ZnO films for surface acoustic wave microfluidic devices

SAW devices were fabricated on c-axis oriented ZnO films grown on Si substrates. Effects of film thickness on the film microstructure and acoustic frequencies were studied. Both Rayleigh and Sezawa mode waves were detected on the devices, and their resonant frequencies were found to decrease with increase in film thickness.

Persistent photoconductivity in Hf-In-Zn-O thin film transistors

Passivated Hf-In-Zn-O (HIZO) thin film transistors suffer from a negative threshold voltage shift under visible light stress due to persistent photoconductivity (PPC). Ionization of oxygen vacancy sites is identified as the origin of the PPC following observations of its temperature- and wavelength-dependence. This is further corroborated by the photoluminescence spectrum of the HIZO. We also show that the gate voltage can control the decay of PPC in the dark, giving rise to a memory action. © 2010 American Institute of Physics.

Instability in threshold voltage and subthreshold behavior in Hf-In-Zn-O thin film transistors induced by bias-and light-stress

Electrical bias and light stressing followed by natural recovery of amorphous hafnium-indium-zinc-oxide (HIZO) thin film transistors with a silicon oxide/nitride dielectric stack reveals defect density changes, charge trapping and persistent photoconductivity (PPC). In the absence of light, the polarity of bias stress controls the magnitude and direction of the threshold voltage shift (Δ VT), while under light stress, VT consistently shifts negatively. In all cases, there was no significant change in field-effect mobility. Light stress gives rise to a PPC with wavelength-dependent recovery on time scale of days. We observe that the PPC becomes more pronounced at shorter wavelengths. © 2010 American Institute of Physics.

The CU-HTK Mandarin broadcast news transcription system

This paper discusses the development of the CU-HTK Mandarin Broadcast News (BN) transcription system. The Mandarin BN task includes a significant amount of English data. Hence techniques have been investigated to allow the same system to handle both Mandarin and English by augmenting the Mandarin training sets with English acoustic and language model training data. A range of acoustic models were built including models based on Gaussianised features, speaker adaptive training and feature-space MPE. A multi-branch system architecture is described in which multiple acoustic model types, alternate phone sets and segmentations can be used in a system combination framework to generate the final output. The final system shows state-of-the-art performance over a range of test sets. ©2006 British Crown Copyright.