A thin, lightweight, flexible solar cell is developed that maximizes the power-to-mass ratio under AM0 illumination and has a competitive efficiency after typical high energy electron irradiation. The inverted metamorphic triple junction (IMM3J) solar cells with Ga 0.51 In 0.49 P/GaAs/Ga 0.73 In 0.27 As subcells are grown on GaAs substrates and have a total epitaxy thickness of about 10 μm. After epitaxial growth, the inverted layer stack is metallized, with the metal serving as back-contact, back reflector and support layer for the ultra-thin solar cells before the GaAs substrate is separated by an epitaxial lift-off (ELO) process. The nondestructive nature of the ELO process makes multiple reuses of the GaAs substrate possible. The solar cell structure is optimized for maximum EOL efficiency, that is, after 1-MeV electron irradiation with a fluence of 1 Â 10 15 cm À2 , by means of simulations that include the irradiation induced defects in the various semiconductor alloys. Assuming realistic charge carrier lifetime in the materials, we predict a near-term efficiency potential for the IMM3J ELO of 30.9% under AM0 illumination before and 26.7% after irradiation. Several IMM3J ELO solar cells with an area of approximately 20 cm 2 from different development stages were tested under AM0 illumination. The newest solar cells (generation III) with a mass density of only 13.2 mg/cm 2 reach conversion efficiencies of 30.2% at 25 C. The resulting power-to-mass ratio of 3.0 W/g for the bare solar cell is one of the highest published ratios. After irradiation, a conversion efficiency of 25.4% was measured for "generation II" devices under AM0 illumination, which corresponds to a power-to-mass ratio of 2.6 W/g. IMM3J ELO solar cells from "generation I" were also tested for mechanical stability as "de-risking" test of this new cell technology. No degradation of the cell performance was found after dipping the cell in liquid N 2 and then heating up to 25 C for five times, despite of strong deformation of the flexible cell during the temperature cycle.

Substrate-based GaAs solar cells having a dense Au/Cu front contact grid with 45% surface coverage were exposed to accelerated life testing at temperatures between 200 and 300 • C. TEM analysis of the front contacts was used to gain a better understanding of the degradation process. During accelerated life testing at 200 • C only intermixing of the Au and Cu in the front contact occurs, without any significant influence on the J-V curve of the cells, even after 1320h (55 days) of accelerated life testing. At temperatures ≥ 250 • C a recrystallization process occurs in which the metals of the contact and the GaAs front contact layer interact. Once the grainy recrystallized layer starts to approach the window, diffusion via grain boundaries to the window and into the active region of the solar cells occurs, causing a decrease in V oc due to enhanced non-radiative recombination via Cu trap levels introduced in the active region of the solar cell. To be a valid simulation of space conditions the accelerated life testing temperature should be < 250 • C in future experiments, in order to avoid recrystallization of the metals with the GaAs contact layer.

The development of flexible, thin-film, and high-efficiency III-V solar cells enables the design of new flexible, lightweight solar arrays for space applications. A requirement for these solar panels is the replacement of the rigid coverglasses by a flexible shielding layer. In this work, three candidate materials based on commercially available polyimides and synthesized polysiloxanes for such a shielding layer are compared with respect to their ease of synthesis, transparency. Polysiloxanes based on methyltrimethoxysilane (MTMS) based siloxane (MBS) showed the best reproducibility in synthesis of layers of the required thickness of about 300 mm with sufficient transparency and was therefore selected for further analysis. It was demonstrated that the MBS material could be doped with Ce to increase the radiation hardness. Showing virtually no loss of volatile condensable components in outgassing tests it can be concluded that the properties MBS are found suitable for further space qualification testing. V C 2016 Wiley Periodicals, Inc. J. Appl.Polym. Sci. 2016, 133, 43661.

High efficiency, thin-film Epitaxial Lift-Off (ELO) III-V solar cells offer excellent characteristics for implementation in flexible solar panels for space applications. However, the current thin-film ELO solar cell design generally includes a copper handling and support foil. Copper diffusion has a potentially detrimental effect on the device performance and the challenging environment provided by space (high temperatures, electron and proton irradiation) might induce diffusion. It is shown that heat treatments induce copper diffusion. The open-circuit voltage (V oc ) is the most affected solar cell parameter. The decrease in V oc can be explained by enhanced non-radiative recombination via Cu trap levels in the middle of the band gap. The decrease in V oc is found to be dependent on junction depth. In all Cu cells annealed at T ≥ 300 ○ C signs of Cu diffusion are present, which implies that a barrier layer inhibiting Cu diffusion is necessary. Electron radiation damage was found to have no influence on Cu diffusion.

In this study accelerated ageing testing (AAT), J-V characterization and TEM imaging in combination with phase diagram data from literature are used to assess the potential of Ti, Ni, Pd and Pt as diffusion barriers for Au/Cu-based metallization of III-V solar cells. Ni barriers show the largest potential as at an AAT temperature of 250 • C both cells with 10 and 100 nm thick Ni barriers show significantly better performance compared to Au/Cu cells, with the cells with 10 nm Ni barriers even showing virtually no degradation after 7.5 days at 250 • C (equivalent to 10 years at 100 • C at an E a of 0.70 eV). Detailed investigation shows that Ni does not act as a barrier in the classical sense, i.e. preventing diffusion of Cu and Au across the barrier. Instead Ni modifies or slows down the interactions taking place during device degradation and thus effectively acts as an 'interaction' barrier. Different interactions occur at temperatures below and above 250 • C and for thin (10 nm) and thick (100 nm) barriers. The results of this study indicate that 10-100 nm thick Ni intermediate layers in the Cu/Au based metallization of III-V solar cells may be beneficial to improve the device stability upon exposure to elevated temperatures.