Thermal Stability of Organoclays: Effects of Duration and Atmosphere

Nov 15, 2010 - Heating periods necessary to destroy iodide sorption capacity of the quaternary (alkyl) ammonium and phosphonium modified bentonites we...
0 downloads 0 Views 1022KB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 9311–9317

Thermal Stability of Organoclays: Effects of Duration and Atmosphere of Isothermal Heating on Iodide Sorption A. MELESHYN* AND B. RIEBE Institute for Radioecology and Radiation Protection, Leibniz Universita¨t Hannover, Herrenha¨user Strasse 2, 30419 Hannover, Germany

Received March 29, 2010. Revised manuscript received October 14, 2010. Accepted November 4, 2010.

Heating periods necessary to destroy iodide sorption capacity of the quaternary (alkyl) ammonium and phosphonium modified bentonites were determined using iodide sorption batches. For this purpose, prior to the batches the studied organoclays were isothermally heated in air in the temperature ranges of 110-180 °C and 160-300 °C, respectively. The temperature dependence of the heating periods was found to follow the Arrhenius relationship, which allowed a determination of Arrhenius parameters for the reaction leading to the loss of the iodide sorption capacity of a bentonite modified by CP+ (cetylpyridinium), BE+ (benzethonium), CTMA+ (cetyltrimethylammonium), or TPP+ (tetraphenylphosphonium) surfactant. At 160°C,thethermalstabilityoftheiodidesorptioncapacityofTPP+bentonite is much higher than that of the second most stable CTMA+-bentonite (80 days vs 5 days). However, the obtained Arrhenius parameters predict that CTMA+-bentonite becomes the most stable one as the heating temperature decreases to 40 °C with iodide sorption still available for ∼12000 years as compared to ∼8000 years for TPP+-bentonite. Heating of the organoclays in a N2-atmosphere ( NO3- > Br- > Cl- (7). This sequence may be related either to the correspondingly decreasing hydration energies of these anions (7) or to their decreasing ability to shield the positive charged groups of neighboring organic cations (10). Differently than for natural clays, however, much less is known about the long-term stability of organoclays. In particular it is not clear how long their iodide sorption capacity will persist in the near-field of a final repository for heat producing radioactive waste. The majority of previous reported studies of the thermal stability of organoclays have been carried out either by heating the samples at a constant rate of 1 up to 20 °C/min over a large temperature range (as a rule from room temperature up to 600-1000 °C) (e.g., refs 12 and 14-23) or by heating isothermally at various temperatures in the range 150-240 °C for 10 min (e.g., refs 22 and 24). The application of such heating protocols is related to a specific issue of the required organoclay stability during the desirable organoclay-based industrial fabrication of polymer-clay nanocomposites at polymer melt processing temperatures. These studies have suggested nucleophilic substitution (in the presence of chloride anions, which become coadsorbed on clay during its organic modification with surfactant chloride salt) and Hoffman elimination as the predominant reaction pathways of thermal decomposition of quaternary ammonium surfactants (14, 22). Alkenes, amines, and chloroalkanes are main products of the thermal decomposition reaction with chloroalkanes (mainly chloromethane) identified as the major volatile compound (14, 18, 19, 22, 24). The mechanism behind the related surfactant mass loss has been shown to be independent of the temperature decrease in the range 240-170 °C, which was accompanied by a decrease of the loss rate by an order of magnitude (22). Although previous studies have strongly contributed to an improved understanding of the processes occurring in organoclays as a result of heating, there are still many unanswered questions remaining on this research subject. On one hand, both heating protocols discussed above have not provided an insight into the equilibrium properties of the heated organoclays, because the surfactant mass loss by organoclays is still far from reaching a steady state even after 10 min of isothermal heating (22, 24). For a comparison, equilibration of quaternary alkylammonium surfactant structure has been shown to require at least 6 up to 20 h at 22 °C (1-2 h at 32 °C) on mica (25) or 16 h up to 5 days at room temperature on montmorillonite (9, 26, 27). On the other hand, the studies discussed above have been aimed at improving the thermal stability of clay-modifying surfactants, which in case of smectitic clays are intercalated in the VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9311

interlayer space and aggregated on the external surface of smectite tactoids. The present study shows that fulfilling of this aim does not necessarily result in an improvement of some other properties of interest as, e.g., anion sorption. Moreover, our results suggest that a choice of clay-modifying surfactant resulting in a higher stability of the anion sorption at a specified heating temperature may become unfavorable at lower heating temperatures. In the preceding experimental studies (28, 29), an isothermal pretreatment for 72 h has resulted in only a negligible (if any) decrease of iodide sorption capacity in the range 40-140 °C and in its strong decrease at 160-200 °C. In another study (30), a decrease of iodide sorption capacity has been found after an isothermal pretreatment for 70 h at 140 °C. Differently from these studies, the present work aims to accurately determine heating times necessary to destroy the iodide sorption capacity of organoclays as a result of isothermal heating. Knowledge of these times is a prerequisite for an assessment of long-term performance of organoclays in final repositories of radioactive wastes. For this purpose, time evolution of the reaction leading to the loss of iodide sorption capacity has to be recorded in intervals small enough with respect to the unknown total reaction duration. Accordingly, initial “trial and error” determinations of the total reaction duration at the highest studied temperatures were used to estimate the durations at lower temperatures based on a reasonable assumption that the dependence of the reaction rate on temperature follows an Arrhenius relationship. In an additional series of experiments, the influence of the pretreatment atmosphere (air and N2) on the total reaction duration was studied to account for the strong deficit of molecular oxygen in the final radioactive waste repository emerging several years after its closure (1).

2. Materials and Methods Bentonite (labeled MX-80) was used as obtained from Amcol Specialty Minerals (Winsford, Great Britain). According to producer, it contains >90 wt % of Wyoming-type montmorillonite with the crystallochemical formula (Na,Ca)0.33(Al, Fe1.67Mg0.33)Si4O10(OH)2 (Al/Fe and Fe(III)/Fe(II) ratios of ∼6.9 and ∼8.3, respectively) and several wt % of feldspar and biotite. A CEC of 0.86 ( 0.02 meq g-1 was measured according to the silver-thiourea method (31). For organic modification, 20 g of bentonite was dispersed in 1 L of aqueous solution prepared from doubly deionized water (18 MΩ cm) and chloride salts of organic cations in an amount corresponding to 200% CEC of the bentonite amount to be suspended. Organobentonites were then obtained by stirring for 24 h, filtering (Schleicher & Schuell Nr. 602 H 1/2, pore size 2 µm), rinsing with 1 L of doubly deionized water (using a dropping bottle) to remove excess organic cations, and freeze-drying. Cetylpyridinium (CP+, C21H38N+) chloride, cetyltrimethylammonium (CTMA+, C19H42N+) chloride, tetraphenylphosphonium (TPP+, C24H20P+), and benzethonium (BE+, C27H42O2N+) (Scheme 1) chlorides as obtained from SigmaAldrich and Molekula Ltd. were used to modify the bentonite. For thermal treatment of CP+-, CTMA+-, TPP+-, and BE+bentonites, heating chambers (FD 115 from Binder, Tuttlingen) with a nominal temperature of 300 °C and a deviation of 0.7 °C (at 70 °C) up to 1.8 °C (at 150 °C) were used. For each pretreatment temperature (110, 120, 130, 140, 150, 160, 170, and 180 °C for CP+-, CTMA+-, and BE+-bentonites as well as 160, 200, 230, 260, and 300 °C for TPP+-bentonite), multiples of identical organobentonite samples were prepared in order to perform a time series between one and up to 300 days. They were filled into glass test tubes (diameter of 2.5 cm, total volume of ∼35 mL) or, additionally, spread in a thin layer on crystallization dishes (diameter of 22.3 cm) and placed in heating chambers. Following preset increasing pretreatment periods, samples were taken out of the heating 9312

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

SCHEME 1. Chemical Structures of Surfactants Used To Prepare Organobentonites

chambers and sealed (after heat dissipation) for subsequent use in sorption batches. In order to investigate a possible influence of a time lag between the thermal pretreatment of the organobentonites and the start of sorption batches, a total of five (CTMA+) to seven (CP+ and BE+) time series of pretreatment at 110 °C and two time series (CP+ and BE+) of pretreatment at 120 and 130 °C were carried out. Variation of the time lag in these time series was in the range between one day and a maximum of 125 days (110 °C), 622 days (120 °C), or 669 days (130 °C). Due to a limited number of heating chambers, time series themselves were started not simultaneously but sequentially with time intervals of up to 1.5 years between single time series (organobentonites for different series were prepared at correspondingly different times). In an additional thermal experiment, heating chambers were operated in a glovebox in N2-atmosphere ( 10 (34), which are significantly higher than the measured pH values. Measurements of the chloride content in CP+-bentonite before and after a three-day thermal pretreatment at 170 °C were carried out using micro X-ray fluorescence instrument (Eagle µProbe, Ro¨ntgenanalytik Messtechnik). Fourier transform infrared (FTIR) spectra for organobentonites were recorded with a Bruker Vertex 70 FTIR spectrometer equipped with a diamond attenuated total reflectance (ATR) accessory (Harrick, MVP-Pro-Star) in the range of 4000-400 cm-1 with a resolution better than 1 cm-1. To record transmission FTIR spectra, the KBr pressed disk technique was additionally used with 1 mg of sample dispersed in 300 mg of KBr. For each sample, 128 scans were accumulated.

3. Results and Discussion 3.1. Effect of Pretreatment Temperature on Iodide Sorption. The effect of the thermal pretreatment on the iodide sorption capacity of CP+-, BE+-, CTMA+-, and TPP+-bentonites is demonstrated by the iodide sorption data in Figure 1. These data reveal that upon decrease of the pretreatment temperature from 180 to 150 °C, the pretreatment period τloss necessary to zero the iodide sorption capacity increases from ∼0.5 day to 2-4 days for CP+- and BE+-bentonites as well as from ∼1 day to 10 days for CTMA+-bentonite. The period τloss increases to 45-55 days for CP+- and BE+bentonites upon a further decrease of the pretreatment temperature to 120 °C and becomes even longer (∼70 days) for CTMA+-bentonite after pretreatment at 130 °C. Eventually, in the longest carried out time series with clays pretreated at 110 °C, τloss is found to equal 100-110 days for CP+- and BE+-bentonites, whereas it could not be determined for CTMA+-bentonite with iodide sorption capacity decreased by only ∼60% within 300 days allocated for this time series (Figure 1C). Importantly, although TPP+-bentonite shows by far the highest thermal stability in the studied temperature range, it also shows by far the lowest iodide sorption (cf. Figures 1A-1C and 1D for 160 °C). It should be noted that despite similar values of τloss for CP+- and BE+-bentonites, the decrease of the iodide sorption capacity within the period τloss/2 equals only 10-30% for CP+-bentonite as compared to that of 70-80% for BE+bentonite. It is also notable that a replacement of the pyridinium group (in CP+) by the trimethylammonium group (in CTMA+) leads to an increase by a factor of 2.5-3 of the τloss values for a CTMA+-bentonite as compared to CP+bentonite at pretreatment temperatures of 130-170 °C and to an apparently even stronger increase at 110 °C. Furthermore, the iodide sorption data for the time series at 110-130 °C (Figure 1) show good reproducibility. It can be therefore concluded that the iodide sorption capacity of the studied clays does not depend on the length of latency period between the thermal pretreatment and the onset of the iodide sorption reaction, which varied in the range from one day to about two years. (Note that besides differing in the length of latency period, the time series at a given temperature were carried out with organoclays prepared at very different times during the three-year study period.) Plotting the obtained ln τloss-1 values as a function of the reciprocal pretreatment temperature 1/T in accordance with the Arrhenius equation ln τloss-1 ) ln A0 - Ea/RT, where R is

FIGURE 1. Iodide sorption (%) on (A) CP+-, (B) BE+-, (C) CTMA+-, and (D) TPP+-bentonites from 0.01 M KI solution as a function of the duration of thermal pretreatment (data for pretreatment temperatures >140 °C are given in insets). A 100% iodide sorption corresponds to a loading of 25.2 mg iodide per g clay. Each sorption series was carried out in triplicate, and the corresponding measurement uncertainty was smaller than the symbols representing the measured value. Therefore, error bars are shown only for the cases that several sorption series were carried out for the given temperature (110-130 °C for CP+- and BE+-bentonites as well as 110 °C for CTMA+-bentonite) as discussed in the text and represent standard deviations of the average value calculated from several sorption series. The decision limit is denoted by the shaded area. VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9313

TABLE 1. Iodide Sorbed (%) on CP+- and CTMA+-Bentonites in Dependence on Fraction q (%) of the Exposed Sample Surface, the Temperature, and the Atmosphere of a Thermal Pretreatment at 160 °C for 4 Days 25 °C (air) 160 °C (air) 160 °C (N2)

CP+ q1 (q2)a

BE+ q1 (q2)a

CTMA+ q1 (q2)a

99 4 (6) 33 (28)

89 0 (0) 13 (11)

96 9 (12) 86 (26)

q1 ∼26% and q2 ∼56% (diameters of 2.5 and 22.3 cm as well as filling heights of ∼2.4 and ∼0.1 cm for a test tube and a crystallization dish, respectively). a

FIGURE 2. Natural logarithms of the reciprocal observed pretreatment periods τloss, which are necessary to zero the iodide sorption capacity of CP+-, BE+-, and CTMA+-bentonites as well as of TPP+-bentonite (inset) as functions of the reciprocal pretreatment temperature 1/T. Straight lines represent results of linear fits (with correlation coefficients equal to -1 for all organobentonites) to the observed data. The corresponding equations are given in the legend. the gas constant, and the constant A0 and the activation energy Ea can be calculated from a linear fit as shown in Figure 2. Application of the Arrhenius equation to the obtained data is justified considering that the correlation coefficient R equals -1, which indicates a perfect negative linear relationship. The linear fits yielded Arrhenius constant values of 105.1, 108.2, 109.5, or 1010 s-1 and activation energies of 99, 112, 121, or 129 kJ/mol for the reaction leading to the loss of the iodide sorption capacity of TPP+-, CP+-, BE+-, or CTMA+-bentonites, respectively. Although a detailed characterization of the reaction taking place in the organobentonites as a result of the thermal pretreatment is beyond the scope of the present study, interpretation of the obtained IR and ICP-AES data does give insight into the mechanism behind this reaction. The thermal pretreatment of the studied organobentonites was noticed to be accompanied by a deposition of reaction products on the previously cleaned air exhaust louver of the heating chamber. In contrast to the IR spectra of untreated CP+bentonite and CPCl powder, the IR spectrum of the reaction products shows a strong absorption band at 758 cm-1 similarly to the earlier observed band assigned to chloroalkanes (19), which are reportedly produced as a result of thermal decomposition of clays modified with quaternary alkyl ammonium ions (14, 18, 22). This assignment is supported by a strong decrease of the chloride content from 3.2 ( 0.9 mmol/g for untreated CP+-bentonite to 0.8 ( 0.2 mmol/g for CP+-bentonite heated for three days at 170 °C. Moreover, the IR spectrum of the volatile reaction products shows strong absorption bands at 2850 and 2920 cm-1 characteristic for C-H stretching of methyl(ene) groups. A formation of volatile chloroalkane compounds is believed to occur as a result of nucleophilic attack of a R4N+ moiety by a chloride ion (14, 18, 19, 22). This formation can be suggested as the most reasonable explanation for the observed decrease of chloride content, as it has been found to dominate the decomposition pathway of the clays modified with quaternary alkyl ammonium ions in the case of chloride availability (22). Similarly, a formation of volatile benzyl chloride compounds (22) can be proposed to explain the decreases of the iodide sorption capacity in the case of BE+bentonite. The lower thermal stability of CP+- and BE+bentonites as compared to CTMA+-bentonite manifested by changes in iodide sorption (Figure 1) is very likely related to the far more higher reactivity rates of the elecrophilic aromatic 9314

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

groups (pyridinium in CP+ and phenyl in the beta position of nitrogen in BE+) than of the methyl group in CTMA+ (22). Furthermore, despite the loss of organic substance, which was manifested by a strong decrease of the intensities of characteristic IR absorption bands at 1459, 1476, 1512, 2876, 2903, and 2955 cm-1, we observed no deposition of reaction products on the air exhaust louver of the heating chamber for BE+-bentonite. This can be explained by the formation as a result of heating of BE+-bentonite of benzyl chloride (22), which can be assumed to be more volatile than the long-chain chloroalkanes (14, 18) formed in the case of CP+and CTMA+-bentonites as discussed above. Summarizing, the observed decreases of the iodide sorption capacity of CP+-, BE+-, and CTMA+-bentonites after thermal pretreatment most likely occur as a result of the loss of exchangeable chloride accompanying the formation of volatile chlorinated compounds. The IR analyses of thermally pretreated CP+-bentonite additionally revealed the development of an absorption band at 1710-1713 cm-1, which was not observed for BE+- and CTMA+-bentonites and is characteristic for CdO stretching of carbonyl groups (35, 36). This observation indicates that a Hoffmann elimination reaction may occur along with the nucleophilic substitution reaction or even predominate during the decomposition of heated organobentonites, at least in the case of the CP+ surfactant. This is in agreement with results of a previous study (21) and with the present observation of a strong pH decrease for CP+-bentonite (cf. Section 2) as a result of thermal treatment. The CdC bond of unsaturated hydrocarbons produced in the Hoffmann elimination reaction can be attacked by a radical (OH or O atom) to produce CdO bond according to the recently proposed mechanism of hydrocarbon oxidation (37). The production of CdC bonds in the thermally treated CP+- and CTMA+-bentonites is manifested by the presence of strong absorption bands at 1377-1378, 1393-1398, 1446-1449, and 1466 cm-1, which are characteristic for CH deformation modes of the CH3-CH-CH3 group (38), in IR spectra from reaction products deposited on the air exhaust louver of the heating chamber. The absence of these bands in the heated bentonites as well as the absence of a CdO band in the heated CTMA+-bentonite indicates that only smaller, volatile unsaturated hydrocarbons are produced in the latter clay, whereas larger, nonvolatile (at the applied temperatures of 110-180 °C) unsaturated hydrocarbons can also be produced in CP+-bentonites. 3.2. Effect of Pretreatment Atmosphere on Iodide Sorption. The latter findings prompted an investigation of a possible oxygen effect on the thermal stability of the CP+-, CTMA+-, and BE+-bentonites. It revealed that at the pretreatment temperature of 160 °C the iodide sorption increases by ∼30% for CTMA+-bentonite and by up to ∼90% for CP+and BE+-bentonites as a result of the change of the pretreatment atmosphere from air to N2 (Table 1). No oxygen effect on iodide sorption on TPP+-bentonite was detected. The observation that the oxygen effect is stronger for CP+-

bentonite as compared to CTMA+-bentonite is in line with the above-discussed observation of CdO band in the heated CP+-bentonite. However, the presence of the oxygen effect for CTMA+-bentonite strongly indicates that oxygen may catalyze the nucleophilic substitution reaction in organobentonites. The thermal pretreatment of the clay samples at 160 °C in air and in N2-atmosphere was carried out using the same type of test tubes. Similarly as in air, the treated samples showed changes in color intensity along the test tube profile with the strongest changes at the sample-atmosphere interface. The most likely reason for these color changes is the formation of organic molecules containing more than ∼8-10 conjugated double bonds necessary for visible light absorption (39). Although the color changes were not accompanied by changes in the iodide sorption capacity, an additional thermal pretreatment experiment was carried out, in which clay samples of the same weight as used with test tubes were spread in a thin layer on crystallization dishes. The most unexpected result of this experiment is the dependence of the thermal stability of the iodide sorption capacity on the fraction of the sample surface exposed to the N2-atmosphere observed for CTMA+-bentonite (Table 1): An increase by a factor of ∼2 in the exposed sample area led to a decrease by ∼60% in the iodide sorption of the pretreated clays. This observation indicates that thermal stability of the iodide sorption capacity might be governed by some process, which is inhibited at the clay-atmosphere interfaces in the pore space of the clay sample and occurs at the clayatmosphere interfaces exposed directly to a large atmosphere reservoir. Accordingly, evaporation of organic compounds, as discussed previously (24), or water, which has been shown to be present at least up to 180 °C in organoclays (23, 40) and up to 300 °C in nonmodified clays (14, 40), can be suggested as such a process. Indeed, evaporation occurs only at the exposed surface, whereas it is inhibited in the pore space due to adsorption on the pore-forming mineral surfaces. Molecules are forced to diffuse through the pore space toward the exposed surface in order to evaporate as also suggested in earlier studies (18, 20). The results of additional experiments, which are to be reported in detail elsewhere, showed in support of this suggestion that the thermal stability of the iodide sorption capacity increases as the pore space of the clay sample is reduced upon compaction. Thus, the most important finding from the experiments in N2-atmosphere is that the reaction leading to a decrease of the iodide sorption on organobentonites as a result of thermal pretreatment is strongly inhibited upon a decrease of both, the molecular oxygen content in the contacting atmosphere and the clay surface exposed to the latter. The iodide sorption of CP+-, BE+-, or CTMA+-bentonite after a four-day thermal pretreatment at 160 °C in N2-atmosphere corresponds to that at 140-145 °C in air (cf. Table 1 and Figure 1). It can be assumed that this inhibition effect depends only on the oxygen content itself and not on the applied temperature. Accordingly, the observed increase of the thermal stability of the organobentonites is equivalent to the corresponding left-hand shift by 15-20 °C (20 °C for CTMA+-bentonite) of the observed points and the fitted straight lines in the Figure 2. 3.3. Extrapolation of the Observed Data to Lower Pretreatment Temperatures. The Arrhenius constants and activation energies obtained as discussed for Figure 2 can be used to predict time periods during which the iodide sorption on the studied organobentonites will persist even though strongly decreased because of exposure to temperatures e100 °C (Table 2). The minimum temperature relevant for disposal equals ∼40 °C, which is the temperature characteristic for a depth of ∼1 km due to geothermal energy (1). Still, even lower temperatures are considered in Table 2 in order to

TABLE 2. Time Periods (in Years), within Which CP+-, BE+-, CTMA+-, and TPP+-Bentonites (in Powder Form) Will Be Able To Sorb Iodide in Air, Calculated Using Arrhenius Parameters As Discussed for Figure 2 T, °C

CP+

BE+

CTMA+

TPP+

100 90 80 70 60 50 40 30 20 10

0.9 2.3 7 20 ∼70 ∼230 ∼900 ∼3600 ∼1.7 × 104 ∼8 × 104

0.9 2.5 8 26 ∼90 ∼360 ∼1500 ∼7000 ∼3.6 × 104 ∼2 × 105

3.9 12 42 ∼150 ∼600 ∼2500 ∼1.2 × 104 ∼6 × 104 ∼3.5 × 105 ∼2.3 × 106

18 44 110 ∼290 ∼830 ∼2500 ∼8000 ∼2.8 × 104 ∼5.5 × 104 ∼4.5 × 105

account for the time periods becoming relevant upon the decreasing molecular oxygen supply as discussed in the previous section. This is justified considering that the molecular oxygen in a geological repository will be consumed within several years or at most a few tens of years after the repository closure (1). The actual temperature in the near-field of a geological repository of radioactive waste depends on the type and amount of waste, which is stored in a waste canister, as well as on the distance from the waste canister and time. According to the amounts of heat-producing radioactive waste (per waste canister) discussed currently in the literature, a temperature in the range between 50 and 70 °C (vitrified waste) or between 70 and 90 °C (spent fuel) will be characteristic for geotechnical barrier at distances beyond 0.75-1 m from the canister in the period between 10 and 1000 years after the waste deposition (1). More specifically, temperatures of 90 °C, 80 °C, 70 °C, 60 °C, and 50 °C will be reached after 20-30, 600, 1000, 2000, and 5000 years, respectively (41). The extrapolated values predict that between ∼0.7 (CP+) and ∼18 years (TPP+) would be necessary to carry out an experimental investigation of the thermal stability of the studied organobentonites at a pretreatment temperature of 100 °C (Table 2). For a temperature of 90 °C, which is relevant for disposal, the time requirement for the experiments increases to ∼2 (CP+), ∼12 years (CTMA+), or even ∼44 years (TPP+). At a temperature of 40 °C, which can be assumed to be the minimum temperature characteristic in a repository for nonheat-producing waste, the iodide sorption capacity of CTMA+-bentonite is predicted to persist within ∼12000 years as compared to ∼600 years for CP+-bentonite and ∼1500 years for BE+-bentonite. Importantly, although TPP+-bentonite shows the highest thermal stability at temperatures above ∼50 °C, it becomes the second most thermally stable organoclay at 40 °C. This is due to a much lower than for CTMA+-bentonite activation energy for the reaction leading to the loss of the iodide sorption capacity as discussed above (cf. corresponding slopes of the fitted straight lines in Figure 2). A similar change in the thermal stability sequence occurs for CP+- and BE+-bentonites at ∼90 °C (Table 2). The increase in the thermal stability of organoclays in the absence of molecular oxygen is equivalent to the shift of the values calculated for air to lower temperatures by at least 10 °C in a conservative approach (or 20 °C as actually observed). Accordingly, the most thermally stable CTMA+-bentonite will retain the iodide sorption capacity within 42 (or 150), 150 (or 600), 600 (or 2500), 2500 (or 1.2 × 104), 1.2 × 104 (or 6 × 104), and 6 × 104 (or 3.5 × 105) years at temperatures of 90 °C, 80 °C, 70 °C, 60 °C, 50 °C, and 40 °C, respectively. A comparison of the latter predicted values and the above-discussed temperature evolution in the near-field of a repository (1, 41) reveals that CTMA+-bentonite follows or overlaps this VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9315

temperature evolution even if the conservative approach is applied. The application of the actually observed correction suggests that iodide sorption capacity of CTMA+-bentonite will be largely preserved for the disposal relevant temperature and time periods. It should be stressed that the present study considered only two out of several parameters relevant for the possible geological repositories of radioactive waste: temperature and oxygen content. The bulk dry density of the bentonite backfill is another important parameter and should equal at least 1.75 g cm-3 (1) as compared to the density of ∼0.3 g cm-3 of the bentonite powders considered in the present study. Its increase exerts a positive influence on the thermal stability of the organobentonites as mentioned in section 3.2 and will be reported in detail elsewhere. As discussed above, the thermal stability of the iodide sorption capacity is very likely governed by evaporation processes. The increased pressure of about 10-12 MPa at the suggested disposal depth of ∼1 km strongly raises the boiling point for involved molecules (to ∼374 °C at 10 MPa for water). It might be therefore expected to lead to a much higher thermal stability of the iodide sorption capacity as compared to that in the present experiments at the laboratory pressure of 0.1 MPa. The positive effects of the increased density and pressure, if comparable with the observed positive effect of oxygen absence, can extend the applicability of CTMA+-bentonite for iodide sorption to time periods equivalent to those predicted for air and temperatures e10 °C (g2.3 million years, Table 2). Furthermore, the iodide sorption on CP+-bentonite has been found to be reduced by only 5% or 50% after the exposure to cobalt-60 γ radiation amounting in an accumulated dose of 0.66 or 4 MGy, respectively (32). The dose rate of cobalt-60 γ radiation has been shown to be attenuated by a factor of ∼20 in a bentonite layer with a thickness of 1 m and a density of 0.64 g cm-1 (32). This value and the γ radiation dose rate of ∼0.035 Gy h-1 at the canister surface at the time of canister emplacement (1) can be used to estimate that the γ radiation doses of 0.66 and 4 MGy will be accumulated in the organobentonites at the distance of 1 m from the canister only after ∼4 × 104 and 2.6 × 105 years, respectively. Considering that (i) the bentonite backfill should have a much higher density of 1.75 g cm-3 (1), (ii) most radionuclides relevant for geological disposal are characterized by γ radiation of lower energy than that of cobalt-60 (1.17 and 1.33 MeV), (iii) the total gamma activity within the canister decreases by about 3 orders of magnitude within 104-105 years due to radioactive decay (1), the above estimates of the long-term radiation stability of the organobentonites should be considered as very conservative ones. Summarizing, organobentonites if applied in geological repositories for radioactive waste can provide a long-term retardation of mobile anionic radionuclides. Their application is proposed to be restricted to the distances beyond 0.75-1 m from the canister with heat-producing waste to strongly reduce thermal or radiation damage to the modifying surfactants, whereas nonmodified bentonite should be used for backfilling purposes at the closer distances as proposed previously (1).

Acknowledgments We thank Prof. Dr. C. Vogt (Institute of Inorganic Chemistry) for giving us access to a µXRF instrument and M. Azeroual for carrying out µXRF measurements. We thank Dr. S. Dultz (Institute of Soil Science) for the CEC determination. We are also grateful to Dr. C. Bunnenberg for fruitful discussions and his help in preparation of the experiments in N2atmosphere and to Mrs. G. Erb-Bunnenberg for committed and skillful lab assistance. We thank three anonymous reviewers for very helpful comments. Financial support of 9316

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 24, 2010

Bundesministerium fu ¨ r Wirtschaft und Technologie under the contract No. 02 E 10025 is greatly acknowledged.

Literature Cited (1) Johnson, L.; Schneider, J.; Zuidema, P.; Gribi, P.; Mayer, G.; Smith, P. Project Opalinus Clay: Safety Report; Technical Report 02-05; Nagra, Wettingen (Switzerland), 2002. (2) Tournassat, C.; Gaucher, E. C.; Fattahi, M.; Grambow, B. On the mobility and potential retention of iodine in the CallovianOxfordian formation. Phys. Chem. Earth 2007, 32, 539–551. (3) Glaus, M.; Mu ¨ ller, W.; Van Loon, L. R. Diffusion of iodide and iodate through Opalinus Clay: Monitoring of the redox state using an anion chromatographic technique. Appl. Geochem. 2008, 23, 3612–3619. (4) Bors, J. Sorption of radioiodine in organo-clays and -soils. Radiochim. Acta 1990, 51, 139–143. (5) Bors, J.; Dultz, S.; Riebe, B. Retention of radionuclides by organophilic bentonite. Eng. Geol. 1999, 54, 195–206. (6) Kaufhold, S.; Pohlmann-Lortz, M.; Dohrmann, R.; Nu ¨ esch, R. About the possible upgrade of bentonite with respect to iodide retention capacity. Appl. Clay Sci. 2007, 35, 39–46. (7) Behnsen, J.; Riebe, B. Anion selectivity of organobentonites. Appl. Geochem. 2008, 23, 2746–2752. (8) Teng, B. L. G.; Greenland, D. J.; Quirk, J. P. Adsorption of alkylammonium cations by montmorillonite. Clay Miner. 1967, 7, 1–17. (9) Xu, S.; Boyd, S. A. Alternative model for cationic surfactant adsorption by layer silicates. Environ. Sci. Technol. 1995, 29, 3022–3028. (10) Xu, S.; Boyd, S. A. Cationic surfactant sorption to a vermiculitic subsoil via hydrophobic bonding. Environ. Sci. Technol. 1995, 29, 312–320. (11) Chen, Y. L.; Chen, S.; Frank, C.; Israelachvili, J. Molecular mechanisms and kinetics during the self-assembly of surfactant layers. J. Colloid Interface Sci. 1992, 153, 244–265. (12) He, H.; Zhou, Q.; Martens, W. N.; Kloprogge, T. J.; Yuan, P.; Xi, Y.; Zhu, J.; Frost, R. L. Microstructure of HDTMA+-modified montmorillonite and its influence on sorption characteristics. Clays Clay Miner. 2006, 54, 689–696. (13) Meleshyn, A. Cetylpyridinium aggregates at the montmorilloniteand muscovite-water interfaces: A Monte Carlo study of surface charge effect. Langmuir 2009, 25, 6250–6259. (14) Xie, W.; Gao, Z.; Pan, W.-P.; Hunter, D.; Singh, A.; Vaia, R. Thermal degradation chemistry of alkyl quaternary ammonium montmorillonite. Chem. Mater. 2001, 13, 2979–2990. (15) Davis, R. D.; Gilman, J. W.; Sutto, T. E.; Callahan, J. H.; Trulove, P. C.; De Long, H. C. Improved thermal stability of organically modified layered silicates. Clays Clay Miner. 2004, 52, 171–179. (16) Xi, Y.; Ding, Z.; He, H.; Frost, R. L. Structure of organoclays an X-ray diffraction and thermogravimetric analysis study. J. Colloid Interface Sci. 2004, 277, 116–120. (17) He, H.; Ding, Z.; Zhu, J.; Yuan, P.; Xi, Y.; Yang, D.; Frost, R. L. Thermal characterization of surfactant-modified montmorillonites. Clays Clay Miner. 2005, 53, 287–293. (18) Edwards, G.; Halley, P.; Kerven, G.; Martin, D. Thermal stability analysis of organo-silicates, using solid phase microextraction techniques. Thermochim. Acta 2005, 429, 13–18. (19) Cervantes-Uc, J. M.; Cauich-Rodriguez, J. V.; Vazquez-Torres, H.; Garfias-Mesias, L. F.; Paul, D. R. Thermal degradation of commercially available organoclays studied by TGA-FTIR. Thermochim. Acta 2007, 457, 92–102. (20) Bellucci, F.; Camino, G.; Frache, A.; Sarra, A. Catalytic charringevolatilization competition in organoclay nanocomposites. Polym. Degrad. Stab. 2007, 92, 425–436. (21) Stoeffler, K.; Lafleur, P. G.; Denault, J. Thermal decomposition of various alkyl onium organoclays: Effect on polyethylene terephtalate nanocomposite’s properties. Polym. Degrad. Stab. 2008, 93, 1332–1350. (22) Cui, L.; Khramov, M. D.; Bielawski, C. W.; Hunter, D. L.; Yoon, P. J.; Paul, D. R. Effect of organoclay purity and degradation on nanocomposite performance, Part 1: Surfactant degradation. Polymer 2008, 49, 3751–3761. (23) Kooli, F. Thermal stability investigation of organo-acid-activated clays by TG-MS and in situ XRD techniques. Thermochim. Acta 2009, 486, 71–76. (24) Shah, R. K.; Paul, D. R. Organoclay degradation in melt processed polyethylene nanocomposites. Polymer 2006, 47, 4075–4084. (25) Lamont, R. E.; Ducker, W. A. Surface-induced transformations for surfactant aggregates. J. Am. Chem. Soc. 1998, 120, 7602– 7607.

(26) Greenland, D. J.; Quirk, J. P. Adsorption of 1-n-alkyl pyridinium bromides by montmorillonite. Clays Clay Miner. 1962, 9, 484– 499. (27) Slade, P. G.; Gates, W. P. The swelling of HDTMA smectites as influenced by their preparation and layer charges. Appl. Clay Sci. 2004, 25, 93–101. (28) Riebe, B.; Dultz, S.; Bunnenberg, C. Temperature effects on iodine adsorption in organo-clays. I. Influence of pretreatment and sorption temperature. Appl. Clay Sci. 2005, 28, 9–16. (29) Riebe, B.; Bunnenberg, C. Influence of temperature pretreatment and high-molar saline solutions on the adsorption capacity of organo-clay minerals. Phys. Chem. Earth 2007, 32, 581–587. (30) Maes, N.; de Canniee´re, P.; Sillen, X.; van Ravestyn, L.; Put, M. DS 2.6 Migration of iodide in backfill materials containing an anion getter; Restricted contract report, SCK CEN-R-3896, CCHO-98/332-KNT909791501; 2004. (31) van Reeuwijk, L. P. Procedures for soil analysis; ISRIC Technical Paper 9; Wageningen (Netherland), 1993. (32) Jeschke, F. Einfluss von Gamma-Strahlung auf die Sorptionsfa¨higkeit von Organotonen fu ¨ r Iod. Diploma Thesis, Leibniz Universita¨t Hannover, 2007. www.zsr.uni-hannover.de/ arbeiten/dipljesc.pdf (accessed November 11, 2010). (33) Michel, R.; Kirchhoff, K. Nachweis-, Erkennungs- und Vertrauensgrenzen bei Kernstrahlungsmessungen. Fachverband fu ¨r ¨ V-Verlag: Ko¨ln, 1999. Strahlenschutz e. V., TU

(34) Brookins, D. Eh-pH diagrams for geochemistry; Springer: Heidelberg, 1988; p 15. (35) Kolb, V. M.; Stupar, J. W.; Janota, T. E.; Duax, W. L. Abnormally high IR frequencies for the carbonyl group of semicarbazones of the benzaldehyde and acetophenone series. J. Org. Chem. 1989, 54, 2341–2346. (36) Lo´pez Pasquali, C. E.; Herrera, H. Pyrolysis of lignin and IR analysis of residues. Thermochim. Acta 1997, 293, 39–46. (37) Taatjes, C. A.; Hansen, N.; Mcllroy, A.; Miller, J. A.; Senosiain, J. P.; et al. Enols are common intermediates in hydrocarbon oxidation. Science 2005, 308, 1887–1889. (38) Pistarino, C.; Finocchio, E.; Romezzano, G.; Brichese, F.; Di Felice, R.; Busca, G. A Study of the catalytic dehydrochlorination of 2-chloropropane in oxidizing conditions. Ind. Eng. Chem. Res. 2000, 39, 2752–2760. (39) Atkins, P.; De Paula, J. Physical Chemistry, 8th ed.; Oxford University Press: 2006; p 398. (40) Dultz, S.; Riebe, B.; Bunnenberg, C. Temperature effects on iodine adsorption on organo-clay minerals. II. Structural effects. Appl. Clay Sci. 2005, 28, 17–30. (41) Ho¨kmark, H.; Claesson, J. Use of an analytical solution for calculating temperatures in repository host rock. Eng. Geol. 2005, 81, 353–364.

ES100993C

VOL. 44, NO. 24, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9317