Langmuir 2007, 23, 8909-8915
8909
Quantitative Lateral Force Microscopy Study of the Dolomite (104)-Water Interface Steven R. Higgins,*,† Xiaoming Hu,† and Paul Fenter‡ Department of Chemistry, Wright State UniVersity, Dayton, Ohio 45435, and Argonne National Laboratory, CHM/200, 9700 South Cass AVenue, Argonne, Illinois 60439 ReceiVed February 16, 2007. In Final Form: June 8, 2007 The friction and lateral stiffness of the contact between an atomic force microscopy (AFM) probe tip and an atomically flat dolomite (104) surface were investigated in contact with two aqueous solutions that were in equilibrium and supersaturated with respect to dolomite, respectively. The two aqueous solutions yielded negligible differences in friction at the native dolomite-water interface. However, the growth of a Ca-rich film from the supersaturated solution, revealed by X-ray reflectivity measurements, altered the probe-dolomite contact region sufficiently to observe distinct friction forces on the native dolomite and the film-covered surface regions. Quantitative friction-load relationships demonstrated three physically distinct load regimes for applied loads up to 200 nN. Similar friction forces were observed on both surfaces below 50 nN load and above 100 nN load. The friction forces on the two surfaces diverged at intermediate loads. Quantitative measurements of dynamic friction forces at low load were consistent with the estimated energy necessary to dehydrate the surface ions, whereas differences in mechanical properties of the Ca-rich film and dolomite surfaces were evidently important above 50 nN load. Attempts to fit the quantitative stiffness-load data using a Hertzian contact mechanical model based on bulk material properties yielded physically unrealistic fitting coefficients, suggesting that the interfacial contact region must be explicitly considered in describing the static and dynamic contact mechanics of this and similar systems.
Introduction Carbonate mineral surfaces represent a significantly reactive and ubiquitous geochemical system, particularly when these surfaces form solid-water interfaces. The carbonate minerals in general, and particularly calcite (CaCO3) and dolomite (CaMg(CO3)2), are significantly more reactive in terms of surface area normalized dissolution rates than many common oxide and silicate minerals,1 thereby making carbonates particularly active materials in terms of interactions with natural waters. Furthermore, carbonate minerals are known for their tendency to incorporate a number of divalent metals into the solid phase2 giving these minerals the ability to sorb, from natural waters, otherwise mobile heavy metal ions. Understanding the fundamental processes involved in ion and molecular incorporation is key to advancement of interfacial models covering diverse problems including these environmentally related systems and those pertinent to device development in sensors and nanotechnology. An important component in such advancement must be the ability to characterize interfaces in terms of structure and composition with atomicscale resolution. Lateral force microscopy (LFM) studies of mineral surfaces have emerged recently due to the instrumental sensitivity to changes in surface chemistry and atomic structure as seen through measured friction forces. Combined with the inherently high spatial resolution of LFM (100 nN), the measured friction forces were again indistinguishable between the two surfaces. This observation may be explained by the onset of surface wear under the tip as the underlying bulk dolomite is exposed upon removal of the film. The friction trend becomes less smoothly varying with load in Figure 4a, suggesting that the contact area could have undergone abrupt changes with increased load. To verify this, a series of friction loops were taken at incrementally higher normal load. The results from one such experiment are given in Figure 5 showing a large-scale AFM image taken after performing the series of friction loops. A linear pattern of 5-6 pits can be observed in the image. These pits resulted from the application of the highest load values (>120 nN) during the friction loops. The size of the contact region in AFM measurements is rarely known due to the lack of control of the tip size and shape. This unknown places serious limitations on the ability to compare data across experiments where the tip is usually changed. The combination of friction and lateral stiffness measurements provides a measure of the contact shear strength, an intensive property of the contact independent of the contact area for a given tip geometry (e.g., spherical). Carpick et al.26 showed this combination to yield the following expression for contact shear strength assuming a spherical tip
τ)
64G*2f πk2cont
(7)
The data in Figure 4a and b were combined using this expression to yield the contact shear strength as a function of applied load (see Figure 6). The most significant observation in Figure 6 is the approximately constant, and large value for the shear strength above 20 nN applied load. At nearly 103 GPa, the computed shear strength is about 3 orders of magnitude larger than that determined by Carpick et al.26 in similar studies using a silicon nitride cantilever/tip on mica in humid air. It is also noteworthy that the ktot values obtained in this work are generally about 10-fold smaller than those obtained in the mica studies.26
AFM Study of Dolomite-Water Interface
Langmuir, Vol. 23, No. 17, 2007 8915
studies10,22 indicated a decrease in the relative friction (compared with friction on dolomite) on films grown from solutions with a lower Ca/Mg ratio than used in the current study. From the results presented above, the sensitivity of LFM to Ca and Mg compositional changes at the native inorganic interfaces appears to be limited to approximately full unit changes in the metal ion ratio (e.g., 1:1 vs 2:1 Ca/Mg film composition). Therefore, trace element incorporation into surface layers is not likely to yield visible signals in the LFM unless further surface chemical modification methods are developed to amplify the effects of trace impurities on the lateral forces.
Conclusions Figure 6. Contact shear strength versus normal load data obtained on the dolomite (104) surface (open circles) and on the film-covered surface (dark circles) while the sample was immersed in dolomitesupersaturated (SI ) 2.3) solution. The contact shear strength was calculated using eq 9 and the data in Figure 4 and assuming bulk values for the shear modulus and Poisson’s ratio of Si and dolomite were appropriate for calculating G*.
The large shear strength observed in the present work is indicative of a nonlubricated contact if, for comparative purposes, the water film in the mica studies behaved as a contact lubricant, suggesting that the tip dehydrates the dolomite surface during scanning. This same suggestion was also supported by predictions for the experimental friction force based on hydration enthalpies of the ions. The fact that τ varies with load at low load indicates a transition at the contact around 20 nN, above which τ is nominally independent of load. The load dependence could be associated with a continuous change of the molecular structure of the contact, such as changes in the number of hydration layers separating the probe from the dolomite and/or the properties of the hydration layer itself such as the molecular areal density, molecular orientation, and bulk density. The mechanisms associated with friction contrast observed from compositionally distinct surface regions clearly involve a complex contribution from physicochemical properties and experimental variables. The results reported here do suggest that the structural aspects of the surface film have a key role in governing relative friction measurements. Films containing significant strain, as in the Ca-rich film studied here, are generally expected to display higher friction when the load is sufficiently high to form a direct tip-film contact but not at loads high enough to induce wear. What remains to be investigated is the impact of film composition on film strain. Previous FFM
The above investigations yielded new information on the origin of friction force differences previously observed at dolomitewater interfaces. The variations in friction forces on Ca-rich film and dolomite surface regions were explained with a basic model involving the deformation and penetration of the tip into the interfacial region. The quantitative friction-load relationships obtained in this work demonstrated dynamic friction forces consistent with the energy necessary to dehydrate the surface ions with differences in the film and dolomite friction correlating with displacements of the calcium and magnesium ions away from their ideal locations at the dolomite-water interface. The physically unrealistic values for the AFM probe radius obtained through application of the Hertz contact model demonstrated that the basic model reliance on bulk material properties is not appropriate for the interfacial contact in the present system, further supporting the suggestion that inclusion of the interfacial structural properties and surface hydration layer into contact models is important. Future studies aimed at describing the role of the interfacial water structure in the tribochemical properties of similar systems could lead to significant advancement in contact mechanical models having widespread relevance in fields from mineral-fluid geochemistry to micro- and nanoelectromechanical systems. Acknowledgment. The authors gratefully acknowledge the financial support of this work by the Chemical Sciences, Geosciences and Biosciences Division, Basic Energy Sciences, Office of Science, Department of Energy, the National Science Foundation, Geoscience Division, and the Wright State University Office of Research and Sponsored Programs. The authors also thank Dr. Zhan Zhang and Dr. Changyong Park for their assistance with X-ray experiments and Prof. Neil Sturchio for his helpful suggestions for improving the original manuscript. LA700467Q
