6

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 


 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

N
E
W
S
L
E
T
T
E
R


 


CAMP Professor Ming-Cheng Cheng Models Heat Flow in SOI Devices

CAMP Professor Ming-Cheng Cheng and his group are studying thermal flow in silicon-on-insulator (SOI) devices in static and dynamic situations. The objectives of this work are to understand static and dynamic heat flow in SOI devices and to develop efficient and accurate thermal models for these devices taking into account self-heating. The study will provide efficient tools for microelectronic circuit simulation accounting for self-heating effects on the electronic characteristics of SOI devices. In addition, information on the temperature profile in devices and on heat flow to interconnects will lead to a more accurate prediction of device/interconnect reliability, delay time and power consumption.

The silicon-on-insulator CMOS technology will soon dominate the high performance microelectronic ICs and will play a crucial role in the wireless industry as well. Although providing superior performance to bulk MOSFETs, SOI MOSFETs suffer from serious self-heating effects due to the buried oxide. This induces some crucial cooling and reliability issues in SOI devices/interconnects.

Fig. 9. Cross section of an SOI MOSFET. tsi, tbox, td and tsub are the thickness of the Silicon film, buried oxide, isolation oxide and Silicon substrate, respectively


An example of heat flow modeling in SOI devices is illustrated in Figures 9 - 11. An SOI structure is given in Figure 9 with isolation oxide indicated by a length of Lox on both sides of the silicon film. Temperature variations along the x direction at y = td and tbox are given in Figure 10 and 2D temperature contours in the isolation oxide for x < 0 are shown in Figure 11. In Figure 10 different values of interconnect thermal resistance are used. Rth is the SOI thermal resistance measured between the film and the substrate, and Rint is the thermal resistance accounting for heat flow from the source, gate or drain to the interconnects.





Fig. 10. Temperature variation along the x direction in the SOI MOSFET at Vgs= 1.5V and Vds=2.4V. Dash/dot lines are from Atlas device simulation coupled with the heat flow equation. The solid lines are from the developed analytic thermal model.

 

 

 

As can been seen, in addition to the high temperature at the channel-drain junction, large temperature variation is observed in the silicon film. This indicates that different amounts of heat flowing out of the gate and drain/source to interconnects are expected. An analytical heat flow model developed by Professor Cheng's group is able to provide very reasonable results compared to those obtained from the rigorous device simulation coupled with the heat flow equation.


Fig. 11. Temperature contours with 6 degree increments in the isolation oxide for -1mm< x< 0. Vgs=1.5V and Vds=2.4V. The substrate temperature To = 300K.

For more information about Professor Ming-Cheng Cheng and his research, you may call him at 315-268-7735 or send email to mcheng@clarkson.edu.

CAMP Professors Regel and Wilcox Model Crystal Growth

For many years, CAMP Professors Liya L. Regel and William R. Wilcox have been combining theoretical modeling with their experimental research on growth of semiconductor crystals, metal alloys, and diamond films. Some of the resulting papers are shown at http://www.clarkson.edu/~regel/papers.htm. Modeling methods have included FORTRAN computer code written by their students, the computer algebra system Maple, FLUENT flow-modeling software, and HYSYS chemical engineering software. A few recent projects are summarized below.

New Method for Deposition of Diamond Films

Professor Regel invented a new method for deposition of polycrystalline diamond films using a closed chamber containing hydrogen gas at about 0.1 atmosphere and an electrically heated graphite rod. With the assistance of CAMP Professor Goodarz Ahmadi, FLUENT was used to calculate the chemical species produced at equilibrium and the buoyancy-driven convection in the chamber. Figure 12 shows an example of a computed flow field.

Figure 12. Flow field for hydrogen gas in diamond deposition chamber.

 

TOP
PREVIOUS PAGE
NEXT PAGE
INDEX PAGE

Page
6