元素和化合物半导体薄膜的激光退火

时间:2023-05-24 09:07:09 浏览量:0

Although the A1 - Sb system was chosen for this experiment partly because of  the remarkably close melting points of A1  and Sb, we have recently observed a similar laser induced transformation for the  system AlAs, and expect the result to be  quite qeneral for similar systems.


Although comparatively few irradiations  were performed with the ruby laser, and  complications due to poor pulse uniformity and reproductibility arose with the  dye laser, the two types of irradiation  appear to produce more or less the same  effects for the same pulse energy.


Observations by TEM and HRED revealed  no detectable structure changes (≤0.7%  change in amorphous structure factor  up to pulse energies (dye only) of s 100  and ~10 mJ/cm2 for supported and unsupported films, respectively. However  measurements on supported films at this  and lower energies showed both reversible and, with increasing pulse energy,  apparently irreversible changes in film  conductivity. Reversible but very  long duration (up to 107s ! ) photoconductivity effects are known to occur in amorphous elemental semiconductors  /5/, and may possibly be related to a  redistribution of essentially point defects (i. e. dangling bonds) , although  the material remains amornhous.  


At pulse energies exceeding those in  b) crystallised regions appear in the  films for both types of irradiation,  fisure I.Observe firstly that the general form  of these "stars" indicates very rapid  qrowth. Each one is evidently associated with a single nucleation event occuring at the centre followed by rapid  qrain growth outwards and terminating  in a boundary of randomly oriented microcrystals. Note also that the density  of stars, and hence that of the associated nucleation events, is very low  compared to volume or surface atomic  3.ensities (tysically 10-9 even for this  latter) in the films. Yoreover, a subsequent pulse irradiation has no  effect on stars already formed but results in a seemingly random distribution of new ones. Thus the nucleation  event seems to be statistically governed rather than associated with any  particular type of structural or compositional defect. A further interesting feature is that the star diameter  is not only independent of laser power  (dye vs. ruby) but is also essentially  independent of pulse energy. Neither  is it related to the pulse duration  which in any case would imply growth  rates of several m/s even for the dye  laser, a value untypically high for a  covalently bonded material.


3

fig1


There is much current speculation on the mechanisms of laser annealing,  particularly as regards the nucleation  process. It is perhaps self evident that  the first process occuring in a solid  subjected to laser irradiation is optically  induced electronic excitation and that,  at some later time, the energy will have  been transferred to the lattice phonons so  that the solid is hot in a purely classical sense. The points that are unclear are  the time interval involved between these  initial and final. states and thus whether,  depending on hot electron diffusion rates,  the resulting thermal effects can be sufficient to account for the annealing  effects observed, whether the very high  hot electron densities encountered can result in annealing effects quite independently of any lattice heating, and, more  interestingly, whether the annealing  observed differs in any way from that expected and found in a purely "thermal"  anneal.


Published data for A1 and Sb  diffusion in AlSb give activation enerjies  of 1.88 and 1.7 eV respectively and suggest  the possibility of forming AlSb by solid  diffusion through the growing compound  layer at temperatures around 1300 K in lps  For the crystallisation of a-Ge the temperature "necessary" is rather more difficult to estinate but by comparison with  thermal data can be safely assumed to be  above % 700 K.

文件下载请联系管理员: 400-876-8096