There has been much interest in developing a thin-film solar cell because it is lightweight and flexible. The GaAs thin-film solar cell is a top contender in the thin-film solar cell market in that it has a high power conversion efficiency (PCE) compared to that of other thin-film solar cells. There are two common structures for the GaAs solar cell: n (emitter)-on-p (base) and p-on-n. The former performs better due to its high collection efficiency because the electron diffusion length of the p-type base region is much longer than the hole diffusion length of the n-type base region. However, it has been limited to fabricate highly efficient n-on-p single-junction GaAs thin film solar cell on a flexible substrate due to technical obstacles. We investigated a simple and fast epitaxial lift-off (ELO) method that uses a stress originating from a Cr/Au bilayer on a 125-μm-thick flexible substrate. A metal combination of AuBe/Pt/Au is employed as a new p-type ohmic contact with which an n-on-p single-junction GaAs thin-film solar cell on flexible substrate was successfully fabricated. The PCE of the fabricated single-junction GaAs thin-film solar cells reached 22.08% under air mass 1.5 global illumination.

Ge/MoS₂van der Waals heterostructure enables bias-dependent selective detection of visible and near infrared.

As machine vision technology generates large amounts of data from sensors, it requires efficient computational systems for visual cognitive processing. Recently, in-sensor computing systems have emerged as a potential solution for reducing unnecessary data transfer and realizing fast and energy-efficient visual cognitive processing. However, they still lack the capability to process stored images directly within the sensor. Here, we demonstrate a heterogeneously integrated 1-photodiode and 1 memristor (1P-1R) crossbar for in-sensor visual cognitive processing, emulating a mammalian image encoding process to extract features from the input images. Unlike other neuromorphic vision processes, the trained weight values are applied as an input voltage to the image-saved crossbar array instead of storing the weight value in the memristors, realizing the in-sensor computing paradigm. We believe the heterogeneously integrated in-sensor computing platform provides an advanced architecture for real-time and data-intensive machine-vision applications via bio-stimulus domain reduction.

Near‐infrared organic photodetectors (NIR OPDs) comprising ultra‐narrow bandgap non‐fullerene acceptors (NFA, over 1000 nm) typically exhibit high dark current density under applied reverse bias. Therefore, suppression of dark current density is crucial to achieve high‐performance of such NIR OPDs. Herein, cyano (CN) with a strong electron‐withdrawing property is introduced into alkoxy thiophene as a π‐bridge to adjust its optoelectronic characteristics, and the correlation between dark current density and charge injection barrier is investigated. Compared with their motivated NFA (COTH), the novel CN‐substituted NFAs, COTCN and COTCN2, exhibited deeper‐lying highest occupied molecular orbital energy levels and narrower optical bandgap (<1.10 eV), owing to the strong inductive and resonance effect of CN. The dark current and total noise currents are minimized as the number of substituted CN increases because of the larger hole injection barrier. Consequently, PTB7‐Th:COTCN2 exhibited the best shot‐noise limited detectivity (D^*_sh, 1.18 × 10¹² Jones) and total noise detectivity (D^*_n, 1.33 × 10¹¹ Jones) compared with those of PTB7‐Th:COTH (D^*_sh, 2.47 × 10¹¹ Jones and D^*_n, 1.96 × 10¹⁰ Jones).