In external beam radiotherapy quality assurance is carried out on the individual components of the treatment chain. The patient simulating device, planning system and linear accelerators are tested regularly according to set protocols developed by national and international organizations. Even though these individual systems are tested errors can be made in the transfer between systems. The best quality assurance for the system is at the end of the treatment planning chain. In-vivo dosimetry measures the dose to the target volume through indirect measures at the end of the treatment planning chain and is therefore the most likely method for picking up errors which might occur earlier in the chain. Metal Oxide Semiconductor Field Effect Transistors (MOSFETs) have been shown to have a similar error in estimating entrance dose for in-vivo dosimetry to diodes, but no studies have been done clinically with entrance dose in-vivo dosimetry with MOSFETs. The time savings for using MOSFETs makes them preferable to TLD's. Due to their small size and versatility in other applications they are useful as more than dedicated in-vivo dosimetry systems using diodes. Clinical implementation of external beam in-vivo dosimetry would add another use to the MOSFETs without purchasing more specialized equipment. My studies have shown that MOSFETs can be used clinically for external beam in-vivo dosimetry using entrance dose measurements. After the MOSFET measurement system was implemented using a custom built aluminium build up cap clinical measurements were performed. A total of 23 patients and 54 fields were studied. The mean for all clinical measurements was 1.3 %, with a standard deviation of 2.6 %. Results were normally distributed around a mean with skewness and kurtosis as -0.39 and 0.34 respectively. For breasts the mean was 1.8 %, with a standard deviation of 2.7 %. For prostates and hips the mean was 1.3 % with a standard deviation of 2.9 %. These results are similar to studies conducted with diodes and TLD's. From these results one can conclude that MOSFETs can be used for entrance dose in-vivo dosimetry and are no worse than diodes or TLD's in terms of their measurement accuracy. / Thesis (M.Sc.)--School of Chemistry and Physics, 2006.

Graduation date: 2001

Originally presented as the author's thesis (Swiss Federal Institute of Technology), Diss. ETH No. 16228. / Summary in German and English, text in English. Includes bibliographical references (p. 123-132).

The purpose of this work is to develop a new model for LDD n-MOSFET degradation in drain current under long-term AC use conditions for lifetime projection which includes a self-limiting effect in the hot-electron induced device degradation. Experimental results on LDD n-channel MOSFETs shows that the maximum drain current degradation is a function of the AC average substrate current under the various AC stress conditions but not a function of frequency or waveforms or different measurement configurations. An empirical model is constructed for circuit applications. It is verified that the self-limiting in drain current is due to the thermal re-emission of a trapped-hot-electron in the oxide. Results show that self-heating during AC stress releases trapped electrons, which in turn limits the maximum amount of drain current degradation. Moreover, tunneling to and from traps model is employed to visualize the internal mechanism of thermal recovery of electrons under different bias conditions. Although the LDD device structure can reduce the hot electron effect, various processing technologies can also affect the device reliability. A carbon doped LDD device with the first and the second level metal and passivation layer but without any final anneal shows that a significant reduction in the shifts of the threshold voltage of MOSFETs with time can be achieved. However, the long-term reliability projection of nMOSFETs based on DC stress tests alone is shown to be overly pessimistic. / Graduation date: 1992

We have grown the high quality AlGaN/GaN heterostructure by plasma-assisted molecular beam epitaxy. We obtained the mobility of two-dimensional electron gas of the AlGaN/GaN is 9300 cm2/Vs and carrier concentration is 7.9¡Ñ1012 cm-2 by conventional van der Pauw Hall measurement at 77K. The samples made of the AlGaN/GaN heterostructure were patterned to Hall bar geometry with a width of 20£gm by conventional photolithography. After the photolithography, the nanowire was fabricated by the process of focus ion beam (FIB), and the widths of nanowire were reduced to 900 nm, 500 nm, 300 nm, 200nm, 100 nm, 80 nm and 50 nm respectively. The SiO2 layer and Al electrode were deposed on the samples to form nanowired MOSFETs. We have studied the leakage current measurement on the AlGaN/GaN nanowired MOSFETs at 300K. On the 100 nm and 200 nm width of nanowires, we did not observe the leakage current for the gate voltage work range from -2.5 to 3.0 V and from -0.5 to 0.5 V respectively.

Thesis (Ph. D.)--Lehigh University, 2000. / Includes vita. Includes bibliographical references (leaves 132-136).

Thesis (Ph. D.)--Rutgers University, 2009. / "Graduate Program in Electrical and Computer Engineering." Includes bibliographical references (p. 151-156).

Thesis (Ph. D.)--University of Texas at Austin, 2000. / Vita. Includes bibliographical references (leaves 129-145). Available also in a digital version from Dissertation Abstracts.

Thesis (Ph. D.)--University of Texas at Austin, 2002. / Vita. Includes bibliographical references. Available also from UMI Company.

Thesis (M.Phil.)--Hong Kong University of Science and Technology, 2003. / Includes bibliographical references. Also available in electronic version. Access restricted to campus users.