EXPERIMENTAL TRANSFORMATION OF 2M SERICITE INTO A RECTORITE-TYPE MIXED-LAYER MINERAL BY TREATMENT WITH VARIOUS SALTS

Clays and Clay Minerals, Vol. 25, pp. 302-308. Pergamon Press 1977. Printed in Great Britain EXPERIMENTAL TRANSFORMATION OF 2M SERICITE INTO A RECTORITE-TYPE MIXED-LAYER MINERAL BY TREATMENT WITH VARIOUS SALTS KATSUTOSHI TOMITA Institute of Earth Sciences, Kagoshima University, Kagoshima, Japan (Received 21 December 1976) Abstract--Some dehydroxylated sericites were boiled with solutions of various salts. A rectorite-like regular mixed-layer was formed when 2M sericite was treated with solution containing salts such as NaNOa, Na2SO,, CaSO4, CaC12, MgC12 and MgSO4 respectively. A random mica/montmorillonite mixed-layer was formed from 1M sericite. In order to change the 2M sericite into a regularly interstratifled mineral, they are heated to the temperature range of dehydroxylation. The formation of a regularly interstratified mineral from 2M sericite can be explained by the change in the (OH) bond direction after the extraction of the potassium ion. INTRODUCTION EXPERIMENTAL Mica weathering in soils has been attributed to a loss of K and a gain in water. Loss of K changed micas into expanded layer silicates (Jackson and Sherman, 1953). Various chemical methods of extracting interlayer K from mica in the laboratory have been carried out with aqueous salt solutions (Barshad, 1948, 1954; Caill~re et al., 1949; Mortland, 1958; Rausell-Colom et al., 1965; Reichenbach et al., 1969; Mamy, 1970). Interlayer K in muscovite has been extracted by treating the mineral with molten LiNO3 at 300~ (White, 1956, 1958). The use of sodium tetraphenylboron to extract interlayer K from micaceous minerals was initiated by Hanway (1956), De Mumbrum (1959, 1963), Scott et al., (1960), Scott and Reed (1962a, b), Scott and Smith (1967). K depletion of micas has produced more interstratification when the particles are smaller (Scott, 1968). Experimental transformations of 2M sericites into a rectorite-type interstratified mineral have been reported by Tomita and his colleagues using various techniques, including acid treatment of pre-heated sericite (Tomita and Sudo, 1968a, b; Tomita, 1974), treatment of sericite with molten lithium nitrate, either alone or with NaC1 present (Tomita and Sudo, 1971), and sericite heated to 800~ and treated with sodium tetraphenylboron to remove potassium (Tomita and Dozono, 1972). Development of long basal spacings from chlorite by laboratory process was reported by Brindley and Chang (1974). Ross and Kodama (1976) reported a regularly interstratified chlorite-vermiculite after reaction in saturated bromine water on the steambath. The author have succeeded in forming a rectorite-like mineral from pre-heated 2M sericite by boiling with various salts. The procedure is an easy method and its result seems to be pertinent to the discussion of the mechanism of formation of an interstratified clay mineral from 2M sericites. Startin 9 materials and methods In these experiments 1M, 2M 1 and 2M 2 sericites were used as starting materials. Sample description, identification and chemical composition are given in Table 1. Particles less than 2 pm were obtained by sedimentation. All experiments were carried out on the minus 2/~m fraction. Air-dried specimens were heated to 800~ for certain hours, quenched to room temperature, and boiled with solutions containing a salt such as NaNO3, Na2SO4, CaSO4, CaC12, MgC12 and MgSO4. After boiling for several hours, the sample was washed with distilled water many times. The washed sample was dried in air. Preferred particle Table 1. Chemical composition of starting specimens 1 2 3 SiO2 55.20 47.24 47.01 TiO2 0.06 0.38 0.10 A1203 31.33 35.04 38.07 Fe203 l FeO J 1.13' 0.59* 0.65* MnO tr. tr. 0.06 MgO 0.40 0.21 tr. CaO 0.29 0.16 tr. K20 8.11 8.75 9.47 Na20 tr. 1.37 0.19 H20(+) 5.64t 5.52 4.83t H20(- ) -- 0.32 -- P205 0.05 tr. 0.03 Total 102.21% 99.58% 100.41% 1. 1M sericite from Masuda, Gifu Prefecture, Japan (Kanaoka and Kato, 1972). (Analyst: S. Kanaoka.) 2. 2M1 sericite from Goto mine, Nagasaki Prefecture, Japan (Tomita and Sudo, 1968a, b). (Analyst: K. Tomita.) 3:2M2 sericite from Izumiyama, Saga Prefecture, Japan (Kanaoka and Kato, 1972). (Analyst: S. Kanaoka.) * Total iron determined as Fe203. "~ Ignition loss. 302

Experimental transformation of 2M sericite 303 orientation was used for X-ray analysis. Corrections for Lorentz/polarization factors were made on the X-ray data prior to calculating the d-values less than 2 0 = 10 ~ At each position of a peak, intensity was divided by the Lorentz/polarization factor and a corrected line profile was made. The d-values were calculated for the corrected peak. 2M1 sericite RESULTS NaNO3 treatment. A specimen heated to 800~ for was boiled in a solution of 2 N NaNO 3. After 50 hr boiling the specimen was transformed into an interstratified mineral of mica and hydrous mica. An X-ray powder diffraction pattern of the specimen (Fig~are 1) shows distinctly a 22.7A peak which expands to 23.5 A after ethylene glycol saturation. The 22.7/~ peak contracted to 10d~ on heating to 800~ Some rectorite-type mixed-layer mineral is formed in this specimen in addition to the original sericite. Na,SO4 treatment. After 350 hr boiling with a solution of 1 N Na2SO4, a rectorite-type mineral was formed. A 21.6A peak is observed in the X-ray powder diffraction pattern of the specimen (Figure CaCt 2 25.2~ ~ozz 21.~ / CaSO4, / I/ ~tsz, a I n EQ 2r~ o1~ /I 23,9~ EG "A '1'8~ vq_ l..j _. 2e (CuKc~) Jo ~ " 5 10 ' 20 ' - 30* 2e(CuKc~) Figure 1. X-ray powder patterns for the altered specimens from the heated 2M1 sericite with various salt solutions. NaNO3: altered specimen treated with 2 N NaNO 3 for 50 hr; EG: treated with ethylene glycol; Na2SO4: altered specimen treated with 2 N Na2SO4 for 350 hr; CaSO4: altered specimen treated with saturated CaSO4 for 150 hr; CaCla: altered specimen treated with 2 N CaC12 for 300 hr, MgCla: altered specimen treated with 2 N MgC12 for 310 hr; MgSO4: altered specimen treated with 2 N MgSO~ for 80 hr.

304 KATSUTOSHI TOMITA 1). It moved to 22 A by treatment with ethylene glycol, and contracted to 10 on heating to 800~ for 1 hr. CaSO, treatment. The X:ray powder diffraction pattern of a specimen which was boiled with a samrated solution of CaSO, for 150 hr is shown in Figure 1. A long basal reflection of 23.3 is observed. It expanded to 26 A by treatment with ethylene glycol, and contracted to 10 ~ when heated to 800~ for I hr. CaC12 treatment. The X-ray powder diffraction pattern of a specimen which was treated with 2 N CaC12 for 300 hr is shown in Figure I. A 24.2 A basal reflection is observed. It expanded to 25.2/~ on treatment with ethylene glycol, and contracted to 10 A on heating to 800~ MgC12 treatment. An X-ray powder diffraction pattern of a specimen which was formed by boiling in the solution of 2N MgC12 for 310hr is shown in Figure 1. A long basal reflection of 23.9 A is observed. It did not change by treatment with ethylene glycol, and contracted to 10.1 A after heating to 800~ for 1 hr. MgSO4 treatment. A specimen heated to 800~ for 1 hr was boiled with solution of 2 N MgSO4. After 80 hr boiling a rectorite-type mixed-layer mineral was formed. The X-ray powder diffraction pattern of the altered specimen is shown in Figure 1, The pattern shows a sharp 23.6 A peak, which moved to 23.9 A on treatment with ethylene glycol. The peak moved to 10.1 A after heating to 800~ There are some significant differences in the d-values for the altered specimens, except for those treated with Mg containing salts, not only for different cation saturations but also for the same cation when the accompanying anion is varied. These differences are due to differences of the nature of the interstratification of these altered specimens. Unfortunately, the natures of the interstratification of the altered specimens, except for the altered specimen treated with MgSO4, could not be investigated because of the presence of unaltered sericite. The differences of the nature of the interstratification may depend on the reaction time and the density of the salts. The d-value of the specimen treated with MgCl2 showed 23.9 A, and it did not change by treatment with ethylene glycol. The altered specimen may have a structure of regular mica/vermiculite-like mixedlayer. Scanty expansion after ethylene glycol saturation is due to the large layer charges. As the altered specimen treated with MgSO4 contained a very small amount of unreacted sericite, the specimen was investigated in detail. MacEwan's Fourier transform method (1956) was used to deduce the nature of the interstratification of the specimen. The [ F 12 values of a dioctahedral mica layer with IK +, 1H20 in the interlayer was used for the transform. The combined Lorentz-polarization factor function was (1 + cos 2 20)/sin 20. In Figure 2, the result of the transform of the basal reflection is shown, where A represents a mica layer and B a hydrous layer. Peaks Components A B AA AB AAB ABB ABAB Colc.height 0.51 0.49 0.07 G87 0.50 0.48 0.77 W(R) J / I, I I I I I I I I I to 20 30 40 50 R,~ Figure 2. Fourier transform of basal reflections of the altered specimen from the heated 2M1 sericite by treatment with solution of 2 N MgSO, for 80 hr.

all,i b ~_2 I 1 ( I L I I I I I I I I 0 200 400 600 800 1000 *C Figure 3. Differential thermal analysis curves of: (a) unheated 2M 1 sericite; (b) 2M 1 sericite heated to 800~ for 1 hr; (c) specimen altered from the heated sericite by treatment with solution of 2 N MgSO 4 for 80hr. 2.5 Experimental transformation of 2M sericite 305 WAVELENGTH of types AB and ABAB are outstanding and indicate a marked tendency for alternation of two different layers. The differential thermal analysis curves of the altered specimen, the original sericite and the dehydroxylated sericite are shown in Figure 3. The curve of the original sericite has an endothermic peak at about 680~ due to the dehydroxylation of structure water. The curve of the sericite heated to 800~ has no endothermic peak, but the curve of the altered specimen shows an endothermic peak at about 125~ due to loss of adsorbed and interlayer water, and an endothermic peak at about 550~ due to the dehydroxylation of structure water. Thus rehydration and rehydroxylation occurred in the heated sericite when it was boiled in MgSO~ solution. Significant differences between the dehydroxylation temperature for the original sericite and the altered specimen were observed. The lowering of the dehydroxylation temperature for the altered specimen is due to decrease of the particle sizes and formation of beidellite-like expansible layers in the rectorite-like structure. Beidellite has its main dehydroxylation peak in the 3 4 5 8 7 G 9 t0 I'~ ;2 ;4 a b C 40 36 32 28 24 20 18 16 14 12 10 9 8 7 WA VENUM~3ER cm -1 X 100 Figure 4. Infrared absorption spectra of: (a) unheated 2Mr sericite; (b) 2MI sericite heated to 800~ for 1 hr; (c) specimen altered from the heated sericite by treatment with 2 N MgSO4 solution for 80hr,

9 306 KATSUTOSHI TOMITA lo.7x 3.3~ showed a band at about 3640 cm-1 which is due to the OH stretching vibration, bands at 1640 cm-1 and a very broad absorption band between 3500 and 3200 cm-1 which are due to adsorbed water and interlayer water, respectively. These facts agree with the experimental result of differential thermal analysis. 2Mz sericite A specimen heated to 850~ for 2.5 hr was boiled with a solution of 2N NaNO3 for ll4hr. X-ray powder diffraction patterns for the altered specimen after various treatments are shown in Figure 5. The pattern of the altered specimen shows a 22 ]~ reflection which does not change on treatment with ethylene glycol. It contracted to 10.2 ~ after heating to 800~ for 1 hr. X-ray powder diffraction patterns of specimens which were altered from the heated 2M2 sericite by boiling with a solution of a salt such as NazSO4, CaC12 and MgClz also showed long basal reflections. 1M sericite The d-values of (001) reflections for specimens which were altered from the heated IM sericite by boiling with a solution of a salt such as NaNO3, Na2SO4, CaC12, MgC12 and MgSO4 are shown in Table 2 together with reaction time. They did not show any long basal spacings, and showed random mixed-layer structures. I i I I 1 i 5 10 20 30 ~0 5d' 28(CuKod Figure 5. X-ray powder patterns of 2M 2 sericite after various treatments. (a) heated to 800~ for 1 hr; (b) specimen altered from the heated sericite by treatment with 2 N NaNO3 solution for 114 hr; (c) treated with ethylene glycol; (d) heated to 800~ for i hr. 550~ region (Grim and Rowland, t942; Greene- Kelly, 1957). Infrared absorption spectra of the altered specimen, the original sericite and the heated sericite at 800~ are shown in Figure 4. In the heated sample absorption bands related to the hydroxyl groups disappeared. The absorption bands of the altered specimen DISCUSSION Tomita and his colleagues succeeded in forming a rectorite-type interstratified mineral from 2M sericite using various techniques in the laboratory at atmospheric pressure by acid treatment (Tomita and Sudo, 1968a, b; Tomita, 1974), treatment with molten lithium nitrate, either alone or with NaC1 present (Tomita and Sudo, 1971), and treatment with sodium tetraphenylboron (Tomita and Dozono, 1972). Present experiments indicate that a rectorite-type mixedlayer mineral is formed from dehydroxylated 2M sericite by treatment with various salt solutions. In order to change the sericite into an interstratified mineral, the samples have been heated to the temperature range of dehydroxylation. There are three hypotheses which concern mechanisms of formation of a rectorite-type mineral. They are: (1) primary formation from amorphous materials (Iiyama and Roy, 1963), from pyrophyllite or kaolinite (Matsuda and Henmi, 1976) Table 2. d-values of (001) reflections for specimens which were altered from the heated IM sericite by boiling with various salts Reaction d Treated with Heated to Kind of solution time (hr) (A) ethylene glycol 800~ for 1 hr 2 N NaNO 3 195 11.0 11.0~ 10.1 2 N NazSO4 480 11.2 11.3 10.16 2 N CaC1 z 430 10.65 10.25 10.25 2 N MgClz 430 10.9 10.5 10.4 2 N MgSO~ 430 10.3 10.3 10.3

G Experimental transformation of 2M sericite 307 G H~ AI Hf,~ AI A --cleavage ptane /~'50~60' cleovage plane Figure 6. Diagrammatic representation of a mechanism that results in regular interstratification. (A) 2M1 sericite showing the relative positions and orientations of the hydroxyl and interlayer K +. (B) Suggested changes in hydroxyl orientation when K + replaced by hydrated Na +. or rocks; (2) formation from montmorillonite (Shutov et al, 1969; Brindley and Sandalaki, 1963); and (3) formation from mica by leaching of alternate interlayers of potassium ions and hydration (Sudo et al., 1962; Tomita and Sudo, 1968a, b, 1971; Tomita and Dozono, 1972). In the third case, formation of a rectorite-type mixed-layer mineral from mica was confirmed by various experiments of Tomita and his coworkers. A regularly or nearly regularly interstratified mineral was formed from 2M sericite and a random mixed-layer mineral was formed from IM sericite (Tomita, 1974). The present experiment was carried out based on the third hypothesis. In this experiment the mechanism of formation of a regularly interstratified mineral from 2M sericite can be explained by the idea of Norrish (1972) and by Giese's results (1971, 1972). When the orientation of the dipole of the octahedral hydroxyl ions is perpendicular to the cleavage plane the proton is near the interlayer potassium. In this case the repulsion force between K + and the proton of the hydroxyl is large and it is easy to extract interlayer potassium. When the dipole of the hydroxyl ion is changed to an inclined position with respect to the cleavage plane the interlayer K + is in a more electrically negative environment. In other words the interlayer K + becomes more strongly held by the surrounding oxygens, When the K* is removed, the (OH) bond direction in the octahedral layer of 2M sericite will increase from about 15-20 ~ (Figure 6A) (Serratosa and Bradley, 1958) to 50-60 ~ (Figure 6B) with the cleavage plane (Giese, 1971). Norrish proposes that the expansion of alternate interlayer regions caused by the substitution of another hydrated cation for the K + increases the distance between the interlayer cation and the neighboring hydroxyls. This permits the hydroxyls to the reorient themselves more toward the normal to the cleavage plane, thus reducing the proton-proton repulsion. (Figure 6B). When this bond angle increases with the loss of the interlayer potassium there will be a decrease in angle of the (OH) bond of the other hy- B droxyl that is attached to the same octahedral cation, but facing the adjacent interlayer region. The bonding of interlayer potassium is very dependent on hydroxyl orientation and so if the (OH) bond angle is reduced the remaining potassium ions will be bound more strongly. According to Giese (1972), the OH in 1M mica makes an angle of 0 ~ with the mica layer which increases only to 16 ~ when the K + is removed indicating little interaction between the hydroxyl and potassium. From these facts regular interstratification should occur in 2M sericite but not in 1M. Acknowledgements--The author wishes to express his sincere gratitude to Professor N. Oba and Mr. M. Yamamoto of the Kagoshima University for their comments. The author greatly appreciates the help of Dr. S. Kanaoka of the Government Industrial Research Institute at Nagoya, who made some samples available. This research was supported in part by a Grant in Aid for Scientific Research from the Ministry of Education. REFERENCES Barshad, I. (1948) Vermiculite and its relation to biotite as revealed by base exchange reactions, X-ray analysis, differential thermal curves and water content: Am. Miner. 33, 655-678. Barshad, I, (1954) Cation exchange in micaceous minerals: replaceability of ammonium and potassium from vermiculite, biotite and montmorillonite: Soil Sci. 78, 57-76. Brindley, G. W. and Chang, T. S. (1974) Development of long basal spacings in chlorites by thermal treatment: Am. Miner. 59, 152-158. Brindley, G. W. and Sandalaki, Z. (1963) Structure, composition and genesis of some long-spacing mica-like minerals: Am. Miner. 48. 138-148. Caill6re, S., Henin, S. and Guennelon, R. (1949) Transformation experimentale due mica en divers types de min6raux argileux par separation des feuilletes: Compt. Rend. 228. 1741. De Mumbrum, L. E. (1959) Exchangeable potassium levels in vermiculite and K-depleted micas and implications relative to potassium levels in soils: Soil Sci, Soc. Am. Proc. 23, 192-194. De Mumbrum, L. E. (1963) Conversion of mica to vermiculite by i~otassium removal: Soil Sei. 96, 275-276. Giese, R. F. Jr. (1971) Hydroxyl orientation in muscovite as indicated by electrostatic energy calculations: Science 172, 263-264. Giese, R. F. Jr. (1972) In "General discussion of K-exchange in micas": Proc. Int. Clay Conf., Madrid, 1972, pp. 494-495. Greene-Kelly, R. (1957) The Differential Thermal Analysis of Clays, (Edited by Mackenzie R. C., Chapter V), Mineralogical Society, London. Grim, R. E. and Rowland, R. A. (1942) Beidellite: Am. Miner. 27. 801-818. Hanway, J. J. (1956) Fixation and release of ammonium in soils and certain minerals: Iowa State Coll. J. Sci. 30, 374-375. Iiyama, J. T. and Roy, R. (1963) Controlled synthesis of heteropolytypic (mixed-layer) clay mineral: Clays & Clay Minerals 10, 4-22. Jackson, M. L. and Sherman, G. D. (1953) Chemical weathering in soils: Adv. Aoron. 5. 219-318. Kanaoka, S. and Kato, E. (1972) Natures of sericites in various potter's clays: Program of Annual Meetin 9 of the Mineralogical Society of 3apan (Abstract), p. 10. MacEwan, D. M. C. (1956) Fourier transform methods for studying scattering from lamellar systems I. A direct

308 KATSUTOSH1 TOMITA method for analysing interstratified mixtures: Kolloid Z. 149. 96-108. Mamy, J. (1970) Extraction of interlayer K from phologopite, specific effects of cations. Role of Na and H concentrations in extraction solution: Clays & Clay Minerals 18, 157-163. Matsuda, T. and Henmi, K. (1976) A synthesized 25 A regularly interstratified mineral: 20th annual meeting of the Clay Science Society of Japan (Abstract), p.18. Mortland, M. M. (1958) Kinetics of potassium release from biotite: Soil Sci. Soc. Am. Proc. 22. 503-508. Norrish, K. (1972) Factors in the weathering of mica to vermiculite: Proc. Int. Clay Conf., Madrid, 1972, 417432. Rausell-Colom, J. A., Sweatmen, C. B., Wells, C. B. and Norrish, K. (1965) In Experimental Pedology (Edited by Hodsworth, E. G. and Crawford, D. V.), pp. 4-72, Butterworths, London. Reichenbach, H. Graf Von and Rich, C. I. (1969) Potassium release from muscovite as influenced by particle size: Clays & Clay Minerals 17, 23-29. Ross, G. J. and Kodama, H. (1976) Experimental alteration of a chlorite into a regularly interstratified chlorite-vermiculite by chemical oxidation: Clays & Clay Minerals 24, 183 190. Scott, A. D. (1968) Effect of particle size on interlayer potassium exchange in micas: Trans. 9th Cong. Int. Soil Sci. Soc. 2, 64%660. Scott, A. D., Hunziker, R. R. and Hanway, J. J. (1960) Chemical extraction of potassium from soils and micaceous minerals with solutions containing sodium tetraphenylboron I. Preliminary experiments: Soil Sci. Soc. Am. Proc. 24, 191 194. Scott, A. D. and Reed, M. G. (1962a) Chemical extraction of potassium from soils and micaceous minerals with solutions containing sodium tetraphenyboron II. Biotite: Soil Sci. Soc. Am. Proc. 26. 41-45. Scott, A. D. and Reed, M. G. (1962b) Chemical extraction of potassium from soils and micaceous minerals with solutions containing sodium tetraphenylboron III. Illite: Soil Sci. Soc. Am. Proc. 26, 45-48. Scott, A. D. and Smith, S. J. (1967) Visible changes in macro mica particles that occur with potassium depletion: Clays and Clay Minerals, Proc. 15th Nat. Conf. 357 373, Pergamon Press, Oxford. Serratosa, J. M. and Bradley, W. F. (1958) Determination of the orientation of OH bond axes in layer silicates by infrared absorption: J. Phys. Chem. 62, 1164-1167. Shutov, V. D., Drits, V. A. and Sakharov, B. A. (1969) On the mechanism of a postsedimentary transformation of montmorillonite into hydromica: Proc. Int. Clay Conf., 1969, Vol. 1, 523-531. Sudo, T., Hayashi, H. and Shimoda, S. (1962) Mineralogical problems of intermediate clay minerals: Clays and Clay Minerals, 9th National Conf., 378 392, Pergamon Press, Oxford. Tomita, K. (1974) Similarities of rehydration and rehydroxylation properties of rectorite and 2M clay micas: Clays & Clay Minerals 22, 79-85. Tomita, K. and Dozono, M. (1972) Formation of an interstratified mineral by extraction of potassium from mica with sodium tetraphenylboron: Clays & Clay Minerals 20, 225-231. Tomita, K. and Sudo, T. (i968a) Interstratified structure formed from a pre-heated mica by acid treatments: Nature 217. 1043-1044. Tomita, K. and Sudo, T. (1968b) Conversion of mica into an interstratified mineral: Rep. Faculty Sci., Kagoshima Univ. 1. 89 ll9. Tomita, K. and Sudo, T. (1971) Transformation of sericite into an interstratified mineral: Clays & Clay Minerals 19, 263-270. White, J. L. (1956) Layer charge and interlamellar lattice silicates: Clays & Clay Minerals 4, 133-146. White, J. L. (1958) Layer charge and interlamellar expansion in a muscovite: Clays & Clay Minerals 5, 289-294.