When two monolayer materials are stacked with a relative twist, an effective moiré translation symmetry emerges, leading to fundamentally different properties in the resulting heterostructure. As such, moiré materials have recently provided highly tunable platforms for exploring strongly correlated systems1,2. However, previous studies have focused almost exclusively on monolayers with triangular lattices and low-energy states near the Γ (refs. 3,4) or K (refs. 5-9) points of the Brillouin zone (BZ). Here we introduce a new class of moiré systems based on monolayers with triangular lattices but low-energy states at the M points of the BZ. These M-point moiré materials feature three time-reversal-preserving valleys related by threefold rotational symmetry. We propose twisted bilayers of exfoliable 1T-SnSe2 and 1T-ZrS2 as realizations of this new class. Using extensive ab initio simulations, we identify twist angles that yield flat conduction bands, provide accurate continuum models, analyse their topology and charge density and explore the platform's rich physics. Notably, the M-point moiré Hamiltonians exhibit emergent momentum-space non-symmorphic symmetries and a kagome plane-wave lattice structure. This represents, to our knowledge, the first experimentally viable realization of projective representations of crystalline space groups in a non-magnetic system. With interactions, these systems act as six-flavour Hubbard simulators with Mott physics. Moreover, the presence of a momentum-space non-symmorphic in-plane mirror symmetry renders some of the M-point moiré Hamiltonians quasi-one-dimensional in each valley, suggesting the possibility of realizing Luttinger-liquid physics.

Plastic pollution is a pervasive and growing global problem1-4. Chemicals in plastics are often not sufficiently considered in the overall strategy to prevent and mitigate the impacts of plastics on human health, the environment and circular economy5-7. Here we present an inventory of 16,325 known plastic chemicals with a focus on their properties, presence in plastic and hazards. We find that diverse chemical structures serve a small set of functions, including 5,776 additives, 3,498 processing aids, 1,975 starting substances and 1,788 non-intentionally added substances. Using a hazard-based approach, we identify more than 4,200 chemicals of concern, which are persistent, bioaccumulative, mobile or toxic. We also determine 15 priority groups of chemicals, for which more than 40% of their members are of concern. Finally, we examine data gaps regarding the basic properties, hazards, uses and exposure potential of plastic chemicals. Our work maps the chemical landscape of plastics and contributes to setting the baseline for a transition towards safer and more sustainable materials and products. We propose that removing known chemicals of concern, disclosing the chemical composition and simplifying the formulation of plastics can provide pathways towards this goal.

Lunar mare basalts illuminate the nature of the Moon's mantle, the lunar compositional asymmetry and the early lunar magma ocean (LMO)1-3. However, the characteristics of the mantle beneath the vast South Pole-Aitken (SPA) basin on the lunar farside remain a mystery. Here we present the petrology and geochemistry of basalt fragments from Chang'e-6 (CE6), the first returned lunar farside samples from the SPA basin4-7. These 2.8-billion-year-old CE6 basalts8 share similar major element compositions with the most evolved Apollo 12 ilmenite basalts. They exhibit extreme Sr-Nd depletion, with initial 87Sr/86Sr ratios of 0.699237 to 0.699329 and εNd(t) values (a measure of the neodymium isotopic composition) of 15.80 to 16.13. These characteristics indicate an ultra-depleted mantle, resulting from LMO crystallization and/or later depletion by melt extraction. The former scenario implies that the nearside and farside may possess an isotopically analogous depleted mantle endmember. The latter is probably related to the SPA impact, indicating that post-accretion massive impacts could have potentially triggered large-scale melt extraction of the underlying mantle. Either way, originating during the LMO or later melt extraction, the ultra-depleted mantle beneath the SPA basin offers a deep observational window into early lunar crust-mantle differentiation.

Plastic pollution of the marine realm is widespread, with most scientific attention given to macroplastics and microplastics1,2. By contrast, ocean nanoplastics (<1 μm) remain largely unquantified, leaving gaps in our understanding of the mass budget of this plastic size class3-5. Here we measure nanoplastic concentrations on an ocean-basin scale along a transect crossing the North Atlantic from the subtropical gyre to the northern European shelf. We find approximately 1.5-32.0 mg m-3 of polyethylene terephthalate (PET), polystyrene (PS) and polyvinyl chloride (PVC) nanoplastics throughout the entire water column. On average, we observe a 1.4-fold higher concentration of nanoplastics in the mixed layer when compared with intermediate water depth, with highest mixed-layer nanoplastic concentrations near the European continent. Nanoplastic concentrations at intermediate water depth are 1.8-fold higher in the subtropical gyre compared with the open North Atlantic outside the gyre. The lowest nanoplastic concentrations, with about 5.5 mg m-3 on average and predominantly composed of PET, are present in bottom waters. For the mixed layer of the temperate to subtropical North Atlantic, we estimate that the mass of nanoplastic may amount to 27 million tonnes (Mt). This is in the same range or exceeding previous budget estimates of macroplastics/microplastics for the entire Atlantic6,7 or the global ocean1,8. Our findings suggest that nanoplastics comprise the dominant fraction of marine plastic pollution.

The cause of Mars's loss of surface habitability is unclear, with isotopic data suggesting a 'missing sink' of carbonate1. Past climates with surface and shallow-subsurface liquid water are recorded by Mars's sedimentary rocks, including strata in the approximately 4-km-thick record at Gale Crater2. Those waters were intermittent, spatially patchy and discontinuous, and continued remarkably late in Mars's history3-attributes that can be understood if, as on Earth, sedimentary-rock formation sequestered carbon dioxide as abundant carbonate (recently confirmed in situ at Gale4). Here we show that a negative feedback among solar luminosity, liquid water and carbonate formation can explain the existence of intermittent Martian oases. In our model, increasing solar luminosity promoted the stability of liquid water, which in turn formed carbonate, reduced the partial pressure of atmospheric carbon dioxide and limited liquid water5. Chaotic orbital forcing modulated wet-dry cycles. The negative feedback restricted liquid water to oases and Mars self-regulated as a desert planet. We model snowmelt as the water source, but the feedback can also work with groundwater as the water source. Model output suggests that Gale faithfully records the expected primary episodes of liquid water stability in the surface and near-surface environment. Eventually, atmospheric thickness approaches water's triple point, curtailing the sustained stability of liquid water and thus habitability in the surface environment. We assume that the carbonate content found at Gale is representative, and as a result we present a testable idea rather than definitive evidence.

Classical mechanics characterizes the kinetic energy of a particle, the energy it holds due to its motion, as consistently positive. By contrast, quantum mechanics describes the motion of particles using wave functions, in which regions of negative local kinetic energy can emerge1. This phenomenon occurs when the amplitude of the wave function experiences notable decay, typically associated with quantum tunnelling. Here, we investigate the quantum mechanical motion of particles in a system of two coupled waveguides, in which the population transfer between the waveguides acts as a clock, allowing particle speeds along the waveguide axis to be determined. By applying this scheme to exponentially decaying quantum states at a reflective potential step, we determine an energy-speed relationship for particles with negative local kinetic energy. We find that the smaller the energy of the particles-in other words, the more negative the local kinetic energy-the higher the measured speed inside the potential step. Our findings contribute to the ongoing tunnelling time debate2-6 and can be viewed as a test of Bohmian trajectories in quantum mechanics7-9. Regarding the latter, we find that the measured energy-speed relationship does not align with the particle dynamics postulated by the guiding equation in Bohmian mechanics.

An increasing number of scholars, policymakers and grassroots communities argue that artificial intelligence (AI) research-and computer-vision research in particular-has become the primary source for developing and powering mass surveillance1-7. Yet, the pathways from computer vision to surveillance continue to be contentious. Here we present an empirical account of the nature and extent of the surveillance AI pipeline, showing extensive evidence of the close relationship between the field of computer vision and surveillance. Through an analysis of computer-vision research papers and citing patents, we found that most of these documents enable the targeting of human bodies and body parts. Comparing the 1990s to the 2010s, we observed a fivefold increase in the number of these computer-vision papers linked to downstream surveillance-enabling patents. Additionally, our findings challenge the notion that only a few rogue entities enable surveillance. Rather, we found that the normalization of targeting humans permeates the field. This normalization is especially striking given patterns of obfuscation. We reveal obfuscating language that allows documents to avoid direct mention of targeting humans, for example, by normalizing the referring to of humans as 'objects' to be studied without special consideration. Our results indicate the extensive ties between computer-vision research and surveillance.

Mirroring the complex structures and diverse functions of natural organisms is a long-standing challenge in robotics1-4. Modern fabrication techniques have greatly expanded the feasible hardware5-8, but using these systems requires control software to translate the desired motions into actuator commands. Conventional robots can easily be modelled as rigid links connected by joints, but it remains an open challenge to model and control biologically inspired robots that are often soft or made of several materials, lack sensing capabilities and may change their material properties with use9-12. Here, we introduce a method that uses deep neural networks to map a video stream of a robot to its visuomotor Jacobian field (the sensitivity of all 3D points to the robot's actuators). Our method enables the control of robots from only a single camera, makes no assumptions about the robots' materials, actuation or sensing, and is trained without expert intervention by observing the execution of random commands. We demonstrate our method on a diverse set of robot manipulators that vary in actuation, materials, fabrication and cost. Our approach achieves accurate closed-loop control and recovers the causal dynamic structure of each robot. Because it enables robot control using a generic camera as the only sensor, we anticipate that our work will broaden the design space of robotic systems and serve as a starting point for lowering the barrier to robotic automation.

Planets are thought to form from dust and gas in protoplanetary disks, with debris disks being the remnants of planet formation. Aged a few million up to a few billion years, debris disks have lost their primordial gas, and their dust is produced by steady-state collisions between larger, rocky bodies1,2. Tens of debris disks, with sizes of tens, sometimes hundreds, of astronomical units have been resolved with high-spatial-resolution, high-contrast imagers at optical and near-infrared or (sub)millimetre interferometers3,4. They commonly show cavities, ring-like structures and gaps, which are often regarded as indirect signatures of the presence of planets that gravitationally interact with unseen planetesimals2,5. However, no planet responsible for these features has been detected yet, probably because of the limited sensitivity (typically 2-10 MJ) of high-contrast imaging instruments (see, for example, refs. 6-9) before the James Webb Space Telescope. Here we have used the unprecedented sensitivity of the James Webb Space Telescope's Mid-Infrared Instrument10,11 in the thermal infrared to search for such planets in the disk of the approximately 6.4-Myr-old star TWA 7. With its pole-on orientation, this three-ring debris disk is indeed ideally suited for such a detection. We unambiguously detected a source 1.5 arcsec from the star, which is best interpreted as a cold, sub-Jupiter-mass planet. Its estimated mass (about 0.3 MJ) and position (about 52 AU, de-projected) can thoroughly account for the main disk structures.

Lanthanides have shown magnetic memory at both the atomic1,2 and molecular3,4 level. The magnetic remanence temperatures of lanthanide single-molecule magnets can surpass d-transition metal examples5,6, and since 2017, energy barriers to magnetic reversal (Ueff) from 1,237(28) cm-1 to 1,631(25) cm-1 and open magnetic hysteresis loops between 40 K and 80 K have typically been achieved with axial dysprosium(III) bis(cyclopentadienyl) complexes7-17. It has been predicted that linear dysprosium(III) compounds could deliver greater mJ (the projection of the total angular momentum, J, on a quantization axis labelled z) state splitting and therefore higher Ueff and hysteresis temperatures18-21, but as lanthanide bonding is predominantly ionic22,23, so far dysprosium bis(amide) complexes have shown highly bent geometries that promote fast magnetic reversal24,25. Here we report a dysprosium bis(amide)-alkene complex, [Dy{N(SiiPr3)[Si(iPr)2C(CH3)=CHCH3]}{N(SiiPr3)(SiiPr2Et)}][Al{OC(CF3)3}4] (1-Dy), that shows Ueff = 1,843(11) cm-1 and slow closing of soft magnetic hysteresis loops up to 100 K. Calculations show that the Ueff value for 1-Dy arises from the charge-dense amide ligands, with a pendant alkene taking a structural role to enforce a large N-Dy-N angle while imposing only a weak equatorial interaction. This leads to molecular spin dynamics up to 100 times slower than the current best single-molecule magnets above 90 K.

A key virtue of spin qubits is their sub-micron footprint, enabling a single silicon chip to host the millions of qubits required to execute useful quantum algorithms with error correction1-3. However, with each physical qubit needing multiple control lines, a fundamental barrier to scale is the extreme density of connections that bridge quantum devices to their external control and readout hardware4-6. A promising solution is to co-locate the control system proximal to the qubit platform at milli-kelvin temperatures, wired up by miniaturized interconnects7-10. Even so, heat and crosstalk from closely integrated control have the potential to degrade qubit performance, particularly for two-qubit entangling gates based on exchange coupling that are sensitive to electrical noise11,12. Here we benchmark silicon metal-oxide-semiconductor (MOS)-style electron spin qubits controlled by heterogeneously integrated cryo-complementary metal-oxide-semiconductor (cryo-CMOS) circuits with a power density sufficiently low to enable scale-up. Demonstrating that cryo-CMOS can efficiently perform universal logic operations for spin qubits, we go on to show that milli-kelvin control has little impact on the performance of single- and two-qubit gates. Given the complexity of our sub-kelvin CMOS platform, with about 100,000 transistors, these results open the prospect of scalable control based on the tight packaging of spin qubits with a 'chiplet-style' control architecture.

Emergence of universal collective behaviour from interactions within a sufficiently large group of elementary constituents is a fundamental scientific concept1. In physics, correlations in fluctuating microscopic observables can provide key information about collective states of matter, such as deconfined quark-gluon plasma in heavy-ion collisions2 or expanding quantum degenerate gases3,4. Mesoscopic colliders, through shot-noise measurements, have provided smoking-gun evidence on the nature of exotic electronic excitations such as fractional charges5,6, levitons7 and anyon statistics8. Yet, bridging the gap between two-particle collisions and the emergence of collectivity9 as the number of interacting particles increases10 remains a challenging task at the microscopic level. Here we demonstrate all-body correlations in the partitioning of electron droplets containing up to N = 5 electrons, driven by a moving potential well through a Y-junction in a semiconductor device. Analysing the partitioning data using high-order multivariate cumulants and finite-size scaling towards the thermodynamic limit reveals distinctive fingerprints of a strongly correlated Coulomb liquid. These fingerprints agree well with a universal limit at which the partitioning of a droplet is predicted by a single collective variable. Our electron-droplet scattering experiments illustrate how coordinated behaviour emerges through interactions of only a few elementary constituents. Studying similar signatures in other physical platforms such as cold-atom simulators4,11 or collections of anyonic excitations8,12 may help identify emergence of exotic phases and, more broadly, advance understanding of matter engineering.

Capturing particles with low, nanonewton-scale adhesion is an ongoing challenge for conventional air filters1,2. Inspired by the natural filtration abilities of mucus-coated nasal hairs3,4, we introduce an efficient, biomimetic filter that exploits a thin liquid coating. Here we show that a stable thin liquid layer is formed on several filter media that generates enhanced particulate adhesion, driven by micronewton to sub-micronewton capillary forces5,6. Enhanced particle adhesion increases the filtration of airborne particulates while maintaining air permeability, providing longer filter lifetime and increased energy savings. Moreover, strong adhesion of the captured particles enables effective filtration under high-speed airflow as well as suppression of particle redispersion. We anticipate that these filters with thin liquid layers afford a new way to innovate particulate matter filtering systems.

The pursuit of strong yet ductile alloys has been ongoing for centuries. However, for all alloys developed thus far, including recent high-entropy alloys, those possessing good tensile ductility rarely approach 2-GPa yield strength at room temperature. The few that do are mostly ultra-strong steels1-3; however, their stress-strain curves exhibit plateaus and serrations because their tensile flow suffers from plastic instability (such as Lüders strains)1-4, and the elongation is pseudo-uniform at best. Here we report that a group of carefully engineered multi-principal-element alloys, with a composition of Fe35Ni29Co21Al12Ta3 designed by means of domain knowledge-informed machine learning, can be processed to reach an unprecedented range of simultaneously high strength and ductility. An example of this synergy delivers 1.8-GPa yield strength combined with 25% truly uniform elongation. We achieved strengthening by pushing microstructural heterogeneities to the extreme through unusually large volume fractions of not only coherent L12 nanoprecipitates but also incoherent B2 microparticles. The latter, being multicomponent with a reduced chemical ordering energy, is a deformable phase that accumulates dislocations inside to help sustain a high strain hardening rate that prolongs uniform elongation.

Sexual reproduction relies on meiotic chromosome pairing to form bivalents, a process that is complicated in polyploids owing to the presence of multiple subgenomes1. Uneven ploidy mostly results in sterility due to unbalanced chromosome pairing and segregation during meiosis. However, pentaploid dogroses (Rosa sect. Caninae; 2n = 5x = 35) achieve stable sexual reproduction through a unique mechanism: 14 chromosomes form bivalents and are transmitted biparentally, while the remaining 21 chromosomes are maternally inherited as univalents2,3. Despite being studied for over a century, the role of centromeres in this process has remained unclear. Here we analyse haplotype-resolved chromosome-level genome assemblies for three pentaploid dogroses. Subgenome phasing revealed a bivalent-forming subgenome with two highly homozygous chromosome sets and three divergent subgenomes lacking homologous partners, therefore explaining their meiotic behaviour. Comparative analyses of chromosome synteny, phylogenetic relationships and centromere composition indicate that the subgenomes originated from two divergent clades of the genus Rosa. Pollen genome analysis shows that subgenomes from different evolutionary origins form bivalents, supporting multiple origins of dogroses and highlighting variation in subgenome contributions. We reveal that bivalent-forming centromeres are enriched with ATHILA retrotransposons, contrasting with larger tandem-repeat-based centromeres mainly found in univalents. This centromere structural bimodality possibly contributes to univalent drive during female meiosis. Our findings provide insights into the unique reproductive strategies of dogroses, advancing our understanding of genome evolution, centromere diversity and meiotic mechanisms in organisms with asymmetrical inheritance systems.

Climate change threatens global food systems1, but the extent to which adaptation will reduce losses remains unknown and controversial2. Even within the well-studied context of US agriculture, some analyses argue that adaptation will be widespread and climate damages small3,4, whereas others conclude that adaptation will be limited and losses severe5,6. Scenario-based analyses indicate that adaptation should have notable consequences on global agricultural productivity7-9, but there has been no systematic study of how extensively real-world producers actually adapt at the global scale. Here we empirically estimate the impact of global producer adaptations using longitudinal data on six staple crops spanning 12,658 regions, capturing two-thirds of global crop calories. We estimate that global production declines 5.5 × 1014 kcal annually per 1 °C global mean surface temperature (GMST) rise (120 kcal per person per day or 4.4% of recommended consumption per 1 °C; P < 0.001). We project that adaptation and income growth alleviate 23% of global losses in 2050 and 34% at the end of the century (6% and 12%, respectively; moderate-emissions scenario), but substantial residual losses remain for all staples except rice. In contrast to analyses of other outcomes that project the greatest damages to the global poor10,11, we find that global impacts are dominated by losses to modern-day breadbaskets with favourable climates and limited present adaptation, although losses in low-income regions losses are also substantial. These results indicate a scale of innovation, cropland expansion or further adaptation that might be necessary to ensure food security in a changing climate.