Defect related radiative recombination in mono‐like crystalline silicon wafers. Issue 8 (27th April 2017)

• Record Type: Journal Article
• Authors: Olsen, E.; Bergan, S.; Mehl, T.; Burud, I.; Ekstrøm, K. E.; Di Sabatino, M.
• Is Part Of: Physica status solidi. Volume 214:Issue 8(2017)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2017-04-27
• Subjects: defects -- photoluminescence -- radiative recombination -- silicon -- wafers; Solid state physics -- Periodicals; Solids -- Industrial applications -- Periodicals; 530.41
• DOI: 10.1002/pssa.201700124
• Languages: English
• ISSNs: 1862-6300

Abstract

Abstract : We report on studies of sub‐bandgap defect related photoluminescence (DRL) signals originating from radiative recombination through traps in the bandgap of cooled mono‐like silicon wafers. Spectrally resolved photoluminescence (SPL) and multivariate curve resolution (MCR) have been used in combination, to study the behaviour of sub‐bandgap photoluminescence (PL) emissions in wafers cut from different heights in a pilot‐scale mono‐like silicon ingot. No DRL signals were found in the main mono‐like body. Strong defect related sub‐bandgap emissions correlating with heavily dislocated areas, are found directly above some of the seed junctions. The DRL signal exhibits a correlation with the number of axis with small angle misalignment in the junctions of the seeds. The signal conventionally labelled D1 (0.80 eV) decreases with ingot height. A mechanism relating this signal to oxygen is proposed. The signals D3 (0.94 eV) and D4 (1.00 eV) are found to co‐occur, supporting previous studies, and similarly to the D2 (0.87 eV) signal, their strength is found to increase with ingot height. As the content of the transition metal impurities in the ingot is supposed to increase with height, this supports a reported link between the D3 and D4 signals with Fe, as well as a link between D2 and other impurities. An emission previously found in multicrystalline material and labelled D07 (0.70 eV), is found to solely exist as the only DRL signal recorded by us in parasitic crystals, growing into the main mono‐like ingot from the crucible walls. This contradicts the common notion that the D1–D4 signals are strongly related to, and always follow dislocations. Total photoluminescence spectrum (right) and distribution (left) of the PL signal with centre energy 0.70 eV emanating from the parasitic crystals growing into the bulk mono‐like Si crystal from the crucible walls.