Browse Prior Art Database

Modification to Strip Heated Zone Melt Recrystallization

IP.com Disclosure Number: IPCOM000119339D
Original Publication Date: 1991-Jan-01
Included in the Prior Art Database: 2005-Apr-01
Document File: 3 page(s) / 84K

Publishing Venue

IBM

Related People

Evans, PV: AUTHOR [+2]

Abstract

A modification to the zone melt recrystallization (ZMR) process is disclosed. The purpose is to improve the quality of ZMR-processed silicon on insulator films by eliminating temperature inversion at the solid-liquid interface, so avoiding a cellular or dendritic solidification morphology.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Modification to Strip Heated Zone Melt Recrystallization

      A modification to the zone melt recrystallization (ZMR)
process is disclosed.  The purpose is to improve the quality of
ZMR-processed silicon on insulator films by eliminating temperature
inversion at the solid-liquid interface, so avoiding a cellular or
dendritic solidification morphology.

      In strip-heated ZMR, an amorphous or polycrystalline silicon
film sandwiched between two layers of silicon dioxide on a silicon
substrate is passed under a graphite strip heater.  The power to the
heater and its height above the film are chosen such that a molten
zone exists beneath the strip under steady-state conditions. The
resolidified material is single crystal through the film thickness
with (100) texture, but has a number of high angle grain boundaries,
typically a few mm apart, running parallel to the direction of motion
of the film.  However, in-situ observation of the solid-liquid
interface has revealed that, for a given grain, the solid-liquid
interface is invariably cellular or dendritic (1).  Any solute
partitioned between solid and liquid will thus segregate in the
liquid and accumulate in the intercellular regions.  Further, the
cell walls have been observed to comprise trails of dislocations,
clusters of dislocations or low-angle sub-boundaries, depending on
the exact processing conditions (2).

      The cellular front discussed above is not the result of
constitutional supercooling: the interphase boundary in a stationary
film is also observed to be cellular (1).  Rather it arises from the
difference in reflectivity between the molten and crystalline
silicon, which, in turn, affects the efficiency with which...