Kinetics of Hydrothermal Reactions of Minerals in Near-critical and Supercritical Water

400Vol. 74 No. 2ACTA GEOLOGICA SINICAJune 2000Kinetics of Hydrothermal Reactions of Mineralsin Near-critical and Supercritical W aterZHANG Ronghua, HU Shumin and ZHANG XuetongOpen Research Laboratory of Geochemical Kinetics, Chinese Academyof Geological Sciences, 26 Baiwanzhuang Road, Beijing 100037Abstract This work presents new experimental results on the kinetics of mineral dissolution in near-critical andsupercritical water in a temperature range (T) from 25 to 400°C and a constant pressure of 23 MPa. Kinetic experi-ments were carried out by using a flow reactor (packed bed reactor) of an open system. The dissolution rates of albiteand magnetite were measured under these experimental conditions. Na, Al and Si release rates for albite dissolutionin water were measured as a function of the temperature and flow velocity in the reaction system. The maximum re-lease rates of Na, Al and Si of albite dissolution in the hydrothermal flow systems under different flow velocitieswere always obtained at 300°C, that is to say, the maximum albite dissolution rates in the flow systems, regardless ofdifferent flow rates, were repeatedly measured at 300°C. Results indicate a wide fluctuation in albite dissolution ratesoccuring close to the critical point of water. The dissolution rates increased when the temperatures increased from 25to 300°C and decreased when the temperatures increase from 300 to 400°C. At some flow velocities, the dissolutionrates rose as the temperature surpassed 374°C. Albite dissolution was incongruent in water at most temperatures. Itwas only at 300PC that albite dissolution was congruent. The albite dissolution from 25 to 300°C (at 23 MPa) willchange from incongruent to congruent, whereas from subcritical 300 to 400°C (at 23 MPa), the dissolution willchange from congruent to incongruent. The release ratio of AI/Si (or Na/Si) is positive at T<300°C, and it is negativeat T>300°C. The dissolution rates of magnetite in water increased with increasing T until T at the critical point ofwater or around it. The authors believe that this is caused by the wide fluctuations in water properties under the con-ditions from the near-critical to supercritical state.Key words: kinetics, hydrothermal reaction, critical state1 Introductiondissolution rates of many minerals are affected bytemperature, pH and solution composition. Some eIn order to understand the mechanism of rock (or min-ports emphasize the congruent dissolution dominanceeral) and water ( or aqueous solutions) interaction inat lower temperatures (< 100°C) and at conditions farthe lithosphere, it is necessary to test mineral dissolu-from equilibrium (Alekseyev et al, 1993, 1997).tion rates at elevated temperatures and pressures. MostOther reports suggest that the tendency towards con-rock-water interactions take place in the subsurface ofgruent dissolution at 300°C occurs as a result of risingthe Earth at temperatures above 300°C (Xu et al,,temperatures and that the relative difference between1997; Zhang Ligang et al, 1996; Zheng et al., 1995;the rate of ion exchange and hydrolysis diminishesZhou et al, 1997). Therefore, a laboratory under-due to the strong temperature dependence of hydroly-standing of rock-water interaction requires a resis reactions (Helmann et al., 1989; Zhang Ronghua etcreation of the kinetic behaviour of mineral dissolu-al., 1992; Hellmann, 1995). Experimental studies sug-tions in aqueous solutions at temperatures abovegest that the changes in the stoichiometry of mineral300°C in critical and supercritical states.dissolution can be worked out as a function of timePrevious kinetic experiments of mineral dissolutionwhen the mineral is dissolved at constant physical andinvolving feldspar dissolution were carried out atchemical conditions. In these studies, the time evolu-300°C or below 300°C. The results suggest that thetion of the stoichiometery for albite was determined中国煤化工YHCNMH G.Kinetics of Hydrothermal Reactions of Minerals in WaterZhang Ronghua et al.401by the aqueous release rates of sodium (Na), alumi-Note: The continuous flow reactor incorporates anum (Al)， and silicon (Si) measured from the initialcontinuous input of aqueous solutions through a pac k-stages (non-steady state kinetics) to the establishmented mineral bed and a resulting continuous fluid output.of the steady state of kinetic conditions. When disso-Within the packed bed reactor, a transient materiallution rates were continually measured over a periodbalance in a column at the length Z giveof long-term kinetic stability, it was found that con-82C . aC_ aCD,9_-∪+ r=.(1)gruent dissolution occurred at 300°C (Zhang Ronghua,)Z’)t1992; Zhang and Hu, 1996, 1997; Hellmann, 1995;where DL is the axial dispersion coefficient, U is theAlekseyev, 1997).flow rate and r is the reaction rate (mol/min orIn this work, we have observed albite dissolution inmol/min /m2). The output solutions pass through anwater at temperature above 300°C, revealing that aelectric conductivity detector, and a computer monitortransition from incongruent to congruent dissolutiondisplays the continuous variation of the electric con-and a return to incongruent are correlated to tempera-ductivity of the output solutions in the reaction system,ture. This work addresses the question of whether thewhich can tell us whether the system is in a steadywide fluctuations in water properties from a sub-critical state (300- 374°C) to a critical state (374°C) atstate or not. As =0, the system is in a steady22.1 MPa affect the kinetics of mineral dissolutionstate, i.e. the electric conductivity of the outlet fluidsprocesses (Zhang Ronghua et al, 1999).is constant. At this time, the liquid product of the sys-tem is sampled (Zhang Ronghua et al, 1990, 1992,2 Experimental Approach1998). This model reflects mass transfer in an axialdirection in terms of an effective longitudinal diffu-The equipment used during the laboratory experi-sivity superimposed on the plug flow reactor rate. Thisments duplicated that used in previous studies (Zhangequation is considered as a flow reactor equation or aRonghua et al, 1990; Zhang and Hu, 1996) and in-reaction-transport equation.cluded a pressure vessel, a liquid pump, a back pres-The current experiments were carried out undersure regulator, a temperature controller, a heat sourceuniform conditions at temperatures in a range of 25°(furmace), a pressure gauge, an electric conductivityto 400°C, a constant pressure of 23 MPa, and a flowdetector and a computer.rate varying from 0.5 to 3.0 ml/min. Constant pres-Natural albite crystals were obtained from Geologi-sure and temperature were maintained while the flowcal Museum, Beijing. Mineral particles were crushedrate was changed in the isothermal processes. Samplesin a shatter box and sieved to 20 to 40 mesh, and thenof output solutions were taken at constant pressure,the pure albite crystals were hand-picked under thetemperature and flow rate. In order to obtain an aque-microscope.ous sample at a steady state condition, fluid samplesMineral particles were crushed again in the shatterwere normally taken one or two hours after the flowbox and sieved to 20 to 120 mesh. And also, the min-rate, or temperature was changed. The average res i-eral particles were cleaned ultrasonically using ace-dence time in the reactor at each flow rate was meas-tone to remove fine particles, rinsed with distilledured, and then the residence-time distribution func-water and dried at 70-809C. Then surface areas oftion, E(t), and the cumulative residence time function,representative samples were determined by a singleF(t), for the flow reactor can be determined (Denbighpoint (Kr-He) BET method.and Turner, 1984; Zhang and Hu, 1996).Mineral particles were introduced to a pressureFor each aqueous sample, the authors analyzed thevessel, which was mounted vertically. De-ionized andconcentrations of Na, Al and Si. Analysis was alsode-gassed water was derived into the vessel from theconducted of the surfaces of solid (albite) prior to andbottom to the top at different flow rates (0.5- -3following the procedure. The chemical composition ofm/min). Output solutions were sampled and the com-the albite was also analyzed. Through examination ofposition of these solutions was analyzed.the residence中国煤化工and concen-"TYHCNMHG.402Vol. 74 No. 2ACTA GEOLOGICA SINICAJune 2000tration of dissolved material in the input and outputog -(X/S)qX5S0，where X refers to Na or AI, aq refers .solutions, the authors were able to measure the reac-tion rates.to aqueous solution and solid refers to the originalmineral material. The logarithm value > 0 indicates3 Experimental Resultspositive incongruent dissolution; the logarithm value=0 indicates congruent dissolution; and the logarithmThe experimental results are presented in terms of thevalue < 0 indicates negative incongruent dissolution.release rates of Na, Al and Si for albite dissolution.The release ratio (logarithm value) for Na/Si orThe release rates of Na, Al and Si were obtained at theAl/Si in the T range from 25 to 300°C shifts fromfollowing temperature levels: 25, 100, 150, 200, 250,positive to zero. In most cases, Na and Al easily enter300, 350, 374 and 400C, at a constant pressure of 23into aqueous solutions. But in the T range from 300 toMPa and under highly varied hydrodynamic condi-400°C, the release ratio is from zero to negative,tions, i.e. varied flow rates.which indicates that the Si release ratio is higher thanthose of Na and Al. The release ratio = zero indicates4 Release Rates of Na, Al and Sicongruent dissolution. The release ratio of Na/Alchanges from negative to zero, as the temperatureNa, Al and Si release rates in aqueous solutions wereincreases from 25 to 300°C. The release ratio of Na/Almeasured as functions of temperatures and flow ratesshifts from zero to negative when the temperaturein reaction systems. The albite dissolution is describedincreases from 300 to 400°C (Fig. 2).by plotting the output concentrations of dissolvingspecies (Na, Al and Si) against temperature or by6 Dissolution Rate Lawplotting the release rate against temperature at eachflow rate. The maximum release rates of Na, Al and SiAlbite dissolution rate in water has been measured in aof albite dissolution in the flowing hydrothermal sys-flow reactor at temperature up to 300°C and at 13.8tems under different flow velocities were always db-MPa. Based on the relation between the dissolutiontained at 300°C. The dissolution rates increased as therate (release rate for Na, Al and Si) and the concentra-temperature was increased from 25 to 300°C, but c-tion of the output solution, the dissolution rate con-creased from 300 to 400°C. However, under somestant can be worked out.hydrodynamic conditions (flow rate 1.3- 1.6 ml/min.)Albite steady state dissolution rates were measureddissolution rates rose after the temperature exceededin open-flow reactors as a function of concentrations374°C. This suggests that the reaction rates fluctuateof aqueous silica, aluminum and sodium in our ex-widely around the critical temperature and pressureperimental conditions. The measured dissolution ratespoints of water (Fig. 1).in far-equilibrium conditions are consistent with-r=kII al;"(2)5 Congruent and Incongruent Dissolutionwhere -r is the dissolution rate, k stands for the rate .constant, a; is the activity of the aqueous species i, andAlbite dissolution is often incongruent in water atn is the rate order. Under far-equilibrium conditions,temperatures from 25 to 400°C. The current studydissolution rates vary linearly with a;", where a; is theinvestigates the hypothesis that dissolution movesactivities of aqueous silica, aluminum and sodiumtowards congruence at 300°C and 23 MPa, but thataqueous species in solutions. Because of incongruentonly at 300°C is albite dissolution congruent. Thdissolution, the release rate of Si is different from theresults suggest that in a temperature range from therelease rate of Al or release rate of Na in most cases.subcritical 300°C to the supercritical 400C (at 23The n value of the Si -release rate is also different fromMPa) albite dissolution shifts from congruent to in-those of release rates of Al and Na. At temperaturescongruent.from 150 to 300°C, the n of the Si-release rate rangeThedefinition of the release ratio isfrom-0.4 to -中国煤化工-0.7,and forTHCNMHG.Kinetics of Hydrothermal Reactions of Minerals in WaterZhang Ronghua et al.403Si120-P=22 MPa1100 | U=1.4 ml/min合80号0.AlU 40-20-1.Na550100050030400450T(C)550100150200250300350400430Na/Si120U=1.6 m/min10目800.0(b40十200501001502002503003040040)|o S0100150200250300350400450AlNa02-120十\ Si04-U=1.9 ml/min00130-(c2-2A140120 H1.6十55001502002303003504004500t-士+0 SO 100 150200 250300350400 450Fig.2. Distribution of the output concentra-T('C)Fig. 1. The relative release ratios for AI/Si, Na/Si and AINa astions of reaction products Si, Al and Na) as aa function of temperature. Release ratios were measured at 25,at a constant pressure of 23 MPa. (a) Flow rate50, 100, 150, 200, 250, 300, 350, 374 and 400°C at a constantequals to 1.4 ml/min; (b) flow rate equals to 1.6pressure of 23 MPa. The data have been spread apart at eachm1/min; (C) flow rate equals to 1.9 ml/min.temperature to show the variety of the release ratios with thechange of flow rates.Naitis from -0.7 to -2. When T > 300°C, the release NaAlSiz Og +8H2O=Na*+ Al(OH)4+HSiO4(3)rate of Si rose with increasing concentration of aque-Suppose that a reaction is near equilibrium or hasous Si (or Al and Na).two reaction rates in both positive and negative direc-The equilibrium constant for albite dissolution intions, then the approximation of the reaction to equi-water and the output solution saturation state (explibrium or a saturation state of the solution is given bythe Gibbs free energy of the reaction.ressed as the Gibbs free energy, AG,) were determined△Gr= RT In (Q/K)(4) .with respect to the overall reactionwhere R is the中国煤化工rature (K),YHCNMHG.404Vol. 74 No. 2ACTA GEOLOGICA SINICAJune 2000Q is the activity product, calculated by using experi-magnetite in water using a flow reactor in the tem-mental data; and K is the reaction equilibrium constant.perature range from 25 to 400°C and at a constantAs a result, the QGr in the reaction systems rangepressure of 22.1 MPa. The experimental results ofmagnetite dissolution indicate that the maximum dis -from 0 to -19 (kcal/mol), which indicates that thesolution rate occur as the temperature is close tosystem is under a wide range of saturation state con- 300°C. The dissolution rate increases with increasingditions from near-equilibrium to very far from equilib-temperature until 300°C, and decreases with increas -rium. The△Gr values vary with temperatures and flowing temperature (T > 300°C).rates, which causes the systems to change equilibriumIn the process of temperature increase past the criti-conditions (Fig. 3).cal point for water, the dissolution rates of albite andmagnetire experienced wide fluctuation, which was7 Discussionlikely caused by the resulting fluctuations of the prop-erties of water. The properties of water, especially theThe current theory on the mechanism of mineral dis-density and dielectric constant, vary greatly from thesolution involving albite was derived through experi-subcritical region (300 to 374°C) to supercritical emental results gained at 300°C or below 300C. Pre-gion (374 to 400°C) at 22.1 MPa, which likely affectsvious experimental results suggested that the dissolu-the kinetics of mineral dissolution (Johnson and Nor-tion rate of albite and its incongruent dissolution be-ton, 1991; Levelt Sengers, 1993).haviour are affected by temperature, pH and solutionAs water passes from the subcritical to supercriticalcomposition.phase, its density and dielectric constant continue toThis study presents the experimental results of ddecrease. The variation of water dielectric constantbite dissolution in water at temperature ranging fromwill affect the ion exchange at the albite surface. The25 to 400°C and at 22- -23 MPa. It is significant that high dielectric constant of water will cause water mo-the dissolution rate is affected by crossing the criticallecules to break the Si-O bond and Al-O bond in thetemperature and pressure points for water Tc= =3749C silicate framework. The low dielectric constant ofand Pc = 22.1 MPa).water will support the aqueous ion association andWe have also measured the dissolution rates of impede dissolution of albite.t Abit-water system 22 MPaAcknowledgmentsWe would like to thank the Ministryof Science and Technology and the吕Ministry of Land and Resources for目supporting our project. The experi-mental results presented in this paper8were obtained with the financial support of GTB basic research fund16 t9501115, the“Climbing Project' 95Pre-39, G1999043212 and National8tNatural Science Foundation of China20grant 29673008. We also thank Ms Su200100Yanfeng for her work in the labora-T(C)tory.Fig.3. OGr of reaction versus temperature. OG were calculated based on the albitehydrolysis reaction.中国煤化工MHCNMH G.Kinetics of Hydrothermal Reactions of Minerals in WaterZhang Ronghua et al.405Manuscript received March 2000fluorite in flow systems and surface chemistry. Science inedited by Fei Zhenbi and Hao ZiguoChina (Series D), 39(6): 567-575.Zhang Ronghua and Hu Shumin, 1997. Kinetics of mineral ofdissolutions in open-flow systems and non-linear dynamicReferencesbehavior in the fluid/solid interface. In: Proceeding of 30thAlekseyev, V .A., Medvedeva, L.S.. and Prisyagina, N., 1993.IGC 1996, 19:29- -47, V.S.P.Rates of congruent dissolution of feldspars in acid and alka-Zhang Rongh ua, Hu Shumin, Tong J. and Jiang L., 1998.line hydrothermal solutions. Geochim. Int., 30: 26- -35.Book: Mineral-fluid reaction kinetics in open systems. BeiAlekseyev, V.A., Medvedeva, L.S., Prisyagina, N.I, Meshalkin,jing: Science Press (in Chinese with detailed English b-S.,. and Balabin, A.I, 1997. Change in the dissolution ratesstract).of alkali feldspars as a result of secondary mineral precipi-Zhang Ronghua, Posey- Dowty, J,, Hellmann, R., Boresik, M.,tation and approach to equilibrium. Geochim. Cosmochim.Crerar, D., and Hu Shumin, 1990. Kinetics of mineral-waterActa, 61(6): 1125-1142.reactions in hydrothermal flow systems at elevated tem-Denbigh, K.G., and Turner, J.C.R., 1984. Chemical Reactorperatures and pressures, Science in China (Series B) (Eng-Theory. Press Syndicate of Univ. Cambridge.lish edition), 33(9): 1136- 1152.Hellmann, R., Crerar, D.,. and Zhang R., 1989. Albite feldsparZhang Ronghua, Zhang Xuetong and Hu Shumin, 1999. Ki-hydrolysis to 300°C. In: Reactivity of Solids: Proc. lIth S,netics of ore-forming fluids in epithermal systems. Chinese314 -329.Science Bulletin, 44, supp:5- -6.Hellmann, R., 1995. The albite-water system: part i. The time-Zheng Haifei, Xie Hongsen, Xu Yongsheng et al, 1995. Ex-evolution of the stochiometry of dissolution as a function ofperimental study of the influence of hydrous minerals on thepH at 100, 200, and 300°C. Geochim. Cosmochim. Acta,melting behaviours of rocks at high temperatures and pres-59(9): 1669- -1697.sures. Acta Geologica Sinica, 69(4): 326-336 (in ChineseJonhson, J.W., and Norton, D., 1991. Critical phenomena inwith English abstract).hydrothermal systerms: state, thermodynamic, electrostatic,Zhou Wenbin and Rao Bing, 1997. Experimental study of W-and transport proporties of H2O in the critical region, Am. J.ter-rock hydrogen and oxygen isotope exchange in the Xi-angshan uranium ore-field. Geological Review， 43(3):Levelt Stengers, J.M.H，1993. Critical behavior of fluids:322- -327 (in Chinese with English abstract).concepts and applications, in supercritical fluids. In: Kiram,E., and Levelt Sengers, J.M.H. (eds.), Pub. Coop. Nato Sci-About the first authorXu Baolong, Zheng Yongfei, Gong Bing and Fu Bin, 1997.Zhang RonghuaBorn in 1938; graduated fromExperimental studies of oxygen isotope fractions betweenPeking University in 1963 and took graduate course atbrucite and water at low temperatures. Acta Geologicathe Chinese Academy of Geological Sciences in 1966.Sinica, 71(4): 340- -349 (in Chinese with English abstract).As a visiting scholar, he did research work in Univer-Zhang Ligang, Liu Jingxiu, Yu Guixiang and Chen Zhensheng,sity of California during 1981-1984 and worked in1996. H and O isotope study on the water-rock interactionPrinceton University from 1983- 1984，1986- 1987system of the Yingshan (Cu)-Pb-Zn-Ag mine， JiangxiProvince. Acata Geologica Sinica 70(1): 48- -60 (in Chinesend 1989- 1990. As Director of the Open ResearchLaboratory of Geochemical Kinetics Prof. Zhang an-Zhang Ronghua, Hu Shumin and Cretar, D., 1992. Chemicalgages in research on chemical kinetics of minerals onKinetics of Minerals in Hydrothermal Systems and Massopen-flow systems at elevated temperatures and pres-Transfer. Bejjing: Science Press (in Chinese with Englishsures.abstract).Zhang Ronghua and Hu Shumin, 1996. Reaction kinetics of中国煤化工MHCNMH G.