Pergamon Geochimica et Cosmochimica Acta. Vol. 58, No. 6, pp. 1667-1677, 1994 Copyright 0 1994 Elsevier Science Ltd Printed in the USA. All rights reserved 0016-7037/94 $6.00 + .OO LETTER Gibbs free energies of formation at 298 K for imogolite and gibbsite from solubility measurements CHUNMINGSu’ and JAMESB. HARSH’ ‘U.S. Salinity Laboratory, USDA, ARS, 4500 Glenwood Drive, Riverside, CA 9250 1, USA *Department of Crop and Soil Sciences, Washington State University, Pullman, WA 99 164-6420, USA (Received October 19. 1993; accepted in revised form January 20, 1994 ) Abstract-The aqueous solubility of synthetic imogolite at 298 K and 1 bar pressure and at 373, 393, 408, and 423 K and equilibrium vapor pressure was determined in 0.01 M NaCl at two initial pH levels. Samples were run with and without pretreatment with HCl and in the presence and absence of gibbsite or boehmite. At 298 K and 1 bar pressure, dissolution of non-HCl-washed synthetic imogolite at the initial pH values of 1.8 and 2.2, with or without addition of AlC13 and H&O., , approached equilibrium within 335 and 33 days, respectively. Dissolution of HCl-washed imogolite and gibbsite at initial pH 2.5 and 3.0 reached equilibrium within 332 and 487 days, respectively. There was no difference between the log IAP values at equilibrium from non-HCl-washed and HCl-washed imogolite samples. Dissolution of non-HCI-washed synthetic imogolite and gibbsite at an initial pH 2.2 attained equilibrium within 485 days of equilibration, but equilibrium was not reached after 861 days for samples at an initial pH 1.8. Dissolution of HCl-washed imogolite at initial pH values of 2.5 and 3.0 did not reach equilibrium within 766 days. At 373 and 393 K and an initial pH 3.0, dissolution of imogolite and synthetic boehmite reached equilibrium. The calculated Gibbs free energies of formation at 298 K were 2923.79 + 3.38 (synthetic imogolite), -2920.83 + 3.92 (natural imogolite), -1155.06 +- 1.43 (gibbsite), -915.10 + 1.83 (boehmite, extrapolated from elevated temperature), and -920.64 ? 1.4 1 kJ mol-’ (boehmite, from 298 K solubility). The results indicate that synthetic imogolite is more soluble than earlier reports suggest and natural imogolite is less stable than its synthetic counterpart. INTRODUCTION IMOGOLITE DISSOLVES AND gibbsite accumulates in soil en- vironments which favor desilication ( WADA, 1989); however, the conditions under which imogolite forms and persists in soils and sediments are not fully known ( WADA, 1989; HEM- INGWAY and SPOSITO, 1989). This is due, in part, to the fact that few studies have been performed on the thermodynamic stability of synthetic imogolite or its natural counterpart. FARMER et al. ( 1979 ) heated a synthetic imogolite suspension for periods up to 100 days at temperatures of 373-448 K. They detected boehmite and imogolite in the systems heated to 373-4 13 K. Based on the measured silica concentrations, they calculated the stability at 298 K by extrapolation, as- suming the solution to be in equilibrium with boehmite and imogolite. The value of the free energy of formation of im- ogolite (AGT (298 K) = -2926.7 f 4.6 kJ mol-‘) is ques- tionable because the boehmite AGT (298 K) is uncertain, ranging from -9 12.7 to -92 1 .O kJ mol-’ (Table 1). In an- other study, FARMER and FRASER ( 1982) determined the concentrations of dialyzable Si and Al in a suspension of “proto-imogolite.” Assuming equilibrium with imogolite and gibbsite, they calculated an imogolite AGT (298 K) of -2929.7 * 4.6 kJ mol-‘. Although proto-imogolite can be readily transformed to tubular imogolite upon heating, it is not likely to have the same solubility. Furthermore, there was no indication of the free energy value of a poorly crys- talline gibbsite in their suspension. A study of the stability of imogolite in the Bs and BC horizons of a Spodosol indicated that Si in the soil solution was controlled by imogolite and Al by Al(OH)3 in the in- terlayers of phyllosilicates ( DAHLGREN and UGOLINI, 1989). The authors assumed that the free energy of interlayer Al( OH)3 was identical to that of gibbsite; however, the sol- ubility of Al( OH)3 precipitated on 2: 1 clays has been found to be greater than that of macrocrystalline gibbsite (e.g., KIT- TRICK, 1983; HARSHand DONER, 1985). In addition, a recent study indicates that soluble aluminosilicate complexes affect Al and Si speciation in aqueous solutions and may alter pre- viously determined free energies of many aluminosilicates ( BROWNE and DRISCOLL, 1992 ). The relative stabilities of gibbsite and boehmite in near- surface environments has been a controversial subject (MARSHALL, 1964; GARRELS and CHRIST, 1965; KITTRICK, 1969; CHESWORTH, 1972, 1978; DAY, 1976; HEMINGWAY et al., 1978; ANOVITZ et al., 1991; HEMINGWAY et al., 1991). As pointed out by CHESWORTH ( 1972), MARSHALL ( 1964, p. 147) maintained “the stable form at low temperature is gibbsite.” GARRELS and CHRIST (1965, p. 10) also stated that “gibbsite is the stable phase relative to boehmite in dilute aqueous solution, at 25°C and 1 atmosphere total pressure”; however, KITTRICK (1969) and PERYEA and KITTRICK 1667