Inspired by the high specific capacitances found using ultrathin films or nanoparticles of manganese oxides (MnO(x)), we have electrodeposited MnO(x) nanoparticles onto sheets of carbon nanotubes (CNT sheets). The resulting composites have high specific capacitances (C(sp) ≤ 1250 F/g), high charge/discharge rate capabilities, and excellent cyclic stability. Both the C(sp) and rate capabilities are controlled by the average size of the MnO(x) nanoparticles on the CNTs. They are independent of the number of layers of CNT sheets used to form an electrode. The high-performance composites result from a synergistic combination of large surface area and good electron-transport capabilities of the MnO(x) nanoparticles with the good conductivity of the CNT sheets. Such composites can be used as electrodes for lithium batteries and supercapacitors.

The SiOxFy passivation layer created on structure sidewalls during silicon cryoetching is investigated. This SiOxFy passivation layer formation strongly depends on O2 content, temperature and bias. It is a fragile layer, which mostly disappears when the wafer is warmed up to ambient temperature. A mass spectrometer was used to analyze the desorbed species during the warm-up and using this instrument allowed us to find a large signal increase in SiF3+ between −80°C and −50°C. SiF4 etching products can participate in the formation of the passivation layer as it is shown by a series of test experiments. SiF4∕O2 plasmas are used to form a thin SiOxFy layer on a cooled silicon wafer. Thickness and optical index of this thin film can be determined by in situ spectroscopic ellipsometry. It is shown that the passivation layer spontaneously desorbs when the silicon wafer temperature increases in good agreement with the mass spectrometry analysis. Two physical mechanisms are proposed to explain the SiOxFy passivation layer buildup involving either the etching products or the SiFx sites created during etching. In both cases, oxygen radicals react at the surface to form the SiOxFy layer.

No abstract

A Langmuir probe and a microwave interferometer have been combined to measure the electron density of argon glow discharges in the Gaseous Electronics Conference reference reactor [Bull. Am. Phys. Soc. 36, 207 (1991)]. The two techniques indicate the same charge density at 100 mTorr to within 30%. This 30% difference is easily explained by the experimental peculiarities. While the predicted charge densities obtained from the two techniques track one another as the applied rf voltage is varied at constant pressure, they do not track one another as the pressure is varied at constant rf voltage. In fact, the charge densities predicted from the Langmuir probe using Laframboise’s theory (Tech. Rep. 100, Univ. Toronto Inst. Aerospace Study, 1966) are factors of 2 and 4 times lower than those from the interferometer at 250 and 500 mTorr, respectively. It appears that the probe alters the charge density in its vicinity when the probe radius becomes greater than the ion mean free path. The interferometer has also been used to investigate the macroscopic perturbation of the plasma electron density caused by the Langmuir probe. It was found that presence of the probe in a radio-frequency discharge does not appreciably alter the electron density measured by the interferometer at a set rf voltage irrespective of the potential placed upon it. This was not the case for dc discharges in the same apparatus where the volume-averaged electron density could be depleted by as much as 70% when the probe was biased even a few volts above plasma potential.

A much greater number of useful precursors for plasma-enhanced chemical vapor deposition (PECVD) can be dispersed in high vapor pressure solvents than can be put into the vapor phase directly. In order to enable the use of such precursors, the authors investigated a method by which one can directly inject these liquids as microdroplets into low pressure PECVD environments. The solvent evaporates first leaving behind the desired precursor in the gas/plasma. The plasma dissociates the vapor and causes the deposition of a composite film (from precursor, solvent, and plasma gas). The authors made preliminary tests using Fe nanoparticles in hexane and were able to incorporate over 4% Fe in the resulting thin films. In addition, the authors simulated the process. The time required for a droplet to fully evaporate is a function of the background pressure, initial liquid temperature, droplet-vapor interactions, and initial droplet size. A typical evaporation time for a 50μm diameter droplet of hexane is ∼3s without plasma at 100mTorr. The presence of plasma can decrease the evaporation time by more than an order of magnitude. In addition, the model predicts that the temperature of the injected droplet first decreases by evaporative cooling (to ∼180K for hexane); however, once the solvent has fully evaporated/sublimated, the plasma heats any remaining solute. As a result the solute temperature can first fall to 180K, then rise to nearly 750K in less than 1s.