Evaluation of the Minority-Carrier Lifetime of IMM3J Solar Cells under Proton Irradiation Based on Electroluminescence

Abstract

The shortening of the minority carrier lifetime is the main reason for the degradation of the electrical performance of solar cells; therefore, it is particularly important to evaluate the minority carrier lifetime of inverted metamorphic triple junction (IMM3J) GaInP/GaAs/InGaAs solar cells. We evaluate the minority carrier lifetime of each subcell of IMM3J solar cells before and after 2 MeV proton irradiation by the electroluminescence (EL) method. Before proton irradiation, the minority carrier lifetimes of the GaInP, GaAs, and InGaAs subcells were 6.99 × 10⁻⁹ s, 3.09 × 10⁻⁸ s, and 2.31 × 10⁻⁸ s, respectively. After proton irradiation, the minority carrier lifetime of GaInP, GaAs, and InGaAs subcells degraded significantly. When the proton fluence was 2 × 10¹² cm⁻², the minority carrier lifetimes of the GaInP, GaAs, and InGaAs subcells degraded to 1.63 × 10⁻¹⁰ s, 1.56 × 10⁻¹¹ s, and 1.65 × 10⁻¹⁰ s, respectively. These results provide a reference for predicting the degradation of the short-circuit current and open-circuit voltage of each subcell.

1. Introduction

The inverted metamorphic triple junction (IMM3J) GaInP/GaAs/InGaAs solar cells are designed based on the principle of spectral matching and have higher photoelectric conversion efficiency than the lattice-matched triple junction GaInP/GaAs/Ge solar cells [1,2,3,4,5,6]. The AM0 efficiency of IMM3J solar cells can reach 32% or higher, while that of conventional triple junction devices is limited to below 30% [1]. The high-efficiency IMM3J solar cell manufactured by Spectrolab Company has a conversion efficiency of 32.1% (AM0, 1 sun) [2]. The efficiency of the IMM3J solar cell fabricated by Tianjin San’an Optoelectronics Company reached 32% under one sun and AM0 spectrum, which is 5% higher than that of the lattice-matched GaInP/InGaAs/Ge triple-junction solar cell [3]. Recently, our research group cooperated with the Shanghai Space Power Source Institute to manufacture IMM solar cells with an efficiency exceeding 32% [4]. In addition, IMM3J solar cells can be bonded to lightweight and flexible substrates to fabricate lightweight and flexible thin-film solar cells by epitaxial lift-off [5,6].

The space radiation environment contains a variety of charged particles, which can seriously damage IMM3J solar cells [7,8,9]. Compared with electron irradiation, proton irradiation will cause greater damage to IMM3J solar cells. The study found that the maximum power of IMM3J solar cells degraded by 13.7% and 26.3% under 1 MeV electron irradiation and 10 MeV proton irradiation, respectively [9]. Because low-energy protons cause greater damage to solar cells than high-energy protons, it is particularly critical to study the low-energy proton irradiation effect on IMM3J solar cells. In previous studies, we predicted the degradation of the short-circuit current, open-circuit voltage, fill factor, and efficiency of IMM3J solar cells under 2 MeV proton irradiation [4]. Furthermore, the IMM3J solar cell is composed of three subcells in series, the open circuit voltage is equal to the sum of the voltages of the three subcells, and the short-circuit current is determined by the minimum current of the three subcells [9]. Therefore, it is also very important to study the radiation degradation of each subcell to optimize the performance of IMM3J solar cells for space applications.

We know that the main reason for the performance degradation of solar cells is the reduction in minority carrier lifetime by irradiation-induced displacement damage in the active area of solar cells. Protons elastically collide with the lattice atoms of the solar cell, causing them to move out of their normal lattice position, creating displacement defects. These defects act as recombination centers, increasing the probability of carrier recombination and resulting in a shortened minority carrier lifetime, which in turn leads to the degradation of solar cell performance. Therefore, in this work, we intend to evaluate the minority carrier lifetime of each subcell of IMM3J solar cells by using the electroluminescence (EL) spectra before and after proton irradiation to provide a reference for predicting the degradation of the short-circuit current and open-circuit voltage of each subcell. The minority carrier lifetime is determined by time-resolved photoluminescence [10,11]. The minority carrier lifetime can also be obtained indirectly through the minority carrier diffusion length, which is relatively easier to measure [12,13]. In addition, we can also measure the defect introduction rate and the capture cross-section by deep-level transient spectroscopy and then obtain the minority carrier lifetime [14,15]. However, due to the complexity of the structure of multijunction tandem solar cells, it is difficult to evaluate the minority carrier lifetime of all subcells using these methods.

Electroluminescence (EL) and photoluminescence (PL) have been proven to be effective methods to evaluate the minority carrier lifetime of multijunction solar cells. The minority carrier lifetime of conventional lattice-matched 3-junction solar cells has been evaluated by measuring the EL and PL spectra before and after irradiation [16,17,18]. The PL method needs to select different excitation sources to excite each junction cell separately and then obtain the PL spectrum of each junction cell. However, EL can simultaneously excite all subcells of multijunction solar cells and then generate the measurement of the EL spectra of all subcells. Therefore, EL measurements are easier to obtain and more advantageous than PL measurements for multijunction solar cells. In this work, we first measure the EL spectra of each subcell of the IMM3J solar cell under proton irradiation with different fluences, then fit the EL peak intensity with the relative change in the proton fluences to obtain the product of the introduction rate and minority carrier capture cross-section, and finally give the minority carrier lifetime of each subcell.

2. Materials and Methods

IMM3J solar cells have the advantages of light weight, low cost, flexibility, and high conversion efficiency [4,6,9,19], rendering such solar cells very suitable for space applications. In this work, IMM3J solar cells are fabricated at the Shanghai Space Power Institute and mainly consist of three subcells: GaInP top cell, GaAs middle cell, and InGaAs bottom cell. The detailed structure and growth process of the IMM3J solar cells is shown in Ref. [4].

Proton irradiation was performed using a 2 × 1.7 MV tandem accelerator at Peking University. To ensure uniform irradiation of samples, the proton beam is scanned in the vertical (Y) and horizontal (X) directions by the scanner before hitting the sample. The proton irradiation experiment was carried out at room temperature. The proton energy is 2 MeV in this experiment to ensure that the proton can completely pass through the IMM3J solar cells. The proton fluence is 2 × 10¹¹, 8 × 10¹¹, and 2 × 10¹² cm⁻² in this experiment, which mainly refers to our previous proton irradiation experiments on lattice-matched solar cells [20] and other related references [9]. The flux used in the experiment is 2 × 10⁹ cm⁻² s⁻¹, which can ensure that the irradiation time is not too long or too short and that it cannot result in an increase in sample temperature due to irradiation. Figure 1 shows the track distribution of 2 MeV protons in IMM3J solar cells obtained by SRIM simulation [21]. Protons with an energy of 2 MeV can pass through the IMM3J solar cell and cause damage to all subcells.

EL spectra were measured at room temperature before and after proton irradiation. A current density of 60 mA/cm² was injected into the IMM3J solar cell as an excitation source. Due to the difference in the peak position of the emission spectrum, we chose different gratings and detectors to measure the EL spectrum of each subcell. The EL spectra of the GaInP top cell were split by a grating monochromator with a 600 groove/mm, grating blazed at 500 nm, and then detected by a photomultiplier (PMTH-S1-CR131A). The EL spectra of the GaAs middle cell and InGaAs bottom were split by a grating monochromator with a 600 groove/mm, grating blazed at 750 nm, and then detected by a Si photodetector (DSi200). Finally, the detected signal was processed by a lock-in amplifier and then transmitted to the computer to obtain the EL spectrum.

3. Results and Discussion

Figure 2 shows the EL spectra of the GaInP, GaAs, and InGaAs subcells under different proton fluences. The EL spectrum peaks of the GaInP, GaAs, and InGaAs subcells are located at 652 nm, 872 nm, and 1300 nm, respectively. After proton irradiation, the EL spectral intensities of the GaInP, GaAs, and InGaAs subcells all decrease significantly.

Figure 3 shows the variation in the EL peak intensity of the GaInP, GaAs, and InGaAs subcells with proton irradiation fluence. The EL spectral intensities of all subcells decrease with increasing proton fluence. Under the same fluence, the EL spectral intensity degradation of the GaAs middle cell is the largest, that of the InGaAs bottom cell is the second largest, and that of the InGaP top cell is the smallest. This indicates that the GaAs and InGaAs subcells have weaker radiation resistance than the GaInP top cell. This conclusion is consistent with our previous research results [4].

For a given injection current density, the normalized EL peak intensity is given by the radiative efficiency ($\eta$) [16,22]:

$$\eta ={\left(1+\frac{{\tau }_r}{{\tau }_{nr}}\right)}^{-1}$$

(1)

where ${\tau }_r$ is the radiative recombination lifetime, which is independent of the fluence. ${\tau }_{nr}$ is the nonradiative recombination lifetime.

$${\tau }_r=\frac{1}{BN}$$

(2)

$${\tau }_{nr}=\frac{1}{k\sigma \nu \phi }$$

(3)

where B is the probability of radiative recombination, N is the doping concentration, k is the introduction rate of the nonradiative recombination centers, $\sigma$ is the minority carrier capture-cross section, $\nu$ is the thermal velocity of carriers, and $\phi$ is the fluence. Combining Equations (1)–(3), the change in EL peak intensity to the fluence can be fitted by $\eta ={\left(1+\alpha \phi \right)}^{-1}$, with $\alpha =k\sigma \nu /BN$. The fitting results are shown in Figure 3 and are in good agreement with the experimental data. The $\alpha$ and $k\sigma \nu$ on the GaInP, GaAs, and InGaAs subcells are listed in

Table 1. The parameters B, N, $k\sigma v$ and $\alpha$ on GaInP, GaAs and InGaAs subcells.

	GaInP	GaAs	InGaAs
B * (cm3/s)	2 × 10−10	1.5 × 10−10	1.43 × 10−10
N (cm−3)	2 × 1017	2 × 1017	2 × 1017
$\alpha$	$\left(7.54\pm 0.05\right)\times {10}^{-11}$	$\left(1.07\pm 0.04\right)\times {10}^{-9}$	$\left(1.04\pm 0.02\right)\times {10}^{-10}$
$k\sigma \nu$	3.0 × 10−3	3.2 × 10−2	3.0 × 10−3

* Refs. [16,23].

. Table 1 shows that GaInP and InGaAs have the same damage coefficient ($k\sigma \nu$) of minority carrier lifetime, while GaAs has the largest damage coefficient. This indicates that under 2 MeV proton irradiation, the GaAs subcell of the IMM3J solar cell degrades the most and has the weakest radiation resistance.

The effective minority carrier lifetime ${\tau }_{eff}$ is represented as:

$$\frac{1}{{\tau }_{eff}}=\frac{1}{{\tau }_r}+\frac{1}{{\tau }_0}+\frac{1}{{\tau }_{nr}}$$

(4)

where ${\tau }_0$ is the nonradiative recombination lifetime before irradiation. ${\tau }_0$ is calculated by ${\tau }_0=L_0^2/D$ (D is the diffusion coefficient, and $L_0$ is the diffusion length before irradiation). D is equal to 60, 200, and 188 cm² s⁻¹ for the GaInP, GaAs, and InGaAs subcells of the IMM3J solar cell, respectively [16,24]. $L_0$ is equal to 5.21 × 10⁻³ cm, 7.94 × 10⁻³ cm, and 3.63 × 10⁻⁴ cm for the GaInP, GaAs, and InGaAs suncells of the IMM3J solar cell, respectively. Therefore, the ${\tau }_0$ of the GaInp, GaAs, and InGaAs subcells can be calculated to be equal to 4.52 × 10⁻⁷ s, 3.15 × 10⁻⁷ s, and 7.01 × 10⁻¹⁰ s, respectively. Thus far, the effective minority carrier lifetime ${\tau }_{eff}$ can be calculated by Equation (4) and listed in

Table 2. The effective minority carrier lifetime ${\tau }_{eff}$ of GaInP, GaAs, and InGaAs subcells of IMM3J solar cells under different proton fluences.

Parameter	Fluence (cm−2)	GaInP	GaAs	InGaAs
${\tau }_{eff}$ (s)	0	6.99 × 10−9	3.09 × 10−8	2.31 × 10−8
	2 × 1011	1.35 × 10−9	1.55 × 10−10	1.55 × 10−9
	8 × 1011	3.93 × 10−10	3.90 × 10−11	4.09 × 10−10
	2 × 1012	1.63 × 10−10	1.56 × 10−11	1.65 × 10−10

. The minority carrier lifetimes of the GaInP, GaAs, and InGaAs subcells are 6.99 × 10⁻⁹ s, 3.09 × 10⁻⁸ s, and 2.31 × 10⁻⁸ s, respectively, before proton irradiation, such values being of the same order of magnitude as those in the Refs. [16,24,25].

Figure 4 shows the normalized effective minority carrier lifetime of each subcell of IMM3J solar cells as a function of irradiation fluence. The effective minority carrier lifetimes of all subcells decrease with increasing proton fluence. The carrier lifetime of the GaInP subcell degrades by 1–2 orders of magnitude, that of the GaAs subcell degrades by 3–4 orders of magnitude, and that of the InGaAs subcell degrades by 2–3 orders of magnitude. Therefore, the radiation resistance of the GaInP top cell is the best, that of the GaAs middle cell is the worst and that of the InGaAs bottom cell is in between the two. Therefore, we need to strengthen the protection of GaAs middle cells and InGaAs bottom cells of IMM3J solar cells in space applications.

4. Discussion and Conclusions

In this work, we found that the GaAs subcell has the worst radiation resistance among the three subcells of the IMM3J solar cell. However, some previous studies suggest that the InGaAs subcell of IMM3J has the weakest radiation resistance [9,26]. The reason for this may be due to the difference in type and energy of the irradiated particles or the content of In in the InGaAs subcell. The study noted that the radiation resistance of InGaAs with different In contents is different [19]. In addition, different energies and types of particle irradiation produce different defects in solar cells, and the damage caused to them is also different [15].

Usually, for a solar cell, the short circuit current density (J_𝑠𝑐) can be expressed as [6,9]:

$$J_{sc}=\frac{\alpha A{e}^{\alpha x_j}-A/\sqrt{D{\tau }_{eff}}}{1/D{\tau }_{eff}-{\alpha }^2}$$

(5)

where $A=q\alpha F\left(1-R\right)\exp \left(-\alpha x_j\right)$, q is the electron charge, $\alpha$ is the absorption coefficient, F is the flux of incident light, R is the reflectivity, and $x_j$ is the p–n junction depth. It can be seen from Equation (5) that the shortening of the minority carrier lifetime will lead to the degradation of the short-circuit current, which in turn will lead to the degradation of the open-circuit voltage, fill factor, and conversion efficiency. Therefore, we can predict the short-circuit current degradation of each subcell of the IMM3J solar cell by evaluating the minority carrier lifetime before and after irradiation.

Under proton irradiation, the relationship between the degradation of the open circuit voltage of each junction cell and the intensity of the EL spectrum can be derived by the optoelectronic reciprocity relation [27,28].

$$\Delta {\varnothing }_{EL}=\exp \left(\frac{q\Delta V_{oc}}{kT}\right)$$

(6)

where $\Delta {\varnothing }_{EL}$ is the ratio of the EL intensity after irradiation and before irradiation and $\Delta V_{oc}$ is the ratio of the open-circuit voltage after irradiation and before irradiation. According to Equation (6), the change in the open circuit voltage can be predicted by the change in the EL intensity. Therefore, we can also predict the degradation of the open circuit voltage of each subcell through the change in EL intensity before and after irradiation.

5. Conclusions

We evaluate the effective minority carrier lifetime of each subcell of IMM3J GaInP/GaAs/InGaAs solar cells using the EL method and reveal how it varies with proton irradiation fluence. Compared with GaInP subcells, GaAs and InGaAs subcells have poorer radiation resistance and need to be further optimized for space applications.