Methods for Performing Atomic Force Microscopy Imaging of Clay Minerals in a Q U E O U S Solutions

-Three methods were developed that allow for the imaging of any clay mineral in aqueous solutions with atomic force microscopy (AFM). The methods involve fixing the particles onto special substrates that do not complicate the imaging process, but hold the particles sufficiently so that they do not move laterally or float away during imaging. Two techniques depend on electrostatic attraction under circumneutral pH conditions, between the negatively charged clay particles and the high point of zero charge substrate (either aluminum oxide or polyethyleneimine-coated mica) whereas the third technique depends on adhesion to a thermoplastic film. The first electrostatic technique involves a polished single crystal a-AlaO 3 (sapphire) substrate. This was used successfully as a substrate for clay-sized minerals with high permanent layer charge localized on the basal planes (phlogopite and vermiculite) and when the AFM was operated in TappingMode to limit the lateral forces between the probe tip and the particles. However, electrostatic attraction between the sapphire surface and clay minerals such as smectite and kaolinite (low or no permanent layer charge) is not sufficiently strong to adequately fix the particles for imaging, The second electrostatic technique involves a polyethyleneimine-coated mica surface designed to immobilize a larger variety of clay minerals (phlogopite, vermiculite, montmorillonite, and kaolinite), and in this technique weak bonding between the clay and the organic film is also a factor. The third technique, which does not depend on electrostatic attraction, fixes clay particles into the surface of a thermoplastic adhesive called Tempfix. This has proven useful for fixing and imaging relatively large clay particles with well-defined morphology. The Tempfix mount also requires imaging in TappingMode because the Tempfix is relatively soft. Key Words--Atohaic Force Microscopy, CMS Source Clay KGa-1, CMS Source Clay SWy-1, Fluid Cell, Particles, Polyethyleneimine, Sapphire, TappingMode, Tempfix. I N T R O D U C T I O N Atomic force microscopy (AFM) is a powerful technique for the characterization of mineral surface micro topography and reactivi ty (see Nagy and Blum, 1994; Hochel la , 1995). One great strength of A F M is that it characterizes surface micro topography directly in solution. Thus it can track the progress of reactions such as mineral growth, dissolution, and heterogeneous precipitat ion as they occur (Drake e t al. , 1989; Hi l lner e t al. , 1992; D o v e and Hochel la , 1993; Bosbach and Rammensee , 1994; D o v e and Chermak, 1994; Junta and Hochel la , 1994; Bosbach et al. , 1995; Putnis et al. , 1995; Bosbach and Hochel la , 1996; Bosbach et al. , 1996; Grantham and Dove , 1996; Liang et al . , 1996; Junta-Rosso et al. , 1997). However , none of the above real-t ime, in s i tu studies invo lved clay minerals because the standard sample preparat ion techniques used to fix clay particles for A F M characterizat ion in air are inadequate for fluid-cell applications (Dove and Chermak, 1994). Therefore , A F M work on clays is concentrated on observat ions o f microtopography of clay mineral surfaces and the morphology of clay particles, as measured in air (Lindgreen e t al . , 1991; B lum and Eberl, 1992; Garnaes e t aL, 1992; Johnsson et al. , 1992; Sharp e t al. , 1992; Gaber and Brandow, 1993; Blum, 1994; Nagy, 1994; McDan ie l e t al. , 1995; Brady et al . , 1996; Zbik and Smart, 1998). The application of A F M techniques to the study of c lay-mineral surface react ivi ty has mainly invo lved inferences made f rom these e x s i tu observat ions (Dove and Chermak, 1994; Nagy, 1994). Sample preparat ion for imaging clay particles with A F M involves fixing the particles onto some substrate. However , to image under fluid, certain criteria must be met: (1) The substrate surface must be local ly smooth to aid in image interpretation. Not only is it difficult to find part icles on a rough surface, but the more irregular the surface, the more artifacts are int roduced into the imaging process, also. Similarly, accurate measurement o f whole-par t ic le morpho logy requires a wel l -def ined baseline. (2) The substrate surface must be " s t i c k y " with respect to the part icles o f interest. Thus, the part icles must not float away upon introduction of fluid, and they must not be swept away f rom the imaging area by lateral forces caused by the scanning mot ion o f the A F M tip. T a p p i n g M o d e A F M ( T M A F M ) greatly reduces the lateral t ip-sample forces (Johnson, 1995), and requires a less " s t i c k y " substrate than does contact m o d e AFM. (3) The substrate surface cannot be too soft, that is, the A F M tip cannot interact too strongly with the substrate. In this case, Copyright 9 1999, The Clay Minerals Society 573 574 Bickmore, Hochella, Bosbach, and Charlet Clays and Clay Minerals minute portions of the substrate can adhere to the tip and change its shape. T M A F M also reduces the tipsample interaction, and is capable of imaging softer surfaces than contact mode AFM (Hansma et al., 1994). (4) The substrate surface must be relatively insoluble in the solutions used, and also inert to the reactions of interest. In this paper, three techniques suitable for preparing clay-size particles for A F M examination in solution are described, and they generally meet the above criteria. In one technique, clay-size phyllosilicates are fixed on a polished single-crystal et-A1203 (sapphire) substrate that holds clay particles by electrostatic attraction. In another technique, a mica substrate is coated with a monolayer of polyethyleneimine, a cationic polyelectrolyte, which also fixes clay particles by electrostatic attraction. Finally, clay particles are anchored into the surface of a thermoplastic adhesive, Tempfix. Each technique has advantages and disadvantages, but together they allow for the AFM examination of a variety of clay-size particles in aqueous solution. M A T E R I A L S AND METHODS Samples and characterization Four samples were used, two clays (Georgia kaolinite, KGa-1, and Na-montmorillonite, SWy-1, both obtained from The Clay Minerals Society Source Clay Repository and used as received) and clay-size fractions of phlogopite (Kingsmere, Ontario, Canada) and vermiculite (Phalaborwa Igneous Complex, Palabora, South Africa). The phlogopite was prepared by grinding on a fixed diamond wheel, after which the <2-pma size fraction was separated by sedimentation. Vermiculite was soaked for 3 h in 1% HC1 and cleaned in an ultrasonic bath to remove carbonate impurities. Cleavage sheets were then ground with a file, after which the <2-1xm fraction was separated by sedimentation. Chemical analyses of the phlogopite and vermiculite were obtained using the electron microprobe and Xray fluorescence (XRF), respectively. Analyses of KGa-1 and SWy-I standard clay samples were obtained from van Olphen and Fripiat (1979). These analyses were used to calculate permanent charge (see below) values for each sample.