Cookies

We use cookies to ensure that we give you the best experience on our website. By continuing to browse this repository, you give consent for essential cookies to be used. You can read more about our Privacy and Cookie Policy.


Durham e-Theses
You are in:

Electronic conduction in silicon-rich thin films

Buchanan, D.A (1986) Electronic conduction in silicon-rich thin films. Doctoral thesis, Durham University.

[img]
Preview
PDF
6Mb

Abstract

Silicon-rich silicon nitride (SRN) films were grown by low pressure chemical vapour deposition (LPCVD) with excess silicon concentrations varying from 8.8% to 12.8% by varying the reactant gas phase ratio from R(_n) = 4.0 to R(_n) = 0.25. Dichlorsilane and ammonia were the reactant gases and nitrogen was used as the carrier gas. All films were found to be predominantly ε-Si(_3)N(_9) with free silicon crystallites being found in the films with the greatest silicon content. The conduction mechanism at high temperatures and electric fields is due to Poole-Frenkel emission of trapped electrons (holes) from relatively deep defect levels to the conduction (valence) band tails. From a steady-state analysis the effective trap depth, Ф(_t). was found to be approximately 1.1 eV and it decreased slightly with increasing silicon content of the films. Low values of calculated dielectric constant were found and attributed to the build up of space charge near the injecting contact. For thin films ((^.)1000 Å) steady state analysis cannot be considered accurate unless the effects of trapped space charge are taken into account. For positive (negative) applied bias voltages, the flat band shift is in a positive (negative) direction implying a net increase in negative (positive) charge within the SRN film. A logarithmic time dependence is found for the transient flat band shift while for long periods of time (i.e. t , 1s) the current transient was inversely proportional to time. A charge trapping model is presented which predicts a logarithmic increase in the flat band shift with time. The model is based on the assumption that charge carriers, holes for negative bias and electrons for positive bias, tunnel from the silicon valence and conduction bands into the deep defect levelsin the SRN film. Very good agreement was found between the data and the model for low electric fields. At high electric fields, the situation becomes complicated by Poole-Frenkel 're-emission' from the traps which leads to a 'saturation of the flat band shift with time. At high electric fields, the current transient also becomes dominated by Poole-Frenkel emission of trapped carriers. It was also found from the current transients that the calculated dynamic dielectric constant decreases with increasing time reflecting the increase in trapped charge. From the Schottky curves for times of one second. E(_d) was found to vary from 5.09 to 5.26 for excess silicon content 8.8% to 12.3%. For these SRN films the density of trapping centres near the SRN-silicon interface was found to be of the order of 3x10 19om-3.Silicon-rich silicon oxide (SRO) films were grown in a atmospheric pressure CVD reactor using nitrous oxide and silane as the reactants and nitrogen as the main carrier gas. The conduction in silicon-rich oxide (SRO) was investigated using standard I-V techniques. Three models of conduction were investigated. These were a symmetrical Schottky barrier model similar to that used for polycrystalline silicon, a model based on a type of Fowler-Nordheim tunnelling between silicon crystallites in the SRO film and a model based on Poole-Frenkel emission from the silicon crystallites into the conduction band or band tails of the film. It was extremely difficult to assess the true nature of the conduction in the silicon-rich oxide (SRO) films. However, the evidence seems to point towards the Poole-Frenkel mechanism as being responsible for conduction.

Item Type:Thesis (Doctoral)
Award:Doctor of Philosophy
Thesis Date:1986
Copyright:Copyright of this thesis is held by the author
Deposited On:15 May 2013 14:13

Social bookmarking: del.icio.usConnoteaBibSonomyCiteULikeFacebookTwitter