Data on dopant characteristics and band alignment of CdTe cells with and without a ZnO highly-resistive-transparent buffer layer.

Photovoltaic enhancement of cadmium telluride (CdTe) thin film solar cells using a 50 nm thick, atomic-layer-deposited zinc oxide (ZnO) buffer film was reported in "Enhancement of the photocurrent and efficiency of CdTe solar cells suppressing the front contact reflection using a highly-resistive ZnO buffer layer" (Kartopu et al., 2019) [1]. Data presented here are the dopant profiles of two solar cells prepared side-by-side, one with and one without the ZnO highly resistive transparent (HRT) buffer, which displayed an open-circuit potential (Voc) difference of 25 mV (in favor of the no-buffer device), as well as their simulated device data. The concentration of absorber dopant atoms (arsenic) was measured using the secondary ion mass spectroscopy (SIMS) method, while the density of active dopants was calculated from the capacitance-voltage (CV) measurements. The solar cell simulation data was obtained using the SCAPS software, a one-dimensional solar cell simulation programme. The presented data indicates a small loss (around 20 mV) of Voc for the HRT buffered cells.


a b s t r a c t
Photovoltaic enhancement of cadmium telluride (CdTe) thin film solar cells using a 50 nm thick, atomic-layer-deposited zinc oxide (ZnO) buffer film was reported in "Enhancement of the photocurrent and efficiency of CdTe solar cells suppressing the front contact reflection using a highly-resistive ZnO buffer layer" (Kartopu et al., 2019) [1].
Data presented here are the dopant profiles of two solar cells prepared side-by-side, one with and one without the ZnO highly resistive transparent (HRT) buffer, which displayed an open-circuit potential (V oc ) difference of 25 mV (in favor of the no-buffer device), as well as their simulated device data. The concentration of absorber dopant atoms (arsenic) was measured using the secondary ion mass spectroscopy (SIMS) method, while the density of active dopants was calculated from the capacitance-voltage (CV) measurements. The solar cell simulation data was obtained using the SCAPS software, a one-dimensional solar cell simulation programme. The presented data indicates a small loss (around 20 mV) of V oc for the HRT   Value of the data The SIMS can provide arsenic density in CdTe:As at a detection limit of $ 1 Â 10 16 As/cm 3 The C-V curves (1/C 2 vs. V) can be analyzed to estimate the acceptor density in the absorber Ratio of the acceptor density (C-V result) to dopant atom density (SIMS result) provides an estimate of efficiency of dopant activation The good sensitivity of SIMS and C-V methods makes them powerful in investigating dopingrelated issues with CdTe thin film solar cells SCAPS helps to visualize the band-alignment, and to quickly assess influence of various material parameters on the cell performance, guiding experimental solar cell research.
1. Data   Fig. 1a shows the distribution of As dopant atoms within the solar cells' CdTe absorber layer. It shows that less As is incorporated for the sample containing the ZnO HRT buffer film. The net acceptor density (N A ), shown in Fig. 1b, tracks the As profiles in Fig. 1a in that the N A for the cell with HRT buffer is lower ( $ 1.0 Â 10 16 vs. 1.7 Â 10 16 cm -3 ).
Device simulations by SCAPS was carried out to inspect the band-alignment and calculate device parameters. The band-alignment of device structures with and without a 50 nm ZnO buffer film is given in Fig. 2. Small energy spikes are seen to be introduced in the conduction band at the layer interfaces to the ZnO film.
In device simulations, if the N A is kept constant at 1 Â 10 16 cm À 3 , addition of the ZnO layer and associated the spikes in the conduction band near the buffer layer did not change the device V oc . When experimental N A values from Fig. 1b (i.e. 1.0 Â 10 16 and 1.7 Â 10 16 cm -3 for the HRT and no HRT cases, respectively) are used instead, then a V oc loss of $ 22 mV was calculated for the cell with the HRT layer.

Experimental Design, Materials, And Methods
Arsenic profiling in the solar cells was collected with a Cameca IMS-4f secondary ion-mass spectrometer at LSA Ltd. Cs þ ions, at 10 keV energy obtained with 20 nA current, were used as the primary ions. The specimen was 1 Â 1 cm 2 in size, cleaved from the main coupon, and etched in 0.2% Br 2 (in methanol) for 5 s to polish the sample surface and increase the depth resolution. A previously characterized CdTe:As layer (supplied by CSER to LSA Ltd.) was used for calibration.
CV data were collected from the solar cell structures in dark and at room temperature conditions using a Solartron Impedance Analyzer. An ac bias amplitude of 10 mV was applied at 300 kHz whilst the DC bias (V appl ) was swept from -3V to þ1 V. Linear part of the 1/C 2 vs. V plot (near V appl ¼ 0 V) was then analyzed using the procedure described in Ref. [2], to extract the net acceptor density N A .  The SCAPS programme, used to simulate solar cell characteristics, is a one-dimensional solar cell simulation programme and freely available online through its owner, Prof. Marc Burgelman, University of Gent, Belgium [3]. The layer parameters used in these simulations are given in Refs. [1,4].