Published in J. Environ. Qual. 33:1362-1368 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA
TECHNICAL REPORTS
Organic Compounds in the Environment
Sorption and Stability of the Polycyclic Nitramine Explosive CL-20 in Soil
Vimal K. Balakrishnan,
Fanny Monteil-Rivera,
Mathieu A. Gautier and
Jalal Hawari*
Biotechnology Research Institute, National Research Council of Canada, 6100 Royalmount Avenue, Montréal, QC, Canada H4P 2R2
* Corresponding author (Jalal.Hawari{at}cnrc-nrc.gc.ca).
Received for publication October 20, 2003.
 |
ABSTRACT
|
|---|
The polycyclic nitramine CL-20 (2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazaisowurtzitane) is being considered for use as a munition, but its environmental fate and impact are unknown. The present study consisted of two main elements. First, sorption–desorption data were measured with soils and minerals to evaluate the respective contributions of organic matter and minerals to CL-20 immobilization. Second, since CL-20 hydrolyzes at a pH of >7, the effect of sorption on CL-20 degradation was examined in alkaline soils. Sorption–desorption isotherms measured using five slightly acidic soils (5.1 < pH < 6.9) containing various amounts of total organic carbon (TOC) revealed a nonlinear sorption that increased with TOC [Kd (0.33% TOC) = 2.4 L kg–1; Kd (20% TOC) = 311 L kg–1]. Sorption to minerals (Fe2O3, silica, kaolinite, montmorillonite, illite) was very low (0 < Kd < 0.6 L kg–1), suggesting that mineral phases do not contribute significantly to CL-20 sorption. Degradation of CL-20 in sterile soils having different pH values increased as follows: sandy agricultural topsoil from Varennes, QC, Canada (VT) (pH = 5.6; Kd = 15 L kg–1; 8% loss) < clay soil from St. Sulpice, QC, Canada (CSS) (pH = 8.1; Kd = 1 L kg–1; 82% loss) < sandy soil provided by Agriculture Canada (SAC) (pH = 8.1, Kd = approximately 0 L kg–1; 100% loss). The faster degradation in SAC soil compared with CSS soil was attributed to the absence of sorption in the former. In summary, CL-20 is highly immobilized by soils rich in organic matter. Although sorption retards abiotic degradation, CL-20 still decomposes in soils where pH is >7.5, suggesting that it will not persist in even slightly alkaline soils.
Abbreviations: CSS, clay soil from St. Sulpice • QC, Canada • FS, sandy forest soil provided by a local supplier • FSB, sandy forest soil from Boucherville, QC, Canada • GS, sandy garden soil obtained from a local supplier • HPLC, high performance liquid chromatography • Kd, distribution coefficient • KF, Freundlich adsorption coefficient • Koc, organic-carbon-normalized distribution coefficient • Kow, n-octanol–water distribution coefficient • SAC, sandy soil provided by Agriculture Canada • SSL, Sassafras sandy loam • TOC, total organic carbon • VT, sandy agricultural topsoil from Varennes, QC, Canada
|
INTRODUCTION
|
|---|
THE POLYCYCLIC NITRAMINE CL-20 (Fig. 1) is a highly energetic explosive that is currently being considered for large-scale production and military applications. Presently, due to the relative novelty of CL-20, very little is known about its environmental fate. In general, the manufacturing and usage of munitions has resulted in severe contamination of both soils and ground water (Haas et al., 1990; Myler and Sisk, 1991). Like the monocyclic nitramines RDX (hexahydro-1,3,5-trinitro-1,3,5-triazine) and HMX (octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine) (Fig. 1) that have been reported to be toxic to a number of ecological receptors (Yinon, 1990; Talmage et al., 1999), CL-20 was recently found to be toxic to the earthworm (Robidoux et al., 2004). In contrast to currently used munitions, the adverse environmental consequences of widespread usage of CL-20 can be limited by investigating its transport, transformation, and impact in soil.
Recently, we showed that CL-20 readily underwent denitration under alkaline conditions in water (pH = 10) (Balakrishnan et al., 2003) at a faster rate than did the monocyclic nitramines RDX and HMX. From an environmental perspective, an extension of this investigation to natural media that also encompasses the role of sorption–desorption on CL-20 hydrolysis in soil would be useful.
Sorption of organic chemicals to soil is a process that can affect mobility, degradation, and toxicity by reducing availability. A fundamental understanding of sorption and desorption mechanisms is therefore essential for accurate prediction of the fate and impact of organic contaminants in soils and ground water. Several soil properties, including organic carbon content, pH, cation exchange capacity (CEC), and the amount and type of clay may affect the sorption of organic compounds. Of these properties, organic carbon and clay minerals have been shown to have the highest influence on sorption of nonionic molecules (Huang et al., 1998; Weber et al., 1998; Karapanagioti et al., 2001). Previous studies performed with RDX (Leggett, 1985; Sheremata et al., 2001) and HMX (Brannon et al., 2002; Monteil-Rivera et al., 2003) revealed that both nitramines had a very low tendency to sorption and that this sorption was governed by the clay rather than the organic content of soil. To our knowledge, the sorption–desorption behavior of CL-20 in soils has not been reported in the literature.
In this study, we first measured sorption–desorption isotherms for CL-20 with five slightly acidic soils that varied in organic content. Then the extent of sorption on minerals was measured to evaluate the respective contribution of organic matter and minerals in immobilizing the explosive. Finally, we investigated the effect of sorption–desorption on the stability of CL-20 in sterile soils having different pH values.
|
MATERIALS AND METHODS
|
|---|
Chemicals
CL-20 (purity 99.3%, determined by reverse phase high performance liquid chromatography [HPLC]) was obtained from A.T.K. Thiokol Propulsion (Brigham City, UT), as an 
-form (
95% as determined by IR). Silica (Grade 60, 60–200 mesh), Fe(III) oxide (5 µm, >99% pure), kaolinite, and montmorillonite K10 were obtained from Aldrich Chemicals (St. Louis, MO) and used as received. Monoclinic illite shale {(K,H3O)(Al,Mg,Fe)2(Si,Al)4O10[(OH)2,H2O]} was obtained from Ward's Natural Science Establishment (Rochester, NY), ground, and passed through a 2-mm sieve. The pH of the powdered illite (initially = 8.60) was adjusted to 5.4 using HCl in the sorption experiments. All other chemicals were of reagent grade and used without purification.
Soils
Seven soils were used in this study: (i) a Sassafras sandy loam (SSL) (fine-loamy, siliceous, semiactive, mesic Typic Hapludults) sampled in an uncontaminated open grassland on the property of U.S. Army Aberdeen Proving Ground, (Edgewood, MD); (ii) a sandy agricultural topsoil from Varennes, QC, Canada (VT); (iii) a sandy forest soil from Boucherville, QC, Canada (FSB); (iv) a sandy forest soil provided by a local supplier (FS); (v) a sandy garden soil obtained from a local supplier (GS); (vi) a clay soil from St. Sulpice, QC, Canada (CSS); and (vii) a sandy soil provided by Agriculture Canada (SAC). Soils were passed through a 2-mm sieve and air-dried before using. The soils differed with respect to total organic carbon, pH levels, and sand, silt, and clay content (Table 1). A portion of the VT, CSS, and SAC soils was sterilized, as described elsewhere (Sheremata et al., 2001), by gamma irradiation from a 60Co source at the Canadian Irradiation Center (Laval, Quebec) with a dose of 50 kGy over 2 h.
General Sorption–Desorption Procedure
Batch sorption experiments were conducted at ambient temperature (21 ± 2°C) in 16-mL borosilicate centrifuge tubes fitted with Teflon-coated screw caps. For SSL and VT soils, 15 mL of CL-20 solution was added to 2 g of dried soil, while for all other soils and minerals, 10 mL of CL-20 solution and 1.34 g of dry sorbent were used. The tubes were wrapped in aluminum foil and agitated on a wrist-action shaker at 430 rpm for the desired contact time. They were then centrifuged for 30 min at 1170 x g, and the supernatant was filtered through a Millex-HV 0.45-µm filter (Millipore Corp., Billerica, MA). The filtrate collected after discarding the first 3 mL was analyzed by HPLC, as described below.
Desorption experiments were conducted by adding deionized water (15 mL for SSL and VT soils, 10 mL for other sorbents) to the pellet remaining in the tube after removing the supernatant, and agitating the suspensions for the required contact times. Samples were then centrifuged, filtered, and analyzed as described for sorption experiments. The solution volumes that remained in the soil at the end of the sorption and desorption phases were determined by weight and corrections were made to account for CL-20 present in these volumes.
To estimate CL-20 losses during the experiments, sorbed CL-20 was extracted with acetonitrile from the solid recovered after sorption or desorption as described in USEPA SW-846 Method 8330 (USEPA, 1997), modified to use an acidic calcium chloride solution (CaCl2: 5 g L–1; NaHSO4: 0.23 g L–1; pH = 3) in the sedimentation step.
Sorption and Desorption Kinetics
The time needed to establish sorption and desorption equilibria was determined by conducting kinetics experiments on the VT and GS soils. For sorption experiments, an aqueous solution of CL-20 (3 mg L–1) was shaken with soil over a period ranging from 1 h to 14 d. For desorption kinetics experiments, the soils were first shaken with a 3 mg L–1 CL-20 solution for 72 h. After centrifugation, the supernatant was removed and the pellet was contacted with fresh deionized water over periods varying from 1 h to 6 d. Sorbed CL-20 was extracted from the recovered solid using acetonitrile as described above. The kinetics experiments were conducted in triplicate.
Sorption–Desorption Isotherms
Aqueous CL-20 solutions were prepared to give initial CL-20 concentrations ranging from 0.5 to 3.5 mg L–1. Isotherms were measured using SSL, VT, FSB, FS, and GS soils. Sorption experiments were conducted over a period of 65 h for SSL, FS, and VT soils, and 72 h for FSB and GS soils, while desorption experiments were conducted over a period of 24 h (SSL and VT soils) and 48 h (FSB, FS, and GS soils). Following the general procedure described above, six replicates were used for each concentration. Three were extracted with acetonitrile immediately after sorption and the others were extracted after desorption. As a result, triplicate measurements were obtained at each concentration for both the sorption and desorption steps. Blanks containing the same amount of soil and volume of water were subjected to the same test procedure for all soils. No interfering peaks were detected.
All sorption and desorption data were fitted to the linear (Eq. [1]) and Freundlich (Eq. [2]) equations:
 | [1] |
 | [2] |
where x/m is the mass of solute sorbed per unit mass at equilibrium (mg kg–1), C is the aqueous phase concentration of solute at equilibrium (mg L–1), Kd is the distribution coefficient (L kg–1), KF is the Freundlich constant that gives a measure of the adsorbent capacity (mg1–1/n kg–1 L1/n), and 1/n gives a measure of the intensity of sorption. Because the value of KF depends on the value of 1/n for a given sample, KF values could not be compared between different isotherms. The extent of sorption and desorption were compared using the Kd constants, denoted KSd and KDd for sorption and desorption, respectively. Although the Kd values varied with sorbate concentration for samples that exhibited a high level of nonlinearity, we treated the linear model as providing average Kd constants that were representative of the sorption processes for each soil. These average values were used to determine the Koc values (Koc = KSd/foc, where foc represents the fraction of organic carbon).
Behavior of CL-20 in Alkaline Soils
The effect of pH on the stability of CL-20 in soil was studied either by artificially adjusting the pH of VT soil suspensions or by using naturally alkaline soils. In the case of VT soil, the dried soil (1.34 g) was contacted for 2 h with 5 mL of water containing the amount of HCl or NaOH required to obtain a final pH ranging from 3 to 9. After adding 5 mL of an aqueous solution of CL-20 (3.5 mg L–1), the slurries were agitated for 65 h. They were then centrifuged for 30 min at 1170 x g and the pH was measured in the supernatant. The latter was then filtered through a Millex-HV 0.45-µm filter, diluted using acetonitrile acidified to a pH of 3 (with H2SO4), and analyzed by HPLC. Potential CL-20 products (nitrite, nitrate, and formate ions) were quantified in the most alkaline samples as described below. All the soil pellets were extracted with acetonitrile to calculate a total percent recovery of CL-20.
In the naturally alkaline soils, 10 mL of a CL-20 solution (3.5 mg L–1) and 1.34 g of the sterile SAC (pH = 8.1) soil or sterile CSS (pH = 8.1) soil were agitated in borosilicate centrifuge tubes, using a wrist-action shaker at 430 rpm. At various time intervals, tubes were removed, and after centrifugation for 30 min at 1170 x g, the pH was measured in the supernatant and the latter was filtered. The filtrates collected after discarding the first 3 mL were diluted using acidified acetonitrile and analyzed by HPLC. The degradation products of CL-20, nitrite, nitrate, and formate ions were quantified as described below in the 14-d samples as well as in controls (sterile soil and water without CL-20) stirred for 14 d under similar conditions. The soils were extracted with acetonitrile and a total recovery was calculated for CL-20. For comparison, a similar experiment was conducted with sterile VT soil (pH = 5.6).
Analytical Methods
All CL-20 standards and samples were prepared in acidified (pH = 3) mixtures of 50:50 CH3CN to H2O since acidification prevented CL-20 degradation. Dissolved soil organic matter or clay mineral that precipitated from solution at a pH of 3 were separated by centrifugation for 10 min at 16000 x g with an Eppendorf (Hamburg, Germany) Centrifuge 5415D. The supernatant was subsequently analyzed by HPLC.
CL-20 analysis was performed by HPLC using a chromatographic system (ThermoFinnigan, San Jose, CA) composed of a Model P4000 pump, a Model AS3000 injector, including temperature control for the column, and a Model UV6000LP photodiode-array detector. The injection volume was 50 µL. The separation was completed on a Supelcosil LC-CN column (25 cm, 4.6 mm, 5µm; Supelco, Bellefonte, PA) maintained at 35°C. The mobile phase (70% aqueous methanol) was run isocratically at 1 mL min–1 for the entire run time of 14 min. Chromatograms were extracted at a wavelength of 230 nm and quantified using peak areas of external standards.
Nitrate 
and formate (HCOO–) ions were quantified by ion chromatography (IC) equipped with a conductivity detector as described elsewhere (Monteil-Rivera et al., 2004). Analysis of nitrite 
ions was performed colorimetrically after producing the diazonium salt as described in USEPA Method 354.1 (USEPA, 1979).
|
RESULTS AND DISCUSSION
|
|---|
Sorption–Desorption Kinetics
Kinetics experiments were performed using VT (2.3% TOC) and GS (34% TOC) soils to determine the minimum time required to reach sorption and desorption equilibria. The sorption experiments revealed different behavior in both soils (Fig. 2). For GS soil, a rapid sorption occurred with maximal sorption (94%) reached within 1.5 h, while VT soil yielded a slower sorption, with approximately 200 h being required to attain sorption equilibrium (60% sorption). In the case of VT soil, a fast initial adsorption was followed by a slower migration and diffusion of the CL-20 into the soil matrix. These results show that the sorbing sites present in GS soil were more readily accessible than those present in VT soil, suggesting that different types of sites are involved in CL-20 immobilization by soils. No information is yet available on the sorption and desorption kinetics of CL-20 onto soils, but for the monocyclic nitramines RDX and HMX, a rapid sorption was reported to occur in less than 24 h for each compound in different soils, including VT soil (Xue et al., 1995; Myers et al., 1998; Sheremata et al., 2001; Monteil-Rivera et al., 2003). The slower sorption of CL-20 onto the VT soil is indicative of a different sorption mechanism for CL-20 than for RDX and HMX.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 2. Sorption and desorption kinetics for CL-20 with sandy agricultural topsoil from Varennes, QC, Canada (VT) and sandy garden soil obtained from a local supplier (GS). For desorption, y axis represents the amount of sorbed CL-20 that desorbs. Error bars represent the standard deviations of three replicates.
|
|
Desorption equilibria were achieved in less than 1 and 20 h for GS and VT soils, respectively (Fig. 2). As was the case for sorption, desorption of CL-20 from VT soil was probably delayed by the need for the chemical to diffuse through the soil structure.
CL-20 Sorption–Desorption Isotherms
Sorption and desorption isotherms of CL-20 in various soils are presented in Fig. 3 for the SSL, VT, FSB, FS, and GS soils. Since no microbial growth inhibitor was added to the media and nonsterile soils were used, some degradation of CL-20 was observed (2–16% depending on the soil used; Table 2) during the time frame of the sorption–desorption isotherms (about one week). To account for CL-20 degradation, the amount of sorbed CL-20 was determined by extracting the soil with acetonitrile, rather than estimating it by difference from the aqueous concentration.
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 3. Sorption–desorption isotherms for CL-20 in nonsterile Sassafras sandy loam (SSL); sandy agricultural topsoil from Varennes, QC, Canada (VT); sandy forest soil from Boucherville, QC, Canada (FSB); sandy garden soil obtained from a local supplier (GS); and sandy forest soil provided by a local supplier (FS) at ambient temperature. Sorption is denoted by filled symbols and solid lines; desorption by hollow symbols and dashed lines. For clarity, data obtained with soils having a high affinity for CL-20 (top) are presented separately from data obtained with soils having a low affinity for CL-20 (bottom).
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Isotherm parameters, organic-carbon-normalized sorption coefficient (Koc), and recoveries for CL-20 sorption and desorption by various soils.
|
|
Non-linear regression procedures using Origin software (Microcal, 1999) were used for fitting Freundlich and linear (1/n set equal to 1) isotherms to the sorption data. Values of the resulting parameters for the two models along with the r2 values are presented in Table 2 and theoretical curves resulting from the Freundlich model are presented in Fig. 3. All sorbents exhibited nonlinear isotherms, with 1/n values ranging from 0.49 to 0.94. The r2 values show that the Freundlich model fitted the equilibrium sorption data better than did the linear model. Moreover, the linear model was not appropriate for describing FSB and GS soils (r2 < 0.7). The general nonlinearity of the isotherms observed in this study indicates that CL-20 sorption to soil occurs through interactions with different classes of sites having different sorption energies (Weber and DiGiano, 1996).
Table 2 clearly shows that sorption was significantly higher in the FS (Kd = 311 L kg–1), GS (Kd = 187 L kg–1), and FSB (Kd = 37 L kg–1) soils than in the VT (Kd = 15.1 L kg–1) and SSL (Kd = 2.4 L kg–1) soils. The higher Kd (and KF) values obtained in soils with higher organic content suggest that soil organic matter plays a determining role in CL-20 sorption. This result demonstrates a major difference in behavior between CL-20 and the two monocyclic nitramines, RDX and HMX, whose sorption onto soil was previously demonstrated to be independent of organic content (Leggett, 1985; Sheremata et al., 2001; Brannon et al., 2002; Monteil-Rivera et al., 2003). The higher log Kow measured for CL-20, compared with that of RDX and HMX (Fig. 1; Monteil-Rivera et al., 2004), demonstrates a higher hydrophobicity of the former, which is probably responsible for its superior affinity toward soils with high organic content. Numerous relationships based on Eq. [3] have been developed between soil sorption coefficients (Koc) and n-octanol–water coefficients (Kow):
 | [3] |
where m and b are parameters extracted from linear regressions. To predict a Koc value from a Kow value, a reasonable similarity should be ensured between the solute molecule of interest and the family of compounds used to establish the data set on which m and b are based (Sablji
et al., 1995). No correlation is currently available in the literature for nitramines, but by applying the general model given by Sablji et al. (1995) for nonhydrophobic chemicals [defined as chemicals containing atoms other than C, H, and halogens (m = 0.52; b = 1.02)], a Koc value of 104 L kg–1 was calculated for CL-20. The fact that this theoretical value is inferior to all those measured in the present study (see Table 2), coupled with the fact that the Koc values measured varied by almost one order of magnitude, confirms that CL-20 adsorption onto organic matter is not limited to a pure physical distribution process, as suggested by the nonlinear isotherms. Therefore, the type, and not just the amount, of organic matter present in soils is a determining factor in the immobilization of CL-20.
A sorption–desorption hysteresis 
was observed for the SSL and VT soils (Fig. 3, Table 2), while sorption to the FS and GS soils appeared to be fully reversible (KDd approximately equal to KSd). Meanwhile, the large divergence observed on the measurements performed with FSB soil made it hard to ascertain whether or not this soil gave rise to a sorption hysteresis. Hysteresis phenomena can be caused by organic matter or mineral constituents of soil. We assessed the sorption of CL-20 onto silica, ferric oxide, and three different clays (Table 3). Clays have been shown to play a determining role in the sorption of RDX and HMX (Brannon et al., 2002); when HMX was adsorbed on montmorillonite under similar conditions, a KSd value of 15.6 L kg–1 was measured. The data presented in Table 3 demonstrate that CL-20 has a low tendency to adsorb on any of the mineral phases, and that it adsorbs much less on clay than do RDX and HMX. Based on FTIR measurements, Boyd et al. (2001) suggested that nitro-containing compounds are strongly retained by complexation between metals in clay minerals (e.g., K+) and –NO2 groups. With six –NO2 groups, CL-20 contains more sites capable of interacting with such metals compared to RDX (three –NO2) and HMX (four –NO2). However, its polycyclic caged structure makes CL-20 a bulkier compound [approximately 6–7 Å in diameter, as estimated from crystallographic data (Zhao and Shi, 1996)] that is unable to migrate into the clay layers [3–9 Å for a montmorillonite, approximately 5 Å for a smectite (Brindley, 1981)]. Given the small contribution of mineral phases in CL-20 retention, the hystereses observed with VT and SSL soils were probably caused by the organic content of soil.
The full CL-20 recoveries we obtained on sonicating the sorbed soils in acetonitrile ruled out the formation of covalent bonds between CL-20 and organic matter as the reason for the hysteresis. Weber et al. (1998) previously suggested that the form of the soil organic matter played a critical role in the hystereses observed in the sorption–desorption behavior of hydrophobic organic compounds such as phenanthrene. Humic acids, with their large microporosities, allow for fast interaction with water, thus producing a reversible sorption. In contrast, condensed organic matter presents physically rigid zones that interact poorly with water and tend to trap any molecules sorbed within it, producing a sorption–desorption hysteresis (Weber et al., 1998). The hystereses observed in the present study could therefore have arisen from an entrapment of CL-20 in condensed organic matter. However, a matrix change that could occur from the sorption to the desorption step to give access to sites with higher sorbing capacity in the desorption batch should not be excluded.
Behavior of CL-20 in Alkaline Soils
The ability of CL-20 to be significantly sorbed to soils containing organic matter was demonstrated in the previous section. Given the previously established tendency of CL-20 to undergo rapid hydrolysis under alkaline conditions (pH = 10) (Balakrishnan et al., 2003), an investigation into the effect of soils on the hydrolysis process was deemed worthwhile. To generate a pH profile of CL-20 stability in soil, we adjusted the pH of VT soil and contacted it with CL-20 solution for 65 h. Below a pH of 7.5, percent recoveries remained approximately constant at approximately 100%, while above a pH of 7.5, CL-20 degraded (Fig. 4). Analyses of degradation products in the most alkaline soil samples showed that CL-20 disappearance was accompanied by the formation of NO–2 and HCOO–. The fact that degradation was observed in soils with a pH of >7.5 indicates that CL-20 will not persist in natural alkaline soil environments. In aqueous control samples of CL-20 (Fig. 4), degradation was observed at a pH of >7.5 and was complete at a pH of >8.1. The greater extent of degradation observed in the control solutions than in the soils at a pH of >8 suggests that to a certain extent, the presence of soil protects CL-20 against decomposition.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 4. Percent recovery of CL-20 after 65 h of agitation at different pH values and ambient temperature, in presence or absence of nonsterile sandy agricultural topsoil from Varennes, QC, Canada (VT).
|
|
Since CL-20 significantly degraded when we increased the pH of VT suspensions, the behavior of CL-20 was followed over time in two sterile, naturally alkaline soils. After 14 d of contact with sterile SAC and sterile CSS soils (pH = 8.1), 100 and 82% of the CL-20 degraded, respectively, compared with only 8% loss in sterile VT soil (pH = 5.6) (Fig. 5). Clearly, CL-20 is not stable in alkaline soils. Degradation in the naturally alkaline soils produced two equivalents of nitrite for each mole of CL-20 reacted, exactly as observed for the aqueous alkaline hydrolysis of CL-20 at a pH of 10 (Balakrishnan et al., 2003). Despite the equivalent pH values of both soils, CL-20 degraded faster in SAC soil than it did in CSS soil. The SAC soil contains very low amounts of the two most active sorbents in soils (0.1% clay and 0.08% TOC), whereas CSS soil contains clay (44.5%) and organic carbon (0.31%). As a result, CL-20 did not sorb onto SAC soil 
and was thus not protected from degradation. In contrast, the slight sorption to CSS soil (KSd = 1.0 ± 0.3 L kg–1, n = 24) retarded CL-20 hydrolysis. Therefore, sorption retarded the hydrolysis of the energetic chemical but did not prevent it.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 5. Stability of CL-20 in sterile soils having various pH values (clay soil from St. Sulpice, QC, Canada [CSS] and sandy soil provided by Agriculture Canada [SAC], pH = 8.1; sandy agricultural topsoil from Varennes, QC, Canada [VT], pH = 5.6). Error bars represent the standard deviations of three replicates.
|
|
|
CONCLUSIONS
|
|---|
This study reveals that CL-20 can be highly retained by soils. In contrast to the monocyclic nitramines, RDX and HMX, which are mainly adsorbed by the clay fraction of soil, CL-20 affinity for soil is governed by the organic content of soil, and only a small fraction of the energetic chemical is actually bound to the mineral phase. Even though sorption to soils retards abiotic degradation, CL-20 readily decomposes in natural soils where pH is >7.5 suggesting that, unlike RDX and HMX, CL-20 will not be a persistent organic pollutant in even slightly alkaline soils. As a result, the identification of intermediates and final products in the degradation of CL-20 is of paramount importance to fully define the environmental impact of this nitramine explosive. The identification of products resulting from CL-20 decomposition constitutes the focus of ongoing studies in our laboratory.
|
ACKNOWLEDGMENTS
|
|---|
The authors thank A. Corriveau and D. Manno for their technical assistance. V.K. Balakrishnan thanks the National Sciences and Engineering Research Council of Canada and the National Research Council of Canada for a Visiting Fellowship. We also thank Thiokol Propulsion (Brigham City, UT) for providing CL-20. Funding was provided by the U.S. DoD/DoE/EPA Strategic Environmental Research and Development Program (SERDP CP 1256).
|
REFERENCES
|
|---|
- Balakrishnan, V.K., A. Halasz, and J. Hawari. 2003. Alkaline hydrolysis of the cyclic nitramine explosives RDX, HMX, and CL-20: New insights into degradation pathways obtained by the observation of novel intermediates. Environ. Sci. Technol. 37:1838–1843.[Medline]
- Boyd, S.A., G. Sheng, B.J. Teppen, and C.T. Johnston. 2001. Mechanisms for the adsorption of substituted nitrobenzenes by smectite clays. Environ. Sci. Technol. 35:4227–4234.[Medline]
- Brannon, J.M., C.B. Price, C. Hayes, and S.L. Yost. 2002. Aquifer soil cation substitution and adsorption of TNT, RDX, and HMX. Soil Sediment Contam. 11:327–338.
- Brindley, G.W. 1981. Structures and chemical compositions of clay minerals. p. 1–21. In F.J. Longstaffe (ed.) Short course in clays and the resource geologist. Mineralogical Assoc. of Canada, Edmonton, AB.
- Haas, R., I. Schreiber, E. von Low, and G. Stork. 1990. Conception for the investigation of contaminated munition plants. 2. Investigation of former RDX plants and filling stations. Fresenius' J. Anal. Chem. 338:41–45.
- Huang, W., H. Yu, and W.J. Weber. 1998. Hysteresis in the sorption and desorption of hydrophobic organic contaminants by soils and sediments: 1. A comparative analysis of experimental protocols. J. Contam. Hydrol. 31:129–148.
- Karapanagioti, H.K., J. Childs, and D.A. Sabatini. 2001. Impacts of heterogeneous organic matter on phenanthrene sorption: Different soil and sediment samples. Environ. Sci. Technol. 35:4684–4690.[Medline]
- Leggett, D.C. 1985. Sorption of military explosive contaminants on bentonite drilling muds. CRREL Rep. 85-18. U.S. Army Cold Regions Res. and Eng. Lab., Hanover, NH.
- Microcal. 1999. Origin Version 6.0. Microcal, Northampton, MA.
- Monteil-Rivera, F., C. Groom, and J. Hawari. 2003. Sorption and degradation of octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine in soil. Environ. Sci. Technol. 37:3878–3884.[Medline]
- Monteil-Rivera, F., L. Paquet, S. Deschamps, V.K. Balakrishnan, C. Beaulieu, and J. Hawari. 2004. Physico-chemical measurements of CL-20 for environmental applications. Comparison with RDX and HMX. J. Chromatogr. A 1025:125–132.[ISI][Medline]
- Myers, T.E., J.M. Brannon, J.C. Pennington, W.M. Davis, K.F. Myers, D.M. Townsend, M.K. Ochman, and C.A. Hayes. 1998. Laboratory studies of soil sorption/transformation of TNT, RDX, and HMX. Tech. Rep. IRRP-98-8. U.S. Army Corps of Eng., Waterways Exp. Stn., Vicksburg, MS.
- Myler, C.A., and W. Sisk. 1991. Bioremediation of explosives contaminated soil. p. 137–146. In G.S. Sayler, R. Fox, and J.W. Blackburn (ed.) Environmental biotechnology for waste treatment. Plenum Press, New York.
- Robidoux, P.Y., G.I. Sunahara, K. Savard, Y. Berthelot, S. Dodard, M. Martel, P. Gong, and J. Hawari. 2004. Acute and chronic toxicity of the new explosive CL-20 to the earthworm (Eisenia andrei) exposed to amended natural soils. Environ. Toxicol. Chem. 23:1026–1034.[ISI][Medline]
- Sablji, A., H. Güsten, H. Verhaar, and J. Hermens. 1995. QSAR modelling of soil sorption. Improvements and systematics of log Koc vs. log Kow correlations. Chemosphere 31:4489–4514.
- Sheremata, T.W., A. Halasz, L. Paquet, S. Thiboutot, G. Ampleman, and J. Hawari. 2001. The fate of the cyclic nitramine explosive RDX in natural soil. Environ. Sci. Technol. 35:1037–1040.[Medline]
- Talmage, S.S., D.M. Opresko, C.J. Maxwell, C.J.E. Welsh, F.M. Cretella, P.H. Reno, and F.B. Daniel. 1999. Nitroaromatic munition compounds: Environmental effects and screening values. Rev. Environ. Contam. Toxicol. 161:1–156.[Medline]
- USEPA. 1979. Methods for chemical analysis of water and wastes—Nitrogen, nitrite. Method 354.1. USEPA, Washington, DC.
- USEPA. 1997. Nitroaromatics and nitramines by high performance liquid chromatography (HPLC). Method 8330 SW-846 Update III Part 4:1 (B). Office of Solid Waste, Washington, DC.
- Weber, W.J., and F.A. DiGiano. 1996. Energy relationships: Applications to heterogeneous systems. p. 317–421. In Process dynamics in environmental systems. Wiley-Interscience, New York.
- Weber, W.J., W. Huang, and H. Yu. 1998. Hysteresis in the sorption and desorption of hydrophobic organic contaminants by soils and sediments: 2. Effects of soil organic matter heterogeneity. J. Contam. Hydrol. 31:149–165.
- Xue, S.K., I.K. Iskandar, and H.M. Selim. 1995. Adsorption-desorption of 2,4,6-trinitrotoluene and hexahydro-1,3,5-trinitro-1,3,5-triazine in soils. Soil Sci. 160:317–327.
- Yinon, J. 1990. Cyclotrimethylenetrinitramine (RDX). Cyclotetramethylenetetranitramine (HMX). p. 145–170. In Toxicity and metabolism of explosives. CRC Press, Boca Raton, FL.
- Zhao, X., and N. Shi. 1996. Crystal and molecular structures of -HNIW. Chin. Sci. Bull. 41:574–576.
Related articles in JEQ:
- This Issue in Journal of Environmental Quality
JEQ 2004 33: 1177-1182.
[Full Text]
This article has been cited by other articles:
|
|
|
|
 
K. M. Dontsova, S. L. Yost, J. Simunek, J. C. Pennington, and C. W. Williford
Dissolution and Transport of TNT, RDX, and Composition B in Saturated Soil Columns
J. Environ. Qual.,
October 27, 2006;
35(6):
2043 - 2054.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
