J. CHEM. SOC. FARADAY TRANS., 1994, 90(13), 1973-1977 Dissolution of Amorphous Aluminosilicate Zeolite Precursors in Alkaline Solutions Part 2.T-Mechanism of the Dissolution T. Antonic, A. Ci2mek and B. Subotic Ruder BoSkoviC Institute, P.O. Box 1016,41001Zagreb, Croatia The influence of the initial mass of the solid phase and concentration of NaOH on the kinetics of dissolution of an amorphous aluminosilicate precursor having an Si : At molar ratio of 1.178 : 1 has been studied by measuring the change in the aluminium and silicon concentrations in the liquid phase during dissolution at 80°C.Analysis of the results obtained indicated that only the outside surface and not the internal surface (pore system) is relevant in the dissolution process. The process of dissolution is controlled by two reactions: a forward one, whose rate is proportional to the outer surface area, S, of the solid phase exposed to the action of OH-ions from the liquid phase, and a backward one, whose rate is proportional to the surface area, S, and to the concentration of reactive silicate and aluminate species from the liquid phase.In many cases, the dissolution rate of amorphous alumino- silicate precursor is the rate-determining step of the crys- tallization of zeolites from amorphous aluminosilicate gels. 1,2 Hence knowledge of the mechanism and kinetics of disso- lution of the precursor is necessary for a detailed kinetic analysis of zeolite crystallization. Our previous study of the dissolution of differently prepared amorphous aluminosilicate zeolite precursors having various Si : A1 molar ratios, in 0.2 mol dm-’ NaOH solution at 80°C showed that the ratio of molar concentrations of silicon and aluminium in the liquid phase is constant during the dissolution and equal to the molar Si : A1 ratio of the dissolving precursor, i.e.the disso- lution occurred in a congruent fashion.’ Kinetic analysis of the change in Si and A1 concentrations in the liquid phase during the dissolution showed that the rate of dissolution is directly proportional to the surface area, S, of the solid exposed to alkaline solution and under~aturation,~ i.e. Although the results obtained by numerical solution of eqn. (1) were in excellent agreement with measured values for all analysed prec~rsors,~ the abovementioned conclusions were made on the basis of dissolution experiments performed under one set of conditions, i.e.dissolution of 0.5 g of precur- sor in 200 cm’ of 0.2 mol dm-3 NaOH solution at 80°C. For this reason, the aim of this work is the study of other relevant factors such as the effects of starting amount of dissolving precursor and concentration of NaOH solution on the kinetics of dissolution in order to obtain a better understand- ing of the mechanism of the dissolution process. Experimental The amorphous aluminosilicate precursor PAP 1 was used as the dissolving solid phase in all experiments. The precursor PAPl ([Si : All = b = 1.178) was prepared as follows:’ Al(OH), (8.22 g, Baker) was dispersed in 10 cm’ of distilled water and NaOH (10 g) was added.The suspension was stirred with heating until the Al(OH), was dissolved com- pletely. The solution was diluted with distilled water to 100 cm’. Sodium silicate solution was prepared by dilution of 14 cm’ of water-glass stock solution (8.15 wt.% Na,O, 25.9 wt.% SiO,, 65.95 wt.% H20) with distilled water to 100 cm3. The temperature of the sodium aluminate and sodium silicate solutions was then thermostatically controlled at 25 “C before t Part 1 :ref. 3. mixing. Then, 100 CM’ of the sodium silicate solution was poured (within 10 s) into a 600 cm3 plastic beaker containing 100 cm’ of the sodium aluminate solution stirred with a mag- netic stirrer. The system (precipitated gel dispersed in the mother liquor) was stirred for a further 10 min and was then transferred into a stainless-steel reaction vessel preheated at 80°C and mixed at the same temperature with a magnetic stirrer for 1 h.Thereafter, the gel was centrifuged to separate the liquid from the solid phase. After removal of the clear liquid phase above the sediment (wet amorphous solid), the solid phase was redispersed in distilled water and centrifuged repeatedly. The procedure was repeated until the liquid phase above the sediment was at ca. pH 10. The washed solid was dried at 105°C for 24 h. The dried solid was pulverized in an agate mortar. In order to follow the dissolution process, a determined amount (0.3-2 g) of the solid was added into a stainless-steel reaction vessel containing 200 cm3 of stirred 0.125, 0.2, 1 and/or 2 mol dm- NaOH solution preheated at dissolution temperature (80 “C).The reaction vessel was provided with a thermostatically controlled jacket and fitted with a water-cooled reflux condenser and a thermometer. The reaction mixture was stirred with a Teflon-coated magnetic bar (L = 5 cm, q5 = 0.95 cm) driven by a magnetic stirrer at 510 rpm. At various times, td, after the beginning of the dissolution process, 5 cm3 aliquots of the suspension were drawn off (by pipette) to prepare samples for analysis. The point at which the solid was added to the preheated NaOH solution was taken as time zero of the dissolution process. Aliquots of the reaction mixture drawn off at given dissolution times, t,, were centrifuged.The clear liquid phase was used for the analysis of silicon and aluminium concentrations in the liquid phase. Equilibrium (saturation) concentrations C,,(eq) of alu- minium and C,(eq) of silicon, which correspond to the solu- bility of the precursor PAPl at given conditions were determined by measuring the A1 and Si concentrations in the liquid phase at t, = 4 h. The concentrations C,, and C,,(eq) of aluminium and CSi and C,,(eq) of silicon in the liquid phase were measured by colorimetric method^.^,^ Results and Discussion Kinetic analysis of dissolution of various compounds6-’ including amorphous aluminosilicates ‘g2 and zeolites”-’’ indicated that the dissolution is controlled by at least two processes : (i) Forward reaction caused by breaking of surface bonds due to the action of solvent and formation of soluble species that leave the surface of the dissolving solid.The rate, (dC/dt,),, at which the soluble species (solvated ions and/or molecules) leave the solid phase is assumed to be propor- tional to the surface area, S, of the solid exposed to the action of solvent, i.e. (dcldt,), = klS (2) (ii) Backward reaction, i.e. reaction of the soluble species from the liquid phase on/with the surface of the dissolving solid. The conventional kinetic arg~ments'~ applied to a chemically controlled dissolution of solute, A, B, CaAbf(aq)+ bB"-(aq), lead to: (dC/dt,), = -k2 SC" (3) where C = CA or C = C, and n = a + b.Hence, dC/dt, = (dC/dtd)l + (dC/dtd)z = k,S -k, SC" = k,S[C(eq)" -C"] (4) However, eqn. (4)fails to explain the kinetic form: dC/dt, = kdS[C(eq) -c]" (5) used for the analysis of the kinetics of dissolution of many solids7 including zeolites.' '-' Our earlier study of the kinetics of dissolution of zeolites has shown that the dissolution takes place in accordance with the model proposed by Davies and Jones," and hence the kinetics of dissolution can be expressed as:12,13 dC/dtd = kd SICAl(eq) -CA,l[CSi(eq) -cSi Ib (6) Since the power, n, in eqn. (5) is closely related to the surface integration step (e.g. formation of the bonds relevant for the crystal structure of the dissolving solid, by chemical reaction of the species from the liquid phase on the surface of the ~olid),~.' it is certain that the concentration dependence in eqn.(6) is the consequence of the backward step of the disso- lution process. The latter is characterized by a chemical reac- tion between silicate and aluminate anions from the liquid phase on the surface of the dissolving zeolite crystals and the formation of Si-0-Si and Si-0- A1 bonds characteristic of the appropriate type of zeolite. It can be assumed that the above considered reaction between silicate and aluminate anions from the liquid phase is catalysed by the specific ord- ering of Si and A1 atoms on the surface of zeolite crystals. On the other hand, although the mechanism of the forward reac- tion is the same for both dissolution of zeolites and amor- phous aluminosilicates [see eqn.(2)], a linear relationship between log(dC/dt,) and logS[C(eq) -C]" with n = 1 (see ref. 3) indicates that the backward reactions during the disso- lution of amorphous aluminosilicates in hot alkaline solu- tions can be expressed as : (dC/dt,), = k2SC (7) and hence, dC/dt, = k1S - kzSC (8) where C = C,, or C = Csi. Since in equilibrium: k,S = k2 SC(eq) (9) eqn. (8) can be transformed to the form: dC/dt, = k, s[C(eq) -c]= k, S[C(eq) -c] (10) This leads to an assumption that the backward reaction described by eqn. (7) is not controlled by mutual chemical reactions between silicate and aluminate anions from the liquid phase on the surface of the solid phase (gel), but by another type of surface reaction which will be discussed later.J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 The abovementioned considerations show that both the forward [see eqn. (2)] and backward reactions [see eqn. (3) and (7)], are directly proportional to the surface area, S, of the solid exposed to action of solvent. Also, most models of the dissolution processes suggest that the rate of dissolution is directly proportional to S.'*7-8711-1 Some authors have predicted that the surface area, S, can be expressed as a function of mass, mu,of the undissolved solid. Truskinovskiy and Senderov' and later Mydlarz and JonesI6 proposed that the surface area of the solid exposed to solvent is propor- tional to the 2/3 power of mu,i.e.S = G!(mu)2'3 (11) On the other hand, Cook and Thompson2 assumed that owing to the porosity of amorphous aluminosilicate zeolite precursors, the surface area, S, of the solid exposed to solvent is directly proportional to the mass, mu,of the undissolved solid, i.e. S = pm, (12) and hence, the rate of dissolution is also directly proportional to mu.Our previous analysis of the kinetics of dissolution of amorphous aluminosilicate zeolite precursors3 led to the con- clusion that the rate of dissolution is directly proportional to the surface area S [see eqn. (1) and (lo)], and that the surface area can be expressed as : S = a(mJ2i3 = a[mg -mG(L)]2/3 (13) However, our conclusion was made on the basis of a study of the kinetics of dissolution for constant mg ,constant concen- tration of solvent and constant temperature (see Introduction), so that it is necessary to examine the relations between mu,S and dm,(L)/dt, under different conditions in order to confirm the relations expressed by eqn.(1)and (13). Fig. 1 shows the change in the mass, mG(L), of dissolved solid during heating of different amounts, mg,of the precur- sor PAPl in 0.2 mol dm-3 NaOH solution at 80°C. Since the dissolution occurred in a congruent fashion, the mass m,(L) was calculated as:3 wAL) = [CAI M(A1)' + Csi M(Si)11/2 (14) The 'plateau' of the dissolution curves [mG(L) = mG(L),, see Table 11 is determined by the initial amount, mg,of the pre- F a -2.0 -1.6 m I E -1.2 --.n ?.{ 0.8 f O.li0.0 0 10 20 30 40 50 60 t,/min Fig.1 Change in the mass, m,(L), of the dissolved solid during the dissolution of the amorphous aluminosilicate precursor PAP 1 in 0.2 mol dmP3 NaOH solution at 80°C. td is the time of dissolution. The initial mass, m:, of the precursor PAPl was 1 (O),1.65 (a),2 (O), 2.5 (m) and 3 (A) g dmP3. Solid curves represent the m,(L) us. t, functions calculated by numerical values of rn: ,Kd (see Table I) and mG(eq)= 2.18 g dmP3 as appropriate constants. The horizontal dashed lines are explained in the text. J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 Table 1 Numerical values of the 'plateau' concentration, m,(L) = mG(L)&,, and constant K, in eqn. (17) that correspond to the kinetics of dissolution during heating of different amounts, mg , of the precur- sor PAPl in 0.2 mol dm-3 NaOH solution at 80°C 1.00 1.00 0.220 1.65 1.65 0.215 2.00 2.00 0.200 2.50 2.18" 0.225 3.00 2.18" 0.21 1 ~ ~~~~~~~ " mC(L), = mG(eq)= 2.18 g dm-j is the solubility of the precursor PAPl in 0.2 mol dm-3 NaOH solution at 80°C (see Table 1 in ref.3). cursor and its solubility, m,(eq), in 0.2 mol dm-3 NaOH solution at 80°C (see Table 3 in ref. 3). It is evident that m&), = mg for mg < m,(eq) and that m&), = mG(eq) for mg 2 m,(eq) (see Fig. 1 and Table 1). To determine the rela- tions between dm,(L)/dtd, mg -mJL) and mG(eq) -m&), each of the five mG(L) us. td functions (see Fig. 1) was graphi- cally differentiated at the point of constant m&) [and thus constant m,(eq) -mG(L); intersection between each of the five m,(L) us.td functions and dashed horizontal line]. Then, each set of the differentials drn,(L)/dt, [that determined at the points of constant mG(L)] was plotted against the corre- sponding values of mu = mg -mG(L) (Fig. 2) and (mJ2I3 = [mg -mG(L)]2'3, respectively (Fig. 3). Fig. 2 and 3 show that dmG(L)/dtd is not a linear function of mu, but that the rate of dissolution is directly proportional to (mu)213for each con- sidered constant undersaturation, m,(eq) -m&). Hence, where, in accordance with eqn. (1) and (13), the slopes K, of the straight lines in Fig. 3 are given by = kda[mG(eq) -mG(L)l = Kd[mG(eq) -mG(L)l (16) Fig.4 shows that K is a linear function of the under-saturation [mG(eq) -mG(L)], and that the numerical value of the slope Kd (=0.22 g-2/3 min-') is very close to the values of K, (see Table 1) which correspond to the kinetics of disso-lution presented in Fig. 1. The values of Kd presented in I 0.7 I 0.61 .-C E 0.5 -mI5 0.4 -0--.2 0.3tPih 0.2; ELU 0.1 -0.0 I 0.0 0.4 0.8 1.2 1.6 2.0 2.4 ImG -mG(L)l/g dm-3 Fig. 2 dm,(L)/dt, us. mu = m: -m,(L) plots which correspond to the dissolution of the precursor PAPl at different undersaturations : m,(eq) -mG(L) = 1.78 (O),1.58 (a),1.38 (O),1.18 (H)and 0.98 (A) g dm-j. The concentration of NaOH was 0.2 mol dm-3 and the temperature of dissolution was 80 "C.1975 0.7 0.6 m 0.5 w 0,> 0.4 //I/ /I 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 [mG -mG(L)]2/3/g2/3 dm-' Fig. 3 dm,(L)/dt, vs. = [mg -m,(L)]2/3 plots which corre- spond to the dissolution of the precursor PAPl at different under- saturations: m,(eq) -m,(L) = 1.78 (O),1.58 (*), 1.38 (0)and 0.98 (A) g dm-3. The concentration of NaOH was 0.2 mol dm-3 and temperature of dissolution was 80 "C. Table 1 were calculated by the procedure explained earlier.3 The results presented in Fig. 1-4 undoubtedly confirm that the rate of dissolution is directly proportional to the surface area, S, and the undersaturation as expressed by eqn. (1).The linear relationship between dm,(L)/dt, and [mg -m&)] 2/3 and at the same time, the non-linearity between drnG(L)/dtd and [mg -m&)] indicate that only the outside surface (solid/liquid interface) and not the internal surface (pore system) of the particles of the investigated solids is relevant in the dissolution process.Hence, the kinetics of dissolution can be expressed as : Very good agreement between the measured values of mG(L) (symbols in Fig. 1) and the values of mG(L) calculated by numerical solution of eqn. (17) (curves in Fig. l), using the corresponding values of mg, m,(eq) and Kd (see Table I), confirms this conclusion. It is evident from eqn. (8)-(10) that Kd = ak, is directly proportional to the rate constant k, of the backward reaction and that the product &mG(eq) is directly proportional to the rate constant, k,, of the forward reaction of the dissolution process.0 0.36 t1 c /c/ 0.32 ; 0.28 : I.E 0.24; 7 0.20 1 E 0.16 \ r 0.12; 0.00 1' 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 [m,(eq) -m~(L)l/gdmW3 Fig. 4 Values of constant K in eqn. (15) and (16) plotted against the -corresponding undersaturations : &(eq) -mG(L) - 0 0 10 20 30 40 50 60 t,/min Fig. 5 Change in the mass, m,(L), of the dissolved solid during the dissolution of the amorphous aluminosilicate precursor PAP 1 in 0.125 (O), mol dm-3 NaOH solution at 0.2 (a),1 (a)and 2 (0) 80°C. Solid curves represent the m,(L) us. t, functions calculated by numerical solution of eqn. (17) using the corresponding numerical values of mg ,K, and m,(eq) (see Table 2) as appropriate constants. For a better understanding of the reactions which control the dissolution process, the kinetics of dissolution of the pre- cursor PAP 1 in NaOH solutions of different concentrations were analysed by eqn.(17). The very good agreement between the values of mG(L) measured during the dissolution of the precursor PAP1 in NaOH solutions of different concentra- tions at 80°C (symbols in Fig. 5) and the values of m,(L) calculated by numerical solution of eqn. (16) (curves in Fig. 5), using the corresponding values of mg,mG(eq) and Kd (see Table 2), show that the variation in NaOH concentration does not affect the mechanism of dissolution, i.e. the kinetics of dissolution can be in all cases expressed by the differential equation, eqn.(16). The values of mG(eq) were determined experimentally and the values of Kd were calculated from the log[dm,(L)/dt,] us. log[mL -mG(L)]2/3[mG(eq)-m,(L)] plots by the procedure described earlier.3 The data in Table 2 show that the value of Kd does not depend on the NaOH concentration and that the value of m,(eq) increases with the increase of NaOH concentration. Based on these findings, the process of dissolution of amorphous aluminosilicate precursors in NaOH solutions can be explained as follows: The action of OH- ions from the liquid phase on the solid/liquid interface causes breaking of the surface Si-0-Si and Si-0-A1 bonds of the precur- sor and formation of soluble aluminate and silicate species.Owing to agitation of the suspension, soluble silicate and alu- minate species so formed leave the solid/liquid interface and tend to be homogeneously distributed throughout the bulk of the liquid phase. The rate of formation of the soluble silicate and aluminate species is assumed to be proportional to the number of OH- ions that act to the unit surface area of the Table 2 Numerical values of the measured equilibrium amounts, m,(eq), of the dissolved solid and of the constant K, in eqn. (17) that correspond to the kinetics of dissolution of the amorphous alumino- silicate precursor PAPl in NaOH solutions of different concentra- tions (CNaOH)at 80°C; m: is the initial mass of the precursor in the suspension C,,,,/mol dm - mglg dm - m,(eq)/g dm - Kd/g-213 min - 0.125 2.5 1.66 0.185 0.2 3.0 2.18 0.211 1 6.0 5.39 0.205 2 10.0 6.34 0.185 J. CHEM.SOC. FARADAY TRANS., 1994, VOL. 90 precursor. Hence, the rate at which the soluble silicate and aluminate species leave the solid/liquid interface (forward reaction) increases with increasing concentration of NaOH, and is directly proportional to the surface area, S, exposed to the action of OH- ions, as is expressed by eqn. (2). The dependence of the rate of the forward reaction on the concen- tration of NaOH in the liquid phase is determined by the change of S and mG(L) under the given conditions. Previous analysis of the degree of Si polycondensation in the liquid phase showed that the liquid phase contained silicate anions in both monomeric and dimeric forms with predominant monomeric forms.Some of the silicate and aluminate anions from the liquid phase return to the solid/liquid interface, where they react with surface silicon and aluminium ions forming a new surface layer of the solid phase. Phase analysis of the solid residues showed that no phase transformation occurred during the dissolution, i.e. that the solid phase formed by the backward reaction has the same structural properties as the starting amorphous aluminosilicate precur- SO^.^ The constant Si: A1 molar ratio in the liquid phase during the dissolution which is the same as the Si : A1 molar ratio of the solid phase indicates that the solid phase formed in the backward reaction has a chemical composition identi- cal to that of the initial solid phase.Therefore, based on the linear relationship between dmc(L)/dtd and m,(eq) -m,(L), it can be concluded that the silicate and aluminate anions from the liquid phase do not react with one other at the surface of the dissolving solid, but only with the terminal OH- groups of the solid phase forming new Si-0-Si and Si-0-A1 bonds characteristic for amorphous aluminosilicates. Taking into consideration the abovementioned assumptions it can be expected that the rate of the backward reaction is directly proportional to the concentration of the reactive species in the liquid phase (silicate and aluminate anions, expressed as the molar concentrations CAI of aluminium and CSiof silicon) and the ‘concentration’ of terminal OH- groups on the surface of the solid phase (which is assumed to be pro- portional to the surface area, S, i.e.[OH-], = KS), as expressed by eqn. (7). Under the given experimental condi- tions the rate constant Kd z k, is assumed to be dependent only on the kinetic energy of the reacting species, and thus is independent of the concentration of NaOH. The values of K, in Table 2 confirm such an assumption. Otherwise, the value of K, is a function of the properties of the solid phase (chemical comp~sition,~ particulate characteristics) and is assumed to be dependent on experimental conditions (temperature, mode and rate of agitation, design of vessel etc.).It is evident that the rate of forward reaction decreases owing to the decrease of the surface area, S [caused by the decrease of the mass mu,see eqn. (12)] and that the rate of the backward reaction increases owing to the increase of the concentration of reactive species in the liquid phase during the dissolution process. At the moment when both of the reactions have the same rate, the dissolution process achieves the equilibrium expressed by eqn. (9). The validity of eqn. (8), (10) and (17), which are derived on the basis of the above considered model of the dissolution process, for the kinetic analysis of the dissolution process and for the mathematical description of the change in CAI, CSi and m,(L) during the dissolution, indicates that the above considered model is rele- vant for the process of dissolution of dehydrated amorphous aluminosilicate precursors in NaOH solutions.Conclusion The influence of the initial mass, mg,of the solid phase on the kinetics of dissolution of the amorphous aluminosilicate precursor PAPl (Si : A1 = 1.178 : 1) in 0.2 mol dmP3 NaOH J. CHEM. SOC. FARADAY TRANS., 1994, VOL. 90 solution at 80°C was studied. Analysis of the change in the mass, mG(L), of the dissolved precursor showed that the rate of dissolution is directly proportional to the external surface area, S, and undersaturation, as expressed by eqn. (17), i.e. dmG(L)/dtd = KdS[mG(eq) -m&)], where S = V.[mG(eq) -mG(L)]2’3. The simple relation between the rate of disso- lution, Rd = dmG(L)/dtd, and undersaturation, m,(eq) -m,(L), indicated that the dissolution process is controlled by two reactions: (RJl = klS = &mG(eq)S (forward reaction) and (Rd)2 = -k$mG(L) = -Kd Sm,(L) (backward reaction).This model has been evaluated by analysis of the kinetics of dissolution of the precursor PAPl in NaOH solu- tions of different concentrations. The analysis showed that: (i) the rate of the forward reaction is controlled by breaking of the surface Si-0-Si and S-0-A1 bonds owing to the action of OH-ions from solution. The rate of the forward reaction depends in a direct way on the concentration of solvent as indicated by the increase of the value of mG(eq) with increasing concentration of NaOH (see Table 2).(ii) The rate of the backward reaction is controlled by the formation of new Si-0-Si and Si-0-A1 bonds between the silicate and aluminate anions from the liquid phase and surface silicon and aluminium atoms. Under the given conditions, the rate of the backward reaction indirectly depends only on the concentration of NaOH by the change of the concentration of reactive species in the liquid phase, but the rate constant k, x Kd does not depend on the concentration of NaOH (see Table 2). The values of mG(L) calculated by numerical solu-tion of eqn. (17) are in excellent agreement with the experi- mentally obtained values of mG(L). This shows that the kinetics of dissolution can be described mathematically by eqn. (17), and hence confirms the proposed model of disso- lution.Although the above considered model was successfully applied in the simulation of the synthesis of zeolite A from differently aged, as-created expanded aluminosilicate gels,22 it is clear that, at times, this model can be used strictly for mod- elling and simulation of zeolite synthesis from dehydrated aluminosilicate powders. However, it is our opinion that the experimental experience and theoretical considerations arising from this study will be helpful in further studies of the kinetics of dissolution of as-created, expanded aluminosilicate gels. The authors thank the Ministry of Science and Technology of the Republic Croatia for its financial support. Glossary b Molar Si : A1 ratio of the zeolite Concentration of soluble species (A, B, Al, Si) in the liquid phase C(eq) Equilibrium (saturation) concentration of soluble species (A, B, Al, Si) in the liquid phase dC/dtd Differential change in the concentration of soluble species in the liquid phase kl Rate constant of the forward reaction k, = kd Rate constant of the backward reaction 1977 Kd = akd Rate constant of the dissolution process 4 Initial mass (at td = 0) of the solid phase in the suspension mG(eq) Mass of the dissolved solid which corresponds to its solubility mG(L) Mass of the solid phase dissolved up to any time, td mG(L), Mass of dissolved solid which corresponds to the ‘plateau’ of the dissolution curves shown in Fig.1 m” Mass of undissolved solid M(Al), Mass for the precursor PAPl which contains 1 mol of A1 (see Table 1 in ref.3) M(Si), Mass of the precursor PAPl which contains 1 mol of Si (see Table 1 in ref. 3) S Surface area of the solid phase exposed to the action of solvent td Time of dissolution a Proportionality constant in eqn. (1l), (12) and (16) B Proportionality constant in eqn. (12) References 1 L. M. Truskinovskiy and E. E. Senderov, Geokhimiya, 1983, 3, 450. 2 J. D. Cook and R. W. Thompson, Zeolites, 1988,8, 322. 3 T. Antonik, A. Ciimek, C. Kosanovik and B. Subotik, J. Chem. SOC.,Faraday Trans., 1993,89, 1817. 4 M. L. Blanchet and L. Malaprade, Chim. Anal., 1960,42, 603. 5 G. Valence and S. Marques, Chim. Anal., 1967,49, 275. 6 C. H. Bovington and A. L. Jones, Trans. Faraday Soc., 1970,66, 764.7 A. L. Jones and H. G. Linge, Z. Phys. Chem. NF, 1975,95,293. 8 H. Sverdrup, P. Warfwinge and 1. Bjerle, Vatten, 1986,42,210. 9 R. Wolast and L. Chou, in Physical and Chemical Weathering in Geochemical Cycles, ed. A. Lerman and M. Meybeck, Kluwer, Dordrecht, 1988, p. 11. 10 R. G. Compton and K. L. Pritchard, Philos. Trans. R. SOC. London, A, 1990,330,47. 11 A. Ciimek, LJ. Komunjer, B. Subotik, M. Siroki and S. RonEe-vik, Zeolites, 1991, 11, 258. 12 A. Ciimek, LJ. Komunjer, B. Subotik, M. Siroki and S. RonEe-vik, Zeolites, 1991, 11, 810. 13 A. Ciimek, LJ. Komunjer, B. Subotic, M. Siroki and S. RonEe-vic, Zeolites, 1992, 12, 190. 14 G. H. Nancollas and N. Purdie, Q. Rev. (London), 1964,18, 1. 15 C. W. Davies and A. L. Jones, Trans. Faraday SOC., 1955, 51, 812. 16 J. Mydlarz and A. G. Jones, Chem. Eng. Sci., 1989,44, 1391. 17 A. L. Jones, G. A. Madigan and I. R. Wilson, J. Cryst. Growth, 1973,20,93. 18 C. N. Litsakes and P. Ney, Fortsch. Miner., 1985,63, 135. 19 P. Gohar and M. Cournil, Muter. Chem. Phys., 1986, 14,427. 20 H. Sunada, A. Yamamoto, A. Otsuka and Y. Yonezawa, Chem. Pharm. Bull., 1988,36, 2557. 21 D. Elenkov, S. V. Vlaev, I. Nikov and M. Ruseva, Chem. Eng. J., 1989, 41, 75. 22 B. Subotik and J. BroniC, in Proceedings of the Ninth Interna- tional Zeolite Conference, ed. R. von Ballmoos, J. B. Higgins and M. M. J. Treacy, Butterworth-Heinemann, Boston, 1992, p. 321. Paper 3/040391; Received 12th July, 1993