Open Access
Barnette, Anna Lorraine
Graduate Program:
Chemical Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
December 07, 2010
Committee Members:
  • Seong H. Kim, Committee Chair
  • Michael J. Janik, Committee Member
  • Robert M. Rioux, Committee Member
  • David L. Allara, Committee Member
  • Andrew Zydney, Committee Member
  • sum frequency generation vibration spectroscopy
  • cellulose
  • adsorption
  • nanotribology
  • SFG
The use of silicon oxide with its native oxide layer for the fabrication of microelectromechanical systems (MEMS) with contacting sliding parts requires the need for innovative lubrication methods to extend device lifetimes. The most promising method to date involves the equilibrium vapor phase lubrication (VPL) of MEMS using short chain linear alcohols in ambient conditions. Still, some questions remain regarding the effectiveness of this lubrication method, these include (1) whether or not the adsorbed n-alcohol molecules are the primary lubricant and (2) is this lubrication method effective in humid environments. This study investigates the vapor phase lubrication of SiO2 surfaces using short chain linear alcohols, more specifically n-propanol and n-pentanol. Macro-scale ball-on-flat tribometer tests are used to evaluate the lubriciousness of n-pentanol vapor under a series of contact loads/ pressures. Wear reduction of the SiO2 surfaces is achieved when there is complete coverage of the SiO2 surfaces with the adsorbed n-pentanol molecules. This occurs when the partial pressure relative to the saturation pressure (P/Psat) of n-pentanol was kept above 20% P/Psat which corresponds to approximately monolayer coverage of the SiO2 surface. In contrast to the lubricious effect of n-pentanol vapor, water vapor proves to enhance wear of the SiO2 surfaces when compared to dry (low moisture) conditions. This study also demonstrates that the primary lubrication method of the SiO2 surfaces is most likely the adsorbed n-pentanol molecules and not the tribochemical reaction species produced during the sliding contact. Although this reaction species is always present within the wear tested regions, the production of the tribochemical reaction species is enhanced when more severe wear is observed. So, the adsorbed n-pentanol molecules are the primary method of lubrication. The effectiveness of the lubrication method in environments containing water vapor is also investigated. This study is important because in the ambient, water molecules are always present. In order to fully understand how the n-alcohol molecules interact with the SiO2 surfaces under co-adsorption conditions with water molecules, attenuated total reflectance infrared spectroscopy (ATR-IR) and sum frequency generation (SFG) vibration spectroscopy are performed to understand the thickness, composition, and structure of the binary adsorbate layer on SiO2 in the absence of sliding. These studies show that the water and n-alcohol molecules form a bi-layer structure with the water molecules adsorbed to the SiO2 surface and the n-alcohol molecules adsorbed on top of the water molecules. Once water is adsorbed to the SiO2 surface it is difficult to remove. So, the effectiveness of adsorbed n-pentanol molecules in humid environment in lubricating a MEMS sidewall friction tribometer is investigated. This study demonstrates that an environment containing 50% P/Psat n-pentanol can lubricate these MEMS up to a relative humidity of 25% without device failure. Above this, too much water is adsorbed to the SiO2 surface which inhibits device operation. Thus, although adsorbed water is considered detrimental to SiO2 surfaces during sliding contact, its effect can be offset using n-pentanol VPL. Cellulose is considered the most abundant naturally occurring polymer on earth. Because of this, many applications in the biofuels and pharmaceutical industries are under evaluation. Despite studies dating back to the 1940s, the structures of cellulose in its naturally occurring form (I) as well as fabricated polymorphs (II and III) are not completely agreed upon in the literature. Conventional spectroscopic techniques such as Fourier transform infrared spectroscopy (FTIR), Raman spectroscopy, X-ray diffraction (XRD), and solid-state 13C Nuclear magnetic resonance (NMR) spectroscopy are widely utilized in pursuit of understanding the structure of cellulose. Three of these techniques, FTIR, Raman, and NMR, require the separation of cellulose from its surrounding components in biomass. The chemical processes used for this separation can alter the structure of cellulose. XRD cannot completely differentiate between the sub-types of cellulose I (α or β). This study investigates the use of SFG in analyzing crystalline cellulose within biomass samples, which included hemicellulose and lignin, and the different cellulose polymorphs. This study also demonstrates that SFG is able to selectively detect crystalline cellulose within biomass samples as well as quantify the amount of crystalline cellulose present using proper calibration curves. Unlike what is observed in FTIR, Raman, and NMR, only crystalline cellulose generates a true SFG spectrum so separation from the other components in biomass is not required. Also, SFG is able to differentiate between the different polymorphs showing more distinct differences in spectra when compared to FTIR and Raman. These include distinct differences between the sub-types of cellulose I. SFG is a very promising complementary spectroscopic technique for studying crystalline cellulose.