COVER SHEET

Frost, Ray and Butler, Ian (2006) An Overview of the High-Pressure Vibrational Spectra of Clays and Related Minerals. Applied Spectroscopy Reviews 41(5):pp. 449-471. Accessed from http://eprints.qut.edu.au Copyright 2006 Taylor & Francis

1

An Overview of the High-Pressure Vibrational Spectra of Clays and Related Minerals

Ian S. Butler Department of Chemistry, McGill University, 801 Sherbrooke Street West,

Montreal, Quebec, Canada

Ray L. Frost School of Physical and Chemical Sciences, Queensland University of Technology,

Gardens Point Campus, Brisbane, Queensland, Australia Abstract: An overview of the effect of high external pressures on clays and related minerals is presented. The results show that this is an area that will welcome further investigation in the future, especially in view of the importance of clay minerals in the Earths’s mantle and, possibly, even in prebiotic chemistry in the formation of amino acids and the mineralogy of the Martian surface. Keywords: Clays, infrared, Raman, high-pressure, diamond-anvil cell. Received . Accepted . Address correspondence to Ian S. Butler, Department of Chemistry, McGill

University, 801 Sherbrooke St. W., Montreal, Quebec, Canada H3A 2K6. E-mail: ian.butler@mcgill.ca

2

INTRODUCTION Natural and synthetic clays constitute an incredibly wide range of layered silicate minerals. The chemistry and physics of clay mineral surfaces and interfaces is the principal focus of the most recent book on clays, edited by Wypych and Satyanarayana, entitled Clay Surfaces: Fundamentals and Applications” (1). In a comprehensive review by Solin (2) on the properties of clays and their intercalation compounds, which was published in 1997, there was only passing mention of the effect of high pressures on their vibrational spectra. Since then, the behaviour of clays under high external pressures has become of considerable interest to geochemists, geophysicists, mineralogists from both a technological and a scientific point of view.

Clay minerals are important constituents in sedimentary rocks that occur in subduction zones of the Earth’s mantle. These materials may play an important role in influencing water regulation and, possibly, even in triggering deep-focus earthquakes (3). In his review article, Solin classified clays and clay intercalation compounds in terms of the rigidity of the host layers with respect to transverse distortions that occur perpendicular to the planes of the layers (2). On this basis, graphite intercalation compounds, which contain thin layers of monatomic carbon atoms are termed “floppy”, while three-atom thick dichalcogenide layers are much stiffer and interconnected sheets of atoms, such as those typically found in clays, are considered to be “rigid”. The physical properties of individual clays and clay intercalation compounds depend dramatically on the layer rigidity. Considerable research has now been published concerning the formation of clay intercalation compounds, in which various guest species are located between the clay layers, sometimes distorting the individual layers and sometimes not (1,2).

The present review is not meant to be exhaustive - it has been designed to give the

readers an overview of the effects of high pressures on the infrared absorption and Raman scattering spectra of mainly clays, but also some related minerals. As mentioned already, such research has obvious implications in providing more information regarding the nature of the water reservoir in the Earth’s mantle. In fact, some recent research has indicated that the upper mantle is saturated with water in the 20-40 kbar pressure range (3). Clays are usually considered to be analogous chemically and structurally to phyllosilicates in which varying amounts of water have been trapped between the silicate layers together with a wide range of cationic species. They are now being used in countless industrial situations, e.g., ceramics, construction materials, glossy paper making, paint fillers, and drilling muds. Also, they obviously play a vital role in agriculture

CLASSIFICATION OF CLAYS Clay minerals can be classified into five major classes (4,5): (a) Kaolinite Class, which consists of the four polytypes kaolinite, dickite, nacrite and halloysite with the formula Al2SiO5(OH)4.; (b) Montmorillonite/smectite Class, which includes pyrophyllite, talc, vermiculite, sauconite, saponite, nontronite and montmorillonite with the general formula (Ca,Na,H)(Al,Mg,Fe2+,Zn)2(Si,Al)4O10(OH)2

.xH2O; (c) Illite (Clay-mica) Class, which consists essentially of different hydrated forms of muscovite with the general formula (K,H)Al2(Si,Al)4O10(OH)2.xH2O and rectorite ](Na,Ca)Al4(Si,Al)8O20(OH)4

.2H2O]; (d)

3

Chlorite Class, which encompasses a large number of minerals with the general formula X4-6Y4O10(OH,O)8 (X = Al, Fe, Li, Mg, Mn, Ni, Zn and Cr ; Y = Al, Si, B or Fe); and (e) Sepiolite/Palygorskite Class with formulae Mg4Si6O15(OH)2

.6H2O and (Mg,Al)2Si4O10(OH).4H2O, respectively.

The kaolinite structure is composed of silicate (Si2O5) sheets tightly bonded to gibbsite-like [Al2(OH)4] layers (Figure 1). Two scanning electron microscopy (SEM) pictures of kaolinite are shown in Figure 2. In the montmorillonite/smectite class, the gibbsite layer of the kaolinite class can be replaced by a similar brucite-like [Mg2(OH)4] layer and the overall structures consist of triple-decker sandwiches of silicate, gibbsite (or brucite) and silicate layers with water molecules lying between the layers (Figure 3). A SEM picture of montmorillonite is shown in Figure 4. The structure of the third class is similar to that of the montmorillonite/smectite class; it consists of triple layers of silicate, gibbsite and silicate with water molecules and K+ ions trapped between the layers (Figure 5). The structure of the chlorite class consists of essentially a Dagwood-like sandwich – two triple-decker sandwich units made up of silicate-brucite-silicate layers, in between which there is an additional brucite layer together with varying amounts of water (Figure 6). Finally, the sepiolite structure (Figure 7) involves a 1:1 regular interstratification of a mica and a smectite. The palygorskite structure has a similar fibrous morphology to sepiolite.

[Insert Figures. 1-7 here.]

VIBRATIONAL SPECTROSCOPIC MEASUREMENTS AT HIGH PRESSURES High-pressure vibrational spectroscopic studies can be readily accomplished in the laboratory nowadays with the aid of commercially available diamond-anvil cells (DACs) and modern infrared and Raman microprobe spectrometers. The construction of a DAC makes use of a pair of brilliant cut, gem-quality diamonds that can be squeezed together mechanically, while the diamond faces are carefully kept separate from one another by a very thin (~300-μm thick) Inconel or stainless-steel gasket (Figures 8 and 9). Different types of diamonds are used for the high- pressure vibrational spectroscopic measurements:

[Insert Figures 8 and 9 here.]

type-I (low fluorescence) for Raman and type-IIA for infrared work. The samples under investigation are placed, with the aid of an optical microscope, into a hole (~300-μm diameter) drilled in the center of the gasket. The pressures within the DAC can be determined in several ways. For infrared spectroscopy, intimate mixtures of NaNO2 (0.12 wt% of NO2

-) and/or NaNO3 (0.3 wt% of NO3-) in NaBr are often used, since the pressure

dependences of the antisymmetric N-O stretching modes of the NO2- and NO3

- ions at 1279.0 and 1401.3 cm-1, respectively, are well known (6). In the case of Raman spectroscopic measurements, the pressure behavior of the R1 fluorescence band of a small ruby chip is usually used for calibration purposes (7). Pressures approaching several hundreds of kilobars (8) can be measured using these different calibration techniques. In order to ensure a hydrostatic pressure across the faces of the diamonds and to avoid their

4

fracture under pressure, it is normally advisable to use a pressure-transmitting fluid such Nujol or 1:4 methanol:ethanol mixture (9). Complete details of the use of DACs in undertaking vibrational spectroscopic measurements at high external pressures can be found in Ferraro’s 1984 classic book (10). Finally, a typical DAC, which is capable of achieving pressures around 70 kbar and is available commercially from High Pressure Diamond Optics, Tucson, Arizona for about $8,000 US, is shown in Figure 10.

[Insert Figure 10 here.]

For ultra-high pressure measurements, at ~100 kbar and above, such as those required by geochemists studying the physics and chemistry of the Earth’s mantle, much more sophisticated equipment has been developed. One good example is the so-called Paris-Edinburgh pressure cell at the ISIS facility in the Rutherford Appleton Laboratory in the U.K. (11). This cell of this type was used recently in a high-pressure, neutron powder diffraction and synchrotron infrared absorption spectroscopic study of synthetic kaoite hydrogarnet {Ca3Al2[(OD)4]3}, at pressures up to 94 and 98 kbar, respectively (12). The results obtained were in excellent agreement with earlier high-pressure X-ray diffraction and ab initio calculations. The infrared spectra collected for the same sample exhibited discontinuities in the pressure vs. wavenumber plots for both O-H and O-D stretching vibrations at ~50 kbar. The observed discrepancy between the phase transition pressures from the neutron and infrared measurements was attributed to low signal-to-noise levels in the neutron work leading to some of the reflections being masked.

SELECTED EXAMPLES OF HIGH-PRESSURE VIBRATIONAL STUDIES OF CLAYS AND RELATED MINERALS Since the hydrous minerals, gibbsite [α-Al(OH)3] and brucite [Mg(OH)2], play important roles in the structure of clays and the vibrational spectra of both of these materials have been examined under high pressures, this overview will begin with a description of the effects of high pressure on the structures of these two minerals before moving on the clays themselves. Gibbsite Liu et al. (13) have reported the results of their recent room-temperature, high-pressure infrared absorption measurements on gibbsite [α-Al(OH)3] for pressures up to 250 kbar. They identified a phase transition at ~25 kbar and there was also evidence for gradual disordering of the hydrogen sublattice above 150 kbar. The layered structure of gibbsite consists of stacks of AlO6 octahedra in which each oxygen atom is linked to a hydrogen atom. Fifty percent of the hydrogen atoms are involved in intralayer H-bonds, while the other 50% form interlayer H-bonds. Compression is expected to affect the interlayer H-bonds more than the intralayer ones. Some representative high-pressure infrared spectra of gibbsite are shown in Figure 11, while the pressure dependences of the OH stretching and deformation bands are plotted in Figure 12. These researchers attributed the band at

[Insert Figures 11 and 12 here.]

5

3619 cm-1 to the intralayer O-H stretching mode and the two bands at 3395 and 3375 cm-1 to the interlayer O-H stretching modes. The two latter bands, however, are more likely to be associated with H-bonded water molecules. All three bands disappear at the phase transition pressure (~25 kbar) suggesting that the high-pressure phase is of higher symmetry than is low-pressure one, in agreement with the high-pressure synchrotron X-ray diffraction work on gibbsite by the same authors, which was also reported in this paper. As is typical of O-H stretching modes (10), all the gibbsite O-H bands shift under compression to lower wavenumbers reflecting the increasing H-bond strengths. Moreover, as expected, the interlayer O-H bands shifted more than did the intralayer ones. In most cases, the negative pressure dependences observed were above -1.5 cm-1/kbar. In these vibrational experiments, argon gas was used as the pressure medium in an effort to minimize any stress-induced IR band broadening. Above 150 kbar, the O-H stretching bands become highly broadened and even merge together, indicative of a wide distribution of O-H bond lengths and angles in the high-pressure phase. Liu et al. concluded that there was a gradual disordering of the hydrogen sublattice under compression in the 150-250 kbar range. Just as for Co(OH)2 (14), hydrogen disorder under high pressures is believed to contribute to local disorder in the Al-O linkages in the structure of the high-pressure gibbsite polymorph, but it is insufficient to bring about complete amorphization. The high-pressure Raman spectra of gibbsite were first studied by Huang et al. (15) for pressures up to ~230 kbar. They noted the appearance of additional O-H stretching bands above 30 kbar. More recently, Johnston et al. (16) have performed a single-crystal high-pressure Raman study on gibbsite for pressures up to 550 kbar. They concluded that there was a phase transition at around 22 kbar, in good agreement with parallel infrared work already discussed above (13). Brucite The Raman spectra of brucite [Mg(OH)2] has been investigated at pressures up to 366 kbar under non-hydrostatic conditions and up to 197 kbar under quasi-hydrostatic conditions (17). A pressure-induced structural change was detected at 40 kbar when several new Raman bands appeared. One of these new bands increased in intensity over a wide pressure range as the result of a Fermi resonance interaction. Brucite exists as a layered structure, somewhat similar to that of gibbsite in that each magnesium atom is surrounded by a distorted octahedron of oxygen atoms. A plane of magnesium atoms is sandwiched between two planes of oxygen atoms and the O-H bonds are perpendicular to these three planes. Typical Raman spectra of brucite in the lattice and internal mode regions under quasi-hydrostatic conditions up to ~200 kbar are shown in Figure 13. The pressure dependences obtained for the lattice and O-H stretching modes for pressures up to ~200 kbar are illustrated in Figure 14 for both non-hydrostatic and hydrostatic conditions. The Raman band at 360 cm-1 grows at the expense of a neighbouring band at 280 cm-1, suggesting a pressure-induced Fermi resonance interaction, similar to that reported for norbornadiene by Kawai et al. (18). [Insert Figures 13 and 14 here.] (a) Kaolinite Class The 1:1 dioctahedral phyllosilicate kaolinite, Al4[Si4O10](OH)8, is one of the most abundant clay minerals in the Earth’s crust. The silicate layers consistent of sheets of

6

[SiO4]2- tetrahedra linked together in the form of 6-membered silicate rings through common oxygen atoms to sheets of AlO6 octahedra in the form of 4-membered aluminate rings.

When kaolinite, is subjected to unixial pressures greater than 1 kbar (or ground for ~18 min, for example, in a mortar and pestle), there are distinct changes observed in the crystallinity of the clay (19). The differences in particle morphology caused by such mechanochemical activation, e.g., fractures, bending, deformations and rolling of layers, etc., can be detected by X-ray diffraction, infrared and Raman spectroscopy, thermal analysis and electron microscopy (20). Recently, Johnston and his colleagues have reported the first documented case of a pressure-induced phase transformation in the 1:1 phyllosilicate dickite, which is a polytype of kaolinite. This phyllosilicate undergoes a reversible pressure-induced phase transition at 25 kbar (16,21), as evidenced by the changes observed in the ν(OH) region (Figures 15 and 16). Interestingly, above 25 kbar, the individual 1:1 layers remain unaffected, whereas stacking of the individual layers and the interlayer topology are altered significantly.

[Insert Figures 15 and 16 here.]

About five years ago, Benco et al. (22) were the first to use ab initio molecular dynamics simulations to predict the structure and vibrational spectra of kaolinite, including the orientation of the OH groups. Subsequently, Balan et al. reported some similar theoretical calculations (23). Johnston’s group has now employed density functional theory (DFT) to predict the structures of the three naturally occurring polytypes kaolinite, dickite, and nacrite (24). This same research group has also performed DFT calculations to determine for the first time the bulk modulus and elastic constants for kaolinite (25). A bulk modulus value of 230 kbar was predicted. The latest ab initio calculations are by Torsoni et al. (26), who have employed a periodic approach (CRYSTAL03 program and B3LYP function) to optimize the geometry of kaolinite and to predict the vibrational frequencies (including anharmonicity) for all possible OH groups (inner, inner surface and outer surface). The calculated results are in good agreement with the latest structural (27) and vibrational data (28) reported in the literature. Part of the driving force behind these theoretical calculations is the fact that such clay minerals probably played a vital role in prebiotic chemistry involving the generation of amino acids (29). (b) Montmorillonite/smectite Class One of the earliest reports of a high-pressure vibrational study of a natural clay mineral dates back to the early 1990s when Wada and Kamitakahara (30) described the results of their Raman spectroscopic investigation of the pressure dependence of the OH stretching mode in vermiculite [(Mg,Fe2+,Al)3(Si,Al)4O10(OH)2

.4H2O] upon hydration. These authors contended that there was a “stiffening” of the OH stretching vibration, i.e., a shift to lower wavenumbers, with decreasing water content. Moreover, they attributed this dependence to the deeper penetration of the so-called “gallery” cations into the hexagonal pockets of the mineral structure while the gallery structure is collapsing because of the loss of water. Subsequently, Holtz et al. (31) studied the effects of pressure on the Raman

7

spectra of two dioctahedral (margarite and muscovite) and three trioctahedral (talc, biotite and phlogopite), water-free 2:1 clays. These researchers were particularly interested in the behaviour under pressure of the torsional mode at ~100 cm-1 and the OH stretching mode at ~3600 cm-1. The band at 50 cm-1 in margarite is identified as an interlayer shear mode, while the band observed near 700 cm-1 for all the clays is associated with an oxygen-layer breathing mode. The torsional mode for all five clays displayed positive pressure dependences (Table 1). In the case of the OH stretching modes, however, there

[Insert Table 1 here.] was an interesting variation in the pressure dependences – positive dependences were observed for the three trioctahedral clays and negative ones for the two dioctahedral clays. Normally, the application of pressure is expected to lead to bond shortening and a concomitant increase in the force constant of the bond involved (i.e., bond stiffening) and, consequently, an increase in its vibrational wavenumber (10). The “softening” of the OH stretching modes for margarite and muscovite with pressure was attributed to changes in the orientation of the OH protons with respect to the gallery cations in the minerals, so that the Coulombic repulsions between them leads to a lengthening of the OH bonds with increasing pressure.

It should be mentioned that some controlled, low-pressure water (10-6 to 0.99 bar), near-infrared diffuse reflectance (4500-8000 cm-1region) and Raman (600-750 and 2500-2800 cm-1regions) measurements have been reported for the hydration of a synthetic saponite clay mineral (32). The results of this study have provided information, under controlled water-pressure conditions, on the swelling of the clay layers and changes in the interactions of the adsorbed water molecules.

(c) Illite (Clay-mica) Class In the 1990s, three different groups reported the effects of high pressures on di- and trioctahedral micas (33-35). In particular, these researchers studied the compression of muscovite and concluded there was no phase change occurring and only amorphization could be detected at pressures above 200 kbar. Similarly, no phase changes were identified for paragonite {NaAl2[AlSi3O10](OH)2} for pressures up to 40 kbar (36), nor for philogopite {KMg3[AlSi3O10](OH)2 (37). (d) Chlorite Class No phase changes were detected when chlorite was subjected to high pressures (37). However, the H-bonding in the synthetic clinoclore, (Mg5Al)(Si3Al)O10(OH)8, has been investigated by high-pressure Raman spectroscopy at ambient temperature using a moissanite (single-crystal 6H-SiC)-anvil cell (38-41) for pressures up to 265 kbar (42). There are three O-H stretching bands observed at ambient pressure in the 3400-3650 cm-1 region, which are attributed to the H-bonded interlayer OH groups, and a sharp band at 3679 cm-1 associated with stretching of the free OH groups of the talc-like 2:1 layer. Up to ~60 kbar, all four OH stretching bands exhibit linear pressure dependences. At ~90 kbar, there are dramatic discontinuities in the pressure dependences of the interlayer OH stretching modes indicating the onset of significant changes in the H-bonding during a

8

pressure-induced structural change. There is also a second pressure-induced structural transformation during the compression of clinochlore at ~160 kbar. These two structural transformations are completely reversible. (d) Sepiolite While the infrared and Raman spectra of sepiolite and palygorskite clays [43,44] and species adsorbed on its surface [45] have been investigated recently, the effects of high pressure on the vibrational spectra on this mineral class does not yet appear to have been examined. (e) Other Related Minerals Of interest in connection with the effect of pressure on H-bonding in clay minerals is the work of Kagi et al. on kalicinite (KHCO3) (46). They investigated the infrared spectra of this mineral at pressures up to ~63 kbar. Three vibrational modes could be detected, viz., O-H stretching, O-H…O in-plane bending and O-H…O out-of-plane bending at 2620, 1405 and 988 cm-1, respectively. The O-H stretching mode was relatively insensitive to pressure up to the phase-transition pressure of 28 kbar. The two mother modes increased steadily with increasing pressure up to 28 kbar and then remained essentially constant thereafter. The three Raman bands associated with the internal modes of the HCO group changed drastically at ~28 bar suggesting the environment around this group played a role in the phase transition. Experimental data have been reported by Williams et al. (47) for the ambient pressure Raman and high-pressure (up to ~270 kbar) infrared spectra of MgSiO3 perovskite. Four bands were observed in both the Raman (500-250 cm-1 region) and the infrared (800-500 cm-1 region) spectra. The antisymmetric Si-O stretching modes were assigned to the 800 cm-1 region, while octahedral stretching and bending modes were attributed to bands in the 690-370 cm-1 region and lattice vibrations involving the Mg2+ ions were located in the 300-200 cm-1 region. An average mode Grüneisen parameter of 1.36 ±0.15 was calculated from the pressure shifts of the four infrared-active vibrations. From the mode Grüneisen parameters and observed vibrational data, the thermodynamic Grüneisen parameter was estimated as being about 1.9, which compares very favourably with the thermodynamic value of 1.77 that was obtained from the ambient pressure measurements. Mode Grüneisen parameters can be calculated from eq. 1:

γi = (KT/νi)(∂νi/∂P) (1)

where γi is the mode Grüneisen parameter and KT is the isothermal bulk modulus, νi is the i-th frequency and P is the pressure (10). Recently, Hofmeister and Mao (48) have redefined the mode Grüneisen parameter for polyatomic substances, so that more accurate geophysical properties can be inferred from vibrational spectroscopic data. For example, synchrotron high-pressure (up to 100 kbar) infrared spectra have been obtained in the 50-4000 cm-1 region for the chloritoid group minerals, M2Al4O2(SiO4) 2(OH) 4 (M = Fe, Mg), and various thermodynamic properties have been calculated (49).

9

The structural role of H atoms in hydrous magnesium silicates has been inferred in part from infrared and Raman studies at ambient temperature and pressure (50). It is only quite recently that high-pressure measurements been attempted for the infrared spectra of the magnesium hydrous silicates, phase B [Mg12

VISiIVSi3O19(OH)2] and the Raman spectra of phase A [Mg7

IVSi2O8(OH)6] and superhydrous phase B [Mg10

VISiIVSi2O14(OH)4]. Changes in O-H stretching frequencies were found to depend on both the O-H…O bond lengths and bonding angles. For instance, upon application of pressure to phase B, the frequency of the most intense OH band in the infrared spectra decreases to a minimum at 50 kbar and then rises to a broad maximum at around 350 kbar. The frequency of the other intense OH band decreases to a broad minimum near 300 kbar. Some of the H-O-O angles increase from about 12 to 21º at 370 kbar. The relative incompressibility of the [SiO4]2- tetrahedra is consistent with the changes in O-O bond length and H-O-O angle that are calculated from the trends in the OH stretching frequency with pressure. Similar results were obtained from the high-pressure Raman study of the superhydrous phase B. In the case of phase A, there is a linear dependence found for the frequencies of the OH bands with increasing pressure, an observation that is associated with linear hydrogen bonds and unpaired OH groups.

Single-crystal Raman spectra of synthetic, hydrous γ-Mg2SiO4 have been measured in a diamond-anvil cell at room temperature using helium as the pressure-transmitting medium for pressures approaching 565 kbar (51). The five characteristic Raman bands shift continuously with increasing pressure, but there is no evidence of a pressure-induced phase transition. There is, however, a new band at 802 cm-1 (extrapolated to 1 bar) that appears at ~300 kbar, but which disappears upon pressure release. This subtle structural change may possibly be associated with the formation of Si-O-Si linkages and/or a partial increase in the silicon coordination under pressure.

(f) Experimentally Shocked Minerals There is growing interest in studying the vibrational spectra of experimentally “shocked” minerals with the ultimate aim of generating a spectroscopic data bank for use in the examination of planetary surfaces. Some of these minerals occur naturally, as the result of meteors crashing into the Earth’s surface at great velocities. The crater impacts subject the minerals to extremely high pressures. These high pressures can also be simulated in the laboratory. For example, thermal infrared emission and reflectance spectra in the 150-1400 cm-1 region have been reported for experimentally shocked (170-560 kbar) albite- (NaAlSi3O8) and anorthite-rich (CaAl2Si2O8) rocks (52). The principal spectral absorptions observed occur between 100-1250 cm-1 and are associated with the antisymmetric Si-O stretching modes of the tetrahedral [SiO4]2- groups in the materials. The weaker absorptions appearing between 350-700 cm-1 can be attributed to Si-O-Si bending motions. Most of these spectral features persist at high pressures. The infrared spectra of such materials may well be useful in identifying highly shocked calcic plagioclase feldspars. Similarly, some of the same researchers (53) have examined the emission and reflectance infrared spectra in the 350-1400 cm-1 region of shocked and unshocked samples of orthopyroxenite and anorthite; the shocked materials were subjected to pressures in the 170-630 kbar range. These spectra are considered to be quite comparable to the spectra obtained for the meteor impact craters on Mars by remote

10

sensing techniques using the Global Surveyor. The spectra of the shocked feldspars display systematic changes with increasing pressure as the result of the break-up of the linked [SiO4]2- tetrahedra, viz., the weak bands in the 500-650 cm-1 region and a strong band at 1115 cm-1 disappear. The pyroxene spectra exhibited far fewer changes, even at 630 kbar. CONCLUSIONS Pressure-tuning vibrational spectroscopic studies of clay minerals is still a relatively unexplored field that is ripe for further investigation, especially since the behaviour of these minerals under pressure plays such an important role in the Earth’s mantle and, possibly, in the form of intercalation compounds, even in prebiotic chemistry. Moreover, it also seems clear that studies of the vibrational spectra of experimentally shocked minerals will prove to be of immense help in increasing our knowledge of the mineralogy of the Martian and other planetary surfaces. Surprisingly, however, there do not yet appear to be any high-pressure vibrational spectroscopic studies of clay intercalation compounds. Park et al. (54) have synthesized a layered double-hydroxide intercalation compound in order to measure the uniaxial stress that the host layers exert on the photoluminescence of the Sm3+ ion, which is known to be sensitive to the deformation of intercalants. Interestingly, in the absence of external pressure, the uniaxial stress exerted on the Sm3+ ion by the host layers was shown to be ~139 kbar at room temperature. It is hoped that this brief overview of the effect of high pressure on the vibrational spectra of clays and related minerals will prompt an examination of the high-pressure infrared and Raman spectra of interlamellar compounds, such as kaolinite-(CH3)2SO (55). The oxygen atoms of the sulfonyl groups of the (CH3)2SO molecules in this intercalated material are H-bonded to the OH groups of the gibbsite-sheet surface of kaolinite, thereby affording the possibility of investigating the effect of high pressures on the H-bonding. In addition, high-pressure vibrational studies on smectite sorption of nitroaromatic compounds might well throw more light on the fate and transport of such soil contaminants in the environment (56). Furthermore, lawsonite [CaAl2Si2O7(OH)2

.H2O] is a dense, hydrous silicate that is of particular interest to geologists in view of its possible role as a water reservoir at high pressures in the Earth’s mantle. The structure is based on a framework of edge-sharing Al(O,OH) octahedra bridged by Si2O7 units, with the Ca2+ ions being octahedrally coordinated to six nearest neighbor oxygen atoms. The Ca2+ ions are located together with the water molecules in the large cavities within the framework. Pawley has identified several pathways for the breakdown of lawsonite and he concluded that this mineral is stable up to pressures of ~135 kbar (57). It would be worthwhile investigating the infrared and Raman spectra of this material under high pressures in an effort to ascertain whether or not it does release water under compression at ~135 kbar.

11

REFERENCES

1. Wypych, F. and Satyanarayana, K.G. (eds.) (2004) Clay Surfaces: Fundamentals and Applications, Elsevier Academic Press, New York, N.Y., U.S.A.; Vol. l. in a series edited by A. Hubbard entitled Interface Science and Technology.

2. Solin, S.A. (1997) Clays and clay intercalation compounds: properties and physical phenomena, Annu. Rev. Mater. Sci., 27: 89-115.

3. Bolfan-Casanova, N. (2005) Water in the Earth’s mantle. Mineral. Mag., 69: 229-258 and references therein.

4. These classifications are based in part on those given in by Clay Mineral Group at the following website: http://www.galleries.com/minerals/silicate/clays/htm.

5. Poppe, L.J., Paskevich, V.F., Hathaway, J.C., and Blackwood, D.S. (2001) U.S.

Geological Survey Opem-file Report 01-04: A laboratory manual for X-ray powder diffraction.., U.S. Geological Survey, Coastal and Marine Geology Program, Wood Hole Field Centre, Woods Hole, Massachusetts, U.S.A; this report is available at the following website: http://pubs.usgs.gov/of/2001/of01-041/index.htm.

6. Klug, D.D. and Whalley, E. (1983) Nitrite and nitrate ions as infrared pressure

gauges for diamond anvils. Rev. Sci. Intrum., 54: 1205-1208. 7. Mao, H.K., Xu, J., and Bell, P.M. (1986) Calibration of the ruby pressure gauge to

800 kbar under quasi-hydrostatic conditions. J. Geophys. Res., 91: 4673-4676. 8. The two common units of pressure are kilobar (kbar) and gigapascal (GPa), where

1 GPa = 10 kbar. 9. Lemos, V. and Camargo, F. (1990) Effects of pressure on the- Raman spectra of a

1:4 methanol:ethanol mixture. J. Raman Spectrosc., 21: 123-126 and references therein.

10. Ferraro, J.R. (1984) Vibrational Spectroscopy at High External Pressures: The

Diamond Anvil Cell, Academic Press, Inc., New York, N.Y., U.S.A. 11. This piston device can be used for sample volumes of up to 88 mm3 depending on

whether tungsten carbide or diamond anvils are employed and can operate at pressures between ambient and 250 kbar. See the following website: http://www.isis.rl.ac.uk/crystallography/index.htm?content_area=/crystallography/hipr/hipr.htm&side_nav=/crystallography/crystallographySideNav.htm&.

12. Lager, G.A., Marshall, W.G., Liu, Z., and Downs, R.T. (2005) Re-examination of

12

the hydrogarnet structure at high pressure using neutron powder diffraction and infrared spectroscopy. Am. Mineral., 90: 639-644.

13. Liu, H., Hu, J., Liu, Z., Shu, J., Mao, H.-K., and Chen, J. (2004) Phase transition

and compression behaviour of gibbsite under high pressure. Phys. Chem. Mineral, 31: 240-246.

14. Shieh, S.R. and Duffy, T.S. (2002) Raman spectroscopy of Co(OH)2 at high pressures: implications for amorphization and hydrogen repulsion. Phys. Rev. (B), 66,:134301-13407.

15. Huang, E., Lin, J.F., Xu, J. A., Chen, R.J., and Yamanaka, T. (1996) High-pressure

phase transition in Al(OH)3: Raman and X-ray observations. Geophys. Res. Lett., 23: 3083-3086.

16. Johnston, C.T., Wang, S.-L., Bish, D. L., Dera, P., Agnew, S.F., and Kenney III,

J.W. (2002) Novel pressure-induced phase transformations in hydrous layered material. Geophys. Res. Lett., 29: 171-174 (2002).

17. Duffy, T.S., Meade, C., Fei, Y., Mao, H.-K.,and Hemley, R.J. (1995) High-

pressure phase transition in brucite, Mg(OH)2. Am. Mineral., 80: 222-230. 18. Kawai, N.T., Gilson, D.F.R., and Butler, I.S. (1990) Variable- temperature and –

pressure spectroscopic studies of norbornadiene. J. Phys. Chem., 94: 5729-5734. 19. La Iglesia A. and Aznar, A.J. (1996) Crystallinity variations in kaolinite induced

by grinding and pressure treatments. J. Mater. Sci., 31: 4671-4677. 20. Carmody, O., Kristof, J., Frost, R.L. Mako, E., Kloprogge, J. T., and Kokot, S.

(2005) A spectroscopic study of mechanochemically activated kaolinite with the aid of chemometrics. J. Colloid Interface Sci., 287: 43-56. ].

21. Dera, P., Prewitt, C.T., Japel, S., Bish, D.L., and Johnston, C.T. (2003) Pressure-

controlled polytypism in hydrous layered materials. Am. Mineral., 88: 1428-1435. 22. Benco, L., Tunega, D., Hafner, J., and Lischka, H. (2001) Upper limit of the O-

H…O hydrogen bond. Ab initio study of the kaolinite structure. J. Phys. Chem. B, 105: 10812-10817 and references therein.

23. Balan, E., Saitta, A.M., Mauri, F., and Calas, G. (2005) First-principles modeling

of the infrared spectrum of kaolinite. Am. Mineral., 86,: 1321-1330 and references therein.

24. Sato, H., Ono, K., Johnston, C.T., and Yamagishi, A. (2004) First-principle study

on polytype structures of 1:1 dioctahedral phyllosilicates. Am. Mineral., 89: 1581-1585.

25. Sato, H., Ono, K., Johnston, C.T., and Yamagishi, A. (2005) First-principles

13

studies on the elastic constants of a 1:1 layered kaolonite mineral. Am. Mineral., 90: 1824-1826.

26. Tosoni, S., Doll, K., and Ugliengo, P. (2006) Hydrogen bond in layered materials:

structural and vibrational properties of kaolinite by periodic B3LYP approach. Chem. Mater., 18: 2135-2143.

27. Neder, R.B., Burghammer, M., Grasl, T., Schulz, H., Bram, A., and Fiedler, S.

(1999) Refinement of the kaolinite structure from single-crystal synchrotron data. Clays Clay Mineral., 47: 487-494.

28. Franco, F., Pérez-Maqueda. L.A., and Pérez-Rodrìguez, J.L. (2004) The effect of

ultra-sound on the particle size and structural disorder of a well-ordered kaolinite. J. Colloid Interface Sci., 274: 107-117 and references therein.

29. Meng, M., Stievano, L., and Lambert, J.F. (2004) Adsorption and thermal

condensation mechanisms of amino acids on oxide supports: I. Glycine on silica. Langmuir, 20: 914-923.

30. Wada, N. and Kamitakahara, W.A. (1991) Inelastic neutron- and Raman-

scattering of muscovite and vermiculite layered silicates. Phys. Rev. B, 43: 2391-2397.

31. Holtz, M., Solin, S.A., and Pinnavaia, T.J. (1993) Effect of pressure on the Raman

vibrational modes of layered aluminosilicate compounds. Phys. Rev. B, 48: 13312-13317.

32. Rinnert, E., Carteret, C., Humbert, B., Fragneto-Cusani, G., Ramsay, J.D.F.,

Delville, A., Robert, J.-R., Bihannic, I., Pelletier, M., and Mochot, L.J. (2005) Hydration of a synthetic clay with tetrahedral charges: a multidisciplinary experirmental and numerical study. J. Phys. Chem. B., 109: 23745-23759.

33. Catti, M., Ferraris, G., Hull, S., and Pavese, A. (1994) Powder neutron diffraction

study of 2MI muscovite at room temperature and at 2 GPa. Eur. J. Mineral., 6: 171-178.

34. Faust, J. and Knittle, E. (1994) The equation of state, amporphisation and high

pressure phase diagram of muscovite. J. Geophys. Res., 99: 19785-19792. 35. Comodi, P. and Zanazzi, P.F. (1995) High pressure structural study of muscovite.

Phys. & Chem. Mineral., 22: 170-177. 36. Comodi, P. and Zanazzi, P.F. (1997) Pressure dependence of structural

parameters of paragonite. Phys. & Chem. Mineral., 24: 174-180. 37. Hazen, R.M. and Finger, L.W. (1978) The crystal structures and compressibilities

14

of layer minerals at high pressures. II. Phlogopite and chlorite. Am. Mineral., 63: 293-296.

38. Moissanite-anvil cells permit sample volumes 1000 times larger than those

allowed by diamond-anvil cells and, in time, because of its hardness, high heat conductance, low thermal expansion, good thermal stability and low price, this material may become common in high-temperature and -pressure experimental measurements.

39. Xu, J.-A.and Mao, H.-K. (2000) Moissanite: a window for high-pressure

experiments. Science, 290: 783-785. 40. Liu, Z., Xu, J., Scott, H.P., Williams, Q., Mao, H.-K., and Hemley, R.J. (2004)

Moissanite (SiC) as windows and anvils for high-pressure infrared spectroscopy. Rev. Sci. Instrum., 75: 5026-5029.

41. Duan, T.Y., Sun, O., and Zheng, H.F. (2005) Applications of moissanite anvil cell

for Raman spectroscopy under high-temperature and high-pressure. Guang Pu Xue Yu Guang Pu Fen Xi, 25: 902-905. (In Chinese.)

42. Kleppe, A.K., Jephcoat, A.P., and Welch, M.D. (2003) The effect of pressure

upon hydrogen bonding in chlorite: a Raman spectroscopic study of clinochlore to 26.5 GPa. Am. Mineral., 88: 567-573.

43. McKeown, D.A., Post, J.E., and Etz, E.S. (2002) Vibrational analysis of

palygorskite and sepiolite. Clays Clay Minerals., 50: 667-680. 44. Madejova, J. and Komadel, P. (2001) Baseline studies of the Clay Minerals

Society source clays: infrared methods. Clays Clay Minerals, 49: 410-432. 45. Akyuz S., Akyuz, T. and Davies, J.E.D. (1999) FT-IR and FT-Raman

spectroscopic study of surface species of 2-aminopyridine adsorbed by sepiolite from Anatolia. J. . Struct. Chem., 40: 796-801 and references therein.

46. Kagi, H., Nagai, T., Nagai, J.S., Loveday, J.S., Wada, C., and Parise, J.B. (2003)

Pressure-induced phase transformation of kalcinite (KHCO3) at 2.8 GPa and local structural changes around hydrogen atoms. Am. Mineral., 88: 1446-1451.

47. Williams, Q., Jeanloz, R., and McMillan, P. (1987) The vibrational spectrum of

MgSiO3 perovskite. Zero pressure Raman and mid-infrared to 27 GPa. ,J. Geophys. Res., 92: 8116-8128.

48. Hofmeister, A.M. and Mao, H.-K. (2002) Redefinition of the mode Grüneisen

parameter for polyatomic substances and thermodynamic implications. Proc. Nat. Acad. Sci., 99: 559-564.

49. Koch-Müller, M., Hofmeister, A.M., Fei, U., and Liu, Z. (2002) High-pressure IR-

15

spectra and thermodynamic properties of chloritoid. Am. Mineral., 87: 609-622. 50. Hofmeister, A.M., Cynn, H., Burnley, P.C., and Meade, C. (1999) Vibrational

spectra of dense, hydrous magnesium silicates at high pressure. Importance of the hydrogen bond angle. Am. Mineral., 84: 454-464.

51. Kleppe, A.K., Jephcoat, A.P., and Smyth, J.R. (2002) Raman spectroscopic study

of hydrous γ-Mg2SiO4 at 56.5 GPa. Phys. Chem. Minerals, 29: 473-476. 52. Johnson, J.R., Hörz, F., and Staid, M.I. (2003) Thermal infrared spectroscopy and

modelling of experimentally shocked plagioclase feldspars. Am.. Mineral., 88: 1575-1582.

53. Johnson, J.R., Hörz, F., Lucey, P.G., and Christensen, P.R. (2002) Thermal

infrared spectroscopy of experimentally shocked anorthosite and pyroxenite; implications for remote sensing of Mars. J. Geophys. Res. E: Planets, 107: 3-13.

54. Park, R.-R., Park, T.Y., Kim, H., and Min, P. (2002) High pressure measurement

of the uniaxial stress of host layers on intercalants and staging transformation of intercalation compounds. J. Phys.: Condens. Matter, 14: 11687-11700.

55. Johnston, C.T., Sposito, G., Bocian, D.F., and Birge, R.R. (1984) A vibrational

spectroscopic study of the interlamellar kaolinite-dimethyl sulfoxide complex. J. Phys. Chem., 88: 5959-5964.

56. Johnston, C.T., Sheng, G., Teppen, B.J., Boyd, S.A., and de Oliveira, M. (2002)

Spectroscopic study of dinitrophenol herbicide sorption on smectite. Environ. Sci. Technol., 36: 5067-5074.

57. Pawley, A.R. (1978) Crystal refinement of lawsonite. Am. Mineral., 63: 311-315.

16

LIST OF FIGURES

Figure 1. Kaolinite clay structure (adapted from ref. 5). Figure 2. SEM pictures of (a) Australian kaolinite books showing the layer

structure and (b) Hungarian kaolinite single crystal. Figure 3. Typical montmorillonite/smectite clay structure (adapted from ref. 5). Figure 4. SEM picture of montmorillite. Figure 5. Typical illite (clay-mica) clay structure (adapted from ref. 5). Figure 6. Typical chlorite clay structure (adapted from ref. 5). Figure 7. Sepiolite structure (adapted from ref. 5). Figure 8. Schematic drawing of the two diamonds in a DAC. (A) Diamond

supports, (B) two diamond anvils with working surfaces (culets) of typically 0.6 mm and (C) Inconel gasket. (Reproduced with permission from High Pressure Diamond Optics, Tucson, Arizona, U.S.A.)

Figure 9. Schematic layout of a P-series DAC. (Reproduced with permission from

High Pressure Diamond Optics, Tucson, Arizona, U.S.A.) Figure 10. Photograph of a P-series DAC that is capable of reaching pressures of

~70 kbar. Length, 12.5 cm; base length, 2.5 cm; width, 7.5 cm; total height, 9 cm; weight, ~1 kg. (Reproduced with permission from High Pressure Diamond Optics, Tucson, Arizona, U.S.A. For more information, see the following website: http://www.hpdo.com/.)

Figure 11.. Representative high-pressure infrared spectra of gibbsite (Reproduced

with permission from ref. 13.) Figure 12. Pressure dependences of the OH stretching and deformation bands of

gibbsite. (Reproduced with permission from ref. 13.)

Figure 13.Typical Raman spectra of brucite in the lattice and internal mode regions under quasi-hydrostatic conditions up to ~100 kbar. (Reproduced with permission from ref. 17.)

Figure 14. Pressure dependences for the lattice and O-H stretching modes of

brucite for pressures up to ~300 kbar under both non-hydrostatic and hydrostatic conditions. (Reproduced with permission from ref. 17.)

Figure 15. High-pressure, single-crystal Raman spectra of dickite from 0.38 to 6.5 GPa. (Reproduced with permission from ref. 16.)

Figure 16. Proposed assignments for the ν(OH) vibrations of dickite. (Reproduced

with permission from ref. 16.)

17

Figure 1.

18

Figure 2 (a).

19

Figure 2 (b).

20

Figure 3.

21

Figure 4.

22

Figure 5.

23

Figure 6.

24

Figure 7.

25

Figure 8.

26

Figure 9.

27

Figure 10.

28

Figure 11.

29

Figure 12.

30

Figure 13.

31

Figure 14.

32

Figure 15.

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

33

Figure 16.

QuickTime™ and aTIFF (LZW) decompressor

are needed to see this picture.

34

Table 1. Band positions (cm-1) at ambient pressure (pressure dependences, cm-1/kbar) for the dioctahedral and trioctahedral 2:1 clays. (Adapted from ref. 2, with permission.) Clays ν(torsion) ν(OH)

Dioctahedral clays Margarite 115 (0.07) 3625 (-0.19) Muscovite 100 (0.86) 3616 (-0.49) Trioctahedral clays Talc 111 (0.86) 3672 (0.11) Biotite 3666 (1.3) Phlogopite 104 (0.14) 3700