Content area
Full text
Self-assembled monolayers (SAMs) have attracted considerable attention as a model system for many fundamental and technological phenomena (for example, supramolecular assembly, wetting, tribology, and corrosion inhibition) because of both their simplicity and stability (1-3). In particular, SAMs consisting of n-alkane thiols adsorbed on Au(111) have been extensively studied (1, 4-10). Surprisingly, even in this widely studied system, there is little that is directly known about the head group-substrate structure and interaction. Previous studies found that long-chain disulfides and thiols form similar films when grown from solution, and there is strong evidence that the final chemisorbed state of the molecules is the same in these two cases (11). Although we are unaware of any measurements that have in fact determined the structure of the sulfur-substrate bond, a general consensus in the literature has been reached that the bonding state is in the form of an Au-thiolate (1, 11-13).
We performed an in-depth x-ray diffraction study of the CH sub 3 (CH sub 2 ) sub 9 SH (C10) monolayer on Au(111) and found that the results are inconsistent with the generally accepted thiolate bonding structure. The equilibrium structure of the monolayer includes a S-S spacing of 2.2 (character omitted), which strongly implies the existence of a S-S bond. These results demonstrate that internal molecular conformations are an intrinsic component of even these relatively simple SAM systems. Such a quantitative structural understanding is critical to the control and design of more complex molecular assemblies, as well as in developing a general understanding of organic-inorganic interfaces.
These measurements were performed on the Exxon X10A beamline at the National Synchrotron Light Source (NSLS) with the use of a z-axis spectrometer with a wavelength of 1.09 (character omitted) and an incident angle of 1deg. The resolution was set to Delta-Q sub = 0.21 (character omitted) sup -1 and Delta-Q sub z = 0.041 (character omitted) sup -1 , and the data have been normalized to remove resolution effects (14, 15); for convenience, the hexagonal peak intensity has been divided by 3 to account for the multiplicity of C(4 X 2) domains with respect to the hexagonal substrate. The samples were prepared by growth in dilute (==1 mM) solutions of thiol in ethanol as described previously and were annealed in vacuum...