AVS 59th Annual International Symposium and Exhibition
    Energy Frontiers Focus Topic Tuesday Sessions
       Session EN+TF-TuA

Paper EN+TF-TuA11
Non-Radiative Carrier Recombination in InGaAs/GaAsP Strain-Balanced Superlattice Solar Cell

Tuesday, October 30, 2012, 5:20 pm, Room 15

Session: Thin Film, Heterostructured, and Organic Solar Cells
Presenter: T. Aihara, University of Miyazaki, Japan
Correspondent: Click to Email
An inserting of the quantum wells (QWs) to GaAs p-i-n solar cells could be a promising candidate to solve the current matching issue in the multi-junction solar cells[1]. We have successfully obtained the non-radiative recombination process for excitonic and subband absorptions in the GaAs/AlAs multiple QWs (MQWs) by using PPT methods [2]. In this study, we investigate escape, radiative and non-radiative recombination mechanisms of photo-generated carriers in the strain-balanced InGaAs/GaAsP MQWs or superlattice (SL) inserted into GaAs p-i-n solar cell structure to improve the photovoltaic performance. We then evaluated above three processes by using the surface photovoltage (SPV), photoluminescence (PL), and piezoelectric photothermal (PPT) spectroscopies, respectively. A InGaAs/GaAsP MQWs absorbing layer that inserted into GaAs p-i-n junction was composed of 10 stacks of 7.4-nm-thick InGaAs well and 10.8-nm-thick GaAsP barrier. For SL absorbing layer, ultra-thin GaAsP barriers of 3.7 nm thickness with 0.56-nm-thick GaAs buffer were prepared. All the layers were grown by metal-organic vapor phase epitaxy on the GaAs substrate. The PPT detects a heat generated by the non-radiative recombination by the PZT directly attached to rear surface of the sample. Figures 1 and 2 show the temperature change of PPT spectra of MQWs and SL with GaAs thin buffer samples, respectively. For MQWs sample, three peaks were observed and A-peak was concluded to be due to the excitonic transition associated with the electron transition between first electron (e1) and heavy-hole subbands (hh1) in QW. On the other hand, B-peak was concluded to be the electron transition between 1st minibands in conduction and valence bands in SL. As the temperature decreased, peak intensities of A and B increased, whereas corresponding SPV peaks decreased. The temperature dependence of PL, PPT, and SPV signal intensities can be fitted with the Arrhenius equation. Figure 3 shows the fitting results of PPT A (MQWs) and B (SL) and SPV A peaks. As shown in Fig. 3, activation energy of SL was smaller than that of MQWs. This result implied that carrier escape from the QWs was enhanced for the case of SL . References [1] K. W. J. Barnham and G. Duggan, J. Appl. Phys. 67 (1990) 3409. [2] P. Wang et al.: Jpn. J. Appl. Phys. 46 (2007) 6857.