- Email: [email protected]
Am(III) adsorption on oxides of aluminium and silicon: effects of humic substances, pH, and ionic strength
Journal of Colloid and Interface Science 265 (2003) 221–226 www.elsevier.com/locate/jcis
Am(III) adsorption on oxides of aluminium and silicon: effects of humic substances, pH, and ionic strength Zuyi Tao,∗ Weijuan Li, Fuming Zhang, Youqian Ding, and Zhen Yu Radiochemistry Laboratory, College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou 730000, PR China Received 28 June 2002; accepted 20 November 2002
Abstract The adsorption of Am(III) (total concentration 10−9 mol/l) on alumina, silica, and hematite was studied by a batch technique. The effects of pH, ionic strength, and humic substances on the adsorption of Am(III) on alumina and silica were investigated, and the adsorption isotherms of Am(III) on alumina and silica at different pH values were determined. It was found that compared with the adsorption of Am(III) on alumina, the adsorbability of silica on the basis of mass is less, the relative adsorption rate on silica is slower, the sensitivity of adsorption on silica to ionic strength is less, the dependence of adsorption on silica on pH is gentler, and consequently that the adsorption characteristics of Am(III) on alumina and silica are distinctly different. The negative effect of fulvic acid on the adsorption on silica and the positive effect of humic acid on the adsorption on alumina were found. In contrast to the Am(III) adsorption on alumina and silica, a tremendously high adsorbability of Am(III) on hematite was found. The sequence of adsorbabilities of Am(III) on the basis of mass is Fe2 O3 > Al2 O3 > SiO2 . 2003 Elsevier Inc. All rights reserved.
1. Introduction This paper is an extension of some previous papers [1,2]. The effects of pH, ionic strength, and fulvic acid (FA) on sorption and desorption of Eu(III) and Yb(III) on alumina were investigated by using batch technique and radiotracers 152 + 154Eu and 169 Yb, and it was found that the surface hydrolysis model can satisfactorily explain the observations on the bare alumina and that the competition among the complexations of surface free hydroxyl groups and soluble and sorbed fulvic acids can satisfactorily explain the observations on the coated alumina [1]. The effects of pH, ionic strength, and humic substance (HS) on the sorption and desorption of Co(II) on alumina and silica were studied by using batch and column techniques and the radiotracer 60 Co. The distribution coefficients (Kd ), the breakthrough curves, and the displacement curves were determined. It was found that the sorption characteristics of Co(II) on alumina and silica are distinctly different, that the strong chemical bonds are formed between the bare alumina surface or the coated alumina surface and the Co(II), and * Corresponding author.
E-mail address: [email protected] (Z. Tao). 0021-9797/$ – see front matter 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0021-9797(02)00156-X
that a transition from the adsorption to the surface-induced precipitation of Co(II) on the bare alumina surface takes place [2]. The adsorption of Am(III) on geologic and nongeologic media has been studied extensively [3–8], because for the disposal of high level radioactive wastes, storage in a geological repository is visualized, and the total hazards from spent fuel will be dominated by 241Am after some 10 to 103 –104 years after discharge. Waste forms could be subjected to groundwater attack, and Am(III) could be released in the aquifer. Important mechanisms that control mobility and redistribution of Am(III) are precipitation and adsorption processes, and the adsorption of Am(III) is of course highly related to the pH, the ionic strength, and the composition of the aqueous phase and the nature of the solid phase. It was experimentally found in this study that at ini0 tial concentration CAm = 1.1 × 10−9 mol/l and background electrolyte concentration CNaNO3 or CCaCl2 = 0.1 mol/l, the adsorption percents of Am(III) on the natural hematite are always > 99% at pH 4–6 and V /m = 400 and 200 ml/g, where V is the solution volume, and m is the mass of solid. The adsorption percents are too high for us to investigate the Am(III) adsorption characteristics on hematite.
222
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Therefore, this study was focused on two solid phases— alumina and silica—as model solid phase. The effects of pH, ionic strength, and HS were investigated to elucidate the retention mechanisms. The objective of this paper was to determine and compare the adsorption characteristics of Am(III) on alumina and silica utilizing very low concentrations of Am(III). Consequently, the adsorption data obtained here were not due to the precipitation of Am(III).
2. Experimental The chromatographic Al2 O3 , 241Am, and FA from weathered coal used here were described in previous papers [1,2,9]. Natural hematite was a generous gift of the Geology Department of our University and was ground to a grain size < 0.01 mm. The humic acid (HA) used here was extracted from the peat of Lin Tan county (Gansu province) and purified. The ash content of the HA is less than 1%, and the moisture content is 7.6%. Its element composition on a moisture and ash-free basis is N (2.72%), C (53.65%), H (4.59%), S + O (39.04%). The peat was extracted with NaOH solution under an N2 atmosphere, the extract was centrifugalized and acidified to pH 1, and then redissolved with KOH, and recentrifugalized and acidified with HCl + HF. The product was desalted by dialysis. Commercial SiO2 (Spectral pure) was purchased from the Shanghai First Chemical Reagent Factory (China). Part of the silica was treated with 0.01 mol/l NaOH for 24 h to remove small amounts of organic contamination [10]. The silica without NaOH treatment and the silica treated with NaOH were simultaneously used in the studies reported in this paper. Distilled, deionized water without further membrane filtration and deaeration was used. The conditioning and the storage of Al2 O3 , SiO2 , and Fe2 O3 and the procedures of batch experiments were identical to those in previous papers [1,2]. Contact times between the two phases of 25 and 20 h, respectively, were suggested for alumina and silica from kinetics experiments. The formula used to calculate the distribution coefficient (Kd = ratio of Am solid phase concentration (mol/g) to Am solution phase concentration (mol/ml)) was similar to that in the previous paper [1]. Since the adsorption of Am(III) on the wall of polyethylene test tubes cannot be neglected and must be corrected, in the adsorption experiments, after separating the solid phase from the aqueous solutions by centrifugation and removing the bulk of the solutions from the test tubes, the Am(III) adsorbed on the test tube walls was desorbed with 5 ml of 5 mol/l HNO3 in duplicate. Each contact time of test tube walls with HNO3 was 4 h. The inner surface (geometrical) of the test tubes (25 ml, ∅19 × 100 mm) is 58 ± 1 cm2 . The γ -activities of the 10 ml of 5 mol/l HNO3 and of the SiO2 or Al2 O3 were measured. Thus the count rates of adsorbed material on the solid phase and on the test tube wall were obtained [11].
In the adsorption experiments on silica, the solid phase concentrations of Am(III) (Am, mol/g) were obtained from the difference between the introduced count rate and the sum of count rates of the supernatant solution and of the Am(III) adsorbed onto the test tube wall. In the adsorption experiments on alumina, the solid phase concentrations of Am(III) (Am, mol/g) were directly obtained from the count rate of Al2 O3 . It was found that the errors between the introduced count rates and the total count rates (= the remaining count rate of supernatant solution + the count rate of Am(III) adsorbed on the test tube wall + the count rate of Am(III) adsorbed on the solid phase) range from 1 to 6% [11].
3. Results 3.1. Kinetics The results of adsorption kinetics of Am(III) on alumina and silica, respectively, are illustrated in Figs. 1 and 2. It was found that steady states for both silica without NaOH treatment and for that treated with NaOH are simultaneously reached after about 5 h, though the adsorption percent at equilibrium for the silica treated with NaOH is about 2.5 times that for the silica without NaOH treatment. In other words, the relative rates of Am(III) adsorption on both silicas are very similar, and the equilibration times for both silicas are identical. Compared with the relative rate and the equilibration time (25 h) of Am(III) adsorption on alumina, the adsorption of Am(III) on both silicas is a rapid kinetic process. However, compared with the kinetics and the equilibration time (20 min) for Cs+ adsorption on the same alumina [12], the adsorption of Am(III) on silica and alumina is a much slower kinetic process. Consequently, contact times between the two phases of 25 and 20 h, respectively, were used for alumina and silica.
Fig. 1. Variation in adsorption percentage of Am(III) on alumina as a 0 = 1.1 × 10−9 mol/l, function of shaking time. V /m = 500 ml/g, CAm I = 0.1 mol/l NaNO3 , pH 4.9 ± 0.1, T = 25 ◦ C. V , solution volume; 0 , initial concentration of Am(III). s, mass of solid; CAm
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Fig. 2. Variation in adsorption percentage of Am(III) on silica as a 0 = function of shaking time. •: treated silica, V /m = 50 ml/g, CAm −9 ◦ 1.2 × 10 mol/l, I = 0.01 mol/l NaNO3 , pH 3.4 ± 0.1, T = 25 C. ◦: sil0 = 1.7 × 10−9 mol/l, ica without NaOH treatment, V /m = 50 ml/g, CAm I = 0.1 mol/l NaNO3 , pH 3.9 ± 0.1, T = 25 ◦ C.
Fig. 3. Variation in adsorption percentage of Am(III) on alumina as a 0 = 1.1 × 10−9 mol/l, I = 0.1 mol/l NaNO , function of V /m. CAm 3 pH 3.4 ± 0.1, T = 25 ◦ C, 30 h.
3.2. The effect of the ratio of solution volume to mass of solid (V /m, ml/g) The effects of V /m on the adsorption percent of Am(III) on alumina and silica are shown in Figs. 3 and 4, respectively. Figure 3 shows that the adsorption percents on alumina are larger than 99% at V /m = 20, 50, and 100 ml/g and that the adsorption percents respectively decrease to 98% at V /m = 200 ml/g and to 51% at V /m = 500 ml/g. Thus V /m = 500 ml/g was used in the adsorption experiments on alumina. Figure 4 shows that the adsorption percents on the silica without NaOH treatment decrease gradually with increasing V /m from 20 to 200 ml/g and remain practically constant at V /m of 200 and 500 ml/g. Afterward, V /m = 50 ml/g was used in the adsorption experiments on the silica without NaOH treatment. 3.3. The effect of pH In the absence of HS and at both ionic strengths (0.01 and 0.1 mol/l NaNO3 ), the Kd values of Am(III) adsorption on alumina show a typical pH dependence in the pH range 4–8
223
Fig. 4. Variation in adsorption percentage of Am(III) on silica as a function 0 = 9.2 × 10−10 mol/l, I = 0.01 mol/l NaNO , pH 3.4 ± 0.1, of V /m. CAm 3 T = 25 ◦ C, 25 h.
(Fig. 5). The typical pH dependence was also found for the adsorption of Eu(III) and Yb(III) on the same alumina [1], and for the adsorption of Am(III) on α- and γ -Al2 O3 [5,7] and on granite [8]. The steep increase in adsorption is a typical behavior for the adsorption of hydrolyzable transition metal ions on hydrous oxides [13]. In the absence of HS and at ionic strength 0.1 mol/l NaNO3 , the adsorption of Am(III) on both silica without NaOH treatment and that treated with NaOH also shows the increase in Kd values with increasing pH in the pH range 1–5.5 (Fig. 6). However, the log Kd –pH curves for silica without NaOH treatment differ from those for treated silica. At pH < 6.5, the Kd values for the treated silica are larger than the corresponding values for the silica without NaOH treatment; while at pH > 6.5, the opposite occurs. For the silica treated with NaOH, at pH > 5.5, the Kd values decrease with increasing pH (i.e., there is a maximum in the log Kd –pH curves), while for the silica without NaOH treatment, there is no maximum in the log Kd –pH curve and the Kd values increase monotonically in the pH range 1.5–7.2. 3.4. The effect of ionic strength The effects of ionic strength on Am(III) adsorption on alumina and silica are shown in Figs. 7 and 8, respectively. It was found that the Kd values on alumina gradually increase with increasing ionic strength from 0.01 to 2 mol/l NaNO3 , while the Kd values on silica increase from ionic strength 0.001 to 0.01 mol/l NaNO3 and remain practically constant from ionic strength 0.01 to 2 mol/l NaNO3 . In other words, the Kd values on silica are practically independent of the ionic strength from 0.01 to 2 mol/l NaNO3 , while the Kd values on alumina are sensitively dependent on the ionic strength from 0.001 to 2 mol/l NaNO3 , though the Kd values on alumina are obviously larger than the corresponding values on silica.
224
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Fig. 5. Variation in Kd values of Am(III) adsorption on alumina as a 0 = 1.1 × 10−9 mol/l, V /m = 500 ml/g, T = 25 ◦ C. function of pH. CAm
Fig. 6. Variation in Kd values of Am(III) adsorption on silica as a function 0 = 1.2 × 10−9 mol/l, V /m = 50 ml/g, I = 0.1 mol/l NaNO , of pH. CAm 3 T = 25 ◦ C.
3.5. Adsorption isotherms in the absence and presence of HS The isotherms of Am(III) adsorption on alumina at pH 3.5, 4.5, and 4.9 in the absence of FA and at pH 4.5 in the presence of FA are shown in Fig. 9. The isotherms of Am(III) adsorption on silica at pH 1.7, 5.5, and 5.9 in the absence of HA and at pH 6.3 in the presence of HA are shown in Fig. 10. All the isotherms in both adsorption d values of these linear systems are linear. The average K d values isotherms are denoted in Figs. 9 and 10. These K also indicate the strong effect of pH, the positive effect of FA at low pH, and the negative effect of HA at high pH.
4. Discussion The distinct difference in adsorption characteristics between alumina and silica, the obvious difference in log Kd – pH curves between the silicas (Fig. 6), and the tremendously high adsorbabilitity of Am(III) on hematite on the basis of mass in particular imply that the properties of the solid phase surface are primarily controlling the Am(III) adsorption instead of the species of dissolved Am(III) in aqueous solutions. First and foremost we should deal with the difference
Fig. 7. Variation in Kd values of Am(III) adsorption on alumina as a 0 = 1.1 × 10−9 mol/l, V /m = 500 ml/g, function of ionic strength. CAm T = 25 ◦ C.
Fig. 8. Variation in Kd values of Am(III) adsorption on silica without 0 = 1.7 × 10−9 mol/l, NaOH treatment as a function of ionic strength. CAm V /m = 50 ml/g, pH 1.5, T = 25 ◦ C.
in having electrical charges at surfaces of both solid phases of alumina and silica. It is well known that the pHpzc of silica ranges from 1.8 to 3.5 [14,15]. Therefore, at pH < 3, the surface of silica carries positive charges; at pH > 3, the surface of silica carries negative charges; at pH > 3, the surface free ≡SOH concentration would decrease and the surface ≡SO− concentration would increase with increasing pH. The pHpzc of the alumina used here is 7.5 and it was found that the predominant surface species is always ≡SOH and the surface free ≡SOH concentration is an order or several orders of magnitude larger than the concentrations of the other surface species and that the surface free ≡SOH concentration slowly increases from pH 3 to 7.5 and then decreases at pH > 7.5 [16]. Therefore, at pH < 7.5, the surface of alumina carries positive charges; and at pH > 7.5, the surface of alumina carries negative charges. The solubility limiting phases of Am(III) are Am2 (CO3 )3 and possibly Am(OH)3 at low carbonate concentrations and high pH values. The total solubility would have a minimum of about 10−7 –10−8 mol/l in the pH range 7–9 [4]. Pryke and Rees [17] experimentally determined the solubilities of Am(III) in simulated pore water containing Ca2+ , Na+ , Mg2+ , Cl− , SO4 2− , and CO3 2− ions at pH 7–13 and found that the solubilities grad-
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Fig. 9. Isotherms of Am(III) adsorption on alumina. V /m = 500 ml/g, I = 0.1 mol/l NaNO3 , T = 25 ◦ C. •: pH 3.5 ± 0.1, no FA; ×: pH 4.5 ± 0.1, no FA; ◦: pH 4.9 ± 0.1, no FA; : pH 4.5 ± 0.1, initial concentration of FA = 20 mg/l. Kd values are shown in boxes.
Fig. 10. Isotherms of Am(III) adsorption on silica without NaOH treatment. V /m = 50 ml/g, I = 0.1 mol/l NaNO3 , T = 25 ◦ C. •: pH 1.7±0.1, no FA; ×: pH 5.5 ± 0.1, no FA; ◦: pH 5.9 ± 0.1, no FA; : pH 6.3 ± 0.1, initial concentration of HA = 20 mg/l. Kd values are shown in boxes.
ually decrease from 10−5.5 mol/l at pH 7 to 10−8.5 mol/l at pH 9. The total concentrations of Am(III) used here were smaller than the experimental solubilities. Consequently, the adsorption data of Am(III) obtained in this paper are not due to the precipitation of compounds of Am(III). Am3+ carbonate complexes AmCO3 + and Am(CO3 )2 − as well as the hydroxides Am(OH)2+ and Am(OH)2 + may be the predominant species in aqueous solutions in this study. The species distribution plot for the Am(III)–OH– CO3 system at ionic strength I of 0.1 mol/l NaClO4 , and pCO2 = 10−3.5 atm was drawn by Moulin et al. [18], it was found that at pH < 5, Am3+ is almost the sole species; at pH 5–6, Am3+ is the predominant species and Am(OH)2+ is less than 10%; at pH 6–8, Am3+ , Am(OH)2+ , Am(OH)2 + , and Am(CO3 )+ are present; at pH 8–9, Am(OH)2+ , Am(OH)2 + , Am(CO3 )+ , and Am(CO3 )2 − are present. According to the thermodynamic model for adsorption of James and Healy [13], the Gibbs free energy (Gads ) involves not only the coulombic electrostatic energy (Gcoul )
225
but also the chemical bond energy (Gchem ) and the hydration energy (Ghyd). Because the dehydration energy is always positive and deleterious to adsorption of adsorbing ions and because the electrostatic repulsion between the positively charged species of Am(III) in aqueous solutions and the positively charged surface of alumina at pH < 7.5 also is deleterious to adsorption, the sole driving force for Am(III) adsorption on alumina is the Gchem . In other words, strong chemical bonds are formed between the surface of alumina and the Am(III). Although the electrostatic attraction between the positively charged species of Am(III) solutions and the negatively charged surface of silica at pH > 3 is favorable to adsorption, as a whole, the Am(III) adsorption on silica does not appear to be as strong as on alumina. Therefore, the chemical bonds between the surface of silica and the Am(III) must be weaker than those between the surface of alumina and the Am(III). The log Kd –pH curves at ionic strength 0.01 or 0.1 mol/l NaNO3 (Fig. 5) are very similar to that of Yb(III) sorption on alumina at pH < 10 [1]. Thus the surface hydrolysis model [19] would explain the log Kd –pH curve of Am(III) adsorption on the bare alumina (Fig. 5), the availability of surface free ≡SOH groups is an essential requirement of the adsorption process, and the increase in free ≡SOH concentration is the cause of the steep increase in Kd values. In contrast, the log Kd –pH curves of Am(III) adsorption on the bare silica cannot be explained by the surface hydrolysis model, because at pH > 3 the surface free ≡SOH concentration would decrease with increasing pH. The increase in Kd values of Am(III) adsorption on the bare silica with increasing pH can be partially explained by the increase in electrostatic attraction between the positively charged species of Am(III) and the negative charged surface of bare silica at pH > 3. This explanation implies that ion exchange is a significant mechanism for Am(III) adsorption on silica. The slight decrease in Kd of Am(III) adsorption on silica with increasing pH at pH > 6 may be the result of formation of carbonate complex Am(CO3 )2 − in aqueous solutions and/or the dissolution of silica at higher pH. The similar log Kd –pH curves showing Kd values increasing from low values to maximum and then slightly decreasing from maximum value with increasing pH were also found by several other authors [3,4,21]. Depending on the adsorption system on oxides, adsorption of metal ions at oxide–aqueous solution interfaces has been reported to increase, remain the same, and decrease with increasing ionic strength. It was found that the values of Kd of Cs+ adsorption on the alumina used here decrease with increasing ionic strength [12] and that the values of Kd of Eu(III) and Co(II) sorptions on the alumina used here are practically independent of ionic strength [1,2]. Figure 7 shows that the Kd values of Am(III) adsorption on alumina increase gradually from ionic strength 0.001 to 2 mol/l NaNO3 . Figure 8 shows that the Kd values of Am(III) adsorption on the silica without NaOH treatment increase from 0.001 to 0.01 mol/l NaNO3 and then are
226
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
practically independent of ionic strength from 0.01 to 2 mol/l NaNO3 . A promotive effect of ionic strength on Am(III) adsorption on alumina and silica was found. The promotive ionic strength effect may be explained in terms of more effective shielding of highly charged surface species at higher ionic strength and in terms of the formation of ion pairs between the surface species and the background electrolyte anion [22,23]. The promotive effect was observed in the goethite–Cd(II) [24] and in the hematite–Pb(II) [25] adsorption systems for a long time. Since silica carries a negative surface charge at pH > 3, the negative charge allows an electrostatic binding of positively charged species of Am(III) on the silica surface. However, the strength of this binding may be too strong and so the monovalent electrolyte cation, Na+ , cannot compete for the ≡SO− groups. The ionic strength effect is rather complex and it cannot be interpreted properly at present. The isotherms of Am(III) adsorption on alumina (Fig. 9) and on silica (Fig. 10) in the absence and presence of HS are linear over the solid phase concentration ranges, Am(III), 0–1.6 × 10−9 and 0–1.0 × 10−10 mol/g, respectively. Figures 9 and 10 show the strong effects of pH on the Am(III) adsorption on alumina and silica and the positive and negative effects of HS on Am(III) adsorption. On the bare alumina, the average distribution coefficients (K d ) increase from 18 ml/g at pH 3.5 to 2 × 103 ml/g at pH 4.8 under otherwise equal conditions. On the bare silica, the average distribution coefficient (K d ) increase from 28 ml/g at pH 1.7 to 338 ml/g at pH 5.9 under otherwise equal conditions. The positive effect of FA on the Am(III) adsorption on alumina is demonstrated at low pH 4.5 and initial FA concentration of 20 mg/l, whereas the negative effect of HA on the Am(III) adsorption on silica is demonstrated at high pH 6.5 and initial HA concentration of 20 mg/l. The linearity of isotherms implies that the same surface species of Am(III) is formed over the solid phase concentration range. However, it is impossible for us to indicate the structure of this surface species of Am(III) from the macroscopic observations. The positive and the negative effects are consistent with the general consensus that the adsorption of inorganic cations is increased at low pH and decreased at high pH by adding HS [21]. The negative effect of HA on the Am(III) adsorption on silica at pH 6.5 also is consistent with adsorption of organic matter only being measurable at pH values less than 3 [10] and the complexes with soluble HA being formed in aqueous solution [21]. Adsorption of trivalent cations is more complicated than that of bivalent cations. There is need for further systematic study of adsorption of trivalent cations on oxides by using macroscopic and microscopic experiments [26].
5. Conclusions (1) Under aerobic condition (Eh > 300 mV), in the presence of atmosphere and at relatively lower concentra-
tion (∼10−9 mol/l), because Am(III) exists in positively charged species Am3+ , Am(OH)2+ , Am(OH)2 + , Am(CO3 )+ and negatively charged species Am(CO3 )2 − , and these species are in equilibria with each other in liquid phase and these equilibria are shifted with changing pH, many surface reactions are responsible for the overall adsorption of Am(III) on a surface, and the factors affecting the adsorption process are numerous. (2) The effect of humic substance on the adsorption of Am(III) is very important under aerobic condition, in the presence of atmosphere and at relatively lower concentration (∼10−9 mol/l). (3) The adsorption of Am(III) on alumina under aerobic condition is obviously stronger than that on silica, and the sequence of adsorbabilities of Am(III) on the basis of mass is Fe2 O3 > Al2 O3 > SiO2 .
References [1] X.K. Wang, W.M. Dong, X.X. Dai, A.X. Wang, J.Z. Du, Z.Y. Tao, Appl. Radiat. Isotopes 52 (2000) 165. [2] X.K. Wang, W.M. Dong, G. Wang, Z.Y. Tao, Appl. Radiat. Isotopes 56 (2002) 765. [3] B. Allard, G.W. Beall, T. Krajewski, Nucl. Technol. 49 (1980) 474. [4] B. Allard, U. Olofsson, B. Torstenfelt, Inorg. Chem. Acta 94 (1984) 205. [5] L. Righetto, G. Bidoglio, B. Marcandalli, I.R. Bellobono, Radiochim. Acta 44/45 (1988) 73. [6] I.A. Triay, A. Meijer, M.R. Cisneros, G.G. Miller, A.J. Mitchell, M.A. Ott, D.E. Hobart, P.D. Palmer, R.E. Perrin, R.D. Agnilar, Radiochim. Acta 52/53 (1991) 141. [7] V. Moulin, D. Stammose, G. Ouzounian, Appl. Geochem. Suppl. Issue 1 (1992) 163. [8] A. Kitamura, T. Yamamoto, S. Nishikawa, M. Moriyama, J. Radioanal. Nucl. Chem. 239 (1999) 449. [9] W.M. Dong, H.X. Zhang, M.D. Huang, Z.Y. Tao, Appl. Radiat. Isotopes 56 (2001) 595. [10] J.A. Davis, Geochim. Cosmochim. Acta 46 (1982) 2361. [11] W.J. Li, Y.Q. Ding, Z.Y. Tao, J. Radioanal. Nucl. Chem. 250 (2001) 497. [12] T.W. Chu, J.Z. Du, J.R. Lu, Z.Y. Tao, J. Radioanal. Nucl. Chem. 210 (1996) 197. [13] R.O. James, T.M. Healy, J. Colloid Interface Sci. 40 (1972) 65. [14] Z.Y. Tao, H.X. Zhang, J. Colloid Interface Sci. 252 (2002) 15. [15] G.A. Park, Chem. Rev. 65 (1965) 177. [16] T.W. Chu, J.Z. Du, Z.Y. Tao, Adsorption Sci. Technol. 15 (1997) 349. [17] D.C. Pryke, J.H. Rees, Radiochim. Acta 40 (1986) 27. [18] V. Moulin, P. Robouch, P. Vitorge, B. Allard, Radiochim. Acta 44/45 (1988) 33. [19] A.R. Gupta, B. Venkataramani, Bull. Chem. Soc. Jpn. 61 (1988) 1357. [20] M. Norden, J.H. Ephraim, B. Allard, Radiochim. Acta 65 (1994) 265. [21] Z.Y. Tao, T.W. Chu, J.Z. Du, X.X. Dai, Y.J. Gu, Appl. Geochem. 15 (2000) 133. [22] J. Lützenkirchen, J. Colloid Interface Sci. 195 (1997) 149. [23] M. Kosmulski, J. Colloid Interface Sci. 195 (1997) 395. [24] I. Lamy, M. Djafer, M. Terce, Water, Air, Soil Pollution 57–58 (1991) 457. [25] S. Ardizzone, L. Formaro, J. Lyklema, J. Electroanal. Chem. 133 (1982) 147. [26] Th. Rabung, Th. Stampf, H. Geckeis, R. Klenze, J.I. Kim, Radiochim. Acta 88 (2000) 711.
Am(III) adsorption on oxides of aluminium and silicon: effects of humic substances, pH, and ionic strength Zuyi Tao,∗ Weijuan Li, Fuming Zhang, Youqian Ding, and Zhen Yu Radiochemistry Laboratory, College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou 730000, PR China Received 28 June 2002; accepted 20 November 2002
Abstract The adsorption of Am(III) (total concentration 10−9 mol/l) on alumina, silica, and hematite was studied by a batch technique. The effects of pH, ionic strength, and humic substances on the adsorption of Am(III) on alumina and silica were investigated, and the adsorption isotherms of Am(III) on alumina and silica at different pH values were determined. It was found that compared with the adsorption of Am(III) on alumina, the adsorbability of silica on the basis of mass is less, the relative adsorption rate on silica is slower, the sensitivity of adsorption on silica to ionic strength is less, the dependence of adsorption on silica on pH is gentler, and consequently that the adsorption characteristics of Am(III) on alumina and silica are distinctly different. The negative effect of fulvic acid on the adsorption on silica and the positive effect of humic acid on the adsorption on alumina were found. In contrast to the Am(III) adsorption on alumina and silica, a tremendously high adsorbability of Am(III) on hematite was found. The sequence of adsorbabilities of Am(III) on the basis of mass is Fe2 O3 > Al2 O3 > SiO2 . 2003 Elsevier Inc. All rights reserved.
1. Introduction This paper is an extension of some previous papers [1,2]. The effects of pH, ionic strength, and fulvic acid (FA) on sorption and desorption of Eu(III) and Yb(III) on alumina were investigated by using batch technique and radiotracers 152 + 154Eu and 169 Yb, and it was found that the surface hydrolysis model can satisfactorily explain the observations on the bare alumina and that the competition among the complexations of surface free hydroxyl groups and soluble and sorbed fulvic acids can satisfactorily explain the observations on the coated alumina [1]. The effects of pH, ionic strength, and humic substance (HS) on the sorption and desorption of Co(II) on alumina and silica were studied by using batch and column techniques and the radiotracer 60 Co. The distribution coefficients (Kd ), the breakthrough curves, and the displacement curves were determined. It was found that the sorption characteristics of Co(II) on alumina and silica are distinctly different, that the strong chemical bonds are formed between the bare alumina surface or the coated alumina surface and the Co(II), and * Corresponding author.
E-mail address: [email protected] (Z. Tao). 0021-9797/$ – see front matter 2003 Elsevier Inc. All rights reserved. doi:10.1016/S0021-9797(02)00156-X
that a transition from the adsorption to the surface-induced precipitation of Co(II) on the bare alumina surface takes place [2]. The adsorption of Am(III) on geologic and nongeologic media has been studied extensively [3–8], because for the disposal of high level radioactive wastes, storage in a geological repository is visualized, and the total hazards from spent fuel will be dominated by 241Am after some 10 to 103 –104 years after discharge. Waste forms could be subjected to groundwater attack, and Am(III) could be released in the aquifer. Important mechanisms that control mobility and redistribution of Am(III) are precipitation and adsorption processes, and the adsorption of Am(III) is of course highly related to the pH, the ionic strength, and the composition of the aqueous phase and the nature of the solid phase. It was experimentally found in this study that at ini0 tial concentration CAm = 1.1 × 10−9 mol/l and background electrolyte concentration CNaNO3 or CCaCl2 = 0.1 mol/l, the adsorption percents of Am(III) on the natural hematite are always > 99% at pH 4–6 and V /m = 400 and 200 ml/g, where V is the solution volume, and m is the mass of solid. The adsorption percents are too high for us to investigate the Am(III) adsorption characteristics on hematite.
222
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Therefore, this study was focused on two solid phases— alumina and silica—as model solid phase. The effects of pH, ionic strength, and HS were investigated to elucidate the retention mechanisms. The objective of this paper was to determine and compare the adsorption characteristics of Am(III) on alumina and silica utilizing very low concentrations of Am(III). Consequently, the adsorption data obtained here were not due to the precipitation of Am(III).
2. Experimental The chromatographic Al2 O3 , 241Am, and FA from weathered coal used here were described in previous papers [1,2,9]. Natural hematite was a generous gift of the Geology Department of our University and was ground to a grain size < 0.01 mm. The humic acid (HA) used here was extracted from the peat of Lin Tan county (Gansu province) and purified. The ash content of the HA is less than 1%, and the moisture content is 7.6%. Its element composition on a moisture and ash-free basis is N (2.72%), C (53.65%), H (4.59%), S + O (39.04%). The peat was extracted with NaOH solution under an N2 atmosphere, the extract was centrifugalized and acidified to pH 1, and then redissolved with KOH, and recentrifugalized and acidified with HCl + HF. The product was desalted by dialysis. Commercial SiO2 (Spectral pure) was purchased from the Shanghai First Chemical Reagent Factory (China). Part of the silica was treated with 0.01 mol/l NaOH for 24 h to remove small amounts of organic contamination [10]. The silica without NaOH treatment and the silica treated with NaOH were simultaneously used in the studies reported in this paper. Distilled, deionized water without further membrane filtration and deaeration was used. The conditioning and the storage of Al2 O3 , SiO2 , and Fe2 O3 and the procedures of batch experiments were identical to those in previous papers [1,2]. Contact times between the two phases of 25 and 20 h, respectively, were suggested for alumina and silica from kinetics experiments. The formula used to calculate the distribution coefficient (Kd = ratio of Am solid phase concentration (mol/g) to Am solution phase concentration (mol/ml)) was similar to that in the previous paper [1]. Since the adsorption of Am(III) on the wall of polyethylene test tubes cannot be neglected and must be corrected, in the adsorption experiments, after separating the solid phase from the aqueous solutions by centrifugation and removing the bulk of the solutions from the test tubes, the Am(III) adsorbed on the test tube walls was desorbed with 5 ml of 5 mol/l HNO3 in duplicate. Each contact time of test tube walls with HNO3 was 4 h. The inner surface (geometrical) of the test tubes (25 ml, ∅19 × 100 mm) is 58 ± 1 cm2 . The γ -activities of the 10 ml of 5 mol/l HNO3 and of the SiO2 or Al2 O3 were measured. Thus the count rates of adsorbed material on the solid phase and on the test tube wall were obtained [11].
In the adsorption experiments on silica, the solid phase concentrations of Am(III) (Am, mol/g) were obtained from the difference between the introduced count rate and the sum of count rates of the supernatant solution and of the Am(III) adsorbed onto the test tube wall. In the adsorption experiments on alumina, the solid phase concentrations of Am(III) (Am, mol/g) were directly obtained from the count rate of Al2 O3 . It was found that the errors between the introduced count rates and the total count rates (= the remaining count rate of supernatant solution + the count rate of Am(III) adsorbed on the test tube wall + the count rate of Am(III) adsorbed on the solid phase) range from 1 to 6% [11].
3. Results 3.1. Kinetics The results of adsorption kinetics of Am(III) on alumina and silica, respectively, are illustrated in Figs. 1 and 2. It was found that steady states for both silica without NaOH treatment and for that treated with NaOH are simultaneously reached after about 5 h, though the adsorption percent at equilibrium for the silica treated with NaOH is about 2.5 times that for the silica without NaOH treatment. In other words, the relative rates of Am(III) adsorption on both silicas are very similar, and the equilibration times for both silicas are identical. Compared with the relative rate and the equilibration time (25 h) of Am(III) adsorption on alumina, the adsorption of Am(III) on both silicas is a rapid kinetic process. However, compared with the kinetics and the equilibration time (20 min) for Cs+ adsorption on the same alumina [12], the adsorption of Am(III) on silica and alumina is a much slower kinetic process. Consequently, contact times between the two phases of 25 and 20 h, respectively, were used for alumina and silica.
Fig. 1. Variation in adsorption percentage of Am(III) on alumina as a 0 = 1.1 × 10−9 mol/l, function of shaking time. V /m = 500 ml/g, CAm I = 0.1 mol/l NaNO3 , pH 4.9 ± 0.1, T = 25 ◦ C. V , solution volume; 0 , initial concentration of Am(III). s, mass of solid; CAm
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Fig. 2. Variation in adsorption percentage of Am(III) on silica as a 0 = function of shaking time. •: treated silica, V /m = 50 ml/g, CAm −9 ◦ 1.2 × 10 mol/l, I = 0.01 mol/l NaNO3 , pH 3.4 ± 0.1, T = 25 C. ◦: sil0 = 1.7 × 10−9 mol/l, ica without NaOH treatment, V /m = 50 ml/g, CAm I = 0.1 mol/l NaNO3 , pH 3.9 ± 0.1, T = 25 ◦ C.
Fig. 3. Variation in adsorption percentage of Am(III) on alumina as a 0 = 1.1 × 10−9 mol/l, I = 0.1 mol/l NaNO , function of V /m. CAm 3 pH 3.4 ± 0.1, T = 25 ◦ C, 30 h.
3.2. The effect of the ratio of solution volume to mass of solid (V /m, ml/g) The effects of V /m on the adsorption percent of Am(III) on alumina and silica are shown in Figs. 3 and 4, respectively. Figure 3 shows that the adsorption percents on alumina are larger than 99% at V /m = 20, 50, and 100 ml/g and that the adsorption percents respectively decrease to 98% at V /m = 200 ml/g and to 51% at V /m = 500 ml/g. Thus V /m = 500 ml/g was used in the adsorption experiments on alumina. Figure 4 shows that the adsorption percents on the silica without NaOH treatment decrease gradually with increasing V /m from 20 to 200 ml/g and remain practically constant at V /m of 200 and 500 ml/g. Afterward, V /m = 50 ml/g was used in the adsorption experiments on the silica without NaOH treatment. 3.3. The effect of pH In the absence of HS and at both ionic strengths (0.01 and 0.1 mol/l NaNO3 ), the Kd values of Am(III) adsorption on alumina show a typical pH dependence in the pH range 4–8
223
Fig. 4. Variation in adsorption percentage of Am(III) on silica as a function 0 = 9.2 × 10−10 mol/l, I = 0.01 mol/l NaNO , pH 3.4 ± 0.1, of V /m. CAm 3 T = 25 ◦ C, 25 h.
(Fig. 5). The typical pH dependence was also found for the adsorption of Eu(III) and Yb(III) on the same alumina [1], and for the adsorption of Am(III) on α- and γ -Al2 O3 [5,7] and on granite [8]. The steep increase in adsorption is a typical behavior for the adsorption of hydrolyzable transition metal ions on hydrous oxides [13]. In the absence of HS and at ionic strength 0.1 mol/l NaNO3 , the adsorption of Am(III) on both silica without NaOH treatment and that treated with NaOH also shows the increase in Kd values with increasing pH in the pH range 1–5.5 (Fig. 6). However, the log Kd –pH curves for silica without NaOH treatment differ from those for treated silica. At pH < 6.5, the Kd values for the treated silica are larger than the corresponding values for the silica without NaOH treatment; while at pH > 6.5, the opposite occurs. For the silica treated with NaOH, at pH > 5.5, the Kd values decrease with increasing pH (i.e., there is a maximum in the log Kd –pH curves), while for the silica without NaOH treatment, there is no maximum in the log Kd –pH curve and the Kd values increase monotonically in the pH range 1.5–7.2. 3.4. The effect of ionic strength The effects of ionic strength on Am(III) adsorption on alumina and silica are shown in Figs. 7 and 8, respectively. It was found that the Kd values on alumina gradually increase with increasing ionic strength from 0.01 to 2 mol/l NaNO3 , while the Kd values on silica increase from ionic strength 0.001 to 0.01 mol/l NaNO3 and remain practically constant from ionic strength 0.01 to 2 mol/l NaNO3 . In other words, the Kd values on silica are practically independent of the ionic strength from 0.01 to 2 mol/l NaNO3 , while the Kd values on alumina are sensitively dependent on the ionic strength from 0.001 to 2 mol/l NaNO3 , though the Kd values on alumina are obviously larger than the corresponding values on silica.
224
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Fig. 5. Variation in Kd values of Am(III) adsorption on alumina as a 0 = 1.1 × 10−9 mol/l, V /m = 500 ml/g, T = 25 ◦ C. function of pH. CAm
Fig. 6. Variation in Kd values of Am(III) adsorption on silica as a function 0 = 1.2 × 10−9 mol/l, V /m = 50 ml/g, I = 0.1 mol/l NaNO , of pH. CAm 3 T = 25 ◦ C.
3.5. Adsorption isotherms in the absence and presence of HS The isotherms of Am(III) adsorption on alumina at pH 3.5, 4.5, and 4.9 in the absence of FA and at pH 4.5 in the presence of FA are shown in Fig. 9. The isotherms of Am(III) adsorption on silica at pH 1.7, 5.5, and 5.9 in the absence of HA and at pH 6.3 in the presence of HA are shown in Fig. 10. All the isotherms in both adsorption d values of these linear systems are linear. The average K d values isotherms are denoted in Figs. 9 and 10. These K also indicate the strong effect of pH, the positive effect of FA at low pH, and the negative effect of HA at high pH.
4. Discussion The distinct difference in adsorption characteristics between alumina and silica, the obvious difference in log Kd – pH curves between the silicas (Fig. 6), and the tremendously high adsorbabilitity of Am(III) on hematite on the basis of mass in particular imply that the properties of the solid phase surface are primarily controlling the Am(III) adsorption instead of the species of dissolved Am(III) in aqueous solutions. First and foremost we should deal with the difference
Fig. 7. Variation in Kd values of Am(III) adsorption on alumina as a 0 = 1.1 × 10−9 mol/l, V /m = 500 ml/g, function of ionic strength. CAm T = 25 ◦ C.
Fig. 8. Variation in Kd values of Am(III) adsorption on silica without 0 = 1.7 × 10−9 mol/l, NaOH treatment as a function of ionic strength. CAm V /m = 50 ml/g, pH 1.5, T = 25 ◦ C.
in having electrical charges at surfaces of both solid phases of alumina and silica. It is well known that the pHpzc of silica ranges from 1.8 to 3.5 [14,15]. Therefore, at pH < 3, the surface of silica carries positive charges; at pH > 3, the surface of silica carries negative charges; at pH > 3, the surface free ≡SOH concentration would decrease and the surface ≡SO− concentration would increase with increasing pH. The pHpzc of the alumina used here is 7.5 and it was found that the predominant surface species is always ≡SOH and the surface free ≡SOH concentration is an order or several orders of magnitude larger than the concentrations of the other surface species and that the surface free ≡SOH concentration slowly increases from pH 3 to 7.5 and then decreases at pH > 7.5 [16]. Therefore, at pH < 7.5, the surface of alumina carries positive charges; and at pH > 7.5, the surface of alumina carries negative charges. The solubility limiting phases of Am(III) are Am2 (CO3 )3 and possibly Am(OH)3 at low carbonate concentrations and high pH values. The total solubility would have a minimum of about 10−7 –10−8 mol/l in the pH range 7–9 [4]. Pryke and Rees [17] experimentally determined the solubilities of Am(III) in simulated pore water containing Ca2+ , Na+ , Mg2+ , Cl− , SO4 2− , and CO3 2− ions at pH 7–13 and found that the solubilities grad-
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
Fig. 9. Isotherms of Am(III) adsorption on alumina. V /m = 500 ml/g, I = 0.1 mol/l NaNO3 , T = 25 ◦ C. •: pH 3.5 ± 0.1, no FA; ×: pH 4.5 ± 0.1, no FA; ◦: pH 4.9 ± 0.1, no FA; : pH 4.5 ± 0.1, initial concentration of FA = 20 mg/l. Kd values are shown in boxes.
Fig. 10. Isotherms of Am(III) adsorption on silica without NaOH treatment. V /m = 50 ml/g, I = 0.1 mol/l NaNO3 , T = 25 ◦ C. •: pH 1.7±0.1, no FA; ×: pH 5.5 ± 0.1, no FA; ◦: pH 5.9 ± 0.1, no FA; : pH 6.3 ± 0.1, initial concentration of HA = 20 mg/l. Kd values are shown in boxes.
ually decrease from 10−5.5 mol/l at pH 7 to 10−8.5 mol/l at pH 9. The total concentrations of Am(III) used here were smaller than the experimental solubilities. Consequently, the adsorption data of Am(III) obtained in this paper are not due to the precipitation of compounds of Am(III). Am3+ carbonate complexes AmCO3 + and Am(CO3 )2 − as well as the hydroxides Am(OH)2+ and Am(OH)2 + may be the predominant species in aqueous solutions in this study. The species distribution plot for the Am(III)–OH– CO3 system at ionic strength I of 0.1 mol/l NaClO4 , and pCO2 = 10−3.5 atm was drawn by Moulin et al. [18], it was found that at pH < 5, Am3+ is almost the sole species; at pH 5–6, Am3+ is the predominant species and Am(OH)2+ is less than 10%; at pH 6–8, Am3+ , Am(OH)2+ , Am(OH)2 + , and Am(CO3 )+ are present; at pH 8–9, Am(OH)2+ , Am(OH)2 + , Am(CO3 )+ , and Am(CO3 )2 − are present. According to the thermodynamic model for adsorption of James and Healy [13], the Gibbs free energy (Gads ) involves not only the coulombic electrostatic energy (Gcoul )
225
but also the chemical bond energy (Gchem ) and the hydration energy (Ghyd). Because the dehydration energy is always positive and deleterious to adsorption of adsorbing ions and because the electrostatic repulsion between the positively charged species of Am(III) in aqueous solutions and the positively charged surface of alumina at pH < 7.5 also is deleterious to adsorption, the sole driving force for Am(III) adsorption on alumina is the Gchem . In other words, strong chemical bonds are formed between the surface of alumina and the Am(III). Although the electrostatic attraction between the positively charged species of Am(III) solutions and the negatively charged surface of silica at pH > 3 is favorable to adsorption, as a whole, the Am(III) adsorption on silica does not appear to be as strong as on alumina. Therefore, the chemical bonds between the surface of silica and the Am(III) must be weaker than those between the surface of alumina and the Am(III). The log Kd –pH curves at ionic strength 0.01 or 0.1 mol/l NaNO3 (Fig. 5) are very similar to that of Yb(III) sorption on alumina at pH < 10 [1]. Thus the surface hydrolysis model [19] would explain the log Kd –pH curve of Am(III) adsorption on the bare alumina (Fig. 5), the availability of surface free ≡SOH groups is an essential requirement of the adsorption process, and the increase in free ≡SOH concentration is the cause of the steep increase in Kd values. In contrast, the log Kd –pH curves of Am(III) adsorption on the bare silica cannot be explained by the surface hydrolysis model, because at pH > 3 the surface free ≡SOH concentration would decrease with increasing pH. The increase in Kd values of Am(III) adsorption on the bare silica with increasing pH can be partially explained by the increase in electrostatic attraction between the positively charged species of Am(III) and the negative charged surface of bare silica at pH > 3. This explanation implies that ion exchange is a significant mechanism for Am(III) adsorption on silica. The slight decrease in Kd of Am(III) adsorption on silica with increasing pH at pH > 6 may be the result of formation of carbonate complex Am(CO3 )2 − in aqueous solutions and/or the dissolution of silica at higher pH. The similar log Kd –pH curves showing Kd values increasing from low values to maximum and then slightly decreasing from maximum value with increasing pH were also found by several other authors [3,4,21]. Depending on the adsorption system on oxides, adsorption of metal ions at oxide–aqueous solution interfaces has been reported to increase, remain the same, and decrease with increasing ionic strength. It was found that the values of Kd of Cs+ adsorption on the alumina used here decrease with increasing ionic strength [12] and that the values of Kd of Eu(III) and Co(II) sorptions on the alumina used here are practically independent of ionic strength [1,2]. Figure 7 shows that the Kd values of Am(III) adsorption on alumina increase gradually from ionic strength 0.001 to 2 mol/l NaNO3 . Figure 8 shows that the Kd values of Am(III) adsorption on the silica without NaOH treatment increase from 0.001 to 0.01 mol/l NaNO3 and then are
226
Z. Tao et al. / Journal of Colloid and Interface Science 265 (2003) 221–226
practically independent of ionic strength from 0.01 to 2 mol/l NaNO3 . A promotive effect of ionic strength on Am(III) adsorption on alumina and silica was found. The promotive ionic strength effect may be explained in terms of more effective shielding of highly charged surface species at higher ionic strength and in terms of the formation of ion pairs between the surface species and the background electrolyte anion [22,23]. The promotive effect was observed in the goethite–Cd(II) [24] and in the hematite–Pb(II) [25] adsorption systems for a long time. Since silica carries a negative surface charge at pH > 3, the negative charge allows an electrostatic binding of positively charged species of Am(III) on the silica surface. However, the strength of this binding may be too strong and so the monovalent electrolyte cation, Na+ , cannot compete for the ≡SO− groups. The ionic strength effect is rather complex and it cannot be interpreted properly at present. The isotherms of Am(III) adsorption on alumina (Fig. 9) and on silica (Fig. 10) in the absence and presence of HS are linear over the solid phase concentration ranges, Am(III), 0–1.6 × 10−9 and 0–1.0 × 10−10 mol/g, respectively. Figures 9 and 10 show the strong effects of pH on the Am(III) adsorption on alumina and silica and the positive and negative effects of HS on Am(III) adsorption. On the bare alumina, the average distribution coefficients (K d ) increase from 18 ml/g at pH 3.5 to 2 × 103 ml/g at pH 4.8 under otherwise equal conditions. On the bare silica, the average distribution coefficient (K d ) increase from 28 ml/g at pH 1.7 to 338 ml/g at pH 5.9 under otherwise equal conditions. The positive effect of FA on the Am(III) adsorption on alumina is demonstrated at low pH 4.5 and initial FA concentration of 20 mg/l, whereas the negative effect of HA on the Am(III) adsorption on silica is demonstrated at high pH 6.5 and initial HA concentration of 20 mg/l. The linearity of isotherms implies that the same surface species of Am(III) is formed over the solid phase concentration range. However, it is impossible for us to indicate the structure of this surface species of Am(III) from the macroscopic observations. The positive and the negative effects are consistent with the general consensus that the adsorption of inorganic cations is increased at low pH and decreased at high pH by adding HS [21]. The negative effect of HA on the Am(III) adsorption on silica at pH 6.5 also is consistent with adsorption of organic matter only being measurable at pH values less than 3 [10] and the complexes with soluble HA being formed in aqueous solution [21]. Adsorption of trivalent cations is more complicated than that of bivalent cations. There is need for further systematic study of adsorption of trivalent cations on oxides by using macroscopic and microscopic experiments [26].
5. Conclusions (1) Under aerobic condition (Eh > 300 mV), in the presence of atmosphere and at relatively lower concentra-
tion (∼10−9 mol/l), because Am(III) exists in positively charged species Am3+ , Am(OH)2+ , Am(OH)2 + , Am(CO3 )+ and negatively charged species Am(CO3 )2 − , and these species are in equilibria with each other in liquid phase and these equilibria are shifted with changing pH, many surface reactions are responsible for the overall adsorption of Am(III) on a surface, and the factors affecting the adsorption process are numerous. (2) The effect of humic substance on the adsorption of Am(III) is very important under aerobic condition, in the presence of atmosphere and at relatively lower concentration (∼10−9 mol/l). (3) The adsorption of Am(III) on alumina under aerobic condition is obviously stronger than that on silica, and the sequence of adsorbabilities of Am(III) on the basis of mass is Fe2 O3 > Al2 O3 > SiO2 .
References [1] X.K. Wang, W.M. Dong, X.X. Dai, A.X. Wang, J.Z. Du, Z.Y. Tao, Appl. Radiat. Isotopes 52 (2000) 165. [2] X.K. Wang, W.M. Dong, G. Wang, Z.Y. Tao, Appl. Radiat. Isotopes 56 (2002) 765. [3] B. Allard, G.W. Beall, T. Krajewski, Nucl. Technol. 49 (1980) 474. [4] B. Allard, U. Olofsson, B. Torstenfelt, Inorg. Chem. Acta 94 (1984) 205. [5] L. Righetto, G. Bidoglio, B. Marcandalli, I.R. Bellobono, Radiochim. Acta 44/45 (1988) 73. [6] I.A. Triay, A. Meijer, M.R. Cisneros, G.G. Miller, A.J. Mitchell, M.A. Ott, D.E. Hobart, P.D. Palmer, R.E. Perrin, R.D. Agnilar, Radiochim. Acta 52/53 (1991) 141. [7] V. Moulin, D. Stammose, G. Ouzounian, Appl. Geochem. Suppl. Issue 1 (1992) 163. [8] A. Kitamura, T. Yamamoto, S. Nishikawa, M. Moriyama, J. Radioanal. Nucl. Chem. 239 (1999) 449. [9] W.M. Dong, H.X. Zhang, M.D. Huang, Z.Y. Tao, Appl. Radiat. Isotopes 56 (2001) 595. [10] J.A. Davis, Geochim. Cosmochim. Acta 46 (1982) 2361. [11] W.J. Li, Y.Q. Ding, Z.Y. Tao, J. Radioanal. Nucl. Chem. 250 (2001) 497. [12] T.W. Chu, J.Z. Du, J.R. Lu, Z.Y. Tao, J. Radioanal. Nucl. Chem. 210 (1996) 197. [13] R.O. James, T.M. Healy, J. Colloid Interface Sci. 40 (1972) 65. [14] Z.Y. Tao, H.X. Zhang, J. Colloid Interface Sci. 252 (2002) 15. [15] G.A. Park, Chem. Rev. 65 (1965) 177. [16] T.W. Chu, J.Z. Du, Z.Y. Tao, Adsorption Sci. Technol. 15 (1997) 349. [17] D.C. Pryke, J.H. Rees, Radiochim. Acta 40 (1986) 27. [18] V. Moulin, P. Robouch, P. Vitorge, B. Allard, Radiochim. Acta 44/45 (1988) 33. [19] A.R. Gupta, B. Venkataramani, Bull. Chem. Soc. Jpn. 61 (1988) 1357. [20] M. Norden, J.H. Ephraim, B. Allard, Radiochim. Acta 65 (1994) 265. [21] Z.Y. Tao, T.W. Chu, J.Z. Du, X.X. Dai, Y.J. Gu, Appl. Geochem. 15 (2000) 133. [22] J. Lützenkirchen, J. Colloid Interface Sci. 195 (1997) 149. [23] M. Kosmulski, J. Colloid Interface Sci. 195 (1997) 395. [24] I. Lamy, M. Djafer, M. Terce, Water, Air, Soil Pollution 57–58 (1991) 457. [25] S. Ardizzone, L. Formaro, J. Lyklema, J. Electroanal. Chem. 133 (1982) 147. [26] Th. Rabung, Th. Stampf, H. Geckeis, R. Klenze, J.I. Kim, Radiochim. Acta 88 (2000) 711.