Nathaniel J. Carter

Generalized current-voltage analysis and efficiency limitations in non-ideal solar cells: Case of Cu2ZnSn(SxSe1−x)4 and Cu2Zn(SnyGe1−y)(SxSe1−x)4

Detailed electrical characterization of nanoparticle based Cu2ZnSn(SxSe1−x)4 (CZTSSe) and Cu2Zn(SnyGe1−y)(SxSe1−x)4 (CZTGeSSe) solar cells has been conducted to understand the origin of device limitations in this material system. Specifically, temperature dependent current-voltage analysis has been considered, with particular application to the characterization of solar cells with non-ideal device behavior. Due to the presence of such non-ideal device behavior, typical analysis techniques—commonly applied to kesterite-type solar cells—are found to be insufficient to understand performance limitations, and an analysis methodology is presented to account for the non-idealities. Here, the origin of non-ideal device behavior is chiefly considered in terms of electrostatic and band gap potential fluctuations, low minority carrier lifetimes, temperature dependent band edges, high surface/bulk recombination rates, and tunneling enhanced recombination. For CZTSSe and CZTGeSSe, the main limitations to improved dev...

Generalized quantum efficiency analysis for non-ideal solar cells: Case of Cu2ZnSnSe4

Detailed quantum efficiency (QE) analysis of a nanoparticle-based Cu2ZnSnSe4 (CZTSe) solar cell has been conducted to understand photogenerated carrier collection in the device. Specifically, voltage-dependent analysis has been considered to characterize both diffusion limitations and recombination limitations to carrier collection. Application of a generalized QE model and corresponding experimental and analytical procedures are presented to account for non-ideal device behavior, with specific consideration of photogenerated charge trapping, finite absorber thickness, back-surface recombination, and recombination of photogenerated carriers via interface, space-charge-region limited, and/or band tail limited recombination mechanisms. Analysis of diffusion limited collection results in extraction of the minority carrier diffusion length, mobility, back surface recombination velocity, and absorption coefficient. Additionally, forward bias QE measurements afford analysis of the dominant recombination mechanism for photogenerated carriers. For the analyzed CZTSe device, diffusion limitations are not expected to play a significant role in carrier collection in forward bias. However, voltage-dependent carrier collection, previously identified to contribute to open-circuit voltage limitations, is attributed to high recombination rates via band tail states/potential fluctuations in forward bias. A consideration of the assumptions commonly applied to diffusion length, band gap, and band tail extraction is also discussed.

Device comparison of champion nanocrystal-ink based CZTSSe and CIGSSe solar cells: Capacitance spectroscopy

Capacitance spectroscopy has been used to compare charge carrier and defect properties of champion nanocrystal-ink based CZTSSe and CIGSSe solar cells, with efficiencies reported here at 9.2% and 14.2%, respectively. Differences in energy level, frequency/temperature response, and contributions to bulk conductivity have been identified for the different materials. Due to these differences, contributions to the free carrier density have been associated with a single defect for CIGSSe, while associated with two defects in CZTSSe. Additionally, carrier freeze-out out at low temperatures has been identified for both devices, contributing to increasing series resistance at low temperatures as determined from the bulk conductivity. In addition to differences in defect formation, CZTSSe has been characterized with a reduced Vbi when compared to CIGSSe.

The role of interparticle heterogeneities in the selenization pathway of Cu–Zn–Sn–S nanoparticle thin films: a real-time study

Real-time energy dispersive X-ray diffraction (EDXRD) analysis has been utilized to observe the selenization of Cu–Zn–Sn–S nanoparticle films coated from three nanoparticle populations: Cu- and Sn-rich particles roughly 5 nm in size, Zn-rich nanoparticles ranging from 10 to 20 nm in diameter, and a mixture of both types of nanoparticles (roughly 1 : 1 by mass), which corresponds to a synthesis recipe yielding CZTSSe solar cells with reported total-area efficiencies as high as 7.9%. The EDXRD studies presented herein show that the formation of copper selenide intermediates during the selenization of mixed-particle films can be primarily attributed to the small, Cu- and Sn-rich particles. Moreover, the formation of these copper selenide phases represents the first stage of the CZTSSe grain growth mechanism. The large, Zn-rich particles subsequently contribute their composition to form micrometer-sized CZTSSe grains. These findings enable further development of a previously proposed selenization pathway to account for the roles of interparticle heterogeneities, which in turn provides a valuable guide for future optimization of processes to synthesize high quality CZTSSe absorber layers.

Analysis of temperature-dependent current-voltage characteristics for CIGSSe and CZTSSe thin film solar cells from nanocrystal inks

Thin film solar cells with CIGSSe and CZTSSe absorber layers fabricated from nanocrystal inks represent economically scalable technologies for alternative sources of energy. Although these two materials share similar properties important to functioning as a photovoltaic absorber, lab scale CIGSSe devices have achieved power conversion efficiencies 1.5 to 2 times higher than their CZTSSe counterparts. In the current work, CIGSSe and CZTSSe devices similarly processed from nanocrystal inks and exhibiting efficiencies of 14.2% and 9.2%, respectively, are characterized by temperature-dependent current-voltage (IVT) analysis to reveal limitations to CZTSSe device performance compared to CIGSSe.

Characterization of nanocrystal-ink based CZTSSe and CIGSSe solar cells using voltage-dependent admittance spectroscopy

Voltage-dependent admittance spectroscopy has been applied nanocrystal-ink based CZTSSe and CIGSSe solar cells to understand the origins of admittance signatures characterized for devices fabricated from this method; devices considered here have demonstrated champion device efficiencies of 9.0% and 14.2%, respectively. Two admittance signatures characterized for CZTSSe have been attributed to two bulk defect levels; however, a single admittance signature characterized for CIGSSe is expected to be due to a Schottky barrier at the back contact. Additionally, punch-through due to the free carrier freeze-out at low temperatures only occurs in CZTSSe, which suggests a shallower defect contributing to the carrier density in CIGSSe than CZTSSe. Results are compared with that reported for CIGSSe and CZTSSe devices fabricated from other processing techniques.

High efficiency Cu 2 ZnSnS 4 nanocrystal ink solar cells through improved nanoparticle synthesis and selenization

In this work we present the improved efficiency of nanocrystal ink based Cu2ZnSn(S, Se)4 (CZTSSe) solar cells to 9.15%. CZTSSe devices prepared from nanocrystal inks are known for the presence of an unintended amorphous/fine-grained layer (unsintered layer) near the back contact. We demonstrate the ability to reduce the proportion of the unsintered layer in the final film through improved nanocrystal synthesis techniques and tailored thermal annealing in selenium atmosphere (selenization). Interestingly, our selenization process has not led to similar device performance in films prepared from particles using other recipes from the literature.

Current-voltage analysis of band tail effects in CZTSSe through numerical simulation

Band tails in CZTSSe have been commonly attributed as the cause of the low open circuit voltages observed in many CZTSSe solar cells. Various simplified models have been proposed in literature to account for these band tail effects. However, in order to create a physically accurate numerical model for CZTSSe, it is necessary to build a more sophisticated band tail model such as those commonly used in simulation for other highly disordered materials such as a-Si. In this paper, a detailed numerical device model will be used to fit experimental data taken from multiple measurement techniques (i.e., multi-probe) to characterize the band tail distribution. The results of this model will be compared to analytical solutions derived for the band tail distribution for CZTSSe. Developing such a complete model is an important step in addressing the causes of low open circuit voltages, which limit the performance of these devices.