Samira Khelifi

Investigation of defects by admittance spectroscopy measurements in poly (3-hexylthiophene):(6,6)-phenyl C61-butyric acid methyl ester organic solar cells degraded under air exposure

Electrical transport properties of poly (3-hexylthiophene) (P3HT) (6,6)-phenyl C61-butyric acid methyl ester (PCBM) solar cells, with and without encapsulation, have been investigated and analyzed using admittance spectroscopy and capacitance voltage measurements at different temperatures. The admittance spectroscopy clearly reveals two defect states with activation energies of 53 and 100 meV, and a concentration ten times higher in the unencapsulated sample. These defects seem to have a strong effect on the charge transport and the solar cell performance when they are present with a high concentration, since they lead to a decrease of the mobility and also the short-circuit current and the efficiency. The origin of these defects has been assigned to reaction of the blend with O2 which is also known to induce p-type doping in pure P3HT. In an attempt to understand the effect of these defects on the organic solar cell performance, modeling and simulation were carried out using the effective medium layer mo...

Spectral current-voltage analysis of kesterite solar cells

Current–voltage analysis using different optical band pass filters has been performed on Cu2ZnSnSe4 and Cu2ZnSn(S,Se)4 thin-film solar cells. When using red or orange light (i.e. wavelengths above 600 nm), a distortion appears in the I–V curve of the Cu2ZnSnSe4 solar cell, indicating an additional potential barrier to the current flow in the device for these conditions of illumination. This barrier is reduced when using a Cu2ZnSn(S,Se)4 absorber. Numerical simulations demonstrate that the barrier visible under red light could be explained by a positive conduction band offset at the front interface coupled with compensating defects in the buffer layer.

Signature of a back contact barrier in DLTS spectra

The DLTS signal induced by a back contact barrier is studied both theoretically and through experiments on model circuits. A nonideal back contact is modeled either by a resistor and a capacitor, or by a germanium diode inversely polarized with respect to the junction diode. Depending on the back contact properties, this may result in a positive or negative capacitance transient. For these model circuits the capacitance transient time constants and amplitudes are studied as a function of voltage pulse height and compared with signals originating from emission and slow capture from a defect level. These two origins of DLTS signals present very different properties, which opens possibilities to distinguish between them.

Defect distributions in thin film solar cells deduced from admittance measurements under different bias voltages

The voltage dependence of the derivative of the capacitance to (the logarithm of) the measurement frequency is investigated. Relations describing this dependence are derived for the influence of carrier freeze out, of a defect distribution, and of a back contact barrier. The validity of these relations is investigated with numerical simulations.Considering the extraction of the defect density from capacitance–frequency measurements, the extension of existing formulas to different bias voltages leads to an improved accuracy and the possibility to investigate spatial non-uniformities while preserving a direct link between the defect level energy and the apparent defect density. This is illustrated with voltage dependent admittance measurements of thin film Cu(In,Ga)Se2-based solar cell devices.

KCN Chemical Etch for Interface Engineering in Cu2ZnSnSe4 Solar Cells

The removal of secondary phases from the surface of the kesterite crystals is one of the major challenges to improve the performances of Cu2ZnSn(S,Se)4 (CZTSSe) thin film solar cells. In this contribution, the KCN/KOH chemical etching approach, originally developed for the removal of CuxSe phases in Cu(In,Ga)(S,Se)2 thin films, is applied to CZTSe absorbers exhibiting various chemical compositions. Two distinct electrical behaviors were observed on CZTSe/CdS solar cells after treatment: (i) the improvement of the fill factor (FF) after 30 s of etching for the CZTSe absorbers showing initially a distortion of the electrical characteristic; (ii) the progressive degradation of the FF after long treatment time for all Cu-poor CZTSe solar cell samples. The first effect can be attributed to the action of KCN on the absorber, that is found to clean the absorber free surface from most of the secondary phases surrounding the kesterite grains (e.g., Se0, CuxSe, SnSex, SnO2, Cu2SnSe3 phases, excepting the ZnSe-based phases). The second observation was identified as a consequence of the preferential etching of Se, Sn, and Zn from the CZTSe surface by the KOH solution, combined with the modification of the alkali content of the absorber. The formation of a Cu-rich shell at the absorber/buffer layer interface, leading to the increase of the recombination rate at the interface, and the increase in the doping of the absorber layer after etching are found to be at the origin of the deterioration of the FF of the solar cells.

Accuracy of defect distributions measured by bias dependent admittance spectroscopy on thin film solar cells

Thin film solar cells have achieved efficiencies up to 20%. Despite these excellent results, the understanding of the underlying mechanisms and the influence of defects on their performance is still incomplete. In thin film solar cells often defect level distributions are present rather than discrete defects. These distributions can be calculated from admittance measurements, however several assumptions are needed which hinder an exact defect density determination. By performing the measurements under different bias voltage conditions the accuracy of the method can be improved and assessed. This is illustrated with measurements on a flexible thin film Cu(In, Ga)Se2-based (CIGS) solar cell.

Can a multivalent defect be mimicked by several Shockley-Read-Hall-like defects?

The origin of defects present in a device is not always clear. Hence, in electronic device simulation, the influence of defects with more than two different charge states (multivalent defects) is often modeled as a set of defects with only two possible charge states [Shockley–Read–Hall (SRH)-like defects] which follow the SRH statistics. This paper investigates under which circumstances this procedure is allowed, and provides means to check the equivalence between the multivalent and SRH-like description in a fast and efficient way. The procedures are verified simulating a thin film solar cell structure.

Grading and metastable effects in admittance spectroscopy of CIGS-based solar cells

Cu(In, Ga)Se2-based (CIGS) solar cells have achieved efficiencies up to 20%. Despite these excellent results, the understanding of the underlying mechanisms and the influence of defects on their performance is still incomplete. The determination of the energetic position of the defects and of their density of states is important. Admittance spectroscopy is an adequate technique for this. By varying the external voltage during the measurement, the spatial position where the defect distribution is sensed can be varied. However, the application of external biases can lead to metastable effects in the absorber and therefore to defect relaxation and changes in the doping distribution. Hence, it is important to separate between the effects caused by metastable changes and the change in sensing position of the admittance spectroscopy measurement. This can be achieved by varying the applied voltage during the creation of the metastable state and the measurement itself independently or simultaneously. Admittance spectroscopy under different bias voltage conditions performed on a flexible CIGS-based solar cell are presented and assessed.