Optical Second-Harmonic Generation Study of ... - ACS Publications

We have examined the properties of exfoliated and restacked MoS2-malachite green (MG) inclusion compounds to provide insight into the MG-MoS2 ...
0 downloads 0 Views 265KB Size
Langmuir 2006, 22, 8995-9001

8995

Optical Second-Harmonic Generation Study of Exfoliated MoS2-Malachite Green Inclusion Materials Greg Cetnarowski and Gary W. Leach* Department of Chemistry, Simon Fraser UniVersity, Burnaby, BC V5A 1S6 Canada ReceiVed February 17, 2006. In Final Form: August 1, 2006 We have examined the properties of exfoliated and restacked MoS2-malachite green (MG) inclusion compounds to provide insight into the MG-MoS2 interactions that characterize these materials. The results of X-ray diffraction experiments indicate that MG included into the restacked structure adopts a flat orientation approximately parallel to the MoS2 sheets. Second-harmonic generation experiments conducted on the exfoliated and restacked materials provide information regarding the averaged orientation of the MG. At low MG coverage, our results support the X-ray diffraction findings, and yield large averaged orientation angles, consistent with a flat orientation of MG between the MoS2 layers. However, as the MG coverage is increased, the SHG results indicate averaged MG orientations that are much more upright, consistent with the expulsion of excess MG from the layers to the outside of the restacked crystallites. Together with X-ray diffraction and adsorption isotherm data, our SHG results provide a model for the exfoliation, adsorption, and subsequent restacking of these MG-based inclusion materials and demonstrate the utility of nonlinear optical techniques as probes of these interesting layered structures.

Introduction Metal dichalcogenides comprise a class of layered materials that possess diverse structural and electronic properties.1,2 Their physical properties span the range of metal, semiconductor, insulator, and superconductor, depending on the nature of the metal and chalcogen atoms. An important feature of these materials is that they crystallize in a quasi two-dimensional layered structure which imparts substantial anisotropy to many of their properties. In addition, this layered structure facilitates the insertion (intercalation) of foreign atoms and molecules into these materials, thus allowing a convenient method for altering the structure and electronic properties. Molybdenum disulfide (MoS2) is one of the most stable and versatile members of this family of layered materials. Supported MoS2 is widely used in the petroleum industry in a variety of formulations as a catalyst for hydrodesulfurization (HDS).3 Due to its availability and relatively low cost, MoS2 has also found application as a solid lubricant,4 an electrode material for solidstate batteries, and more generally as a material for host-guest chemistry. Due to its electronic structure, MoS2 generally resists attempts to incorporate guest species between its layers. However, treatment with very strong reducing agents such as n-butyl lithium can result in Li ions being inserted into the layered MoS2 structure.5 The resulting LiMoS2 exhibits a remarkable ability to exfoliate in water through the redox reaction

LiMoS2 + H2O f (MoS2)single-layers + LiOH + H2(g) resulting in a colloidally dispersed suspension of single layers.6,7 These layers can be recovered and “restacked” upon filtration, * To whom correspondence should be addressed. E-mail: [email protected]. (1) Physics of New Materials; Fujita, F. E., Ed.; Springer Series in Materials Science; Springer: New York, 1998; Vol. 27. (2) Hughes, H. P.; Starnberg, H. I. Electron Spectroscopies Applied to LowDimensional Structures. In Physics and Chemistry of Materials with LowDimensional Structures; Kluwer Academic Publishers: Boston, 2000; Vol. 24. (3) Dungey, K. E.; Curtis, M. D.; Penner-Hahn, J. E. J. Catal. 1998, 175, 129. (4) Spalvins, T. J. Vac. Sci. Technol., A 1987, 5, 212. (5) Dines, M. B. Mater. Res. Bull. 1975, 10, 287. (6) Joensen, P.; Frindt, R. F.; Morrison, S. R. Mater. Res. Bull. 1986, 21, 457. (7) Gee, M. A.; Frindt, R. F.; Joensen, P.; Morrison, S. R. Mater. Res. Bull. 1986, 21, 543.

centrifugation, or precipitation with relative ease and have permitted the encapsulation or “inclusion” of a wide variety of guest species, including neutral organic molecules, polymers, metal cations, and oxides.7-11 Not surprisingly, the ability of these materials to encapsulate such a large range and diversity of species has stimulated much interest in the structure of the exfoliated and restacked material12-16 and promise for a number of applications including water treatment, hazardous waste removal, and as a new form of HDS catalyst. While the potential environmental and economic benefits to such new technologies can be readily appreciated, the specific host-guest interactions required for effective encapsulation are not entirely well understood. Here, we present results of experiments on the exfoliation and restacking of MoS2 in the presence of the cationic dye malachite green (MG) in order to extend the current level of understanding of its adsorption and encapsulation abilities. We have previously employed MG as a probe molecule to investigate the orientation, aggregation, and interaction of species at the solid/air interface via optical second-harmonic generation techniques. In the current study we investigate the adsorption interactions of MG to single layers of exfoliated MoS2 and the properties of the resultant restacked inclusion materials. We employ the techniques of X-ray diffraction, electron microscopy, atomic force microscopy, and optical second-harmonic generation to characterize the exfoliated and restacked MoS2 and the MG-based inclusion materials. Experimental Section Exfoliated samples of molybdenum disulfide (Aesar) were obtained by lithium intercalation following conventional literature protocols. (8) Divigalpitiya, W. M. R.; Frindt, R. F.; Morrison, S. R. Science 1989, 246, 369. (9) Kanatzidis, M. G.; Bissessur, R.; DeGroot, D. C.; Schindler, J. L.; Kannewurf, C. R. Chem. Mater. 1993, 5, 595. (10) Zhou, X.; Yang, D.; Frindt, R. F. J. Phys. Chem. Solids 1996, 57, 1137. (11) Kosidowski, L.; Powell, A. V. Chem. Commun. 1998, 2201. (12) Miremadi, B. K.; Morrison, S. R. Surf. Sci. 1986, 173, 605. (13) Dungey, K. E.; Curtis, M. D.; Penner-Hahn, J. E. Chem. Mater. 1998, 10, 2152. (14) Heising, J.; Kanatzidis, M. G. J. Am. Chem. Soc. 1999, 121, 638. (15) Heising, J.; Kanatzidis, M. G. J. Am. Chem. Soc. 1999, 121, 11720. (16) Gordon, R. A.; Yang, D.; Crozier, E. D.; Jiang, D. T.; Frindt, R. F. Phys. ReV. B 2002, 65, 125407.

10.1021/la060461j CCC: $33.50 © 2006 American Chemical Society Published on Web 09/15/2006

8996 Langmuir, Vol. 22, No. 21, 2006 Briefly, 1 mL of 2.5 M n-butyl lithium (Aldrich) in hexane was reacted with 0.1 g of the disulfide under an argon atmosphere for 24 h in hexane. It has been demonstrated that, under these conditions of excess lithium, the resulting LixMoS2 has x ∼ 1.6,15 The hexane was then removed, and upon addition of 50 mL of water, exfoliation of the LixMoS2 occurred spontaneously. Formation of the “bare” restacked material was induced following extensive washing of the aqueous suspension and centrifugation at 3000 rpm for 10 min. Exfoliated single layers of MoS2 are able to adsorb a wide variety of species. Careful adjustment of the pH of the aqueous suspension allows one to control the net charge on both the MoS2 planes and plane edges, providing some measure of control over which species are adsorbed and the subsequent structure of the resulting MoS2 inclusion material following flocculation and “restacking” of the exfoliated sheets. Two types of sample preparation were employed here. For the purposes of the X-ray diffraction and optical SHG studies, the MG-based inclusion materials were obtained following addition of stoichiometric amounts of solid malachite green (Aldrich, 99%) to the freshly exfoliated and water-washed material. Care was taken to ensure that the added MG was completely soluble in the water-based suspension. This step was followed by extensive additional washing to ensure both a constant final pH of the aqueous suspension by removal of excess of LiOH produced in exfoliation and that any MG detected in the subsequently restacked material was adsorbed and associated with the exfoliated MoS2 layers prior to restacking. To obtain the adsorption isotherm data, the MG-based inclusion materials were prepared by addition of stoichiometric amounts of solid malachite green (Aldrich, 99%) to the freshly exfoliated material followed by a pH adjustment to 5.5 using hydrochloric acid. The constant final pH of the aqueous suspension was necessary to facilitate the removal of excess of LiOH produced in exfoliation and to provide control of the (pH-dependent) electronic structure of malachite green that is adsorbed to the exfoliated sheets. To quantify the degree of MG adsorption, UV-vis absorption spectra were used to obtain the concentration of MG left unadsorbed by the exfoliated MoS2 following filtration of the MoS2 suspension (see below). The absorption spectra were obtained from a Cary 3E (Varian) spectrophotometer. These optical absorption measurements of MG loading were corroborated by thermogravimetric analysis (TGA) obtained at scan rates of 2 °C/min. TGA was performed with a TGA50 Shimadzu thermogravimetric analyzer. Atomic force microscopy (AFM) studies were carried out using a TM microscopes Explorer scanning probe microscope system with an 8 µm Z-linearized dry scanner. AFM measurements were taken to assess the overall quality of the films, to determine film thickness, and to assess the sizes and shapes of the MoS2 crystallites. X-ray diffraction experiments were carried out using a RAPID (Rigaku) X-ray diffractometer with a copper target (λ KR ) 1.542 Å) and an image plate detector. Transmission electron microscopy experiments were performed with an FEI Tecnai STEM 200 keV field emission STEM with a Lorentz lens and biprism, high angular annular dark field (HAADF) image filtering, energy-dispersive (Xray) spectroscopy (EDS), and a CCD detector. TEM studies were carried out on multilayer films deposited directly onto carbon-coated copper grids via drop- and dip-cast methods. Molecular mechanics calculations were carried out using a HyperChem Lite software package version (2.0) (see below). Optical SHG measurements were carried out using the regeneratively amplified output of a mode-locked titanium:sapphire oscillator. The detailed description of this optical setup17 and its application to the nonlinear optical response of malachite green has been previously reported.18 Briefly, this system is capable of producing 100 fs pulses with an energy of 1 mJ, at a repetition rate of 1 kHz and nominal wavelength of 800 nm. Since pulses of such high peak intensity can lead to photochemical damage of the films under investigation, only a fraction (typically ∼12/1), the SHG measurements provide large averaged orientation angles, in the neighborhood of ∼70-75°, indicating that the molecules lie flat, with their planes almost parallel with the MoS2 layers. At high MG concentrations, (MoS2/MG molar ratios < ∼5/1), the SHG analysis provides orientation angles of approximately 20-25°, consistent with a much more upright orientation. At intermediate MG coverage, the averaged orientation angles lie between these two ranges. We have employed molecular modeling techniques to provide a connection between our results observed from the XRD studies and those from SHG studies. This correspondence is displayed in Figure 7. Molecular mechanics calculations provide a measure of the MG molecular dimensions and, combined with the tilt angle information from the SHG measurements, provide an expected MoS2 layer separation. This separation, obtained indirectly from the SHG data, is compared to the layer separation extracted directly from the XRD data in Figure 7. A measure of the volume occupied by MG within the MoS2MG inclusion material can be extracted from the XRD data. The XRD data of Figure 5 show that the layer separation of the MoS2 in the absence of MG is ∼6.2 Å (2θ ) 14.4°). The layer separation of the MG-based inclusion material is ∼11.5 Å (2θ ) 8.1°). This

Figure 8. Chemical structure of malachite green (MG) (left) and schematic representation of the arrangement of MG within the MoS2MG layered structure (right) inferred from molecular modeling.

result implies a host interlayer expansion of 5.3 Å upon inclusion of the MG. MG, a triphenylmethane-based indicator, has dimensions in its plane in excess of this value. The only way that the MoS2 can accommodate the MG with such an interlayer expansion is with the MG molecular plane at a large tilt angle with respect to the layer normal, or roughly parallel to the MoS2 layers. This situation is depicted in Figure 8. Our SHG measurements at low MG loadings support this idea and predict an interlayer separation of approximately 11.5 Å.29 Figure 7 shows that there is good agreement between the layer separations obtained directly from the XRD measurements and indirectly from the SHG analysis at low MG loadings. This agreement seems to persist until a MoS2/MG molar ratio of ∼15/1 is reached, at which point, further decreases in the molar ratio yield substantially smaller MG tilt angles and predict correspondingly larger layer separations. The departure from agreement between the XRD- and SHGbased layer separations can be understood by examination of the adsorption isotherm shown in Figure 4. As described previously, the isotherm shows the appearance of a plateau consistent with the onset of monolayer adsorption at MG/MoS2 values of approximately 0.12 (MoS2/MG molar ratios of approximately 8/1) and consistent with a strong driving force for MG adsorption. According to the XRD and molecular modeling studies, the subsequent restacking of the exfoliated MoS2 host material (now decorated with a monolayer of MG on both sides of its plane), results in a layered structure incorporating one monolayer of MG between its sheets, and it does so in such a way that the guest molecule adopts an orientation with its molecular plane roughly parallel with the MoS2 layers. This step must involve the expulsion of MG from the layered structure during the restacking process. It is possible that expelled MG resides at the polar edge sites of the MoS2 planes or otherwise decorates polar adsorption sites on the exterior of the restacked crystallites. In so doing, the MG adopts an averaged orientation consistent with its adsorption in unrestricted, polar environments,18 i.e., with a more upright orientation characterized by interaction of its polar dimethylamino substituents with a polar interface. The resulting SHG data at high MG loadings would then be expected to reflect both signal contributions from MG in flat orientations between the MoS2 sheets and contributions from expelled MG residing outside of the layered structure. Our SHG results are consistent with this behavior. At low MG loadings, the SHG results indicate MG orientations consistent with a flat geometry and residence within the layered structure. As the MG concentration is increased, (MoS2/MG molar ratios