Search

Various solar cells and materials have great potential as future generation solar cells and materials including crystalline Si. Si tandem solar cells such as III-V/Si tandem, II-VI/Si tandem, Chalcopyrite/Si tandem and Perovskite/Si tandem solar cells are expected to have great potential of space and terrestrial application because of high efficiency, light weight and low cost potential. Quantum and Nano structures such as MQWs ‘(Multi Quantum Wells) and QDs (Quantum Dots) are also expected as high-efficiency solar cells. However, there are some problems to be solved. Because point defects in crystalline Si, high density of dislocations and interface recombination in Si tandem solar cell materials and interface recombination in MQWs and QDs affect on solar cell properties, understanding and controlling those defects are very important for realizing higher efficiency. This paper presents analytical results for high efficiency potential of single crystalline Si solar cells, Si tandem solar cells, and MQW and QD solar cells. Although 25.6% efficiency has been attained with hetero-junction and back contact structure and by using bulk minority-carrier lifetime of around 10 msec, high efficiency of more than 27.5% is expected to be realized by increasing bulk lifetime of more than 100 msec Si. Really, high bulk lifetime of around 100 msec has been obtained by the MCZ (Magnetic-field-applied Czochralski) method and decreasing carbon concentration to less than 1014 cm-3 in Si. Understanding behavior of point defects such as Cs-Oi and so forth in Si and decreasing those defects density are suggested for realizing higher bulk lifetime and higher efficiency Si solar cells of more than 27.5%. High efficiency potential of Si tandem solar cells such as III-V/Si tandem, II-VI/Si tandem, Chalcopyrite/Si tandem and Perovskite/Si tandem solar cells is also analyzed by using measure of external radiative efficiency: 43% for III-V/Si 3-J(Junction) tandem, 38% for III-V/Si 2-J tandem, 33% for II-V/Si 2-J tandem, 32% for Chalcopyrite/Si tandem, and 32% for Perovskite/Si tandem solar cells. Effects of dislocations upon minority-carrier lifetime in III-V/Si tandem solar cells are also presented. Our recent result for mechanically stacked InGaP/GaAs//Si 3-junction tandem solar cell with efficiency of 27.4% is shown. Although MQW and QD solar cells are expected as high efficiency solar cells, one of big problems is higher interface recombination near MQWs and QDs because of their lattice mismatching. Lattice mismatch (Δa/a0) dependence of interface recombination velocity (SI) in MQW and QD solar cells is analyzed. Open-circuit voltage drop by interface recombination in MQWs and QDs and their efficiency potential is discussed.

Abstract We demonstrated MOCVD (Metal Organic Chemical Vapor Deposition) GeSn growth co-doped with Sb. The Sb doping accelerated the Sn incorporation in the GeSn epitaxial film. The Sn and Sb concentrations were 6.6% and 0.5%, respectively. We consider the surfactant effect by the Sb as well as atmospheric H2 suppress the surface migration and therefore the precipitation of Sn. 1. Introduction GeSn is a promising material for large scale optoelectronic integrated devices on the Si platform [1]. It is reported that GeSn with Sn concentration more than approximately 10% has direct bandgap and emits light efficiently. In addition, GeSn is also expected to improve the performance of MOSFET, when it is used as a material for the channel or source/drain. In order to enjoy the high performance of various GeSn based devices, it is necessary to achieve both high Sn concentration and high crystallinity. However, it has been difficult to increase Sn concentration, because the solid solubility limit of Sn is approximately 1%. Higher Sn concentration than the solubility limit has been demonstrated by MBE at extremely low growth temperature; however, the low temperature growth generally results in the poor crystal quality. Donor or acceptor doping in the GeSn layer has not been reported so much yet, although it should be indispensable for the device application. In this study, we examined MOCVD GeSn growth with Sb doping and found that the high concentration Sn and Sb doping can be achieved simultaneously. 2. Experiments In our experiments, we used tertiarybutylgermane (t-C4H9GeH3) and tetra-ethyl tin (C2H5)4Sn as MOCVD precursors with H2 carrier gas. The Sb is doped from the atmosphere with residual triisopropyl-antimony [(i-C3H7)3Sb]. These newly employed precursors are significantly safer than conventional Ge2H6 and SnH4 because of their non-pyrophoric and non-explosive characteristics. We used Ge(001) as the substrate. The deposition was carried out in 30 Torr atmosphere for 120 min at 320-360°C. The injection rate of the Ge and Sn precursors were controlled between approximately 3.5 × 10−6 and 1.4 × 10−4 mol/min. H2 supply was at 20-40 sccm. The Sn and Sb concentrations were measured by the combination of RBS and SIMS. TEM observation was also carried out for the evaluation of the crystal quality. 3. Results and Discussion Figure 1 shows the depth profiles of Sn and Sb concentration in the epitaxial film. Here, the (Sn+Sb) profile was determined by RBS first. Then, the Sn/Sb ratio was measured by SIMS. The Sn concentration close to the epitaxial interface may be more than 10% and then decreases to the constant of approximately 6.6%, while the Sb concentration is quite uniform along the depth at approximately 0.5%. The surface segregation for Sn and Sb should be the artifacts of the measurement. No obvious defects or precipitations were found by the TEM observation. The Sn concentration is much higher than that obtained without Sb doping; therefore, we believe the Sb doping promoted the Sn incorporation in the film. It is also remarkable that Sb was doped up to such high concentration without precipitation at this relatively high growth temperature. We have verified the existence of H2 in the MOCVD atmosphere promoted the Sn incorporation in the GeSn film, probably due to the surfactant effect. Sb might bring the similar surfactant effect for the GeSn growth. The H2 suppress the Sn surface migration therefore suppress the precipitation [2]. It has been also reported that Sb during the InGaAsN growth lowers surface energy and suppress both surface migration and island formation [3]. The surfactant effects provided by H2 and Sb, in which the surface migration and therefore the precipitation were suppressed similar to the case of the low temperature growth, might promote the Sn and Sb incorporation beyond the solubility limits. 4. Conclusions In this study, we have demonstrated high concentration Sn and Sb incorporation in the epitaxial Ge grown by MOCVD. The Sn and Sb concentrations were 6.6% and 0.5%, respectively. These high concentration may be due to the surfactant effect of both H2 and Sb. References [1] S. Wirths, et al. Nature photon. 9, 88 (2015). [2] T. Asano, et al. Jpn. J. Appl. Phys. 54, 04DH15 (2015). [3] X. Yang, et al. IEEE Photon. Technol. Lett. 12(2), 128 (2000). Figure 1

Introduction Ge1-xSnx is recognized as one of the post scaling materials, which can be used as a channel material and a stressor for Ge MOSFETs. For instance, it was reported that a hole mobility of pMOSFET with Ge1-xSnx channel was improved comparing to the pure Ge channel [1]. Moreover, Ge1-xSnx can be used as an optical devices because the band structure changes from indirect to direct as Sn composition increasing more than 10 at.%. Ge1-xSnx is a very attractive material, and therefore often fabricated by epitaxial growth on the Ge substrate using CVD technique. However, the precursors used for CVD so far have a safety problem. GeH4 or Ge2H6 for Ge and SnH4 for Sn are very dangerous because they have a pyrophoric and explosive nature [2, 3]. In this paper, we succeeded in performing Ge1-xSnx epitaxial film growth by MOCVD (metal-organic chemical vapor deposition) using tertiary-butyl-german (t-C4H9GeH3) and tetra-ethyl-tin ((C2H5)4Sn) which have neither a pyrophoric nor explosive nature. Experiments We synthesized and used t-C4H9GeH3 and (C2H5)4Sn as Ge and Sn precursors, respectively. Ge precursor has a high vapor pressure and can be decomposed at low temperature. Furthermore, it has few residual carbon impurities in the deposited film [4]. Our selected Sn precursor also has high vapor pressure. Moreover, these precursors are sufficiently safe. In this experiment, we used Ge(100) wafer as a substrate. After the chemical cleaning of the substrates using NH4OH, HCl and HF, residual native oxide on the substrate was removed by thermal treatment at 450 ºC in the CVD process chamber. Then, the precursor vapor was injected into the chamber. The precursor supply was controlled by the flow rate of N2 carrier gas, and the temperatures of precursor bottles and pressure. In the experiment, the injected rates were estimated to be 1.4E-04 mol/min for both Ge and Sn precursors. The deposition temperature and chamber pressure were 360 ºC and 30 torr. The deposited films were investigated by transmission electron microscopy (TEM; JEM-2100, JEOL Ltd.) and X-ray diffractometry (XRD; Rigaku SmartLab). Results and Discussion Figure 1 (a) shows cross-sectional TEM (XTEM) image, here the film thickness is approximately 50 nm. Figure 1 (b) is the close-up image with enhanced brightness at the epitaxial interface indicated by the rectangle in Fig. 1 (a). No obvious defect was observed at the interface and in the film. The electron diffraction patterns taken from the Ge1-xSnx film (Fig. 2 (a)) and Ge substrate (Fig. 2 (b)) showed the same crystalline orientations, implying successful epitaxial growth. Figure 3 shows the X-ray diffraction reciprocal space mapping (XRD-2DRSM) around -2-24 diffraction. The lattice constant was expanded approximately 0.6 % in the epitaxial film from the Ge substrate. This corresponds to approximately 2 at.% Sn incorporation in the film, which exceeds solid solubility limit of 1 at.% reported in the literature [5]. In summary, we have succeeded in the Ge0.98Sn0.02 heteroepitaxial film growth on the Ge substrate by MOCVD for the first time. References [1] S. Gupta et al., in IEDM Tech. Digest., 398 (2011). [2] B. Vincent et al., Appl. Phys. Lett., 99 152103 (2011). [3] R. A. Soref et al., J. Mater. Res., 22 3281 (2007). [4] H. Machida et al., Jpn. J. Appl. Phys., 49 05FF06 (2010). [5] F. A. Trumbore, J. Electrochem. Soc., 103 (11) 597 (1956).

The correlation between recombination properties and detailed misorientation angles at small angle grain boundaries SA-GBs in multicrystalline silicon (mc-Si) after metal contamination were evaluated. After metal contamination, EBIC contrast enhancements at > 1.5o SA-GBs were on the order of Fe/1000 oC > Ni/1000 oC > Ni/600 oC, while at < 1.5o SA-GBs they were seldom different from any contamination. These results might be attributed to Fe and Ni atoms forming different energy levels in recombination centers and the getting abilities of SA-GBs depending on misorientation angles, i.e., dislocation density at SA-GBs. Many dark spots were observed after Ni/600 oC. After Secco etching, we confirmed that the dark spots corresponded to etch pits. Denuded zones at vicinity of SA-GBs were observed after only Fe/1000 oC. The gettering ability of SA-GBs depends on dislocation density and the difference in recombination properties after metal contamination was affected by the types of metal impurities.

not Available.

not Available.

not Available.

not Available.

not Available.