⇦Chinese Physics Letters, 2017, Vol. 34, No. 7, Article code 076106 Temperature-Dependent Photoluminescence Analysis of 1.0 MeV Electron Irradiation-Induced Nonradiative Recombination Centers in n$^{+}$–p GaAs Middle Cell of GaInP/GaAs/Ge Triple-Junction Solar Cells * Jun-Ling Wang(王君玲), Tian-Cheng Yi(易天成), Yong Zheng(郑勇), Rui Wu(吴锐), Rong Wang(王荣)** Affiliations Key Laboratory of Beam Technology and Materials Modification of Ministry of Education, College of Nuclear Science and Technology, Beijing Normal University, Beijing 100875 Received 2 March 2017 *Supported by the National Natural Science Foundation of China under Grant Nos 11675020, 11375028, 11075018 and 10675023.
**Corresponding author. Email: wangr@bnu.edu.cn Citation Text: Wang J L, Yi T C, Zheng Y, Wu R and Wang R 2017 Chin. Phys. Lett. 34 076106 Abstract The effects of irradiation of 1.0 MeV electrons on the n$^{+}$–p GaAs middle cell of GaInP/GaAs/Ge triple-junction solar cells are investigated by temperature-dependent photoluminescence (PL) measurements in the 10–300 K temperature range. The appearance of thermal quenching of the PL intensity with increasing temperature confirms the presence of a nonradiative recombination center in the cell after the electron irradiation, and the thermal activation energy of the center is determined using the Arrhenius plot of the PL intensity. Furthermore, by comparing the thermal activation and the ionization energies of the defects, the nonradiative recombination center in the n$^{+}$–p GaAs middle cell acting as a primary defect is identified as the E5 electron trap located at $E_{\rm c}-0.96$ eV. DOI:10.1088/0256-307X/34/7/076106 PACS:61.82.Fk, 84.60.Jt © 2017 Chinese Physics Society Article Text Currently, multi-junction solar cells, especially GaInP/GaAs/Ge triple-junction (3J) solar cells with high conversion efficiency and superior radiation resistance have attracted extensive interest in space applications.[1] For space applications, degradation in the performance of solar cells has been observed under electron irradiation,[2] and is directly related to the concentrations of the defects acting as nonradiative recombination centers for low irradiation fluencies (typically below 10$^{16}$ cm$^{-2})$.[3] Our recent work[4] showed that the radiation resistance of GaInP/GaAs/Ge solar cells mainly depends on the n$^{+}$–p GaAs middle cell. For a more efficient use of GaInP/GaAs/Ge triple-junction solar cells in space missions, it is essential to investigate the irradiation induced nonradiative recombination centers of the n$^{+}$–p GaAs middle cell. Photoluminescence (PL) is a powerful tool for the analysis of nonradiative recombination centers in solar cells.[5] In particular, the temperature dependence of PL intensity and position provides valuable information about the nonradiative recombination centers and helps to identify them among all the irradiation-induced defects.[6] Therefore, in this work, to elucidate the behavior of the electron irradiation-induced defects in the n$^{+}$–p GaAs middle cell of GaInP/GaAs/Ge space solar cells, we apply temperature-dependent PL measurements to identify the nonradiative recombination centers induced by 1.0 MeV electron irradiation in the n$^{+}$–p GaAs middle cell. In this study, the n$^{+}$–p GaAs middle cell of GaInP/GaAs/Ge triple-junction solar cells fabricated by the mental-organic chemical vapor deposition (MOCVD) technique are used as the sample. The detailed structure of the solar cells is shown in Fig. 1. The solar cells are irradiated with 1.0 MeV electron beams with the fluence of $5\times10^{14}$ cm$^{-2}$ and a flux density of approximately $5\times10^{10}$ cm$^{-2}$s$^{-1}$, which is low enough to avoid significant sample heating during irradiation.
cpl-34-7-076106-fig1.png
Fig. 1. Schematic structure of the n$^{+}$–p GaAs middle cell (shaded area) of GaInP/GaAs/Ge triple-junction solar cells.
In PL measurements of the GaAs middle cell, a 730 nm (1.70 eV) emission laser line (the power is approximately 100 mW and the beam diameter is approximately 3.0 mm) was used as a typical excitation source. The excitation energy is significantly larger than the band gap of GaAs, leading to strong absorbtion (absorption coefficients $\alpha \approx 10^{5}$ cm$^{-1})$.[7] PL from the p-type base layer of the GaAs solar cells was collected by a lens and then transferred to a grating monochromator with 600 groove/mm grating blazed at 750 nm. The output signal from the monochromator was detected by a Si photodetector and the luminescence was chopped to provide a reference frequency for the lock-in amplifier. To measure the temperature-dependent PL spectra, a closed-cycle cryogenic refrigerator (ARS-4HW) equipped with a digital thermometer controller (Lake Shore, 355 Temperature Controller) was used to control the temperature from 10 K to 300 K, with the temperature step of 10 K and 20 K for the ranges of 10–100 K and 200–300 K, and the temperature step of 10 K from 100 K to 200 K, respectively. The temperature stability is better than 0.1 K. The sample was fixed on the cold head of the cryogenic refrigerator and the thermally conductive adhesive was used to realize thermal conduction between the sample and the cold head.
cpl-34-7-076106-fig2.png
Fig. 2. Typical temperature-dependent PL spectra of the n$^{+}$–p GaAs middle cell of GaInP/GaAs/Ge 3J solar cells irradiated with 1.0 MeV electrons with fluence of $5\times10^{14}$ cm$^{-2}$.
Typical temperature-dependent PL spectra (10 K, 100 K, 160 K, 230 K and 300 K) of the n$^{+}$–p GaAs middle cell of GaInP/GaAs/Ge 3 J solar cells irradiated with 1.0 MeV electrons with the fluence of $5\times10^{14}$ cm$^{-2}$ are shown in Fig. 2. Examination of Fig. 2 shows that the photon energy of the PL peak emission at room temperature (300 K) is 1.38 eV, decreasing by approximately 40 meV relative to the GaAs band gap of 1.42 eV, which is probably due to the In element doping in the 3J solar cells[8] causing a band gap narrowing of the GaAs middle cell. As the temperature increases from 10 K to 100 K, 160 K, 230 K and 300 K, the photon energy of the PL peak decreases by 10 meV, 30 meV, 50 meV and 80 meV, respectively. The red shift of the PL peak position with increasing temperature follows the Varnish equation.[9] Furthermore, the PL intensity also decreases with the increasing temperature. Figure 3 shows all the measured PL intensities of the n$^{+}$–p GaAs middle cell plotted against the inverse temperature in the 10–300 K range in an Arrhenius plot. The data seem to show two separate exponential temperature ranges where the Arrhenius dependence is followed, indicating two thermally activated nonradiative recombination mechanisms. Therefore, the data are analyzed in terms of two thermally active processes ($\alpha$ and $\beta$). The temperature dependence of the PL emission efficiency[10] $\eta (T)$ is given by $$\begin{align} \eta (T)=\,&\Big[1+\kappa_\alpha\exp\Big(-\frac{E_\alpha}{k_{\rm B}T}\Big)\\ &+\kappa_\beta\exp\Big(-\frac{E_\beta}{k_{\rm B}T}\Big)\Big]^{-1},~~ \tag {1} \end{align} $$ where $\kappa_\alpha$ and $\kappa_\beta$ are the ratios of the radiative to nonradiative lifetimes for the $\alpha$ and $\beta$ thermally active processes of nonradiative recombination mechanisms at 300 K, $E_\alpha$ and $E_\beta$ are their respective thermal activation energies, and $k_{\rm B}$ is the Boltzmann constant. The solid line in Fig. 3 represents the fits of the measured data to Eq. (1). From the fitting results, we obtain the parameters $\kappa_\alpha=10$, $\kappa_\beta=10^{16}$, $E_{\alpha}=0.04$ eV and $E_{\beta}=0.96$ eV.
cpl-34-7-076106-fig3.png
Fig. 3. Arrhenius plot of measured temperature-dependent PL intensities of the n$^{+}$–p GaAs middle cell. The solid line represents the fit for the measured data by Eq. (1).
The small value of $\kappa_\alpha$ obtained from the results of $\eta (T)$ fitting indicates that the $\alpha$ center could hardly be considered as an efficient recombination center and its influence on the GaAs solar cell could probably be ignored. Furthermore, the large value of $\kappa_\beta$ indicates that the $\beta$ center is a highly efficient recombination center. The electron traps of the p-type GaAs material induced by 1.0 MeV electron irradiation have previously been investigated,[11] and their ionization energies as deduced from deep-level transient spectroscopy (DLTS) data are summarized in Table 1. As listed in Table 1, E1–E5 are detected in the p-type GaAs material irradiated with electrons. Bourgoin et al.[12] have demonstrated that among these defects, E1–E3 are not deep enough in the gap to be ascribed to recombination centers. Thus the E4 defect and (or) E5 defect are probably the nonradiative recombination centers. According to the multi-center model of semiconductor PL quenching proposed by Schön[13] and Klasens,[14] PL quenching occurs when the activation energy is equal to the ionization energy of the defect in the semiconductor. The analysis of the data obtained following electron irradiation yielded an $\alpha$ defect level close to 0.04 eV and a $\beta$ defect level close to 0.96 eV. Obviously, the $\alpha$ defect is not expected to play a significant role in the operation of the cell, because the nonradiative defects located closer to the mid-gap energy are expected to control the solar cell performance.[12] The $\beta$ defect is consistent with the E5 ($E_{\rm c}-0.96$ eV) that is characterized by the DLTS studies of 1.0 MeV electron irradiation on p-type GaAs. This suggests that the E5 defect acts as a nonradiative recombination center (or the minority carrier capture center) of the p-type GaAs middle cell introduced by electron irradiation.
Table 1. Characteristics of the electron traps induced in the p-type GaAs solar cell by electron irradiation obtained by DLTS.[11] Here $E_{\rm i}$ is the ionization energy of the defect.
Defects E1 E2 E3 E4 E5
$E_{\rm i}$ (eV) 0.045 0.140 0.300 0.760 0.960
We thus obtain that the dominant nonradiative recombination center of the p-type GaAs middle cell is defect E5 that lies at 0.96 eV below the conduction band according to the temperature-dependent PL measurements. This result is consistent with the result of Ref. [15] where the E5 defect was found by room temperature PL measurement to control the radiation response and annealing properties of the p-type GaAs middle cell for 3J solar cells after injection-enhanced annealing. As shown in a previous investigation, to identify the nonradiative recombination centers, the room-temperature PL method was applied to study the injection-enhanced annealing behavior of solar cells[15] or to analyze the capture cross-section of the minority carrier in the solar cells irradiated with electrons at different energies and fluences,[5] whereas the temperature-dependent PL measurements could not only identify the nonradiative recombination centers but also reveal the activation process of the nonradiative recombination centers in a solar cell. In conclusion, the temperature-dependent PL measurement of 1.0 MeV electron-irradiation induced nonradiative recombination center in the n$^{+}$–p GaAs middle cell of GaInP/GaAs/Ge triple-junction solar cells has been studied over a wide temperature range from 10 K to 300 K. The Arrhenius plots of PL intensity are used to determine the thermal activation energy of the defects. By comparing the thermal activation energy and the ionization energy of the defects, the nonradiative recombination center is identified as the E5 electron trap located at $E_{\rm c}-0.96$ eV in the p-type GaAs middle cell. It has been clearly demonstrated that the temperature-dependent PL measurement could better identify the nonradiative recombination center in electron-irradiated solar cells than the room-temperature PL measurement. References Multi-junction III?V solar cells: current status and future potentialRadiation-resistant solar cells for space usePrediction of solar cell degradation in space from the electron?proton equivalenceRadiation Damage Analysis of Individual Subcells for GaInP/GaAs/Ge Solar Cells Using Photoluminescence MeasurementsPhotoluminescence analysis of electron irradiation-induced defects in GaAs/Ge space solar cellsTemperature dependence of defect-related photoluminescence in III-V and II-VI semiconductorsHigh-efficiency InGaP/In0.01Ga0.99As tandem solar cells lattice-matched to Ge substratesTemperature dependence of the energy gap in semiconductorsSpatial distribution of radiation-induced defects in p+-n InGaP solar cellsIrradiation-induced defects in GaAsIrradiation-induced degradation in solar cell: characterization of recombination centresZum Leuchtmechanismus der KristallphosphoreTransfer of Energy Between Centres in Zinc Sulphide PhosphorsPhotoluminescence Analysis of Injection-Enhanced Annealing of Electron Irradiation-Induced Defects in GaAs Middle Cells for Triple-Junction Solar Cells
[1] Yamaguchi M, Takamoto T, Araki K and Daukes N 2005 Sol. Energy 79 78
[2] Yamaguchi M 2001 Sol. Energy Mater. Sol. Cells 68 31
[3] Mbarki M, Sun G C and Bourgoin J C 2004 Semicond. Sci. Technol. 19 1081
[4] Zheng Y, Yi T C, Wang J L, Xiao P F and Wang R 2017 Chin. Phys. Lett. 34 026101
[5] Lu M, Wang R, Yang K and Yi T C 2013 Nucl. Instrum. Methods Phys. Res. Sect. B 312 137
[6] Reshchikov M A 2014 J. Appl. Phys. 115 012010
[7]Anspaugh B E 1996 GaAs Solar Cell Radiation Handbook (California: JPL Publication)
[8] Takamoto T, Agui T, Ikeda E and Kurita H 2001 Sol. Energy Mater. Sol. Cells 66 511
[9] Varnish Y P 1967 Physica 34 149
[10] Romero M J, Araújo D, GarcıíA R, Walters R J, Summers G P and Messenger S R 1999 Appl. Phys. Lett. 74 3812
[11] Pons D and Bourgoin J C 1985 J. Phys. C 18 3839
[12] Bourgoin J C and Zazoui M 2002 Semicond. Sci. Technol. 17 453
[13] Schön M 1942 Z. Phys. A 119 463
[14] Klasens H A 1946 Nature 158 306
[15] Zheng Y, Yi T C, Xiao P F, Tang J and Wang R 2016 Chin. Phys. Lett. 33 056102