Mechanical Properties of the Sodium Montmorillonite Interlayer

Biomacromolecules 2005, 6, 3276-3282

3276

Mechanical Properties of the Sodium Montmorillonite Interlayer Intercalated with Amino Acids Dinesh R. Katti,* Pijush Ghosh, Steven Schmidt, and Kalpana S. Katti Department of Civil Engineering and Construction, North Dakota State University, Fargo, North Dakota 58105 Received May 9, 2005; Revised Manuscript Received June 13, 2005

Nanosized montmorillonite clay dispersed in small amounts in polymer results in polymer nanocomposites having superior engineering properties compared to those of the native polymer. These nanoinclusions are created by treating clay with an organic modifier which makes clay organophilic and results in intercalation or exfoliation of the montmorillonite. The modifiers used are usually long carbon chains with alkylammonium or alkylphosphonium cations. In this work, we have investigated the use of some alternative molecules which can act as modifiers for clay composites using clay for reinforcing a matrix of biopeptides or proteins. Such composites have potential applications in the fields of biomedical engineering and pharmaceutical science. In this work, the amino acids arginine and lysine are used as modifiers. The intercalation and mechanical behavior of the interlayer spacing with these amino acids as inclusions under compression and tension are studied using molecular dynamics simulations. Significant differences in the responses are observed. This work also provides an insight into the orientation and interaction of amino acids in the interlayer under different stress paths. Introduction Montmorillonite belongs to the smectite group of minerals, which are commonly called swelling clays. Montmorillonite is comprised of stacked clay sheets. The interlayer spacing between two of its layers has oxygen atoms in the clay surface facing the interlayer and carrying partial negative charges. Recently, montmorillonite-based swelling clays have been extensively used in the design of polymer-clay nanocomposites (PCN). In PCN, montmorillonite is intercalated with an organic modifier and dispersed in a polymer matrix to form a composite material with significantly improved engineering properties than those of the parent polymer forming the matrix. Specifically, the elastic modulus,1-5 tensile strength, and elongation properties,6-8 thermal resistance and flammability,6-11 are enhanced. In addition, the interlayer in montmorillonite can act as a potential site of polymerization. In the preparation of polystyrene-clay nanocomposites and polypropylene-clay nanocomposites, the interlayer is used as a site for in situ polymerization.6 The degree and nature of polymerization controls the d-spacing of the clay crystals. Higher d-spacing causes more polymer to enter into the interlayer, thus increasing the interaction between the matrix phase and the dispersed phase. In nanocomposites, the interfacial properties play an important role in determining the overall properties of the composites. Besides this, another factor which is important for the enhanced properties of a nanocomposite is the miscibility of the components involved in it. In the synthesis of PCN, the clay is treated with an organic modifier before it is mixed with the polymer. This modifica* To whom all correspondences should be addressed. E-mail: [email protected].

tion causes an increase in the d-spacing,12 which facilitates the entry of the polymer into the clay gallery and, thus, the formation of the nanoinclusion. The modifier also turns the clay organophilic, thus enhancing the miscibility with the polymer. The modifiers used are generally long carbon chain compounds with an alkylammonium or alkylphosphonium cation. Some of the commonly used organic modifiers are 12-aminolauric acid, benzidine,13 aromatic amine,14 dodecyldiamine, etc. In this work we are investigating modifiers which can potentially enhance the miscibility of clay and protein, when the latter is used as the matrix in a nanocomposite. Amino acids or biopeptides, which are building units of proteins, if used as a modifier can be expected to be compatible with a protein/biopeptide matrix. Positively charged amino acids lysine and arginine have a similarity in chemical structure with conventional modifiers with alkylammonium cations. In this work we have used the positively charged amino acids lysine and arginine as modifiers to montmorillonite. The interlayer spacing obtained through a molecular dynamics study with the above amino acids as inclusion can provide a very good estimate of the potential of application of these materials as modifiers The mechanical properties of PCNs are influenced by the mechanical response of the clay interlayer. Thus, the study of the mechanical response of clay with different molecules in the interlayer is very important. The understanding of the behavior of molecules, their spatial arrangements, orientations, deformation, etc. can give a good insight into the interfacial properties and thus provides guidelines for tailoring the properties of a nanocomposite. In this work, the mechanical properties of the montmorillonite interlayer are studied through molecular dynamics

10.1021/bm0503219 CCC: $30.25 © 2005 American Chemical Society Published on Web 08/18/2005

Montmorillonite Intercalated with Amino Acids

Biomacromolecules, Vol. 6, No. 6, 2005 3277

Figure 1. Schematic representation of an amino acid.

simulations with amino acids as inclusion molecules. The positively charged amino acids lysine and arginine are used. These positively charged amino acids are assumed to make complete substitution of the Na+ cations present in the interlayer of sodium montmorillonite (Na-MMT). Stresses of gradually increasing magnitude are applied on the clay segments. The evaluation of the stress-displacement characteristics of the interlayer, the arrangements of the amino acids, and the conformational changes of the side chains (functional groups) of the amino acids are the main focus of this work. Comparative study of the mechanical response of the montmorillonite interlayer with the amino acid is made with that of the interlayer without inclusion molecules (dry montmorillonite). Study of adsorption of biomacromolecules in the form of amino acids and protein in clay has been done in the past;15-20 however, not many attempts have been made to study the behavior of the inclusion molecules inside the clay gallery under tensile and compressive stress paths. In summary, this work deals with the application of positively charged amino acids as organic modifiers for montmorillonite prior to the synthesis of polymer-clay nanocomposites. The focus of this work is studying the mechanics of the interlayer with amino acids as inclusion molecules and commenting on the potential use of amino acids as modifiers. The understanding of the mechanisms of the interaction of biopeptides within the interlayer may also help in designing sensors, where biopeptides with specific functional groups are to be inserted into the interlayer. Model Construction Amino acids are the building units of proteins. These molecules contain one amine (NH2) group and one carboxyl acid group (COOH). The schematic representation of amino acids is shown in Figure 1. The chemical structure of the functional group (R) attached to the R-carbon is often referred to as the side chain. The sodium cations present inside the interlayer of Na-MMT makes the clay layer electronically neutral. In this study, all the cations of the interlayer are assumed to be substituted by positively charged amino acids of same charge as Na+, thus maintaining the overall charge neutrality of the system. Lysine and arginine are the two positively charged amino acids used here. The functional group (R) attached to lysine and arginine are -C4H11N and

Figure 2. Clay-amino acid model before minimization.

-C4H11N3, respectively. The structural coordinates of the amino acids are obtained from the fragment library of the builder module in Insight II.21 The length of lysine and arginine in the direction of extension of the functional group or side chain are 9.50 and 10 Å, respectively. The Na-MMT used in this work has chemical formula of NaSi16(Al6FeMg)O40(OH)8,22 and it has a tetrahedraloctahedral-tetrahedral (t-o-t) structure. The dimension of the unit cell used in this simulation is a ) 5.28 Å, b ) 9.14 Å, and c ) 6.56 Å. The clay structure is made up of two layers, each having 4 unit cells in the X direction and 2 unit cells in the Y direction. In the Z direction they are 1 unit cell thick. The coordinates of this model are obtained from the coordinates of pyrophyllite given by Skipper et al.23 The montmorillonite structure is obtained from pyrophyllite by appropriate isomorphous substitution. In this work each Mg2+ and Fe2+ replaces one of every four Al3+. The charge on Mg2+ is taken as 0.68,24 which is one less than the value of 1.68 for Fe2+ and Al3+. The orientation (often called rotamers) of the functional groups or side chains of the amino acids present on the surface of the protein depends to a large extent on the nature of the molecule present at the closest vicinity. The orientation of lysine and arginine side chains in the presence of clay inorganic is not found in the literature. Therefore in this study, as an initial conformation, the side chains of lysine and arginine are placed in the direction perpendicular to the clay surface. To accommodate this geometry of amino acids inside the interlayer, the initial spacing of the interlayer is assumed as 16.00 Å. The two clay layers with this interlayer spacing and amino acids in the interlayer constitute the total model. The overall dimension of the model stands as 21.12 Å × 18.28 Å × 28.50 Å. The layers of the clay are marked as A and B in Figure 2, and the inclusion molecule in the interlayer in the form of the amino acid is named as C. Three types of models are used in this work. In the first case the interlayer contains two lysines and two arginines, as shown in Figure 2. The second model contains all lysine. To compare the effect of amino acids on the mechanical properties of the interlayer, dry montmorillonite without any inclusion molecules is constructed as our third model. To

3278

Biomacromolecules, Vol. 6, No. 6, 2005

Katti et al.

build a 4 × 2 clay segment, the unit cells are patched together using PSFGEN, a module in NAMD.25 Simulation Details NAMD and VMD26 are used for all our molecular dynamics simulations. The structure is geometrically optimized through energy minimization before performing simulations. The minimization is performed stepwise. First, one of the clay layers, A, is minimized keeping the other layer, B, and the amino acids, C, fixed. Subsequently, the second clay layer, B, is minimized, in a similar fashion, followed by minimizations of amino acids C alone. Finally, the whole structure is minimized. The stepwise minimization with one segment at a time prevents the structure from attaining a metastable state. All minimizations are conducted at 0 K and under vacuum. After the minimization, the temperature and pressure of the system is raised to 300 K and 1 atm, respectively. In all the minimizations and molecular dynamics calculations, the CHARMm27 force field is used. The parameters for the Na-MMT for this force field were found in our earlier work.28,29 A time step of 0.5 fs (10-15) is used in all the simulations. All the simulations conducted with applied force are run for a total time of 200 ps (10-12) or 400 000 steps. The energy versus time plot shows that this number of steps is sufficient for the energy to attain equilibrium. Some simulations were run for 400 ps to verify that the 200 ps time is sufficient for this system of clay and amino acids. The energy plot with the 400 ps runtime shows no difference with that of the 200 ps runtime. Moreover, we did not observe any change in conformation between 200 and 400 ps. The periodic boundary condition (PBC) is used in all the simulations. The particle mesh Ewald (PME) technique is used to calculate the electrostatic interaction between atom pairs. The Lennard-Jones potential is used to calculate the van der Waals interaction. The van der Waals cutoff distance is taken as 13.00 Å. This distance is larger than the distance between the middle of the amino acid segment and the octahedral layer of the clay segment. From the symmetry of the model, this captures the van der Waals interaction between the atoms of the amino acids and the clay layers completely. Theoretically, simulations run under identical conditions should result in the exact same output. In the software used in this study, NAMD, under identical conditions, without the use of the Langevin method30,31 the calculations are deterministic. However, in our work, we have used Langevin dynamics30,31 to control temperature, where the correct temperature is maintained by the addition of friction and random forces. In this method, while assigning velocity to each atom, a random number is generated for every simulation. This methodology can make the result slightly nondeterministic. To make sure that the effects of these errors are minimized, each simulation at the same magnitude of stress is run three times. Simulations are run under 0 pN of force to get a stable starting conformation. Forces of gradually increasing magnitude as 10, 50, 100, and 150 pN are applied on the surface oxygen atoms of segments A and B as marked in Figure 2.

Figure 3. Clay-amino acid model after minimization.

Considering the plan area of the segments, the equivalent stresses are calculated as 0.001, 0.50, 1.00, and 1.50 GPa, respectively. The coordinates of the same atoms are used to determine the change in thickness of the interlayer spacing and clay layers on the application of the stresses. Both compressive and tensile stress paths are used in this study. Results and Discussion After minimization of the clay structure containing two lysine and two arginine molecules in the interlayer, the interlayer spacing is found to decrease from 16.00 to 8.00 Å. Initially, the functional groups of the amino acids were placed perpendicular to the clay surface. However, after minimization the functional groups are found to be oriented in the direction parallel to the clay surface. The minimized structure is shown in Figure 3. The interlayer spacing for dry montmorillonite is 4.12 Å. Thus, introduction of the amino acid shows an increased interlayer spacing. The variation of interlayer spacing of montmorillonite with lysine and arginine as the inclusion molecule (Figure 1) under applied compressive stress is shown in Figure 4. Initially, at 0 GPa applied stress the interlayer spacing is 8.00 Å, and under an applied stress of 1.50 GPa it reduces to 7.10 Å. The interlayer strain is plotted against the stress applied in Figure 5, and it shows a nonlinear behavior up to the maximum stress of 1.50 GPa. At this stress level the strain attained is 0.11. The corresponding characteristics of the interlayer under the tensile stress path are shown in Figures 6 and 7. Under the maximum tensile stress of 1.50 GPa, the strain observed is only 0.02, which is appreciably less than its corresponding value in compression. The clay-amino acid, clay-clay, and amino acid-amino acid interactions involved introduce nonlinearity in the interlayer response characteristics. The plots for all the above figures are obtained by adding a trend line to the data points. The slope of these trend lines provides a good estimate of the stiffness of the interlayer. To determine the slope of these



Figure 4. Interlayer spacing-stress characteristics of clay in compression.

Figure 7. Stress-strain characteristics of the interlayer in tension. Table 1. Stress-Strain Values Used to Determine the Slope of the Trend Line in (a) Compression and (b) Tension, Respectively stress (GPa)

strain

0.00 0.40 0.80 1.20

0.00 0.0344 0.079 0.096

0.00 0.40 0.80 1.20

0.00 0.0035 0.008 0.0135

Slope (GPa)

average slope (GPa)

(a) Compression 11.628 8.969 23.529

15

(b) Tension

Figure 5. Stress-strain characteristics of the interlayer in compression.

114.285 88.89 72.727

with that of the solid. The relationship can be represented as σv ) civ iv

Figure 6. Interlayer spacing-stress characteristics of clay in tension.

trend lines we have divided the range of stress into three parts, as 0-0.40, 0.40-0.80, and 0.80-1.2 GPa. Three straight line segments are formed by joining the extreme points on the trend lines corresponding to these stress ranges. The average slope of these three lines is assumed to represent the slope of the nonlinear trend line. The values of stress and strain used to calculate the stiffness of the interlayer for compression and tension are given in Table 1, parts a and b, respectively. The slope of these lines can be used to draw an analogy between the stress-strain characteristics of this interlayer

92

(1)

where σv, civ, and iv are the applied stress, interlayer modulus, and interlayer strain along the swelling axis. For the compressive stress path civ ) 15 GPa and for the tensile path civ ) 92 GPa. The response of interlayer is expected to change with change in characteristics of inclusion molecules (biopeptides or proteins) inside the interlayer. The unfolding nature of proteins, the number of charged and polar amino acids exposed to the clay surface, the presence of metal ions in proteins, and the solvent around the protein are some of the features which could appreciably affect the interlayer response of clay. In this study, both amino acids used, lysine and arginine, are of unit positive charge and consist of almost the same number of atoms. As a part of this study it is intended to find if the stress-strain characteristics of the interlayer are dependent on the amino acids used. To conduct this study both the arginines are replaced by lysine, thus conserving the charge balance. Simulations are conducted under compression in a similar way as before. The stressstrain characteristics of the interlayer with all four Na+ atoms of original Na-MMT replaced by lysine are shown in Figure 8. From this figure it is found that the stiffness of the interlayer civ ) 13 GPa is almost the same as that obtained earlier (15 GPa) with arginine and lysine in compression.

3280


Figure 8. Stress-strain characteristics of the interlayer with only lysine in compression.

Figure 9. Variation of the average amino acid thickness with stress in compression.

The mechanical response as a result of changing amino acid molecules of the same charge is marginally different. The two main components which influence the overall response of the interlayer are the amino acid thickness and the silica-amino acid interaction zone. Our previous work28,29 indicates that the deformation of clay sheets in this range of stresses is very small and could be considered rigid. The distance between the extreme atoms of the amino acids in the direction of the measurement of the interlayer spacing (perpendicular to the clay surface) is defined in this work as the amino acid thickness. This thickness of the amino acid is affected by different types of movements of amino acids such as in-plane, out-of-plane bending, torsion, rolling, contraction of angles and dihedrals, etc. The region bound by the silica surface of the clay segments and the amino acids is called the interaction zone. These zones are marked as Z1 and Z2 in Figure 2. In obtaining the response of the interaction zones, the average of the two zones Z1 and Z2 is considered. Mainly the van der Waals and electrostatic interaction between the organic amino acids and the inorganic clay silica tetrahedral layer determine the deformation of the interaction zone. The changes in thickness of the amino acid for lysine and arginine under compression and tension are shown in Figures 9 and 10, respectively. But besides outof-plane bending there are other types of movements such as torsional movement, rolling, and contraction in angles and dihedrals which can reduce the amino acid thickness. These

Katti et al.

Figure 10. Variation of the average amino acid thickness with stress in tension.

mechanisms are more predominant in tension than in compression, and as a result, with stress significant variation of the amino acid thickness is observed (Figure 10). It is observed that in compression the amino acid thickness changes almost uniformly and the magnitude of change observed is very small. It appears that in compression, mainly the silica-amino acid interaction zone (Z1 and Z2) contributes to the overall interlayer strain. However, in tension the response is a result of the changes in the amino acid thickness (and thus the interaction zone). Snapshots of the amino acids at different magnitudes of stress in compression and tension are analyzed to understand the deformation mechanisms. Example snapshots of lysine taken at 1.00 GPa of stress are shown in Figure 11. In compression it is observed that in most of the cases the side chains of the amino acids rotate in the X-Y plane (i.e., the plane on which it lies initially). In a few cases the side chains are observed to experience some torsion, but bending of the side chain is not observed. The backbone of the chain is found to remain in the X-Y plane in almost all cases. In tension, however, the side chain backbones are observed to undergo out-of-plane bending. The amino acids are found to be rolling over and tipping out of the X-Y plane. The initial configuration of the amino acids before application of stress is applied to the X-Y plane. The out-of-plane bending results in increased amino acid thicknesses, since the thickness in our work is measured as the distance between the two extreme atoms as defined earlier. In both compression and tension the hydrogen atoms are found to be orienting toward the oxygen atoms of the clay surface. Therefore, the major differences in the thickness of the clay-silica interaction zone are observed due to the outof-plane bending and large dihedral angle changes of the side chains under tension. From the above observations, it can be concluded that in tension the amino acids show more tendency to come closer to the clay layer at the cost of their conformational changes. In tension, possibly the amino acid molecules adjust themselves in a fashion so that the van der Waals attractive forces and the electrostatic attractive forces are maximized, thereby minimizing the total energy and making the overall system most stable. This behavior of amino acids could be a possible explanation to the stiffer response of the interlayer in tension than in compression.



Figure 11. Snapshots of lysine at (a) 0 pN, (b) 100 pN compression, (c) 100 pN tension.

Figure 12. Stress-strain characteristics of the clay interlayer of dry montmorillonite in compression.

Figure 13. Stress-strain characteristics of the clay interlayer of dry montmorillonite in tension.

It is realized that the presence of some inclusions in the interlayer influences its mechanical properties. To get a quantitative comparison of the modulus of the interlayer in the presence and absence of amino acids, simulations were conducted on our third model, which is dry montmorillonite without any inclusion in the interlayer. The stress-strain responses of the interlayer in compression and tension are shown in Figures 12 and 13, respectively. For montmorillonite without any inclusion, the civ values obtained in compression and tension are 25 and 30 GPa, respectively. When the results are compared with those of the montmorillonite with amino acids in the interlayer, which are 13 and 79 GPa, respectively, it is found that the presence of amino acids causes the interlayer to behave weaker in compression and stiffer in tension. The nature of the overall interaction between the clay layer and the amino acids is therefore different in compression and tension. For a given

magnitude of applied external stress for the three cases (1) tension with amino acids, (2) compression with amino acids, (3) tension with inclusion molecules (dry montmorillonite), the additional energy introduced into the system is the same. However, we find that in the case of tension with amino acids, the large conformational changes observed appear to consume a significant portion of this energy compared to the compression case resulting in a smaller change in the interlayer spacing. This can be further seen when the interlayer spacing of the two cases, tension with amino acids and tension without inclusions, for the same stress are compared. Here again we see larger changes in the interlayer spacing for the dry montmorillonite case since all additional energy is consumed overcoming the nonbonded interactions between the clay sheets. When the interlayer spacing of the dry montmorillonite case in compression is compared with that of the clay with amino acids case we see a larger deformation of the interlayer in the latter case because (1) the nonbonded interactions between the clay sheets is smaller because of the larger separation and (2) the energy expended due to conformational changes in the case of compression is small. One possible explanation of the behavior observed in tension is the orientation and the adjustments of the angles and dihedrals of the amino acids as found in the snapshots in Figure 11. The behavior of the amino acids (lysine) is almost similar in the case of both models. But in this Figure 11, snapshots from the montmorillonite model with arginine + lysine are shown. From the above observations it is found that the clay and amino acids show an attraction for each other. In compression the direction of the application of force is the same as the direction of this attractive force and thus results in larger strain. However, in tension, these attractive forces are opposite to the direction of the application of tensile forces hence resulting in a smaller strain. This leads to a higher modulus in tension than in compression. Conclusions In this work we have conducted a steered molecular dynamics study of a new clay nanocomposite system. This new nanocomposite consists of amino acids arginine and lysine intercalated between montmorillonite interlayers. Our simulations indicate that the clay amino acids interlayer is about 3 times stiffer under tension as compared to compression. On the other hand, dry montmorillonite shows similar stiffness under tension and compression. The fundamental mechanism of deformation during tension and compression is intrinsically different. Stress-strain behavior of this clay-

3282


amino acid interlayer is predominantly linear up to 1.50 GPa. On the basis of the result of our simulations, it appears that the use of amino acids as modifiers is favorable for intercalation as well as from a mechanical standpoint. This study is a first step toward the potential use of biomacromolecules as modifiers in clay nanocomposites. Acknowledgment. The authors acknowledge the use of the computational resources at the NDSU Center for High Performance Computing (CHPC) and Biomedical Research Infrastructure Network (BRIN). Author P.G. acknowledges support from ND EPSCOR (Grant No. 8486). The hardware and software support for NAMD at NDSU provided by the NDSU Center for high performance computing and Dr. Gregory Wettstein is acknowledged. References and Notes (1) Pramanik, M.; Srivastava, S. K.; Biswas, K. S.; Bhowmick, A. K. J. Appl. Polym. Sci. 2003, 87, 2216. (2) Maiti, P.; Yamada, K.; Okamoto, M.; Ueda, K.; Okamoto K. Chem. Mater. 2002, 14, 4654. (3) Liu, X.; Wu, Q.; Berglund, L. A.; Lindberg, H.; Fan, J.; Qi, Z. J. Appl. Polym. Sci. 2003, 88, 953. (4) Wu, Y. P.; Jia, Q. X.; Yu, D. S.; Zhang, L. Q. J. Appl. Polym. Sci. 2003, 89, 3855. (5) Zhang, G.; Jiang, C.; Su, C.; Zhang, H. J. Appl. Polym. Sci. 2003, 89, 3159. (6) Wang, D.; Wilkie, C. A. Polym. Degrad. Stab. 2003, 80, 171. (7) Lim, S. T.; Lee, C. H.; Choi, H. J.; Jhon, M. S. J. Polym. Sci, Part B: Polym. Phys. 2003, 41, 2052. (8) Park, H. M.; Lee, W. K.; Park, C. Y.; Cho, W. J.; Ha, C. S. J. Mater. Sci. 2003, 38, 909. (9) Chen, G. X.; Hao, G. J.; Guo, T. Y.; Song, M. D.; Zhang, B. H. J. Mater. Sci. Lett. 2002, 21, 1587. (10) Ma, C. C. M.; Kuo, C. T.; Kuan, H. C.; Chiang, C. L. J. Appl. Polym. Sci. 2003, 88, 1686.

Katti et al. (11) Pramoda, K. P.; Liu, T.; Liu, Z.; He, C.; Sue, H. J. Polym. Degrad. Stab. 2003, 81, 47. (12) Ray, S. S.; Okamoto, M. Prog. Polym. Sci. 2003, 28, 1539. (13) Chen, T. K.; Tien, Y. I.; Wei, K. H. Polymer 2000, 41, 1345. (14) Liang, Z. M.; Yin, J.; Xu, H. J. Polymer 2003, 44, 1391. (15) Friebele, E.; Shimoyama, A.; Ponnamperuma, C. J. Mol. EVol. 1980, 269. (16) Nicola, T. W.; Vickers, P. J.; Mann, S. J. Mater. Chem. 1997, 7 (8), 1623-1629. (17) Violante, A.; de Cristofaro, A.; Rao, M. A.; Gianfreda, L. Clay Miner. 1995, 30, 325-336. (18) Yuan, Q.; Wei, N.; Wang, Z.; Wang, G.; Duan, X. Clays Clay Miner. 2004, 52 (1), 40-46. (19) Yu, C. H.; Newton, S. Q.; Miller, D. M.; Teppen, B. J.; Scha¨fer, L. Struct. Chem. 2001, 12, 393-398. (20) Muller, M.; Reiho, T.; Lunkwitz, K. Sci. Eng. 2001, 84, 22. (21) Insight II; Accelrys Inc., 2000. (22) VanOlphen, H., Fritpiat, J. J., Eds. Data Handbook for Clay Materials and other Nonmetallic Minerals; Pergamon Press: New York, 1979. (23) Skipper, N. T.; Chang, F.-R.; Sposito, G. Clays Clay Miner. 1995, 43 (3), 294. (24) Teppen, B. J.; Rasmussen, K.; Bertsch, P. M.; Miller, D. M.; Schafer, L. J. Phys. Chem. B 1997, 101, 1579. (25) Kale´, L.; Skeel, R.; Bhandarkar, M.; Brunner, R.; Gursoy, A.; Krawetz, N.; Phillips, J.; Shinozaki, A.; Varadarajan, K.; Schulten, K. J. Comput. Phys. 1999, 151, 283 http://www.ks.uiuc.edu/Research/ namd/. (26) Humphrey, W.; Dalke, A.; Schulten, K. J. Mol. Graphics 1996, 14 (1), 33 http://www.ks.uiuc.edu/Research/vmd/. (27) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J.; Swaminathan, S.; Karplus, M. J. Comput. Chem. 1983, 4, 187-217. (28) Katti, D. R.; Schmidt, S.; Ghosh, P.; Katti, K. S. Clays Clay Miner. 2005, 53 (2), 171-178. (29) Schmidt, S.; Katti, D. R.; Ghosh, P.; Katti, K. S. Langmuir, in press. (30) Feller, S. E.; Zhang, Y.; Pastor, R. W.; Brooks, B. R. J. Chem. Phys. 1995, 103, 4613-4621. (31) Martyna, G. J.; Tobias, D. J.; Klein, M. L. J. Chem. Phys. 1994, 101, 4177-4187.

BM0503219

Mechanical Properties of the Sodium Montmorillonite Interlayer

Recommend Documents