VLSI Processing of Amorphous Silicon Alloy p-i-n Diodes for Active Matrix Applications
- PDF / 1,281,838 Bytes
- 5 Pages / 417.6 x 639 pts Page_size
- 78 Downloads / 199 Views
VLSI PROCESSING OF AMORPHOUS SILICON ALLOY P-I-N DIODES FOR ACTIVE MATRIX APPLICATIONS 3. MCGILL, V. CANNELLA, Z. YANIV, P. DAY AND M. VIJAN Ovonic Display Systems, Inc., 1896 Barrett Street, Troy, Michigan 48084* ABSTRACT A number of new amorphous silicon alloy microelectronic devices, including LCD active matrix displays, linear image sensors, and thin film These multilayer computer memories, have been developed in our company. applications rely heavily on the quality of the intrinsic semiconductor as well as its ability to withstand the many processing steps used in a modern In this paper, we present electrical data on photolithographic process. amorphous silicon alloy p-i-n diodes after such a process. These devices have an active area of 20m x 20pm defined using standard photolithographic techniques and etched using a dry etch process. These diodes are characterized by ideality factors (n) of 1.4 and extrapolated reverse The diodes exhibit nearly 10 2A/cm2. saturation current densities of 101 orders of magnitude rectification at ± 3V and the reverse bias current 2 In pulsed density remains below 10- A/cm for reverse bias voltages of -15V. forward bias, these diodes can be operated at current densities greater than 300A/cm2 . Thin film amorphous silicon diodes moreover have the advantage that varying the thickness of the intrinsic layer allows the optimization of parameters such as the capacitance per unit area, the reverse bias current density and the forward bias conductance per unit area. We find that these devices are fully compatible with state of the art VLSI processing techniques and are suitable for applications in integrated circuit structures, for example rectification devices in microelectronic arrays and isolation devices in display matrices. INTRODUCTION Active matrix liquid crystal displays depend upon switching devices to replace the threshold characteristics of the liquid crystal material and as the means to differentiate between off and on voltages on the pixels. Switching devices which have been used to achieve this include three terminal TFT devices [1] and two terminal devices such as p-i-n diodes [2,3] and MIM structures [4], for example. At present, a number of problems still exist with TFT technology which makes it difficult to achieve high yield and stability in large area displays. These difficulties include electrical shorts at intersections of crossing lines, large area uniformity of insulator, gate insulator interface control, threshold instability [5,6] also many mask levels are required. Drawbacks also exist for some two terminal device configurations, for example MIM devices have low on-to-off current ratios [1] and variations in the threshold of the devices have made it difficult to generate a useful grey scale capability. The back-to-back diode configuration utilizes only the reverse bias dependence of the diodes which is characterized by soft breakdown phenomena at high fields. As a result of this, stability, repeatability, and uniformity of device operation becomes a problem.
Data Loading...
