Relationship between Raman crystallinity and open-circuit voltage in microcrystalline silicon solar cells

Abstract

A series of nip-type microcrystalline silicon (mc-Si:H) single-junction solar cells has been studied by electrical characterisation, by transmission electron microscopy (TEM) and by Raman spectroscopy using 514 and 633 nm excitation light and both top-and bottomillumination. Thereby, a Raman crystallinity factor indicative of crystalline volume fraction is introduced and applied to the interface regions, i.e. to the mixed amorphous-microcrystalline layers at the top and at the bottom of entire cells. Results are compared with TEM observations for one of the solar cells. Similar Raman and electrical investigations have been conducted also on pin-type mc-Si:H single-junction solar cells. Experimental data show that for all nip and pin mc-Si:H solar cells, the open-circuit voltage linearly decreases as the average of the Raman crystallinity factors for top and bottom interface regions increases.

Microcrystalline silicon solar cells: effect of substrate temperature on cracks and their role in post‐oxidation

“…In parallel, the crystallinity increases for the 3 first temperatures investigated and leads to a decrease of V oc in the solar cells performances. The relation between the crystallinity and the V oc is already well known [31]. An optimization of Raman crystallinity is hence necessary to improve the electrical performance of the solar cells deposited at higher temperatures in order to benefit to both increase of FF and high V oc .…”

“…A larger substrate temperature induces an increased Raman crystallinity of the solar cells, with a saturation to 50-55% at 2708C (average value as measured from the glass and cell side with 633 nm laser). The losses of V oc observed in J(V) curves (see Figure 5b) can be related to the increase of crystallinity [31]. However, crystallinity cannot be held responsible for the variations of silicon density (i.e., number of cracks).…”

Determination of Raman emission cross-section ratio in hydrogenated microcrystalline silicon

“…This sample was chosen as it presents a microstructure consisting of a mix between amorphous and microcrystalline phases. Raman spectra were obtained in a commercial Renishaw Raman microscope in the back-scattering configuration using both a HeNe laser (excitation wavelength 633 nm) and an Ar laser (excitation wavelength 514 nm) in the bifacial mode [1]. The three peaks were deconvoluted with a commercial software (GRAMS) assuming a Gaussian shape.…”

“…The low-wavenumber tail of this peak (around 510 cm À1 ) is attributed to the defective but yet crystalline part of the nanocrystals. Integrated intensities of both phases (I a and I c , respectively) can be easily evaluated from the Raman spectra, and their ratio yields the so-called 'Raman crystallinity factor' / c [1]. This value is not identical with the actual crystalline volume fraction X c which involves the additional parameter y in the following way [2,3]: X c = I c /(I c + yI a ) where y is the 'Raman emission cross-section ratio'.…”

Kinetics of creation and of thermal annealing of light-induced defects in microcrystalline silicon solar cells

“…24 Bi-facial micro-Raman spectroscopy was performed with a 18 mW HeNe laser excitation beam ͑633 nm͒ using a long focal length objective for the measurements on the glass side. 25 Micro-Raman spectroscopy was used to evaluate the average crystallinity factor c for the intrinsic layer. Three Raman peaks were measured and deconvoluted as Gaussian peaks: one at 480 cm −1 ͑TO mode in amorphous silicon͒, one at 510 cm −1 ͑nanocrystals͒, and the third one at 520 cm −1 ͑TO mode in bulk crystalline silicon͒.…”