⇦Chinese Physics Letters, 2017, Vol. 34, No. 2, Article code 026101 Radiation Damage Analysis of Individual Subcells for GaInP/GaAs/Ge Solar Cells Using Photoluminescence Measurements * Yong Zheng(郑勇)1,2, Tian-Cheng Yi(易天成)1,2, Jun-Ling Wang(王君玲)1,2, Peng-Fei Xiao(肖鹏飞)1,2, Rong Wang(王荣)1,2** Affiliations 1Key Laboratory of Beam Technology and Materials Modification of Ministry of Education, College of Nuclear Science and Technology, Beijing Normal University, Beijing 100875 2Beijing Radiation Center, Beijing 100875 Received 25 October 2016 *Supported by the National Natural Science Foundation of China under Grant Nos 10675023, 11075018, 11375028 and 11675020, and the Specialized Research Fund for the Doctoral Program of Higher Education under Grant No 20120003110011.
**Corresponding author. Email: wangr@bnu.edu.cn Citation Text: Zheng Y, Yi T C, Wang J L, Xiao P F and Wang R 2017 Chin. Phys. Lett. 34 026101 Abstract The radiation damage of three individual subcells for GaInP/GaAs/Ge triple-junction solar cells irradiated with electrons and protons is investigated using photoluminescence (PL) measurements. The PL spectra of each subcell are obtained using different excitation lasers. The PL intensity has a fast degradation after irradiation, and decreases as the displacement damage dose increases. Furthermore, the normalized PL intensity varying with the displacement damage dose is analyzed in detail, and then the lifetime damage coefficients of the recombination centers for GaInP top-cell, GaAs mid-cell and Ge bottom-cell of the triple-junction solar cells are determined from the PL radiative efficiency. DOI:10.1088/0256-307X/34/2/026101 PACS:61.82.Fk, 84.60.Jt © 2017 Chinese Physics Society Article Text Recently, multi-junction (MJ) solar cells, especially the GaInP/GaAs/Ge triple-junction solar cells, have attracted broad interests in space mission, owing to their high conversion efficiency and superior radiation resistance.[1] To utilize the GaInP/GaAs/Ge solar cells more efficiently in space mission, the degradation induced by electron and proton irradiations in the performance of solar cells has been investigated.[2] However, the capability to estimate the particle-induced degradation on each individual subcells is important for improving radiation resistance of GaInP/GaAs/Ge triple-junction solar cells, although the investigation in subcells is very difficult because of their complex structure. Therefore, it is essential to monitor not only damage of the entire triple-junction cell, but also that of each individual cell which composes it. As a matter of fact, the degradation of solar cells is mainly due to displacement damage effects caused by ion irradiation, and the recombination centers induced by displacement damage reduce the diffusion and lifetime of the minority carriers.[3] The displacement damage dose ($D_{\rm d}$) approach has provided a mean to investigate the degradation of GaInP/GaAs/Ge solar cells.[4] Here $D_{\rm d}$ calculations were obtained by multiplying the fluence values by appropriate non-ionizing energy loss (NIEL) values linearly.[5] Thus $D_{\rm d}$ is used to analyze the lifetime damage of the solar cells. Photoluminescence (PL) spectroscopy is known to be one of the most sensitive techniques for analyzing both the intrinsic and extrinsic properties of semiconductors. This method has the advantages of non-destructive high-efficiency simple laser source supply. In addition, luminescence is indirectly correlated with the radiation damage of each subcell, and independently of the others.[6] In this work, we apply the PL to analyze the irradiation damage of individual subcells for the GaInP/GaAs/Ge triple-junction solar cells. The samples used in this study were $20\times20$ mm$^{2}$ GaInP/GaAs/Ge triple-junction solar cells fabricated using the metal organic chemical vapor deposition (MOCVD). The detailed structure of the solar cells was shown in Ref. [7]. To assess the radiation damage of these cells, 1.0, 1.8, and 11.5 MeV electrons as well as 1.0 MeV proton irradiations were performed. The irradiation fluence was ranging up to $5\times10^{14}$ cm$^{-2}$ for electrons and $1\times10^{10}$ cm$^{-2}$ for protons, with a flux density of about $5\times10^{10}$ cm$^{-2}$s$^{-1}$ and $5\times10^{8}$ cm$^{-2}$s$^{1}$ for electrons and protons, respectively, low enough to avoid significant sample heating during the irradiations. The electrons have ranges which were long enough to pass through the whole samples, but the protons stop in the Ge bottom-cell.[8] Table 1 lists the irradiation parameters for the samples used in this study. PL spectra from each subcell were obtained by selective tuning of the wavelength of the excitation lasers. We used the 532 nm green solid state laser, the 730 nm red diode laser and the 1064 nm infrared solid state laser as the selective excitation sources for PL from the GaInP top-cell, GaAs mid-cell and Ge bottom-cell, respectively. The typical excitation power was about 75 mW for the green solid state laser, 100 mW for the red diode laser and 250 mW for the infrared solid state laser, and the laser beam diameters were about 3.0, 3.0 and 5.0 mm, respectively. PL in the GaInP top-cell was dispersed by one grating monochromator with 600 grooves/mm grating blazed at 500 nm and was detected by a photomultiplier, that in the GaAs mid-cell was dispersed by another grating monochromator of the same type with 600 grooves/mm grating blazed at 750 nm and detected by a Si photodetector, and that in the Ge bottom-cell was dispersed by the other grating monochromator with 150 grooves/mm grating blazed at 2000 nm and detected by a PbS photodetector. The detected signal was processed with a lock-in technique, and then was transferred to a computer for data processing.
Table 1. The irradiation parameters for the samples used in this study.
$E$ (MeV) $\varphi$ (cm$^{-2}$) $E_{\rm nl}$ (MeVcm$^{-2}$/g) $D_{\rm d}$ (MeV/g)
GaInP GaAs Ge GaInP GaAs Ge
e$^{-}$ 1.0 5$\times$10$^{14}$ 3.50$\times$10$^{-5}$ 2.66$\times$10$^{-5}$ 1.20$\times$10$^{-5}$ 1.75$\times$10$^{10}$ 1.33$\times$10$^{10}$ 6.00$\times$10$^{9}$
1.8 5$\times$10$^{14}$ 5.20$\times$10$^{-5}$ 4.16$\times$10$^{-5}$ 2.02$\times$10$^{-5}$ 2.60$\times$10$^{10}$ 2.08$\times$10$^{10}$ 1.01$\times$10$^{10}$
11.5 5$\times$10$^{13}$ 1.03$\times$10$^{-4}$ 9.36$\times$10$^{-5}$ 7.68$\times$10$^{-5}$ 5.15$\times$10$^{9}$ 4.68$\times$10$^{9}$ 3.84$\times$10$^{9}$
1$\times$10$^{14}$ 1.03$\times$10$^{10}$ 9.36$\times$10$^{9}$ 7.68$\times$10$^{9}$
p$^{+}$ 1.0 1$\times$10$^{10}$ 5.80$\times$10$^{-2}$ 5.30$\times$10$^{-2}$ 4.92$\times$10$^{-2}$ 0.58$\times$10$^{9}$ 0.53$\times$10$^{9}$ 4.92$\times$10$^{8}$
cpl-34-2-026101-fig1.png
Fig. 1. PL spectra of the GaInP top-cell for GaInP/GaAs/Ge triple-junction solar cells unirradiated and irradiated with electrons and protons.
cpl-34-2-026101-fig2.png
Fig. 2. PL spectra of the GaAs mid-cell for GaInP/GaAs/Ge triple-junction solar cells unirradiated and irradiated with electrons and protons.
Figure 1 shows the PL spectra from GaInP top-cells unirradiated and irradiated by electrons and protons at room temperature. The spectra, as shown in Fig. 1(b), are after 1.0 MeV protons irradiation with a fluence of $1\times10^{10}$ cm$^{-2}$, 11.5 MeV electrons irradiation to a fluence of $5\times10^{13}$ and $1\times10^{14}$ cm$^{-2}$, 1.0 MeV electrons irradiation with a fluence of $5\times10^{14}$ cm$^{-2}$, and 1.8 MeV electrons irradiation with a fluence of $5\times10^{14}$ cm$^{-2}$, respectively. It is noted that the emission band is the broad peak feature centered at around 660 nm (1.88 eV), which has a full width at half maximum (FWHM) of approximately 20 nm. The PL intensity has a fast degradation after irradiation, which is similar to the results from the electroluminescence study in irradiated GaInP/GaAs/Ge solar cells.[9] In addition, the degradation of PL intensity for the GaInP top-cell increases with $D_{\rm d}$ (listed in Table 1). The PL spectra detected from GaAs mid-cells are presented in Fig. 2. In Fig. 2, it can be seen that a peak is located at around 890 nm, i.e., 1.40 eV, which has a decrease of about 20 meV compared with a conventional GaAs cell (1.42 eV) with approximately 30 nm FWHM. In recent 3J solar cells, about 1% In is sometimes doped into the GaAs mid-cells for lattice-matching between the GaAs mid-cell and the Ge bottom-cell.[10] This doping causes band gap narrowing of the GaAs mid-cell. A similar degradation of PL intensity is presented in Fig. 2. Furthermore, the degradation of PL intensity for GaAs mid-cells seems to be dramatically faster than that for GaInP top-cells. The PL intensity is close to zero when irradiated by 11.5, 1.8 and 1.0 MeV electrons with the fluences of $1\times10^{14}$ cm$^{-2}$, $5\times10^{14}$ cm$^{-2}$ and $5\times10^{14}$ cm$^{-2}$, especially. For GaAs mid-cells, the higher the $D_{\rm d}$ is, the greater the degradation of the PL intensity is.
cpl-34-2-026101-fig3.png
Fig. 3. PL spectra of the Ge bottom-cell for GaInP/GaAs/Ge triple-junction solar cells unirradiated and irradiated with electrons and protons.
Figure 3 shows the PL spectra from Ge bottom-cells. In Fig. 3, the spectra reveal another peak is located at around 2005 nm (0.62 eV) with an FWHM of approximately 35 nm. About a 40 meV decrease compared with a conventional Ge cell (0.66 eV) contributes to reducing the band gap of the solar cell for obtaining a higher efficiency.[11] The PL intensity has a relatively slower degradation after irradiation, especially irradiated by 1.0 MeV protons with a fluence of $1\times10^{10}$ cm$^{-2}$. We also find that the PL intensity decreases as $D_{\rm d}$ increases. In PL spectroscopy, it is clear that the PL intensity has a fast degradation after irradiation. The degradation of PL intensity for GaAs mid-cell seems to be faster than that for GaInP top-cell, and that for Ge bottom-cell is relatively slower. For each subcell, moreover, as $D_{\rm d}$ increases, the PL intensity decreases. The observed degradation of PL intensity is a result of the generation of nonradiative recombination centers under irradiation found in Ref. [9]. For a given level of the laser power, the PL intensity is given by radiative efficiency $\eta$, $$\begin{align} \eta =\Big(1+\frac{\tau _{\rm r} }{\tau _{\rm nr}}\Big)^{-1},~~ \tag {1} \end{align} $$ where $\tau _{\rm r}$ and $\tau _{\rm nr}$ are the radiative and nonradiative recombination lifetimes, respectively. Since the total recombination lifetime of material ($\tau$) under low-level excitation conditions is determined by $1/\tau =1/{\tau _{\rm r}}+1/{\tau _{\rm nr}}$, the characteristics of the PL emission trend, following the functional form of Eq. (1), show evidence that a strong degradation of the recombination lifetime, occurs in the subcells.[8] In the presence of the irradiation-induced defects, the luminescence efficiency can be expressed as[12] $$\begin{align} \eta =\Big(1+\frac{\tau _{\rm r} }{\tau _{\rm 0nr}}+\tau _{\rm r} k_{\rm r} D_{\rm d}\Big)^{-1},~~ \tag {2} \end{align} $$ where $\tau _{\rm 0nr}$ is the initial nonradiative recombination lifetime. The radiative recombination lifetime $\tau _{\rm r}$ is equal to $1.0\times10^{-8}$ s for the GaInP top-cell, $1.7\times10^{-8}$ s for GaAs mid-cell and $1.6\times10^{-4}$ s for the Ge bottom-cell,[12] respectively, and $k_{\rm r}$ is the lifetime damage coefficients of the recombination centers. The solid squares in Fig. 4 illustrate the normalized PL intensity as a function of $D_{\rm d}$ from the GaInP top-cell, the GaAs mid-cell and the Ge bottom-cell, respectively. Notably, the normalized PL intensity decreases as $D_{\rm d}$ increases for the Ge bottom-cell. A similar degradation behavior is shown for the GaInP top-cell and the GaAs mid-cell, respectively. It is also known that with the same irradiation parameter $D_{\rm d}$, the degradation of PL intensity is most obvious for the GaAs mid-cell, the second for the GaInP top-cell and the least for the Ge bottom-cell. Moreover, both the correlations between the degradation of normalized PL intensity and $D_{\rm d}$ for each subcell of GaInP/GaAs/Ge solar cells shown in Fig. 4 can be well fitted to Eq. (2).
cpl-34-2-026101-fig4.png
Fig. 4. Dependence of the normalized PL intensity with $D_{\rm d}$ for electron and proton irradiated samples captured from each subcell for GaInP/GaAs/Ge triple-junction solar cells.
Table 2. The values of $k_{\rm r}$ in individual sub-cells for GaInP/GaAs/Ge triple-junction solar cells.
Cell GaInP GaAs Ge
top-cell mid-cell bottom-cell
$k_{\rm r}$ (g$\cdot$s$^{-1}$MeV$^{-1}$) 2.1$\times$10$^{-2}$ 6.7$\times$10$^{-1}$ 1.0$\times$ 10$^{-6}$
The fitting results of parameters $k_{\rm r}$ are listed in Table 2. From Table 2, it is seen that the lifetime damage coefficients of the recombination centers for the GaAs mid-cell is determined to be 6.7$\times$10$^{-1}$ g$\cdot$s$^{-1}$MeV$^{-1}$, which is in good agreement with the analysis of GaAs photoluminescence in irradiated GaAs/Ge solar cells.[13] The value of $k_{\rm r}$ for the Ge bottom-cell is approximately a factor of four orders of magnitude lower than that for the GaInP top-cell and five orders of magnitude lower than that for the GaAs mid-cell. It is suggested that the damage of the recombination centers in the GaAs mid-cell is the most serious, secondly for the GaInP top-cell and thirdly for the Ge bottom-cell of the GaInP/GaAs/Ge triple-junction solar cells. In summary, the photoluminescence technique has been carried out to monitor the radiation damage of the electron- and proton-irradiated subcells for the GaInP/GaAs/Ge triple-junction solar cells. The PL spectra of three individual subcells have been acquired. Moreover, the PL intensity has a fast degradation after irradiation, and the degradation of PL intensity increases as $D_{\rm d}$ increases. Finally, the normalized PL intensity varying with $D_{\rm d}$ has been discussed in detail, and the value of the lifetime damage coefficients of the recombination centers for each subcell is obtained from the PL radiative efficiency. References Multi-junction solar cells and novel structures for solar cell applicationsRadiation-resistant solar cells for space useEffects of 0.28–2.80MeV proton irradiation on GaInP/GaAs/Ge triple-junction solar cells for space useApplication of CL/EBIC-SEM Techniques for Characterization of Radiation Effects in Multijunction Solar CellsSpace degradation of multijunction solar cells: An electroluminescence studyHigh-efficiency InGaP/In0.01Ga0.99As tandem solar cells lattice-matched to Ge substratesDefect Analysis of Monolithic GaInP/GaInAs/Ge Triple-Junction Solar Cells by Luminescence TechnologyAnalysis of multijunction solar cell degradation in space and irradiation induced recombination centersPhotoluminescence analysis of electron irradiation-induced defects in GaAs/Ge space solar cells
[1] Yamaguchi M 2002 Physica E 14 84
[2] Yamaguchi M 2001 Sol. Energy Mater. Sol. Cells 68 31
[3]Anspaugh B E 1996 Solar Cell Radiation Handbook (California: JPL Publication)
[4]Summers G P, Walters R J, Xapsos M A and Burke E A 1994 IEEE Proc. 1st World Conference on Photovoltaic Energy Conversion (Waikoloa, Hawaii) 2 2068
[5]Messenger S R, Burke E A, Walters R J, Warner J H, Summers G P and Lorentzen J R 2007 19th Space Photovoltaic Research and Technology Conference p 8
[6]Makham S, Zazoui M and Bourgoin J C 2004 Moroccan J. Condens. Matter 5 181
[7] Wang R, Liu Y H and Sun X F 2008 Nucl. Instrum. Methods Phys. Res. Sect. B 266 745
[8] Maximenko S I, Messenger S R, Cress C D and Freitas J A 2010 IEEE Trans. Nucl. Sci. 57 3095
[9] Zazoui M and Bourgoin J C 2002 Appl. Phys. Lett. 80 4455
[10] Takamoto T, Agui T, Ikeda E and Kurita H 2001 Sol. Energy Mater. Sol. Cells 66 511
[11] Lin S H and Cheng T H 2014 ECS Trans. 64 27
[12] Zazoui M, Mbarki M, Aldin A Z, Bourgoin J C, Gilard O and Strobl G 2003 J. Appl. Phys. 93 5080
[13] Lu M, Wang R, Yang K and Yi T C 2013 Nucl. Instrum. Methods Phys. Res. Sect. B 312 137