Due to spectral sensitivity effects, using a single standard spectrum leads to a large uncertainty when estimating the yearly averaged photovoltaic efficiency or energy yield. Here we demonstrate how machine learning techniques can reduce the yearly spectral sets by three orders of magnitude to sets of a few characteristic spectra, and use the resulting proxy spectra to find the optimal solar cell designs maximizing the yearly energy production. When using standard conditions, our calculated efficiency limits show good agreement with current photovoltaic efficiency records, but solar cells designed for record efficiency under the current standard spectra are not optimal for maximizing the yearly energy yield. Our results show that more than 1 MWh m−2 year−1 can realistically be obtained from advanced multijunction systems making use of the direct, diffuse, and back-side albedo components of the irradiance.

it is desirable to incorporate GaInAs QWs into a single-junction GaAs solar cell to extend the range of photon absorption from 1.41 eV toward the optimum bandgap of ≈1.34 eV for the global solar spectrum. [3] Since GaInAs is compressively strained with respect to GaAs, stress builds up successively with each quantum well grown until it exceeds the critical thickness for the stack, leading to plastic relaxation and the formation of dislocations, and accompanied by a significant loss in voltage due to excessive dislocation density. [4] We address the accumulated stress by using a strain-balance technique, by which the compressively strained GaInAs is compensated with tensile GaAsP [5] that results in nearly zero net stress exerted on the semiconductor material at the top and bottom of the stack and hence no structural defects. Under one-sun solar illumination conditions, a typical strain-balanced GaAsP/GaInAs QW solar cell is usually limited by Shockley-Read-Hall (SRH) recombination arising from the relatively wide depletion width [6] and composition of the intrinsic region. [7] To ensure a high fill-factor it is necessary to maintain full depletion across the QW stack at the maximum power point [8] and to ensure the host PV cell itself can attain high performance. If the cell can be fabricated with a low defect density and good carrier collection efficiency, the resulting cell will out-perform a baseline GaAs solar cell. Strained cells with a small number of GaInAs wells and GaAs barriers have been demonstrated previously, achieving 28.3% under solar concentration [6] and 26.3% under 1-sun AM1.5G. [9] But the total thickness was insufficient to fully absorb the light and boost the photocurrent to achieve an overall efficiency gain compared to a baseline GaAs cell. The difficulty in growing a stress-free QW stack limited the number of wells that were used, and therefore limited the additional absorption and photocurrent. For low-indium Ga 1−x In x As alloys, an optically thick film is achieved at ≥2.5 µm, so that a considerable number of wells must be included, easily exceeding the critical thickness of the compressively strained well material. Thus, the compressive strain in the well must be balanced by tensile strain in the barrier layers such as GaAsP or GaInP. [10] Monitoring in situ wafer curvature during epitaxial growth has helped ensure that this balance is achieved. [11] Several groups have demonstrated increases in absorption, but usually accompanied by substantial High-efficiency solar cells are essential for high-density terrestrial applications, as well as space and potentially vehicle applications. The optimum bandgap for the terrestrial spectrum lies beyond the absorption range of a traditional dual junction GaInP/GaAs cell, with the bottom GaAs cell having higher bandgap energy than necessary. Lower energy bandgaps can be achieved with multiple quantum wells (QWs), but such a pathway requires advanced management of the epitaxial growth conditions in order to be useful. Strain-balanced GaAsP/GaInAs Q...

No abstract