• PDF / 172,782 Bytes
• 6 Pages / 612 x 792 pts (letter) Page_size
• 30 Downloads / 225 Views

DOWNLOAD

REPORT

---

Increasing the Lateral Resolution of Scanning Spreading Resistance Microscopy R.J. Kline1,2, J.F. Richards2, and P.E. Russell1 Analytical Instruments Facility, North Carolina State University, Raleigh, NC 27695 2 Materials Technology Department, Intel Corporation, Santa Clara, CA 95052 1

ABSTRACT This paper discusses problems inherent to scanning spreading resistance microscopy (SSRM) and ways to correct them to increase the resolution of two-dimensional dopant profiling. Specifically this paper looks into issues related to the probe-silicon contact and the damaged surface layer created by the sample preparation technique. Degradation of the measured dopant profile was observed when the probe scanned over the nitride spacers. Attempts to reduce the required contact pressures to increase the lifetime and effectiveness of the probes are addressed. The force required for SSRM was successfully reduced after the damage layer was partially removed by isotropic etching. INTRODUCTION As device dimensions in the semiconductor industry continue to shrink, many issues related to ultrashallow junctions begin to be increasingly important. To maintain proper device performance as device dimensions shrink, the junction depth along with the gate lengths must reduce proportionally. Smaller features require tighter control on process parameters to maintain minimal tolerances. Tighter process control requires an increase in the required measurement accuracy of previous process generations. Additionally new parameters, which previously were not typically measured, are now very important to maintaining the new tolerances. Future device structures will contain increasingly complicated profiles including many two-dimensional features added to alleviate performance problems arising from the reduced dimensions. The NTRS 1998 roadmap calls for a lateral resolution of 2nm with a dopant accuracy of 4% for the 130nm technology and for 1.5nm with a dopant accuracy of 3% for the 100nm technology.1 Present techniques such as scanning capacitance microscopy and selective chemical etching have a lateral resolution of 10-20nm with an accuracy on the order of 30-50%.2 In addition, any successful techniques must have a large enough dynamic range to measure dopant concentrations from 1015 cm-3 up to 3x1020 cm-3. As transistors approach the 130nm technology, critical dimension tolerances reach several nanometers. The technology computer aided design (TCAD) models used in previous generations are no longer sufficient to provide the needed accuracy in simulating the complex interactions of future process technologies. Many variables and factors previously ignored become significant and must be incorporated into new models. The complexity of these new models requires careful testing with real data before they are implemented. These models frequently need to be calibrated with data as well. The data must be obtained from measurements with metrology accuracy greater than the critical dimension tolerances; otherwise the accuracy of the models is limited by th