Perovskite/silicon tandem solar cells have emerged as promising candidates for next-generation photovoltaic technology due to their ultra-high power conversion efficiency (PCE)1-3. However, the mechanical stress generated during repeated environmental stress cycles remains a critical challenge for flexible perovskite/silicon tandem solar cells, leading to interfacial delamination and device degradation. In this work, we propose a dual-buffer-layer strategy with a stress-release mechanism to synergistically mitigate ion bombardment during subsequent sputtering deposition and enhance interfacial adhesion while preserving efficient charge extraction. The loose SnOx buffer layer, engineered by adjusting the purging time of atomic layer deposition, can dissipate strain energy, whereas the compact SnOx layer can ensure robust electrical contact. Based on this dual-buffer-layer, the flexible tandem solar cell, constructed on a 60-micron thick ultra-thin silicon bottom cell, achieves a certified PCE of 33.4% on 1-cm2 area, and an certified PCE of 29.8% on wafer-sized area of 260-cm2 with a power-per-weight of up to 1.77 W/g. The modified tandem solar cells demonstrate good durability, retaining over 97% of their initial power conversion efficiencies after 43000 bending cycles under a maximum curvature radius of around 40 mm in air, and around 97% after thermal cycling testing (-40 °C to 85 °C) for 250 cycles.

Quantum error correction (QEC) [1,2] is essential for the realization of large-scale quantum computers [3,4]. However, due to the complexity of operating on the encoded 'logical' qubits [5,6], understanding the physical principles for building fault-tolerant quantum devices and combining them into efficient architectures is an outstanding scientific challenge. Here we utilize reconfigurable arrays of up to 448 neutral atoms to implement the key elements of a universal, fault-tolerant quantum processing architecture and experimentally explore their underlying working mechanisms. We first employ surface codes to study how repeated QEC suppresses errors [6,7], demonstrating 2.14(13)x below-threshold performance in a four-round characterization circuit by leveraging atom loss detection and machine learning decoding [8,9]. We then investigate logical entanglement using transversal gates and lattice surgery [10-12], and extend it to universal logic through transversal teleportation with 3D [[15,1,3]] codes [13,14], enabling arbitrary-angle synthesis with polylogarithmic overhead [5,15]. Finally, we develop mid-circuit qubit re-use [16], increasing experimental cycle rates by two orders of magnitude and enabling deep-circuit protocols with dozens of logical qubits and hundreds of logical teleportations [17-20] with [[7,1,3]] and high-rate [[16,6,4]] codes while maintaining constant internal entropy. Our experiments reveal key principles for efficient architecture design, involving the interplay between quantum logic & entropy removal, judiciously using physical entanglement in logic gates & magic state generation, and leveraging teleportations for universality & physical qubit reset. These results establish foundations for scalable, universal error-corrected processing and its practical implementation with neutral atom systems.

Flexible solar cells have a transformative potential for niche applications, yet face fundamental challenges in simultaneously achieving high power conversion efficiency (PCE), extreme mechanical resilience and operational stability1-4. Herein, we demonstrate a certified 33.6%-efficient flexible perovskite/crystalline silicon (c-Si) tandem solar cell with a record open-circuit voltage (Voc) of 2.015 V, rivaling its rigid counterpart. The flexible tandem retains 91% of its initial PCE after 5,000 cycles under a bending radius (Rb) of 17.6 mm, and demonstrates exceptional operational and damp-heat (DH) stability, featuring a T80 lifetime exceeding 2,000 hours under continuous illumination and retaining 90% of its initial PCE after 1,000 hours DH test. This advancement is enabled by implementation of the reactive-plasma-deposited (RPD) cerium and hydrogen co-doped indium oxide (ICO:H) recombination layer (RL) that promotes self-assembled monolayers (SAMs) coverage and interfacial charge transfer, and in-situ annealed zinc-doped indium oxide (IZO) front transparent electrode with significantly enhanced optoelectronic and mechanical properties.

Social capital-the trust and cohesion within communities-has been linked to mental health, yet longitudinal evidence remains scarce. Here we show that neighborhood-level personal trust predicts the incidence of severe mental illness in a large, population-based cohort in Stockholm County, Sweden. Among 1.47 million Swedish-born residents followed over 15 years, higher personal trust at baseline was associated with reduced rates of non-affective psychotic disorder and bipolar disorder without psychosis over the follow-up period, but only among individuals of Swedish or European heritage. In contrast, the same exposure increased incidence rates among those of North African, Middle Eastern or Sub-Saharan African heritage. Political and welfare trust showed no consistent associations. These findings suggest that social capital may confer mental health benefits or risks depending on one's own social position, highlighting the need for nuanced public mental health strategies that consider structural and cultural contexts in promoting mental wellbeing.

The biological underpinnings of self-harm in young people are unclear. Self-harm often serves to regulate emotions, and electrodermal activity (EDA) is a well-established biomarker of emotional arousal, which is physiologically related to emotion regulation. A quasi-experimental case control study using predefined groups was conducted. Three groups of young people (16-25 years; n = 180) with different self-harm histories were recruited: no self-harm history (n = 62), self-harm ideation last year with no enaction (n = 51) and self-harm enaction last year (n = 67). EDA was measured during three tasks: an auditory tones habituation task, a psychosocial stress task and an emotional images task. Those in the self-harm enaction group elicited a heightened EDA response (hyperreactivity) across two tasks, specifically a slower habituation rate to auditory tones and higher EDA during the psychosocial stress task compared to other groups. High levels of non-response during the emotional images task limited analyses. These findings expand our understanding of the biomarkers for self-harm, specifically emotional arousal in young people who self-harm. Specifically, they suggest that those with a history of self-harm exhibit a heightened electrodermal response to both stressful and non-stressful stimuli compared to those who have no history of self-harm and those who have only thought about self-harm.

Understanding the neurobiological underpinnings of weight gain could reduce excess mortality and improve long-term trajectories of psychiatric disorders. Using brain scans from healthy individuals (n = 1,504), we trained a model to predict body mass index (BMI) and applied it to individuals with schizophrenia (n = 146), clinical high-risk states for psychosis (n = 213) and recent-onset depression (ROD, n = 200). We computed BMIgap (BMIpredicted - BMImeasured), interrogated its brain-level overlaps with schizophrenia and explored whether BMIgap predicted weight gain at the 1-year and 2-year follow-ups. Schizophrenia (BMIgap = 1.05 kg m-2) and clinical high-risk individuals (BMIgap = 0.51 kg m-2) showed increased BMIgap and individuals with ROD (BMIgap = -0.82 kg m-2) showed decreased BMIgap. Shared brain patterns of BMI and schizophrenia were linked to illness duration, disease onset and hospitalization frequency. Higher BMIgap predicted future weight gain, particularly in younger individuals with ROD, and at 2-year follow-up. Here we show that BMIgap can serve as a potential brain-derived measure to stratify at-risk individuals and deliver tailored interventions for better metabolic risk control.

Depression consists of heterogeneous symptoms that can occur in hundreds of possible combinations. However, intervention studies commonly operationalize depression as a homogeneous condition. Here we adopt a symptom-level approach to test the effects of the selective serotonin reuptake inhibitor sertraline on depression and anxiety symptoms and to test their associations. Using data from the PANDA randomized controlled trial, we use network models to estimate the effects of sertraline at different time points (contemporaneous networks at 2, 6 and 12 weeks) and across time (temporally lagged networks). Results show that sertraline has beneficial effects on core depression and anxiety symptoms as early as after 2 weeks of treatment, counteracted by detrimental effects on somatic symptoms of depression. This intricate pattern of treatment effects is typically masked when measuring depression on a single dimension. Focusing on individual symptoms of depression and anxiety may shed light on the nature, effectiveness and timing of antidepressant action.

Cognitive impairments are a major contributor to psychosocial dysfunction in major depressive disorder, yet mechanistically selective treatments targeting these impairments are lacking. Here, in line with a precision medicine approach, we evaluated guanfacine immediate release (GIR), an α2A receptor agonist, as a novel treatment aimed at enhancing cognitive control circuit function and behavioral performance in a neurobiologically defined subtype of depression, the cognitive biotype NCT04181736. This biotype was prospectively identified based on impairments in both cognitive control circuitry and associated behavioral performance. Seventeen participants with major depressive disorder meeting these prospective criteria completed 6-8 weeks of GIR treatment (target dose of 2 mg per night), consistent with our preregistered per-protocol analysis plan. GIR significantly increased activation and connectivity within the cognitive control circuit. The clinical response (defined as a ≥50% reduction on the 17-item Hamilton Depression Rating Scale (HDRS)) was achieved by 76.5%, of patients, exceeding conventional antidepressant response rates, and 84.6% also achieved remission (HDRS score of ≤7). GIR also led to significant improvements in cognitive control performance, global life satisfaction and quality of life. Here we demonstrate both clinical efficacy and circuit target engagement of GIR as a mechanistically selective treatment for the cognitive biotype of depression.