Altering the nucleation of thermally annealed hydrogenated amorphous silicon with laser processing

Abstract

We demonstrate the use of laser processing to affect the nucleation of crystallites in thermally annealed hydrogenated amorphous silicon (a-Si:H) thin films. The influence of film H content and subcrystallization threshold laser fluence are investigated by x-ray diffraction measurements during in situ thermal annealing at 600 °C. All laser-treated films show a reduced incubation time for crystallization compared to as-grown films, with the largest differences exhibited for samples with higher film H and higher laser fluences. These results are consistent with multivacancy annihilation by laser processing, based upon a recently developed model for a nucleation center in a-Si:H.

The effect of the film H content on the crystallization of laser processed and thermally annealed HWCVD a‐Si:H

“…While detailed TEM measurements of such a film, after laser illumination but before annealing, have not yet been performed, the identical baseline values (~ 100 on the y-axis) for both films before crystallization is initiated strongly suggests that laser illumination does not produce crystallites directly, but rather changes the nucleation density such that the laser treated film exhibits a shorter τ 0 . The same baselines (and baseline trends) exhibited with and without laser processing [9] for films with as grown film C H s smaller than that for the present sample (the laser processed films exhibiting longer baselines before an increase in the silicon (111) peak amplitude appears) lends further credence to this hypothesis.…”

“…In a previous publication [9], we revisited the approach of Efremov et al [8], and found that the reduction in τ 0 generated by laserinduced nucleation without crystallization can be explained by multivacancy annihilation, in the context of a recent model describing the nature of a nucleation center [10].…”

The effect of film tensile stress on crystallite nucleation and growth in thermally annealed a-Si:H

“…The possible effect of oxygen incorporation into the film upon annealing, and its potential role in affecting crystallite nucleation and growth in different areas of the film (film edge, away from edge), has previously been addressed. 13,14 After each anneal, the films were examined by optical microscopy using a Nikon Eclipse LV100 Optical Microscope, in the reflectance mode, and the images were collected using NIS-Elements F2.20 software. 13 The crystallites were counted within a 35 lm  35 lm grid area far away from the edge, and in close proximity to either a film edge or a laser grid placed upon the film; in a previous report, 14 the application of this laser grid introduced before film annealing was also shown to provide stress relief.…”

“…In this model, r n and s g are assumed to be time independent after the onset of nucleation and enable a determination of the final grain size. While some differences in final grain size have been reported and have been attributed largely to variations in r n , one of the most striking observations for thermally annealed a-Si:H films has been the wide range in s o , with values extending from less than 1 h to as long as 10-15 h at an anneal temperature of 600 C. [10][11][12][13][14][15][16] Various reasons have been proposed to explain this large variation, including variations in the film H content. Based upon a structural model in which the local film inhomogeneity is defined by the initial H distributions in the as grown films, a recent report 15 proposed that nucleation center sites were the more well ordered regions in the films that did not contain clustered H sites, as probed by H NMR spectroscopy, and that the frequency and sizes of these sites directly controlled the film s o .…”

Solid-phase crystallization of ultra high growth rate amorphous silicon films

“…0.85 mbar. If nuclei were present in the a-Si:H matrix prior to the SPC treatment, the onset in crystallization would be expected to occur at earlier stages, as observed by Spinella et al 32 and Dabney et al 32,33 On the basis of the above-mentioned observations, it can be concluded that the nuclei development during the plasma deposition of a-Si:H in the pressure range of 0.17-0.47 mbar can be excluded. Therefore, a priori nucleation (i.e., a high nuclei density) is not the cause for the decrease in grain size for R* values larger than 0.32.…”