Hossam Elanzeery

Study of alternative back contacts for thin film Cu2ZnSnSe4-based solar cells

Cu2ZnSnSe4 thin film solar cells are usually fabricated on a soda lime glass substrate with a molybdenum (Mo) back contact. It is suspected that degradation in electrical performance occurs due to the formation of a barrier between the absorber and Mo back contact. To overcome such degradation, Titanium Nitride (TiN), Titanium Tungsten (TiW), Chromium (Cr), Titanium (Ti) and Aluminum (Al) deposited on Mo-coated glass substrates are investigated as alternative back contact materials. Physical and electrical characterization as well as photoluminescence measurements are performed. Compositional analysis of the absorber layer on the metallized substrates identifies Mo, TiN and TiW as being the most inert during the formation of Cu2ZnSnSe4. On the other hand, Ti and Cr reacted with Se during selenization, thereby affecting the growth of the absorber, leading to low conversion efficiency. For Al, the absorber layer was etched after the standard potassium cyanide etch, hence, cannot be used as a back contact. The best device efficiencies obtained are 8.8% on TiN, 7.5% on Mo and 5.9% on TiW, respectively. The TiN back contact provides the lowest barrier value of about 15 meV which could be considered as a good ohmic contact.

Potassium Fluoride Ex Situ Treatment on Both Cu-Rich and Cu-Poor CuInSe2 Thin Film Solar Cells

In this paper, we show that CuInSe2 (CIS) absorbers grown under Cu-excess have better collection efficiencies compared to Cu-poor ones. We also show that an ex situ potassium fluoride postdeposition treatment leads to an improvement in VOC for CIS absorbers grown under both Cu-excess and Cu-poor conditions. Additionally, for absorbers grown under Cu-excess, the junction breakdown, which is observed in reverse bias of untreated cells, is removed. This improvement is based mainly on improving the interface of the CIS absorber grown under Cu-excess to the cadmium sulphide buffer layer through moving the dominant recombination from the interface to the bulk. In contrast to observations in the literature, the treated surface is not completely Cu-free.

Space-charge-limited currents in CIS-based solar cells

Non-linear shunts in Cu(In,Ga)Se2 solar cells have been well described mathematically using the model of a space-charge-limited current, but their physical origin remained unclear so far. We study space-charge-limited currents on Cu-rich CuInSe2 (CIS) devices, which represent a very suitable system: the devices always exhibit non-linear shunts with a very pronounced behavior. Here, we demonstrate a fundamental difference in the transport mechanism between the Cu-rich-based device and the conventional Cu-poor one. We discuss the location of a space-charge-limited current by comparing devices containing various component layers with Ohmic contacts. We confirm that Cu-rich CIS and cadmium sulfide layers alone do not create a non-linear shunt. Our experimental results demonstrate that the origin of the non-linear behavior is located at the interface between the absorber and buffer layers. Temperature dependent current-voltage measurements performed on Cu-rich-based CIS devices are discussed in agreement with ...

Correction to “Potassium Fluoride ex-situ Treatment on both Cu-rich and Cu-poor CuInSe2 Thin Film Solar Cells” [Mar 17 684-689]

Presents corrections to the paper, “Potassium fluoride ex situ treatment on both Cu-Rich and Cu-Poor CuInSe2 thin film solar cells,” (Elanzeery, H., et al), IEEE J. Photovolt., vol. 7, no. 2, pp. 684–689, Mar. 2017.

Interdiffusion and Doping Gradients at the Buffer/Absorber Interface in Thin-Film Solar Cells

An accurate determination of the net dopant concentration in photovoltaic absorbers is critical for understanding and optimizing solar cell performance. The complex device structure of multilayered thin-film solar cells poses challenges to determine the dopant concentration. Capacitance-voltage ( C- V) measurements of Cu(In,Ga)Se2 thin-film solar cells typically yield depth-dependent apparent doping profiles and are not consistent with Hall measurements of bare absorbers. We show that deep defects cannot fully explain these discrepancies. We instead find that the space charge region capacitance follows the model of a linearly graded junction in devices containing a CdS or Zn(O,S) buffer layer, indicating that elemental intermixing at the absorber/buffer interface alters the dopant concentration within the absorber. For absorbers covered with MgF2, C- V measurements indeed agree well with Hall measurements. Photoluminescence measurements of Cu(In,Ga)Se2 absorbers before and after deposition of a CdS layer provide further evidence for a significant reduction of the near-surface net dopant concentration in the presence of CdS. We thus demonstrate that interdiffusion at the absorber/buffer interface is a critical factor to consider in the correct interpretation of doping profiles obtained from C- V analysis in any multilayered solar cell and that the true bulk dopant concentration in thin-film devices might be considerably different.