Session Mo-B1

TFT and Large Area Electronics I

Chair: Karim S. Karim, University of Waterloo

Mo-B1.1 (invited) 14:00–14:30

Oxide TFTs for Displays and Imaging

Arokia Nathan

Department of Engineering, Cambridge University, Cambridge CB3 0FA, UK

Amorphous oxide semiconductors are known for their optical transparency and high electron mobility even when processed at room temperature, making them a promising candidate for the next-generation thin film transistor (TFT) technology. The oxide transistor shows superiority in terms of process simplicity and cost, and stable device behaviour in the dark. While its non-uniformity over large areas is comparable to that of thin film silicon transistors, its photo-instability at low wavelengths can be an issue due to persistence in photoconductivity. This talk will discuss progress and issues related to oxide semiconductor TFTs addressing applications related to displays. In particular, we show how the material can be tuned for imaging applications. We review the dark- and wavelength-dependent photo-instability in oxide transistors and its origins, along with its unique density of states profile, including the role of oxygen vacancies, and the relative dominance of trap-limited conduction and percolation transport.

Mo-B1.2 14:30–14:50

Towards a Digital Radiology Roadmap

John A. Rowlands (1) and Wei Zhao (2)

(1) Thunder Bay Regional Research Institute, Thunder Bay, ON, Canada

(2) State University of New York at Stony Brook, NY 11733 USA

The emergence of digital x-ray detectors based on a-Si TFT technology 20 years ago has revolutionized the practice of radiology. It has done so by making systems of unprecedented capabilities which are well positioned to be integrated into hospital PACS (Picture Archiving and Communication Systems). The time spent on the diagnosis of an individual patient by the introduction of PACS has been reduced by a factor of five or more by increasing the efficiency of collecting and organising image information before presentation to the radiologist. In the last few years CMOS wafer scale devices have become feasible and with a cost similar to TFT based systems. The clinical requirements for x ray systems are extremely varied. Currently each imaging task needs an x ray system optimized for that particular task. Thus the general purpose x-ray rooms in hospitals are used for orthopedics, abdominal and chest imaging, but still a separate room must be devoted to mammographic x ray imaging for breast cancer screening. This is very costly and not realistic in a small clinic or remote location. However, it is now possible to conceive of devices of reasonable cost which are dose efficient for all the tasks x-rays will be called on to perform in the clinic. This is possible because we are now reaching the technological capabilities where the detector which heretofore was mapped on a pixel basis one to one with the display can have all its capabilities programmed. That is, the task optimization can take place in a system comprising the detector with fine enough pixels and bit depth adequate for any application combined with an optimizable x-ray to light phosphor converter layer and sophisticated image processing tuned to the image task at hand. Then the system has capabilities which can be improved over the lifetime of the system by selective replacement of components during routine service along with the updating of software. For such an innovation-driven approach to be economically feasible requires the cost of the critical components to be reduced by one order of magnitude. This is possible if the decrease in costs of consumer electronics with their gigantic market can be passed onto the much smaller (in terms of units) market for medical systems. If the medical detector components can be made essentially identical to those used in displays, then this can be achieved. Thus lobbying to merge the display industry roadmap with the digital radiology roadmap is needed. Other components above and beyond the CMOS or TFT backplanes are needed in addition if displays are to be made into multitasking devices. To reduce cost, the amplification function needs to be performed with a continuous (i.e. non-pixelated) layer or layer. Two possible approaches have been demonstrated to be feasible: liquid crystals and avalanche multiplication. The liquid crystal approach will be dealt with in another paper. Here we will concentrate on the use of avalanche multiplication in a-Se (amorphous selenium), a mechanism discovered and made practical for use in imaging devices by our colleagues G. Juška and K. Tanioka. The properties of this material are, as far as we are aware, uniquely suited to the task of making multi-tasking detectors. The physics as to why a-Se has these unique properties is the subject of the talk.

Keywords: a-Si:H, a-Se, glass, TFTs, CMOS, liquid crystals, task independent imaging, phosphor layers, digital radiography

Mo-B1.3 14:50–15:10

High Mobility Thin Film Transistors Based on Zinc Oxynitride Semiconductors

Joon Seok Park, Hyun-Suk Kim, Tae Sang Kim, Eok Su Kim, Kyoung Seok Son, Jong-Baek Seon, Sunhee Lee, Seok-Jun Seo, Sun-Jae Kim, Myungkwan Ryu, Seong-Ho Cho, and Youngsoo Park

Compound Device Lab, Samsung Advanced Institute of Technology, Giheung-Gu, Yongin-Si, Republic of Korea

Recent needs for high performance devices in the flat panel display industry have triggered the search for high mobility semiconductors, as potential substitutes for the universal material to date; silicon. In the past few decades, the fabrication of active matrix liquid crystal display (AMLCD) panels has involved the integration of thin film transistors (TFTs) based on amorphous silicon (a-Si). Although the latter is still the semiconductor of choice in the current AMLCD market, its relatively low electron mobility (< 1 cm2/Vs) imposes limits on the maximum display size that can be achieved without compromising the image resolution and/or operating frequency. For example, ultrahigh definition (4000 x 2000 pixels) televisions larger than 100 inches, operating at refresh rates higher than 480 Hz, cannot be realized at present using a-Si TFT arrays. This is because it is difficult for the latter to convey fast electrical current to the pixel capacitors located at large physical distances away from the signal input, owing to substantial RC delay.

In this work, high mobility TFT devices that incorporate zinc oxynitride (ZnON) semiconductors are presented. ZnON films can be deposited onto glass substrates by reactive magnetron sputtering using a zinc metal target. The fabrication of ZnON-based devices is compatible with that of conventional a-Si TFTs, which makes ZnON a promising next-generation semiconductor for display applications. It is found that the most critical parameter affecting the electrical properties is the ratio of nitrogen to oxygen gas flow rates during ZnON growth. Regardless of the final composition, all ZnON films are n-type, and the electron mobility increases as the layers become rich in nitrogen. Microstructure analyses indicate the nitrogen-rich films consist of zinc nitride (Zn3N2) nanocrystals embedded in an amorphous Zn-O-N matrix, while additional oxygen induces a mixture of nanocrystalline Zn3N2 and zinc oxide (ZnO) phases. The high mobility compositions of ZnON allow the fabrication of TFT devices with field effect mobility values exceeding 100 cm2/Vs, however for practical use the off current levels and subthreshold swing need to be reduced. In that regard, the addition of cations such as gallium (Ga) is found to be effective, since the dopant acts as a carrier suppressor in the host material. Hence, realistic devices with field effect mobility near 50 cm2/Vs are routinely obtained.

Besides the high mobility, the electrical characteristics of ZnON devices are somewhat similar to those of a-Si TFTs formerly reported in the literature. While the semiconductor is n-type, the elevated off current levels imply the presence of minority hole carriers. The threshold voltage (Vth) varies with the sweep range of the gate voltage (Vg), which is a phenomenon analogous to that observed in a-Si devices. It is anticipated that the trapping of charge carriers at the gate insulator-semiconductor interface is the dominant mechanism for such shifts in Vth. The creation of electrical defects associated with bond-breaking at the interface is also expected to be detrimental concerning device stability, and further studies are underway to elucidate the electrical behavior of ZnON-based TFTs.

Keywords: zinc oxynitride, oxide semiconductor, high mobility, AMLCD, AMOLED

Mo-B1.4 15:10–15:30 Cancelled

Three Dimensional Thin Film Integrated Circuits using Atomic Layer Deposition

Feyza B. Oruç (2) and Ali K. Okyay (1,2)