 2. UV–vis absorption spectra of aqueous solutions of RDX (63.5 lM) and ADN (74.8 lM).

1110

L.R. Qadir et al. / Chemosphere 50 (2003) 1107–1114

Fig. 4. UV absorption spectra of filtrates derived from an aqueous pH 12 microheterogeneous medium containing RDX (127 lM) and 5 mg/ml of Al as a function of sonication time; percentage of degradation is monitored at 236 nm.

Fig. 3. Percentage of RDX (127 lM) decomposed after a 60min ultrasonic treatment, as a function of dispersed solid and pH. Degradation effectiveness is markedly enhanced for sonochemical reaction with dispersed Al(0) powder under basic pH. Replicate error is 8%.

similar chemical species may be generated in situ when sonicating an aqueous microheterogeneous medium containing Al powder with a Ti-alloy horn in a glass vessel. Sonication of aqueous media containing any of these solids yields less than 11% breakdown of RDX after 60 min of sonication, regardless of suspension density and even under alkaline pH (see Fig. 3). In addition, RDX did not degrade after exposing aqueous RDX solution (127 lM) to an as-received (i.e., electrochemically uncharged) strip of Al(0) foil for 60 min without sonication. 3.2. Sonochemistry of RDX As determined by UV absorption, RDX is stable to sonication for 60 min in an aqueous solution. Despite the expectation of ‘‘incineration in a bubble’’ due to the extreme local temperatures that arise upon cavitation, ultrasonic treatment alone does not generate and sustain conditions sufficient to degrade RDX, even with the addition of either OH or suspended Al powder. A prior study of RDX and TNT degradation using ultrasound also indicated minimal degradation of RDX (Sierka, 1984).

When a 127 lM RDX solution, 10 mM in NaOH, is sonicated in combination with Al(0) powder at a suspension density of 5 mg/ml, 74% of the compound degrades within the first 20 min of sonication (Fig. 4). No Al powder visibly remains in the reaction vessel after 60 min of treatment. In contrast, RDX neither decomposes upon stirring alkaline RDX solutions with Al powder in the absence of sonication nor is the Al powder digested. These results indicate that RDX is degraded by the sonochemistry created by the combination of OH , Al powder, and ultrasound. At a lower suspension density of Al powder (0.5 mg/ ml), the concentration of RDX decreases by only 40% after 60 min of sonication, which also indicates that metallic Al is a reactant. Suspensions containing more than 5 mg/ml of Al powder show neither significantly improved efficiency nor an increased extent of degradation. This leveling off could arise for two reasons: (1) high suspension densities of solid would be expected to lower the effectiveness of cavitational disruption of the passive oxide layer of aluminum oxide on the aluminum; and (2) at pH 12, because of Eq. (1), an insufficient concentration of OH may be present at high suspension densities of Al powder to react at the stoichiometry necessary to form aluminate: AlðOHÞ3 þ OH  AlðOHÞ 4

ð1Þ

3.3. Sonochemistry of ADN Experiments with ADN indicate that it behaves similarly to RDX under sonochemical treatment. Insignificant levels of degradation are obtained when ADN is sonicated for 60 min in the absence of NaOH and Al powder or in the presence of non-Al(0) solids. However,

L.R. Qadir et al. / Chemosphere 50 (2003) 1107–1114

2.0

-

-

a

NO 2

NO 3

0 min

1.6

1 0 min

0 min

2 0 min

1.2

4 0 min 6 0 min

0.8

5 min

86%

intensity (au)

Absorbanc e

1111

degradation

0.4 0.0 210

245

280

315

350

385

10 min

20 min

420

Wavelength / nm 30 min

1.0

b

Absorbanc e

0.8

40 min

0.6 410

0.4

402

398

Fig. 6. N 1s X-ray photoelectron spectra of evaporated filtrate derived from aliquots sampled during ultrasonic treatment of  NO 2 þ NO3 =10 mM NaOH/5 mg/ml of Al powder. N 1s intensity is normalized to that of Na 1s.

0.2 0.0

406

Binding energy (eV)

0

10

20

30

40

50

60

Sonication Time / min Fig. 5. (a) UV–vis absorption spectra of filtrate sampled during ultrasonic treatment of 151 lM ADN/10 mM NaOH/5 mg/ml of Al powder; (b) change in absorbance at 286 nm as a function of sonication time.

sonicating a solution of 151 lM ADN/10 mM NaOH/5 mg/ml of Al powder for 60 min produces an 86% loss in ADN concentration (Fig. 5a and b). In alkaline solutions with Al powder and a lower concentration of ADN (50 lM), 100% degradation is achieved within 20 min of sonication. As seen with RDX, increasing the suspension density of aluminum powder above 5 mg/ml does not improve the effectiveness of degradation. 3.4. Sonochemistry of nitrite/nitrate The relatively monotonic and clean loss of the UV absorption features for RDX and ADN (as seen in Figs. 4 and 5a) indicate that the decomposition products are either not UV active or are active in a region masked by UV absorption by OH . Nitrate and nitrite ions, which are possible RDX and ADN decomposition products, are UV active, but in the same wavelength region as  OH . The stability of NO 2 or NO3 to ultrasound (as analyzed by XPS) was explored using the same conditions developed for the RDX and ADN decomposition studies.

Sonicating aqueous solutions of NaNO2 and NaNO3 for 60 min in the absence of base or Al powder produces no significant change in either their UV absorption or N 1s X-ray photoelectron spectra (which exhibit binding energies consistent with assignment as N in NO 2 and NO (Wagner et al., 1979)). This lack of direct ultra3 sonic reaction of nitrate, and especially nitrite, is in agreement with the RDX and ADN controls in pure water or at pH 12 in which no significant sonochemical reactivity occurs. When a pH 12 solution of NaNO3 /NaNO2 is sonicated in the presence of 5 mg/ml of Al powder, the total nitrogen content in the filtrates decreases as a function of sonication time (Fig. 6). On the basis of the N 1s peak ratios of nitrite to nitrate, the nitrite species undergoes reaction first. The ultimate disappearance of NO 2 and NO species during sonication in the presence of OH 3 and Al indicates that nitrogen-containing product species do not remain in the liquid, and escape from the reaction vessel in the form of nitrogenous gases. XPS analysis of filtrate derived from RDX ultrasonic reactions exhibits no N 1s signal above background. 4. Proposed reaction mechanisms Sonochemistry involves complex, coupled physical phenomena and chemical reactions that are not completely understood, which makes it difficult to identify

1112

L.R. Qadir et al. / Chemosphere 50 (2003) 1107–1114

the precise mechanisms that govern the ultrasonically induced breakdown of the energetic materials RDX and ADN in the presence of OH and Al powder. Currently there is not even a mechanistic understanding of the conflagration reactions involving RDX and similar energetic materials (Chakraborty et al., 2000). Prior decomposition studies show that NO 2 is the primary byproduct of shock-induced and photolytic decomposition of RDX (Owens and Sharma, 1980), as well as of alkaline hydrolysis (Heilmann et al., 1996). 4.1. Sonochemically induced reactions In sonolysis, the lifetime of ultrasonically generated chemical radicals exceeds the lifetime of the cavitating bubble, which permits radical-driven chemical reactions to be triggered (Thompson and Doraiswamy, 1999). During cavitational implosion, pure water yields H atoms and OH radicals upon the thermal dissociation of water vapor, Eq. (2)––these radicals are strong reductants and oxidants and induce many of the sonochemical reaction pathways described in the literature (Heilmann et al., 1996; Hoffmann et al., 1996): H2 O  H þ OH

ð2Þ

However active these ultrasonically derived reactants have proven to be with other organic compounds, the complete lack of ultrasonic reactivity in the absence of Al with either aqueous or alkaline solutions of RDX,  ADN, NO 2 , and NO3 indicates that an ultrasoundactivated water/radical mechanism is insufficient to explain our results. 4.2. Aluminum(0) as reducing agent The reducing power of Al metal is difficult to harness due to its passive oxide. Ultrasound can activate highly reactive metal surfaces that are otherwise protected by a passive oxide coating (Preece and Hansson, 1981; Souza et al., 1998; Thompson and Doraiswamy, 1999), because of asymmetric cavitation near extended liquid–solid interfaces (Mead et al., 1976). When a deformed cavity collapses, it emits high-velocity microjets of liquid (nearing 100 m/s) that crack the passive oxide layer to expose a fresh metal surface that is briefly available for reductive chemistry, including electron transfer to organic substrates (Luche, 1994). Previous research has demonstrated that mechanical agitation, even of the same power consumption, does not comparably break the passive oxide layer (Carvalho et al., 1995; Hagenson and Doraiswamy, 1998). It is already established that selective reduction of nitrate and nitrite by Al(0) occurs under alkaline conditions (Murphy, 1991) to generate nitrogenous gasphase products: NOx , NH3 , and N2 (Moon and Pyun,

1999) (Eqs. (3)–(5)). This chemistry is accompanied by dissolution of aluminum metal in the presence of hydroxide to form water-soluble aluminate (Eq. (1)): 3NO 3 þ 2Al þ 3H2 O ! 3NO2 ðgÞ þ 2AlðOHÞ3

ð3Þ

 NO 2 þ 2Al þ 5H2 O ! NH3 þ 2AlðOHÞ3 þ OH

2NO 2

þ 2Al þ 4H2 O ! N2 ðgÞ þ 2AlðOHÞ3 þ 2OH

ð4Þ 

ð5Þ The ultrasonic reactivity of microheterogeneous alkaline media containing nitrate and nitrite and Al powder indicates that cavitation does indeed generate fresh metal surfaces to induce reductive chemistry. Reductive denitration of RDX and ADN with ultrasonically generated Al(0) is in keeping with these classes of reactions. 4.3. Aluminum hydride mechanism We also explored a mechanism involving sonochemically generated aluminum hydride (AlHx ) as a candidate for the decomposition reaction since the Al–H bond has a high reductive potential and sonochemical generation of Al(0) and H should also generate AlHx . Electrogeneration of aluminum hydride in alkaline media has been reported previously (Perrault, 1980). When we exposed an alkaline RDX solution, without sonication, to a hydrided Al foil (as pre-generated by electroreduction), the concentration of RDX decreased by 18% after 60 min and 23% after 120 min. The ability of electrogenerated AlHx to react with RDX indicates that an Al–H mechanism is a plausible pathway in the reaction and that the direct electrochemical generation of Al–H may also be used to reduce nitramines. The fact that a charged foil has a sustained temporal effect on RDX decomposition indicates that sub-surface hydride forms and is available as a pool of reactant over time. 4.4. Mechanistic assessment Ultrasonically induced pyrolysis/thermal decomposition of the energetics is not occurring on the time scale and at the ultrasonic power levels of our experiments. The results reported here demonstrate that ultrasoundexposed Al(0) destroys energetic compounds in an alkaline microheterogeneous medium. The persistence of RDX degradation at electrogenerated aluminum hydride implies that the ultrasonic conditions may also generate this reductant by exposing aluminum metal in the presence of sonochemical reactants, such as H . From our studies, we posit that ultrasound generates fresh Al metal, which then reductively denitrates RDX and ADN, as postulated in Fig. 7. The presence of OH is a necessary component of the mechanistic degradation pathway. Both Al(0)-driven electron-transfer reactions with nitrite/nitrate and elect-

L.R. Qadir et al. / Chemosphere 50 (2003) 1107–1114

1113

Fig. 7. Representative diagram of proposed Al/sonochemical reaction pathway.

rochemical generation of AlHx require alkaline conditions, which substantiates the proposed mechanism in which ultrasonically generated Al(0) reacts with RDX and daughter products such as nitrite and nitrate via reductive chemistry. The lack of a N 1s peak in the XPSanalyzed aliquots confirms that the end products of the ultrasonically treated alkaline microheterogeneous media are in the gas phase. A sonochemical approach has an advantage over conventional treatment methods in that the target molecule needs not be extracted from an aqueous phase, i.e., wastewater streams, as would be required in combustion, or other remediation efforts. These field conditions are also not suitable for electrochemical processes, which can be quenched by impurities. Furthermore, most of the RDX and ADN samples, at concentrations in water approaching saturation, degraded within the first 20 min and all samples were reacted at low temperatures. Ultrasound may be a feasible alternative to previous treatments, although more work is required to determine the efficiency of the sonochemical destruction of energetic materials in aqueous waste upon scale-up.

Acknowledgements This research was supported by the Office of Naval Research. L.R.Q. initiated this project as a participant from La Plata High School (La Plata, MD) in the NRL Science and Engineering Apprentice Program (SEAP). E.J.O.-A. was an ASEE Post-doctoral Associate (1996– 1997) and V.M.C. was an NRC Post-doctoral Associate (1998–1999). The authors extend their thanks to Jeff Hilgert (Branson Ultrasonics) for his loan of an ultrasonicator to establish feasibility and to Doug Elstrodt (Naval Surface Warfare Center, Indian Head MD) for the generous gift of RDX.

References Bard, A.J., Parsons, R., Jordan, J., 1985. In: Standard Potentials in Aqueous Solution. Marcel Dekker, New York, p. 566, 407.

Bhatnagar, A., Cheung, H.M., 1994. Sonochemical destruction of chlorinated-C1 and chlorinated-C2 volatile organic-compounds in dilute aqueous-solution. Environ. Sci. Technol. 28, 1481–1486. Carvalho, L.R.F., Souza, S.R., Martinis, B.S., Korn, M., 1995. Monitoring of the ultrasonic irradiation effect on the extraction of airborne particulate matter by ion chromatography. Anal. Chim. Acta 317, 171–179. Chakraborty, D., Muller, R.P., Dasgupta, S., Goddard, W.A., 2000. The mechanism for unimolecular decomposition of RDX (1,3,5-trinitro-1,3,5-triazine), an ab initio study. J. Phys. Chem. A 104, 2261–2272. Cheung, H.M., Krup, S., 1994. Sonochemical destruction of CFC-11 and CFC-113 in dilute aqueous-solution. Environ. Sci. Technol. 289, 1619–1622. Freedman, D.L., Sutherland, K.W., 1998. Biodegradation of hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX) under nitrate-reducing conditions. Water Sci. Technol. 38, 33–40. Hagenson, L.C., Doraiswamy, L.K., 1998. Comparison of the effects of ultrasound and mechanical agitation on a reacting solid–liquid system. Chem. Eng. Sci. 53, 131–148. Heilmann, H.M., Weismann, U., Stenstrom, M.K., 1996. Kinetics of the alkaline hydrolysis of high explosives RDX and HMX in aqueous solution and adsorbed to activated carbon. Environ. Sci. Technol. 30, 1485–1492. Hoffmann, M.R., Hua, I., H€ ochemer, R., 1996. Application of ultrasonic irradiation for the degradation of chemical contaminants in water. Ultrason. Sonochem. 3, S163–S172. Hundal, L.S., Singh, J., Bier, E.L., Shea, S.D., Comfort, S.D., Powers, W.L., 1997. Removal of TNT and RDX from water and soil using iron metal. Environ. Pollut. 97, 55–64. Kr€ uger, O., Shulze, Th.-L., Peters, D., 1999. Sonochemical treatment of natural ground water at different high frequencies: preliminary results. Ultrason. Sonochem. 6, 123– 128. L€ obbecke, S., Keicher, T., Krause, H., Pfeil, A., 1997. The new energetic material ammonium dinitramide and its thermal decomposition. Solid State Ionics 101, 945–951. Luche, J.L., 1994. Effect of ultrasound on heterogeneous systems. Ultrason. Sonochem. 1, S111–S117. McLellan, W., Hartley, W.R., Brower, M., 1988. Health Advisory for Hexahydro-1,3,5,-tetranitro-1,3,5-triazine. Technical Report PB90-273533; Office of Drinking Water, US Environmental Protection Agency, Washington, DC. Mead, E.L., Sutherland, R.G., Verrall, R.E., 1976. Effect of ultrasound on water in the presence of dissolved gases. Can. J. Chem. 54, 1114–1120. Moon, S.-M., Pyun, S.-I.J., 1999. Effects of applied anodic potential and pH on the repassivation kinetics of pure

1114

L.R. Qadir et al. / Chemosphere 50 (2003) 1107–1114

aluminium in aqueous alkaline solution. Solid-State Electrochem. 3, 104–110. Murphy, A.P., 1991. Chemical removal of nitrate from water. Nature 350, 223–224. Owens, F.J., Sharma, J., 1980. X-ray photoelectron-spectroscopy and paramagnetic-resonance evidence for shockinduced intramolecular bond breaking in some energetic solids. J. Appl. Phys. 51, 1494–1497. Perrault, G.G., 1980. Potentiometric determination of the standard free enthalpy of formation for aluminum trihydride. J. Electrochem. Soc. 127, C137. Preece, C.M., Hansson, I.L.H., 1981. A metallurgical approach to cavitation erosion. Adv. Mech. Phys. Surf. 1, 190–254. Sierka, R.A., 1984. The high-temperature treatment of trinitrotoluene (TNT) and cyclotrimethylene-trinitramine (RDX)

with ozone and ultrasound. Ozone-Sci. Eng. 6, 275– 290. Souza, S.R., Korn, M., de Carvalho, L.R.F., Korn, M.G.A., Taveres, M.F.M., 1998. Multi-parametric evaluation of the chemical ultrasonic irradiation effects in a heterogeneous metal–water system. Ultrasonics 36, 595–602. Suslick, K.S., 1990. Sonochemistry. Science 247, 1439–1445. Suslick, K.S., 1997. Sonoluminescence and sonochemistry. IEEE Ultrasonics Symp. Proc. 1, 523–532. Thompson, L.H., Doraiswamy, L.K., 1999. Sonochemistry: Science and Engineering. Ind. Eng. Chem. Res. 38, 1215– 1249. Wagner, C.G., Riggs, W.M., Davis, L.E., Moulder, J.F., Muilenberg, G.E., 1979. Handbook of X-ray Photoelectron Spectroscopy. Perkin-Elmer Corp. (Physical Electronics Division), Eden Prairie, MN.