Thesis defence M. van Sebille: silicon

03 March 2017 | 10:00
location: Aula, TU Delft
by Webredactie

Silicon nanocrystals embedded in silicon alloys. Promotor: Prof.dr. M. Zeman (EWI).

Direct conversion of light into electricity is one of the most promising approaches to provide renewable energy on a large scale. Solar-cells are devices that use the photovoltaic effect to convert sunlight into electricity. Single-junction solar-cells all suffer from spectral mismatch, reducing their cell’s efficiency. Photons with lower energy than the absorber material’s band gap will be transmitted and photons with higher energy than the band gap will lose the excess energy through thermalization processes as heat. One solution to prevent excessive thermalization is to use multiple absorber materials with varying band gaps. This can be achieved using silicon nanocrystals embedded in a dielectric matrix made of silicon and its compounds with oxygen, nitrogen and carbon. The different band gaps needed for efficient spectral matching can be accomplished by utilizing the size-dependent quantum confinement in nanometer-sized crystals.

Using films containing alternating layers of stoichiometric and silicon-rich silicon alloys allows for the control over the nanocrystal size, limited to the siliconrich layer thickness. Although no clear consensus exists concerning the exact charge carrier transport mechanisms, the total charge transport is expected to be highly dependent on the nanocrystal spacing and the choice of dielectric material. The nanocrystal density in the silicon-rich layers can be controlled by tuning the composition of these layers during deposition. A low silicon content leads to relatively few isolated nanocrystals, and increasing the excess silicon content will eventually lead to clustering of nanocrystals. When the nanocrystal density is too low, the probability of a nearest-neighbor nanocrystal within the transport-distance is too low. In contrast, when the excess silicon content is too high, nanocrystals are so closely spaced that they start clustering, which reduces the quantum confinement in these crystals. This means there is an optimal composition to achieve a limited nanocrystals spacing, while limiting clustering. In chapter 3 we demonstrate an analytical method to optimize the stoichiometry and thickness of multilayer silicon oxide films in order to achieve the highest density of non-touching and closely spaced silicon nanocrystals after annealing. The probability of a nanocrystal nearest-neighbor distance within a limited range is calculated using the stoichiometry of the as-deposited film and the crystallinity of the annealed film as input parameters. Multiplying this probability with the nanocrystal density results in the density of non-touching and closely spaced silicon nanocrystals.

Limited by the nanometer-scale dimensions of nanocrystals, transmission electron microscopy (TEM) is the only direct measurement tool capable of capturing the size and shape of embedded nanocrystals. However, a quick method to measure nanocrystals in TEM images with minimal user input to minimize user bias has been lacking. In chapter 4 we propose a method with minimal bias caused by user input to quickly detect and measure the nanocrystal size distribution from transmission electron microscopy images using a combination of Laplacian of Gaussian filters and non-maximum suppression. We demonstrate the proposed method on bright-field TEM images of an a-SiC:H sample containing embedded silicon nanocrystals with varying magnifications and we compare the accuracy and speed with size distributions obtained by manual measurements, a thresholding method and PEBBLES. Finally, we analytically consider the error induced by slicing nanocrystals during TEM sample preparation on the measured nanocrystal size distribution and formulate an equation to correct for this effect.

To the best of our knowledge, a method to obtain the nanocrystal absorption properties and their density of states from absorption spectra has not been developed yet. In chapter 5 we present a non-destructive measurement and simple analysis method for obtaining the absorption coefficient of silicon nanocrystals embedded in an amorphous matrix. This method enables us to pinpoint the contribution of silicon nanocrystals to the absorption spectrum of nanocrystal containing films. The density of states (DOS) of the amorphous matrix is modeled using the standard model for amorphous silicon while the nanocrystals are modeled using one Gaussian distribution for the occupied states and one for the unoccupied states. For laser annealed a-Si0.66O0.34:H films, our analysis shows a reduction of the nanocrystal band gap from approximately 2.34 to 2.08 eV indicating larger mean nanocrystal size for increasing annealing laser fluences, accompanied by a reduction in nanocrystal DOS distribution width from 0.28 to 0.26 eV, indicating a narrower size distribution.


Embedded silicon nanocrystals can be made by annealing silicon-rich silicon alloy films. Since hydrogen effusion occurs at lower temperatures than phase separation and crystallization, this cannot be avoided, leading to an increased defect density. Reincorporation of hydrogen into the material is considered to be an effective method to reduce the defect density. One option is to combine annealing and hydrogen passivation in a single processing step, by annealing in a H2 containing atmosphere. In chapter 6 we report the effect of hydrogen on the crystallization process of silicon nanocrystals embedded in a silicon oxide matrix. We show that hydrogen gas during annealing leads to a lower sub-band gap absorption, indicating passivation of defects created during annealing. Samples annealed in pure nitrogen show expected trends according to crystallization theory. Samples annealed in forming gas, however, deviate from this trend. Their crystallinity decreases for increased annealing time. Furthermore, we observe a decrease in the mean nanocrystal size and the size distribution broadens, indicating that hydrogen causes a size reduction of the silicon nanocrystals.

More information?
For access to theses by the PhD students you can have a look in TU Delft Repository, the digital storage of publications of TU Delft. Theses will be available within a few weeks after the actual thesis defence. 

© 2017 TU Delft