A storm-chasing campaign will help scientists to understand why some storms yield large, damaging hailstones and others generate no hail at all. Credit: Benjamin Cremel/AFP/Getty Hail regularly pulverizes crops and smashes up homes around the world — but it remains a scientific mystery despite its menace. Researchers still don’t know why some storms make monster hailstones, whereas others drop only harmless fragments of ice. Now meteorologists hope to clear up some of the unknowns about hail, which causes tens of billions of dollars in damage annually in the United States alone. Over a six-week period, dozens of scientists will chase ice-dropping storms across the US Great Plains, in the biggest US study devoted to hail in four decades. “We have so few observations that that any observation of any hailstorm is going to give us new and exciting information,” says Rebecca Adams-Selin, an atmospheric scientist at the consultancy firm Atmospheric and Environmental Research in Lexington, Massachusetts. She is lead principal investigator of the field campaign, known as ICECHIP. Hailstone nurseries Hail forms during strong thunderstorms when upward-flowing winds carry raindrops to higher, colder regions of the atmosphere, where the drops freeze. Depending on how the hailstone is blown around during the storm, it can either grow or shrink before it falls to the ground. But the environment in which this happens is something of a black box. ICECHIP is studying how baby hailstones form and accumulate ice as they move through a thunderstorm, and what conditions create the most damaging hail. The instruments researchers are using include a funnel-and-freezer combination to gather and store hailstones, panels studded with pressure sensors that convert hailstone impacts into kinetic energy, and a drone that appeared in last year’s tornado-chasing film Twisters. Night-time storm chasers stalk their prey on US Plains The results should help people to better prepare for damage from incoming hail, says Ian Giammanco, a meteorologist at the Insurance Institute for Business and Home Safety in Richburg, South Carolina. That might mean designing better hail-resistant roofs, putting hail netting over vulnerable fruit trees or adjusting the angle of solar panels to allow hail to bounce off them. Such measures could help to contain the rising economic toll of hail damage. “This exponential growth in loss is unsustainable, and we’ve got to start figuring out ways to take a bite out of it,” he says. Global warming is likely to make things even worse. Although in a warmer climate thunderstorms are expected to harbour warmer air, which could melt ice as it descends, they are also projected to feature more and stronger updrafts that could give hailstones more chances to form and more time to grow1. Modelling studies suggest that the overall trend will be towards larger and more damaging hailstones, says Victor Gensini2, a meteorologist at Northern Illinois University in DeKalb. Filling the gap However, ICECHIP’s future is not assured. It is funded by the US National Science Foundation, which has been slashing research monies under cost-cutting and ideological directives from the administration of US President Donald Trump. ICECHIP’s leaders say they have the funding for this year’s fieldwork, but follow-up campaigns are now in doubt. This year’s six-week study is costing US$11 million — nearly equivalent to the losses incurred at a peach-tree farm from a hailstorm that hit Florida last month, Gensini notes. The ICECHIP team is currently helping the beleaguered National Weather Service, which can no longer perform some of its crucial weather-spotting work because of funding and staffing cuts made by the Trump administration. Some offices in the region where ICECHIP works have cut back on launching weather balloons, which gather observations that feed into weather forecasting. ICECHIP researchers are launching hundreds of their own weather balloons, and are sharing their data with the weather service. “But, you know, the campaign is only temporary,” says Gensini. It will all be over by the end of June.

Students enter Harvard Yard, on the university’s main campus in Cambridge, Massachusetts.Credit: John Tlumacki/The Boston Globe via Getty As the US government slashes Harvard University’s funding, the damage to research at the school is becoming clearer. Nature has learnt that researchers at the university have lost nearly 1,000 grants worth more than US$2.4 billion. Will US science survive Trump 2.0? Last week, the administration of US President Donald Trump announced the terminations in a press release, but did not specify how many would be targeted or list individual grants. Nature obtained the figures from a variety of sources, including US funding agency employees and an online volunteer tracking effort at grant-watch.us. An e-mail to Harvard from the US National Science Foundation (NSF) lists 193 grants worth nearly $150 million as being terminated, and one from the US Department of Defense (DoD) logs 56 grants worth $105 million. Other cuts are smaller: for instance, the Department of Agriculture and the Department of Housing and Urban Development each terminated three grants. But by far, the largest tranche comes from the US National Institutes of Health (NIH), the world’s largest funder of biomedical science: it is cutting more than 600 grants worth about $2.2 billion over multiple years. The cuts do not include Harvard-affiliated hospitals. Through research grants, the US government funds about 11% of Harvard’s annual $6.4 billion budget, and these cancellations will be devastating, researchers say. “Harvard cannot, even with its vast resources, just make up for this loss of federal funding,” says Joseph Loparo, a biological chemist at Harvard Medical School in Boston, Massachusetts, who lost two NIH grants for studying repair processes in DNA totalling $4.3 million. In the crosshairs Harvard, whose main campus is in Cambridge, Massachusetts, is one of the most prestigious universities in the world — and the wealthiest — with its $53-billion endowment. The university has been a prime target for the Trump administration as it seeks to eradicate what it calls ‘woke’ ideology from US campuses. According to The New York Times, Trump posed the possibility of never paying Harvard its allotment of grant money during a private luncheon on 1 April. “Wouldn’t that be cool?” he asked. On Thursday, the US Department of Homeland Security made the extraordinary announcement that it had cancelled Harvard’s ability to enroll international students — a substantial revenue stream. Today, the university sued, and a US judge quickly placed a temporary freeze on the Trump administration’s policy until a hearing can be held. Although the Trump administration has terminated grants at other research institutions — such as Columbia University in New York City — the cancellations at Harvard are exceptional in scale. The vast majority of the university’s NIH awards have been terminated, for example. And the cuts across multiple agencies include funding for stated priorities of the Trump administration, such as artificial intelligence and quantum physics. A $20-million grant for a quantum materials centre was axed, along with several multimillion-dollar grants for quantum computing. Many of these grants have multi-institution collaborators whose funding situation is unclear. Alan Garber, Harvard’s president, arrives for a graduation ceremony at the university.Credit: Craig F. Walker/The Boston Globe via Getty The Trump team has alleged that Harvard and other universities have fostered an environment of antisemitism. In e-mails it sent justifying the grant terminations, the administration said that Harvard has also engaged in “race discrimination” in admissions. In early April, government officials contacted Harvard and presented it with a list of demands that must be met in order for the university to continue receiving federal money, some of which would give the government oversight of its admissions and hiring practices. Harvard denied the request publicly, saying that it would be a violation of academic freedom. In response, the administration froze the institution’s research grants. Harvard sued on 21 April, arguing that the government was withholding federal funding “as leverage to gain control of academic decisionmaking at Harvard”. When asked for a response to last week’s terminations, a Harvard spokesperson pointed Nature to public comments made by university president Alan Garber on 14 May: “We stand behind our thousands of outstanding faculty, postdoctoral, staff, and student researchers. Together they continue to make revolutionary discoveries, cure illness, deepen our understanding of the world, and translate that understanding into impact … It is crucial for this country, the economy, and humankind that this work continue.” The White House and the DoD did not respond to Nature’s request for comment, and the NSF declined to comment. A spokesperson from the US Department of Health and Human Services, which oversees the NIH, stated that “The HHS is taking decisive action to uphold civil rights and protect taxpayer investments in higher education.” Laboratory losses Scientists at Harvard who have lost funding spoke to Nature about the impact of the terminations on their research. How Trump’s attack on universities is putting research in peril David Charbonneau, an exoplanet researcher at Harvard, had a multi-year NSF grant worth $538,000 terminated with one year to go. The grant funded operation of the Arizona-based Tierras Observatory, which records tiny dips in light from stars — a tell-tale sign of a planet from outside the Solar System passing in front of them. Charbonneau has managed to secure funding from Harvard to pay the postdoctoral fellow who has been operating Tierras for the next year, but beyond that it’s unclear what will happen. “How does cutting a research grant in astrophysics to look for planets orbiting other stars address antisemitism on campus?” Charbonneau asks. Postdocs and students are some of the primary targets of the administration’s cuts. The NSF cut a $43-million grant that included all of Harvard’s funding for the agency’s graduate research fellowship programme. An NSF source who requested anonymity because they are not authorized to speak to the press tells Nature that the fellowships have not been terminated, but that students cannot access the funds unless they transfer to a new institution. “Those fellowships were awarded directly to students, they weren't awarded to Harvard — the students earned that recognition,” says Emily Balskus, a chemical biologist at the university. Three students in her lab have NSF fellowships, and she says that Harvard has committed to financially supporting them in the short term. But the loss of those grants “makes me worry about the future of science”, she says. In 2020, Balskus received the Waterman Award, a prestigious $1-million prize awarded to between one and three young scientists each year, for her research on the chemistry of microorganisms. Winning that recognition motivated her team “to be that much more creative and ambitious with our scientific goals”, she says. Earlier this month, it too was terminated as part of the university-wide purge. Ripple effects Grant terminations at Harvard aren’t just affecting research at the institution itself. They have implications for researchers at other US universities, who collaborate with Harvard scientists and often share grants with them, and for those farther abroad. ‘My career is over’: Columbia University scientists hit hard by Trump team’s cuts Molly Franke, an epidemiologist at the Harvard T.H. Chan School of Public Health in Boston, works with about 20 researchers in Peru studying interventions that could help teenagers with HIV. The grant for this work, worth $2.2 million, was terminated last week. “It’s quite a vulnerable group, and so to terminate a grant like that in the middle is quite unethical,” she says. There is funding until the end of the month, but after that it’s not clear how the work can continue. Franke also had three other NIH grants terminated, totaling $4.4 million. Two of these were for studying drug-resistant tuberculosis, which globally kills about 150,000 people each year. In the long term, the termination of grants like these could have “devastating consequences” for global-health infrastructure, she says. The cuts are particularly shocking to Franke’s international colleagues. “I’ll get news articles from our collaborators in Lesotho or Peru, and they'll say, ‘Is this really true?’,” she adds. “I think they kind of can’t believe that this is happening.”

The year was 2013, and the release of hotly anticipated zombie-apocalypse video game was on the horizon. The game, called The Last of Us, invited players to explore what then seemed a fanciful scenario: a world devastated by a pandemic in which a pathogen kills millions of people. Unlike in many apocalypse fictions, the pathogen responsible wasn’t a bacterium or a virus, but a fungus called Cordyceps that infects humans and takes over their brains. Will bird flu spark a human pandemic? Scientists say the risk is rising The writers at game studio Naughty Dog, based in Santa Monica, California, were inspired by real fungi — particularly Ophiocordyceps unilateralis, known as the zombie-ant fungus. The fungus infects insects and releases chemicals into the animals’ brains to change their behaviour. Ahead of the game’s release, Naughty Dog turned to scientists, including behavioural ecologist David Hughes, a specialist in zombie-ant fungi (he named one after his wife), to field questions from the media about the fungal and pandemic science that inspired the story. Hughes, who is at the Pennsylvania State University in University Park, has since moved to studying climate change and food security. The Last of Us spawned a sequel game in 2020 and a critically acclaimed television show, the second season of which concludes on 25 May on HBO. Hughes spoke to Nature about his experience consulting on the game and why COVID-19 changed our appetite for zombies. What was your involvement with the game? Naughty Dog studios asked me and a few other people who were notable in this space, including psychologists, to talk about whether we could have a global pandemic. Of course, in the intervening period, we all learnt that the answer was yes. They asked us to go around Europe and do a series of lectures to stave off critique and provide support to the idea that infections that jump from one species into another — zoonotic infections — are not only possible, but actually they’re the predominant mechanism by which humans are infected with new parasites that cause disease. I had the good fortune to go to the studios and see the artistry that was involved, meet the team and the voice actors of the of the video game. Behavioural ecologist David Hughes advised the game’s writers. Credit: Huck Institutes of the Life Sciences at Penn State What did you make of the science in the game? I was really impressed by how much the game’s writers got into the science of it and started to understand things like fungi and slime moulds, and just trying to think about the ways in which these organisms do their business. They really took it by themselves and incorporated those elements into the game. I think they were even mail-ordering slime moulds so they could just leave it out on a petri dish and examine it. And you see that throughout the game. And now the TV programme, in the intros, they have these slime balls. The writers were geeky, and understanding fungi is not complex, so they ran with it. Did you play the games? I tried and I failed miserably! I’m just a typical hopeless scientist. A scene from the show showing Cordyceps fungus growing out of the body of a human it has infected. Credit: Liane Hentscher/HBO Is the idea of a Cordyceps pandemic realistic? It is not unrealistic that fungi can infect humans if they come from animals. It is unrealistic to think that they could cause the behavioural changes in humans. The writers took liberties. They had different stages about how the infection changes over time. That’s all fanciful, of course. Looking at the second season of the TV show, it was interesting that they have this communicative nature of the spores or the fungal hyphae. That’s interesting because we know fungi are connected that over many kilometres — for example, the mycorrhizal fungi, which are underneath root systems in trees, do that effectively. Have you been impressed by the science in the TV show? I often find that’s the wrong question, because I don”t think the job of the entertainment industry is to impress scientists. Scientists are highly problematic individuals. It’s called the Carl Sagan effect. The more you popularize science, the less good your science is. It’s an inverse relationship. I think it doesn’t really matter. Science belongs to society, and people should tell stories about that. And, you know, snooty scientists saying, ‘Oh, you didn’t get this exactly right’, — like, who cares? The real fungus Ophiocordyceps unilateralis, which manipulates the brains of ants to ensure its own onward transmission, inspired the game’s writers.Credit: David P. Hughes What was your reaction when the COVID-19 pandemic happened? I told you so! In The Last of Us lectures I talk about the same thing. I said, the problem is not whether we’ll have zombie-ant fungi manipulating humans. It’s not going to happen. The problem is if we lose 5% of our population, and the global economy shuts down, which we saw. Do you think the COVID-19 pandemic changed our appetite for zombie-apocalypse media? It’s very interesting. You build a game about a dystopian future based on a pandemic, you live through a pandemic, and then what’s the relevance of the game or the movie? I think our appetite for being scared by pandemics has receded because we all have PTSD. Or, we don’t have PTSD and realized that some of us just don’t care about other people. So it’s interesting to look at the history of zombie lore. Back in the 1950s and 60s, it was all about nuclear weapons, because we were all collectively fearful of that. And then it moved into diseases, because we had an over-populated society. Then we had a pandemic, and we shrugged and moved on. So the fascinating thing is, The Last of Us is nice, but it’s not what it used to be.

Abstract Hydrogen peroxide (H2O2) is a vital industrial chemical and sustainable energy carrier. However, achieving a simple, efficient and cost-effective synthesis under mild conditions remains an important challenge. Here we show that SnSe nanosheets with Sn vacancies can directly catalyse H2O2 production from H2O and O2 under ambient conditions, without additional energy inputs (for example, light and electricity), cocatalysts or sacrificial reagents. This approach achieves an optimal H2O2 production rate of ~2.6 mmol g−1 h−1 at 40 °C and maintains long-term stable production (~0.3 mmol l−1) in a continuous-flow reactor for over 50 h at room temperature. Experimental and theoretical analyses reveal that this unique thermocatalytic effect arises from a dynamic process involving Sn vacancy defect-induced sequential dissociation of H2O and activation of O2 molecules, along with reversible surface restructuring of the SnSe nanosheets to release H2O2. Our findings offer a notably simple, highly efficient and entirely green strategy for H2O2 production, with broader implications in other catalytic reactions involving water activation. You have full access to this article via Lanzhou Branch of National Science Library, CAS. Download PDF Main Hydrogen peroxide (H2O2), one of the key industrial products, is widely utilized in chemical synthesis, water purification, biological treatment and microelectronic manufacturing1. The global demand for H2O2 is forecasted to increase to 5.7 million metric tonnes by 2028, with a compound annual growth rate of 4.6% (ref. 2). In addition to its current utility, H2O2 is increasingly recognized as a potential high-density energy carrier (3.0 MJ l−1, 60 wt%) in sustainable energy systems, providing a safer and more convenient alternative to hydrogen (H2) in terms of storage and transportation3,4,5. However, the current anthraquinone process for industrial H2O2 production is energy intensive and environmentally detrimental6, prompting extensive research into alternative synthetic routes operating under milder and greener conditions. Among these, the direct synthesis from a mixture of H2 and O2 has emerged as a promising approach7. Nevertheless, this method faces severe constraints imposed by the necessity of Pd-based catalysts and the handling of high-pressure, flammable H2. To address these safety, cost and environmental concerns, producing H2O2 directly from H2O and O2 has been regarded as an ideal solution with 100% atom utilization efficiency. However, this approach remains a formidable challenge due to the inherent stability of water molecules, characterized by the high homolytic cleavage bond dissociation energy (118 kcal mol−1) of O‒H bonds8. Current strategies to overcome this barrier typically require high-energy inputs, such as light, electricity or mechanical force, involving photocatalysis3,9,10,11,12, laser-direct synthesis13, electrocatalysis14,15,16,17 and mechanocatalysis18,19. Moreover, these methods often rely on complex and expensive catalysts, thereby greatly limiting their practical applicability. We report a thermocatalytic route for H2O2 production exclusively from H2O and O2 over a moderate temperature range of 0–60 °C, utilizing SnSe nanosheets (NSs) with Sn vacancies as catalysts, without requiring additional energy input (for example, light and electricity), cocatalysts or sacrificial reagents. By optimizing the concentration of Sn vacancies, we achieved an exceptional H2O2 production rate of ~2.6 mmol g−1 h−1 with Sn0.9Se NSs at 40 °C, placing our method among the top-performing technologies such as photocatalysis. Experimental results and theoretical simulations revealed a unique defect-induced thermocatalytic effect that involves a dynamic process of adsorption–dissociation of water molecules and Sn vacancy-direct surface restructuring of SnSe. This process enables dual-channel synthesis of H2O2 through simultaneous two-electron water oxidation reaction (WOR) and oxygen reduction reaction (ORR) pathways. This simple, mild and efficient thermocatalytic strategy uses earth-abundant and eco-friendly SnSe (refs. 20,21) nanomaterials as catalysts under ambient conditions, demonstrating superior efficiency and economic advantages over existing technologies. Results Thermocatalytic H2O2 production activity The thermocatalytic H2O2 production activity of a series of Sn1−xSe (x = −0.1, 0, 0.05, 0.1 and 0.2) NSs, synthesized via a hot-injection method by adjusting the stoichiometric ratios of the Sn and Se precursors, was evaluated by dispersing the powder catalysts in deionized (DI) water under varying temperatures. The reaction was performed in the dark using a sealed set-up that excludes any contribution from light, as illustrated in Supplementary Fig. 1. A constant temperature was maintained using a water bath to supply thermal energy. Remarkably, the SnSe NSs enabled consistent H2O2 production across a temperature range from 0 °C to 60 °C (Fig. 1a and Supplementary Fig. 2). The yields of H2O2 increased with rising reaction temperature, but production declined after 2 h at 60 °C owing to thermal decomposition of H2O2. Consequently, subsequent tests were conducted at a moderate temperature of 40 °C. As shown in Fig. 1b,c, the H2O2 yield substantially increased with the introduction of more Sn vacancies compared with SnSe NSs. The highest production amount of ~7.8 mmol g‒1 H2O2 was obtained using Sn0.9Se NSs after 3 h of reaction. However, further increasing the defect concentration led to a decline in yield, as in the case for Sn0.8Se NSs. By optimizing the load amount of the Sn0.9Se catalyst, an optimal H2O2 evolution rate of ~2.6 mmol g−1 h−1 was achieved (Supplementary Fig. 3). This efficiency is comparable to that of most leading photocatalytic materials, such as C3N4 (refs. 11,22), ZnIn2S4 (ref. 23), CoOx/Mo:BiVO4/Pd (ref. 24), resorcinol–formaldehyde resins12, tetrakis(4-carboxyphenyl) porphyrin3 and covalent organic frameworks9,25,26,27, as listed in Supplementary Table 1. Notably, SnSe bulk crystals exhibited negligible activity, and Sn1.1Se NS samples with fewer Sn defects displayed an inferior generation rate (~1.2 mmol g−1 h−1) compared with SnSe NSs (Fig. 1c), further demonstrating the vital role of Sn vacancies in this thermocatalytic process. The turnover frequency for H2O2 production using the optimal Sn0.9Se sample was calculated to be 0.533 h−1 (Supplementary Note 1). Although this thermocatalytic reaction route demonstrates a relatively lower turnover frequency compared with conventional direct synthesis methods utilizing gases (H2/O2/CO) under high pressure (0.1–4 MPa)28,29, it is comparable to well-studied photocatalytic routes (Supplementary Table 2)9,30. Importantly, the H2O2 production yields over the Sn0.9Se NSs stayed nearly constant after ten consecutive cycles (Fig. 1d), and the pH value of the reaction medium remained almost unchanged throughout the process (Fig. 1e). Additional characterizations, provided in Supplementary Figs. 4 and 5, confirmed the structural and chemical stability of the catalysts before and after use, demonstrating their excellent durability under mild reaction conditions. Fig. 1: Thermocatalytic H2O2 generation from H2O and O2 using Sn1−xSe NSs under ambient conditions. a, Temperature-dependent H2O2 production using SnSe NSs. b, H2O2 yields versus reaction time for various Sn1−xSe NSs and bulk SnSe single crystals at 40 °C. c, Corresponding H2O2 production rates for different catalysts. d, Cycling catalytic tests using Sn0.9Se NSs at 40 °C. e, Monitoring of the pH value of the reaction solution during catalytic tests with Sn0.9Se NSs over 3 h. f, Long-term H2O2 production in a continuous-flow reactor immobilized with Sn0.9Se NSs at room temperature (26.1–27.5 °C). Error bars represent the s.d. of three replicate tests. Source data Full size image Characterizations of Sn vacancies X-ray diffraction patterns of all samples confirmed a pure Pnma orthorhombic structure (PDF#48-1224) without detectable impurity phases (Supplementary Fig. 6). All Sn1−xSe samples exhibited similar square-shaped NS morphologies, with thicknesses ranging from ~5 to 10 nm and lateral sizes extending up to several micrometres (Fig. 2a and Supplementary Fig. 7). Energy-dispersive X-ray spectroscopy mapping revealed a homogeneous distribution of Sn and Se elements throughout the individual single-crystalline NSs (Supplementary Fig. 7). The elemental compositions of the synthesized series Sn1−xSe NSs were further determined using inductively coupled plasma optical emission spectrometry (Supplementary Table 3), indicating the absence of detectable metal impurities. Using Sn0.9Se as a representative sample, atomic-resolution high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images displayed perfect lattices matching the crystal model of SnSe projected along the [001] zone axis (Fig. 2b), demonstrating the high crystalline quality of the Sn1−xSe NSs. Fig. 2: In situ heating STEM characterizations. a, A low-magnification annular bright-field STEM image of an individual Sn0.9Se NS. b, A high-resolution HAADF-STEM image of the lattice along the [001] axis (scale bar, 1 nm), with superimposed Sn (purple atoms) and Se (green atoms) at each atomic column. c–h, A series of HAADF-STEM images of the same region acquired at 60 °C at 0 s (c), 25 s (d) and 50 s (e), respectively (scale bar, 1 nm). Sn vacancies are marked by yellow circles, based on corresponding quantitative integrated intensity analysis of each atomic column at 0 s (f), 25 s (g) and 50 s (h), respectively. f–h. Panels c–e were Gaussian blurred for better visualization, while the analysis in f–h was generated from raw HAADF-STEM data without Gaussian blur. Source data Full size image To validate the presence of Sn vacancies, we used multiple characterization techniques including X-ray absorption near-edge structure (XANES), X-ray photoelectron spectroscopy (XPS), electron paramagnetic resonance (EPR), Raman spectroscopy and Mott‒Schottky electrochemical measurements. For comparison, bulk SnSe single crystals were synthesized following our previous report31. Unlike the bulk counterparts, the Se K-edge k2χ(k) oscillation curve of the SnSe NSs exhibited oscillation damping at 4–10 Å−1 (Extended Data Fig. 1a). Fourier transforms of the extended X-ray absorption fine structure (EXAFS) spectra (Extended Data Fig. 1b), along with fitting results (Supplementary Fig. 8 and Supplementary Table 4), showed reduced coordination numbers and increased disorder degrees for the Se–Sn scattering paths at ~2.6 Å and ~3.3 Å in the SnSe NSs compared with the bulk sample. These results indicate the presence of abundant dangling bonds on the SnSe NS surface and the distortion of the surface structure with a substantial amount of Sn vacancy defects, which are prevalent in SnSe crystals owing to their low formation energy32. Similar phenomena have been observed in other materials, including oxyhydroxide NSs33 and CeO2 ultrathin sheets34. In addition, the Se K-edge XANES spectra of Sn1−xSe NS samples shifted towards lower energy levels with increasing x values (Supplementary Fig. 9), indicating a slightly reduced valence state of Se and a weakened Se–Sn bonding interaction. This was further validated by XPS characterizations (Supplementary Fig. 10), which showed that the binding energy peaks of Se 3d shifted to lower energies with increasing defect concentration. EPR spectra of all samples exhibited a single Lorentzian line signal at g = 2.002 (Extended Data Fig. 1c). While the SnSe bulk crystal displayed a minimal signal, the peak intensity of Sn1−xSe NSs progressively increased with higher x values, suggesting the elevated concentration of Sn vacancies (Extended Data Fig. 1d). Further evidence was provided by Raman spectroscopy (Extended Data Figs. 1e and f) and ultraviolet-visible-near infrared (UV–Vis-NIR) absorption analysis (Supplementary Fig. 11). Mott‒Schottky measurements revealed that the SnSe NSs are p-type semiconductors (Supplementary Fig. 12), a characteristic typically attributed to Sn vacancies acting as acceptors35. From the slope of the straight portion of the plot, the defect density (ND) of the SnSe NSs was estimated to be 5.97 × 1016 cm−3, nearly five times greater than that of the bulk crystals (1.24 × 1016 cm−3). These characterizations collectively substantiate the existence of Sn vacancies in SnSe NSs and highlight their distinct structural and electronic properties compared with bulk SnSe crystals. To directly observe the existence and migration of Sn vacancies, in situ environmental STEM measurements were performed on Sn0.9Se NSs. Real-time quantitative integrated intensity analysis was performed on the basis of a series of HAADF-STEM images acquired under heating at 60 °C, as displayed in Fig. 2c–h. The integrated intensity map (Fig. 2f–h) showed the existence of numerous point vacancies (highlighted by yellow circles) within the lattice (Fig. 2c–e). These point defects are attributed to the formation of Sn vacancies based on the Se-rich stoichiometry in Sn0.9Se. Notably, the positions of these defects shifted between acquisitions, indicating the diffusion of Sn vacancies at moderate temperatures. While beam damage during imaging is inevitable, its impact on Sn vacancy generation and migration in the Sn0.9Se NSs is probably minimal. Intensity analysis demonstrated no trend of increasing Sn vacancies or sputtered atoms accumulating on the surface, suggesting that beam damage does not strongly affect the formation of Sn vacancies. In addition, the resistance of Sn atoms to sputtering implies that the diffusion of Sn vacancies is less reliant on beam irradiation. The observed dynamic migration behaviour of Sn vacancies in SnSe NSs under ambient conditions plays a crucial role in the thermocatalytic H2O2 production reactions, as discussed in the following sections. Mechanism of thermocatalytic H2O2 production To investigate the reaction mechanism, we conducted catalytic tests in the absence of O2. The reaction was carried out in a sealed glass chamber, which was first evacuated and then filled with Ar. As shown in Extended Data Fig. 2a, only negligible amounts of H2O2 were detected, with no increase observed over 3 h. This result clearly demonstrates that the WOR cannot proceed without O2, highlighting its essential role in the continuous production of H2O2. The dual-channel reaction mechanism was further confirmed by isotopic labelling experiments using H218O suspensions and purging with 18O2 gas, as displayed in Extended Data Fig. 2b,c. These experiments verified the simultaneous involvement of the ORR and WOR pathways in the thermocatalytic synthesis of H2O2. Moreover, as the tri-n-octylphosphine (TOP) surface ligand was introduced during the hot-injection synthesis of colloidal SnSe NSs, it is important to evaluate its influence on catalytic performance. Fourier-transform infrared spectra of SnSe NSs (Supplementary Fig. 13) showed that the vibration peaks associated with TOP ligands were well preserved after catalytic reactions, suggesting the stability of the ligand under mild reaction conditions. To determine whether the ligand contributes to catalytic performance, we synthesized SnSe NSs via a surfactant-free aqueous solution route36, avoiding the introduction of any organic ligands (Supplementary Fig. 14). These ligand-free SnSe NSs exhibited similar H2O2 production activity compared with the TOP-SnSe NSs, indicating that the surface ligand has a negligible influence on catalytic performance. However, different from SnSe NSs with TOP ligand, a tiny amount of Se metal was detected in the ligand-free catalysts after the reaction, indicating slight oxidation during the thermocatalytic process. These findings suggest that the surface ligand may play a protective role, preventing surface oxidation of SnSe NSs during the reaction. To further elucidate the reaction pathways of the thermocatalytic process, multiple in situ spectroscopic measurements were conducted. Operando Se K-edge XANES spectra of Sn0.9Se NSs were recorded to monitor changes in the Se oxidation state and local coordination structure during catalytic reactions at 40 °C (Fig. 3a). Upon interaction with water, an evident intensity enhancement of the white-line peak (that is, the 1s–4p dipole transition) was observed, indicating electron redistribution in the Se 4p orbital. This phenomenon is probably related to the formation of Se‒H bonds, where the overlap between the 1s orbital of the H atom and the 4p orbital of the Se atom increased the density of unoccupied states with 4p character and promoted the 1s–4p dipole transition, similar to a previous report37. This interpretation was further corroborated by simulated Se K-edge XANES spectra (Fig. 3b), assuming that the H atom occupied the (0.5, 0.75, 0) lattice site in SnSe (calculations for H atom absorption at different sites are provided in Supplementary Fig. 15). These results indicate that Se sites with abundant adjacent Sn vacancies are highly active in capturing H atoms, thereby facilitating the water activation process on the catalyst surface. Fig. 3: Multiple operando spectroscopic measurements. a, Normalized Se K-edge XANES spectra of SnSe NSs before and during catalytic reactions in DI water at 40 °C over 3 h. b, Calculated Se K-edge XANES spectra for SnSe and H-absorbed SnSe, with the H atom occupying the (0.5, 0.75, 0) lattice site. c, In situ DRIFTS spectra of the Sn0.9Se catalyst–water interface collected at different time intervals at 40 °C under O2 for 50 min. d, Time-resolved two-dimensional Raman mapping of Sn0.9Se NSs powder. e, An in situ interfacial Raman mapping image of Sn0.9Se NSs in O2-saturated water during the reaction process at 40 °C over 2 h. f, Corresponding in situ interfacial Raman spectra during the reaction, along with Raman spectra of Sn0.9Se NS powder acquired before and after catalytic reactions. Source data Full size image In situ diffuse reflectance infrared Fourier-transform spectroscopy (DRIFTS) was used to identify key reactive intermediates involved in the reaction (Fig. 3c). Notably, the peak at 2,860 cm−1, corresponding to O–H bonding, increased gradually over time, demonstrating the formation of H2O2 (refs. 9,38). Concurrently, the absorption peaks associated with *OOH (1,190 cm−1) and •O2− (1,228 cm−1) resonances progressively intensified, confirming the activation of O2. The observed vibration peaks of *OH (3,740 cm−1) and *H (1,528 cm−1) indicated the adsorption and dissociation of water molecules on the catalyst surface39. These spectral evolutions provide compelling evidence for the simultaneous involvement of the ORR and WOR under mild ambient conditions. Complementary in situ Raman spectra revealed the emergence of two new peaks (252.9 and 495.8 cm−1) after the Sn0.9Se NSs catalyst was immersed in O2-saturated DI water (Fig. 3e). These peaks can be attributed to the formation of Se∙∙∙H–O (ref. 40) and Sn–OH bonds41, respectively. The peak density increased gradually with reaction time, indicating the continuous splitting of water molecules. These findings are consistent with the DRIFTS measurements and XANES simulations. Furthermore, in situ EPR spectroscopy, using 5,5-dimethyl-1-pyrroline N-oxide (DMPO) as a spin-trapping agent, revealed gradually intensified characteristic signals of DMPO-•O2− and DMPO-•OH as the reaction temperature increased from 0 °C to 50 °C (Extended Data Fig. 2d,e). These results confirm that •O2− and •OH are crucial intermediates in the thermocatalytic process. Quantitative analysis further demonstrated that the concentrations of these radicals were notably higher for Sn0.9Se compared with SnSe NSs (Extended Data Fig. 2f), aligning with the observed enhancement in thermocatalytic H2O2 production (Fig. 1b). The dissociation of water molecules poses a significant challenge due to their inherent stability, characterized by the high homolytic cleavage energy of O–H bonds. Typically, water cleavage requires substantial energy inputs, such as light, electricity or mechanical force. In this study, however, water dissociation was achieved at ambient temperature on the surface of Sn vacancy-rich SnSe NSs. In conventional thermocatalytic H2O2 production methods using H2 and O2 gases, the reaction is exothermic and readily occurs at room temperature with Pd-based catalysts42. By contrast, this thermocatalytic reaction, which produces H2O2 directly from H2O and O2, is endothermic and requires external energy input. To quantify the energy requirements, we calculated the thermal energy needed for the conversion process (Supplementary Note 2). As presented in Supplementary Table 5, the thermal energy required to produce H2O2 from H2O and O2 in all experiments was less than 2.5 J (for 100 ml water over 3 h). When this energy is released from 100 ml of water, the temperature change for all experiments is merely less than 0.006 °C. This slight energy release can be readily supplied and compensated by the surrounding ambient environment, enabling the reaction to proceed continuously. Furthermore, such a negligible temperature difference is insufficient to induce a thermoelectric effect in SnSe, which typically requires a temperature exceeding ~100 K to be significant20,21,31. To gain deeper insights into the reaction mechanism of room-temperature thermocatalytic H2O2 synthesis using SnSe NSs at the atomic scale, we systematically conducted density functional theory (DFT) theoretical calculations and ab initio molecular dynamics (AIMD) simulations (for computational details, refer to the Methods, Supplementary Figs. 16–18 and Supplementary Table 6) to clarify the critical role of Sn vacancies in facilitating water dissociation and oxygen activation. We began by performing AIMD simulations on the interfacial structure of water and SnSe with Sn vacancy (denoted as VSn) to study the interaction between the water molecules and the defective SnSe NSs. As shown in Fig. 4a, compared with pristine SnSe, VSn clearly exhibited stronger interactions with water, with more H2O molecules adsorbed near the Sn vacancy. A distinct layer of adsorbed water molecules formed at the VSn–H2O interface, as further revealed by the concentration distribution of water on the material surface throughout the AIMD simulations (Fig. 4b). These simulations indicate that Sn vacancies enhance the hydrophilicity of SnSe, as validated by surface potential measurements in water (Supplementary Fig. 19). Figure 4c summarizes the adsorption energy (\({\Delta E}_{{{{\mathrm{H}}}_{2}{\mathrm{O}}}^{* }}\)) and dissociation energy (\({\Delta E}_{{{\mathrm{H}}}^{* }-{{{\mathrm{OH}}}}^{* }}\)) of water on SnSe with single (VSn) or double (V2Sn) Sn vacancies (H2O and H–OH adsorbed structures are depicted in Supplementary Fig. 20). Notably, Sn vacancies could reduce \({\Delta E}_{{{{\mathrm{H}}}_{2}{\mathrm{O}}}^{* }}\) by approximately 0.250 eV. More interestingly, Sn vacancies significantly lowered \({\Delta E}_{{{\mathrm{H}}}^{* }-{{{\mathrm{OH}}}}^{* }}\), reducing it from 1.220 eV on the SnSe surface to 0.191 eV and 0.131 eV for VSn and V2Sn, respectively. This reduction of \({\Delta E}_{{{\mathrm{H}}}^{* }-{{{\mathrm{OH}}}}^{* }}\) by Sn vacancies was further analysed using the crystal orbital Hamilton population43 calculations of Se and the corresponding adsorbed H pairs. As shown in Supplementary Fig. 21, the pz orbital of Se and the s orbital of H formed a strong bond, without any antibonding interactions below the Fermi level (Ef). The H2O dissociation process was further investigated using the climbing-image nudged elastic band method, as displayed in the calculated minimum energy pathway for H2O dissociation on the surface of VSn (Fig. 4d). The initial state (IS) involved two H2O molecules adsorbed on VSn, and the final state (FS) involved H2O dissociation around the Sn vacancy, where OH was absorbed on Sn and H was absorbed on Se, as shown in Supplementary Fig. 22. The calculated energy barrier (Eb) for H2O dissociation on VSn was found to be 0.76 eV, indicating that this process is feasible at room temperature. This result is further supported by reaction rate (k) calculation using Eyring equation (Supplementary Note 3 and Supplementary Table 7) and previous reported activation energies for water dissociation on metal/metal oxides (Eb = 0.7–0.9 eV) at room temperature44,45. Fig. 4: Theoretical modelling of room-temperature thermocatalytic H2O2 production mechanism. a, A representative snapshot of the interfacial structure of water with SnSe (left) and SnSe with a VSn (right). Purple, green, red and white balls represent Sn, Se, O and H atoms, respectively. VSn is indicated with a black arrow. b, Concentration distribution profiles of water molecules along the surface normal direction. The grey-shaded area denotes adsorbed surface water molecules. c, Calculated adsorption energy (\({\Delta E}_{{{{\mathrm{H}}}_{2}{\mathrm{O}}}^{* }}\)), dissociation energy (\({\Delta E}_{{{\mathrm{H}}}^{* }-{{{\mathrm{OH}}}}^{* }}\)) of water and adsorption energy of oxygen (\({\Delta E}_{{{{\mathrm{O}}}_{2}}^{* }}\)) on pristine SnSe and SnSe with single or double Sn vacancies. d, The calculated minimum energy pathway for H2O dissociation on VSn, with Fermi level of the initial (H2O + VSn) and final states (OH* + HSn). The H2O dissociated configuration is OH absorbs on Sn atom and H absorbs on Se atom in the Sn vacancy (denoted as HSn). Dashed green lines indicate the potential for O2/•O2− and O2/H2O2. e, Free energy diagram of H2O2 production. The inset shows the local surface reconstruction through VSn migration, where the dashed black circles represent VSn. f, A schematic diagram of thermocatalytic H2O2 production from H2O and O2 over the surface of SnSe with Sn vacancies. Full size image Due to the important role of water dissociation in water-involved reactions, this remarkable reduction in \({\Delta E}_{{{\rm{H}}}^{* }-{{\rm{OH}}}^{* }}\) by VSn facilitates the initiation of the synthesis of H2O2 from H2O and O2. To further explore the effect of Sn vacancies on O2 activation, we calculated the adsorption energy of O2 on SnSe, VSn and V2Sn (Fig. 4c). The results indicate that O2 exhibited weaker adsorption compared with H2O molecules. In contrast to H2O molecules, the presence of Sn vacancies further reduced the adsorption strength of O2. This suggests that the primary function of the Sn vacancy active sites is the adsorption and activation of H2O molecules, rather than O2. The weaker binding of O2 ensures that the surface species remain balanced, with H2O molecules being the dominant species undergoing activation at these sites. Notably, the Ef of FS was upshifted from −4.68 eV in the IS to −4.20 eV (Fig. 4d and Supplementary Fig. 23), which is higher than the potential of O2/•O2− and O2/H2O2. This upshift enables electron transfer from SnSe to oxygen to initiate the ORR process. The predicted band gap reduction of SnSe NSs after interacting with H2O molecules (Supplementary Fig. 23) was supported by the redshift of the absorption edge observed via in situ UV–Vis measurements under heating in O2-saturated water, as shown in Supplementary Fig. 24. The free energy diagram of H2O2 production from H2O and O2 on SnSe and VSn is shown in Fig. 4e. After the dissociation of water molecules, the step involving the coupling of oxygen with the adsorbed hydrogen atoms for the formation of H2O2 is spontaneous, with a ΔG of −0.763 eV. Upon completing the ORR, the desorption of OH* from the surface and its subsequent coupling to form H2O2 are critical steps in the reaction pathway. As shown in Fig. 4e, OH* cannot desorb directly from the surface of VSn, as this step requires a high ΔG of 1.137 eV. By contrast, the desorption of OH* from SnSe is more favourable, with a ΔG of 0.595 eV. Constrained AIMD simulations further confirm that the free energy barrier for OH* desorption to form H2O2 is much lower on SnSe (0.72 eV) than on VSn (1.11 eV), as shown in Supplementary Fig. 25. As discussed earlier, in situ STEM observations (Fig. 2) confirmed the migration of Sn vacancies in SnSe lattices at ambient temperature. To provide a theoretical understanding of this phenomenon, we performed DFT calculations (Supplementary Fig. 26), which revealed that the migration energy barrier for Sn vacancies in SnSe is ~0.60 eV, consistent with previous reports46,47. Furthermore, we found that OH absorption has negligible influence on the migration barrier (~0.68 eV for SnSe–OH), indicating that Sn vacancy migration remains feasible after interaction with water at room temperature. In addition, our previous work demonstrated that SnSe (Pnma) crystals exhibit strong anharmonicity, attributed to the lone pair electrons of Sn and the weak interlayer van der Waals interactions31. This intrinsic anharmonicity, coupled with strong phonon–phonon interactions, provides additional energy to Sn vacancies, creating pathways for vacancies to hop between lattice sites even at room temperature. To further validate this, we performed additional DFT calculations to evaluate the impact of the Sn vacancies on the anharmonic effects (Supplementary Note 4 and Supplementary Table 8). Our results show that the Sn vacancy defects introduce additional scattering centres and reduce sound velocity migration, thereby strengthening the anharmonic effects in SnSe (in agreement with prior studies48) and facilitating the migration of Sn vacancies through multiphonon interactions. Furthermore, to examine the potential movement of adsorbed species, we carried out additional DFT calculations to investigate the mobility of OH species near the Sn vacancy. Our calculations show that the energy barrier for OH migration is 0.82 to 0.95 eV (Supplementary Fig. 27). These values are significantly higher than the energy barriers for Sn vacancy migration (0.6 to 0.68 eV). These findings suggest that OH migration is unlikely to occur during reactions at room temperature. This supports the proposed mechanism that Sn vacancy migration, rather than OH mobility, plays a dominant role in the observed surface processes. To directly monitor the structural evolution of Sn-vacancy SnSe NSs, we performed in situ Raman spectroscopy, a technique highly effective for characterizing local structural changes49. As shown in Fig. 3d, the Raman signal of SnSe NSs remained unchanged under heating at 40 °C. In a distinct contrast, after immersing SnSe NSs in O2-saturated water, we observed a notable variation in the intensity ratio of the B3g/Ag(2) modes with respect to the reaction time (Fig. 3e,f). These results clearly indicate a surface reconstruction process of SnSe NSs during the reaction, analogous to vacancy defect-induced surface restructuration observed in electrocatalysts such as NiFe-LDH50. Importantly, SnSe NSs surface returned to its initial state after H2O2 desorption (Fig. 3f). Based on these findings, we propose a defect-direct dynamic surface restructuring mechanism for thermocatalytic H2O2 production from water and oxygen. As illustrated in the inset of Fig. 4e, the OH* desorption is facilitated by localized surface reconstruction of SnSe, driven by the migration of Sn vacancies, which weakens OH* adsorption. Based on the experimental results and theoretical calculations, the mechanism of thermocatalytic H2O2 generation over Sn vacancy-defective SnSe NSs is illustrated in Fig. 4f. The process involves three key steps. (1) H2O dissociation (equation (1)): H2O molecules dissociate at Sn vacancy sites, forming OH* on Sn atoms and H* on Se atoms. (2) Oxygen activation and H2O2 formation: O2 couples with two H* to produce H2O2 (equation (2)); concurrently, through localized surface reconstruction, OH* desorbs from the surface to form H2O2 (equation (3)). (3) Surface recovery: after OH* desorption and further surface reconstruction, the surface of SnSe recovers to its initial state. It is worth noting that excess Sn vacancy defects may impede this reconstruction process of the catalyst to the pristine state, hindering the subsequent H2O2 generation. In summary, the thermocatalytic synthesis of H2O2 from H2O and O2 over Sn vacancy-defective SnSe NSs proceeds via the following steps: $${{\rm{H}}}_{2}{\rm{O}}\to {{\rm{OH}}}^{* }+{{\rm{H}}}^{* }$$ (1) $${{\rm{O}}}_{2}+{2{\rm{H}}}^{* }\to {{\rm{H}}}_{2}{{\rm{O}}}_{2}$$ (2) $$2{{\rm{OH}}}^{* }\to {{\rm{H}}}_{2}{{\rm{O}}}_{2}.$$ (3) Equations (2) and (3) serve as the main routes for H2O2 generation, thereby completing the overall reaction route (2H2O + O2 → 2H2O2). This defect-direct dynamic process allows the continuous evolution of H2O2 from H2O and O2, driven solely by the thermal energy from the ambient environment. To evaluate the universality of the defect-direct thermocatalytic effect observed in SnSe, we synthesized a series of representative group IV–VI compounds, including SnS, PbSe, SnTe and PbTe. All samples exhibited single-crystal phases and similar NS microstructures (Supplementary Figs. 28 and 29). Notably, all synthesized NSs enabled H2O2 generation at 40 °C under O2 flow (Supplementary Fig. 30). Among these, SnTe NSs exhibited the highest catalytic efficiency, with a generation rate of ~0.5 mmol g−1 h−1. This can be attributed to the abundance of intrinsic Sn vacancies in SnTe, resulting from its low vacancy formation energy51. These findings suggest that the defect-directed thermocatalytic effect is probably universal across the group IV–VI series of semiconducting materials, opening new opportunities for exploration in similar material systems. Continuous H2O2 production at room temperature To demonstrate the industrial application potential of this thermocatalysis strategy, we conducted a scaled-up experiment using Sn0.9Se NSs as catalysts. The catalysts were filled into a commercial stainless-steel column integrated with a continuous-flow reactor that excludes any contribution from light (the experimental set-up and working process are detailed in Supplementary Fig. 31 and Supplementary Video 1). Operating at room temperature, the system generated H2O2 solutions with a steady concentration of ~0.3 mmol l−1 within 2 h and retained over 85% of its activity for 50 h (Fig. 1f). Analysis using gas chromatography coupled with mass spectrometry (Supplementary Fig. 32) and 1H-nuclear magnetic resonance (NMR; Supplementary Fig. 33) confirmed the absence of by-products in the synthesized H2O2 solution. This process substantially reduces costs compared with existing technologies and simplifies the set-up for on-site H2O2 generation. Remarkably, the catalysts exhibited slightly higher activity in saline water (~2.7 mmol g−1 h−1) and real seawater (~2.8 mmol g−1 h−1) compared with DI water (Supplementary Fig. 34). These results highlight their versatility and practical applicability of the system across diverse water sources. Conclusions We have discovered and validated a defect-direct thermocatalytic effect in SnSe NSs with Sn vacancies, enabling highly efficient, cost-effective and sustainable H2O2 synthesis. This approach relies solely on water and oxygen at ambient temperature, eliminating the need for additional energy input, cocatalysts or sacrificial reagents. As such, this method offers an almost ideal strategy and significant advantages over existing technologies. Through a combination of experimental and theoretical investigations, we demonstrated that Sn vacancy-direct reversible surface reconstruction in SnSe facilitates the continuous dissociation of water and activation of oxygen molecules through simultaneous dual-channel H2O2 generation. This mechanism is essential for driving the thermocatalytic reaction and ensuring long-term stable performance. Moreover, H2O2 production tests at room temperature utilizing a commercial continuous-flow reactor highlight the potential for eco-friendly H2O2 production on an industrially viable scale. Beyond H2O2 synthesis, we anticipate that this catalytic mechanism will enable broad applications in water activation and other important chemical reactions under ambient conditions. Methods Chemicals Selenium powder (Se, 99.998%), Tin(II) chloride (SnCl2, 99.99%), oleylamine (OLA, 80 − 90%), TOP (90%), hexamethyl disilylamine (HMDS, >98%), potassium iodide (KI, 99.0%), sodium hydroxide (NaOH, 95%) and potassium hydrogen phthalate (C8H5KO4, 99.8%) were purchased from Shanghai McLean Biochemical Technology. Ethanol (≥99.7%), isopropanol (≥99.7%) and methylbenzene (≥99.5%) were purchased from Aldrich. Ammonium molybdate ((NH4)2MoO4·4H2O, >99%) was purchased from Adamas-beta. DI water of Millipore grade was used. High-purity oxygen, nitrogen and argon gases (>99.9995%) were supplied by ALPHAGAZ. Safety precautions must be strictly followed when handling pure oxygen. Synthesis of Sn1−xSe NSs The synthesis of Sn1−xSe NSs followed a previously reported hot-injection method with slight modifications52. The synthesis procedure for SnSe is detailed as follows. First, 0.395 g of Se powder was dissolved in 5 ml of TOP to form a TOP–Se precursor solution. Then, 0.25 ml of the TOP–Se solution was mixed with 1 ml of HMDS to prepare HMDS-TOP–Se solution. Meanwhile, 47.4 mg of SnCl2 was dissolved in 20 ml of OLA and ultrasonicated for 2 min to form a Sn precursor solution, which was subsequently degassed and purged with N2 for 1 h to remove dissolved oxygen. The Sn-precursor solution was heated to 240 °C and maintained for 15 min before injecting the HMDS–TOP–Se solution. Once the mixture turned black, it was further aged at 240 °C for 30 min under N2 atmosphere. After cooling to room temperature, a mixed solution containing 40 ml of isopropanol and 8 ml of methylbenzene was added, and the mixture was then centrifuged at 12,850g for 10 min to collect the black powder. The powder was washed three times with a mixture of isopropanol and ethanol (2:1 volume ratio) by centrifugation at 12,850g for 5 min each time, and finally dried in a vacuum oven at 60 °C for 12 h. By adjusting the amount of SnCl2 according to the designed stoichiometric ratios, a series of Sn1−xSe NSs (Sn1.1Se, Sn0.95Se, Sn0.9Se and Sn0.8Se) were synthesized following the same procedures. Materials characterizations Powder X-ray diffraction analysis was performed on a Bruker D8 Advance diffractometer with a filtered Cu Kα radiation source (λ = 0.154 nm) at 45 kV and 40 mA. Transmission electron microscopy (TEM) imaging and energy-dispersive spectroscopy were conducted using a Thermo Scientific Talos F200i microscope operated at 200 kV. XPS was performed on an ESCALAB 250 Xi spectrometer (Thermo Fisher Scientific) with a monochromatic Al Kα source (hν = 1487 eV), with binding energies calibrated to the C1s peak at 284.6 eV. Inductively coupled plasma optical emission spectrometry measurements were performed using an Agilent 720ES instrument with radio frequency (RF) power of 1.2 kW (powdered samples were microwave-digested with concentrated HNO3 and diluted to 10 ml with 2% HNO3 before inductively coupled plasma analysis). Raman spectra and mappings were recorded using an inVia Qontor Raman spectrometer (Renishaw) equipped with a 532-nm laser as the excitation source. The laser power at the sample surface and exposure time were set to be 1% and 60 s, respectively. UV–Vis-NIR absorption spectra were collected on a PerkinElmer Lambda 1050 + UV/Vis/NIR spectrophotometer in the range of 500–2,500 nm. Fourier-transform infrared spectra were recorded on a Thermo Scientific Nicolet iS50 spectrometer. EPR signals were recorded using a Bruker Biospin A300 spectrometer at a frequency of ~9.85 GHz. Proton NMR measurements were performed on a Bruker Avance II 400 MHz NMR spectrometer using D2O as the solvent. Batch tests for H2O2 production First, 100 ml of DI water was oxygenated by bubbling O2 gas for 30 min in a 250-ml glass beaker. Next, the powdered catalysts (5 mg) were dispersed in the oxygen-saturated DI water by ultrasonication for 30 s. All batch catalytic experiments were conducted at controlled temperatures (0, 20, 40 and 60 °C) in a stainless-steel water bath to ensure uniform temperature distribution without stirring. The beaker was covered with aluminium foil to prevent light exposure and evaporation. During the reaction process, the solution was continuously bubbled by O2 gas. At intervals of every 30 min, 3 ml of the reaction suspension was sampled using a syringe and filtered through a 0.45-μm polyethersulfone filter to remove catalysts. For cycling performance tests, the catalysts were collected by centrifugation at 12,850g for 5 min, washed with DI water and dried in a vacuum oven at 60 °C before reuse. Flow reactor H2O2 production Typically, 0.3 g of Sn0.9Se NSs was mixed with silica sand in a mortar. The resulting mixture was then dry-packed into a stainless-steel liquid chromatography (LC) column (20 cm length, 1 cm inner diameter). This column was connected to an LC-10AT pumping system (Shimazu Scientific) for continuous flow operation. DI water, continuously bubbled with O2 gas, was pumped through the LC column at a flow rate of 1 ml min−1. The temperature of the DI water was monitored in real time using a precision temperature monitoring system. Determine of H2O2 concentration The concentration of H2O2 solution was determined using the classic iodometric method53. Two solutions were prepared: solution A (0.4 M KI, 0.06 M NaOH and 0.1 mM (NH4)2MoO4) and solution B (0.1 M C8H5KO4). For each test, 3 ml of the reaction solution was mixed with 0.5 ml of solution A and 0.5 ml of solution B, and kept for 15 min. The amount of I3− formed was determined by UV–vis spectroscopy at a wavelength of 350 nm. The H2O2 concentration was quantifiably estimated with reference to a preestablished standard curve. XAFS characterizations and analysis Se K-edge X-ray absorption fine structure (XAFS) measurements were conducted in transmission mode at the BL11B beamline of the Shanghai Synchrotron Radiation Facility, operating at 3.5 GeV with a maximum current of 210 mA. Powdered Sn1−xSe samples were mixed with boron nitride and pressed into pellets for XAFS measurements. In situ XAFS measurements were conducted using a custom-designed reaction cell, where the catalyst was dispersed in DI water and maintained at 40 °C, with Se K-edge spectra collected at hourly intervals. The acquired XAFS spectra were initially processed using the Athena module implemented in the IFEFFIT software package54 to obtain normalized XAFS spectra and EXAFS oscillation functions. The EXAFS fittings were carried out using the ARTEMIS module to derive quantitative local structural information around Se atoms in both bulk SnSe and Sn1−xSe NSs. For bulk SnSe, the coordination numbers of the scattering paths were set to nominal values, and key parameters including the amplitude reduction factor (S02), atomic distances (R), Debye–Waller factor (σ2) and energy shift (ΔE0) were fitted. The value of S02, identified as 0.76, was fixed during the fitting of Se K-edge data for SnSe NSs. The k2-weighted EXAFS function χ(k) data were fitted within a k range of 2.8–12.6 Å−1 and an R range of 1.5–3.4 Å. XANES spectra calculations for the Se K edge were performed on the basis of full multiple scattering theory using the FEFF9 program55. The self-consistent field method was used to estimate atomic scattering potentials, with cluster radii fixed at 5 Å for self-consistent field and 8 Å for full multiple scattering to ensure convergence and accuracy of the calculations. The XANES spectra were simulated by considering different lattice sites occupied by hydrogen (H) atoms in the SnSe structure to accurately model the electronic and local atomic environments. In situ HAADF-STEM characterizations The Sn0.9Se NSs were ultrasonically dispersed in anhydrous ethanol and dropped onto carbon-coated copper TEM grids. The HAADF-STEM images were acquired using a JEM-ARM200F aberration-corrected microscope (JEOL), operating at an acceleration voltage of 200 kV. In situ heating was realized using a heating chip and holder purchased from DENSsolutions, Ltd. This set-up allowed real-time observation of structural changes under controlled temperature conditions. To directly visualize Sn vacancy migration in the NSs, quantitative intensity analysis was conducted on a series of temporal Z-contrast ADF-STEM images acquired at 25-s intervals. A circular mask, sized to cover the Sn columns, was utilized. The pixel intensities within the mask were spatially integrated and normalized to track changes in Sn column intensity over time. In situ EPR measurements of free radicals The reactive oxygen species radical capture experiments were conducted using DMPO as a spin-trapping agent for the detection of •O2− and •OH radicals, performed on a Bruker Biospin A300 EPR spectrometer. For DMPO-•OH detection, 10 mg of SnSe NSs were dispersed in 5 ml of an aqueous DMPO solution (100 mmol l−1). To measure DMPO-•O2‒, 10 mg of SnSe NSs were dispersed in 5 ml of DMPO solution (100 mmol l−1, 9:1 of water/ethanol in volume). The suspensions were kept in the dark and maintained at different temperatures (0, 20, 40 and 50 °C) for 30 min before EPR analysis. In situ DRIFTS measurements In situ DRIFTS measurements were conducted using a Thermo Scientific Nicolet iS50 spectrometer equipped with a custom-designed glass reaction cell. A thin gold (Au) film was first deposited onto a silicon crystal. After that, a suspension containing 10 mg of SnSe NSs powder in ethanol was dropped onto the Au-coated silicon crystal, which was then placed in a custom-designed reaction cell with the Au film facing upwards. Subsequently, 20 ml of DI water was introduced into the reaction cell, maintained at 40 °C using a water bath and continuously bubbled with O2 gas. The DRIFTS signals were collected in reflection mode at 1-min intervals to monitor the reaction in real time. In situ Raman measurements A custom-designed reactor was used for the in situ Raman spectroscopy (inVia Qontor, Renishaw) characterizations, with a 532-nm laser as the excitation source. In this set-up, 2 ml of O2-saturated DI water was introduced into the reaction chamber, which was maintained at 40 °C using heated water. The Raman signals were collected in real time using focus-tracking mode to ensure consistent data acquisition throughout the experiments. Isotopic labelling experiments Isotopic labelling experiments were performed according to reported procedures56. For the H2O labelling tests, 2.5 mg of SnSe NSs were first added to 5 ml of an H218O solution (20% v/v in DI water) and maintained at 40 °C for 3 h. After the reaction, the suspension was filtered through a polyethersulfone filter (0.45 μm) to remove the catalyst. Subsequently, 1 ml of the obtained solution was mixed with an equal volume of a preprepared 4-carboxyphenylboronic acid (4-CPB, C7H7BO4) solution (100 μmol l−1). The product of the deborylation of 4-CPB was then analysed using LC–mass spectrometry (Thermo Scientific Q-Exactive). For the O2 labelling tests, 2.5 mg of SnSe NSs were added to 5 ml of DI water and bubbled with 18O2 gas at a flow rate of 5 ml min−1, while maintaining the temperature at 40 °C. The subsequent steps of the experiment were identical to those described for the H218O labelling test. Mott–Schottky measurements Mott–Schottky measurements were conducted using an electrochemical workstation (CHI 760E, Chenhua) in a standard three-electrode configuration. The working electrode was prepared by coating an indium tin oxide substrate with SnSe powder, while a platinum (Pt) plate and an Ag/AgCl electrode were used as the counter and reference electrodes, respectively. A deoxygenated Na2SO4 solution (0.1 M) was used as the electrolyte. Mott–Schottky plots were recorded over a voltage range of −0.05 V to 0.2 V at frequencies of 500 Hz and 1,000 Hz, respectively. Theoretical simulations DFT and AIMD simulations were carried out using the Vienna Ab Initio Simulation Package (VASP)57, using the generalized gradient approximation with the Perdew–Burke–Ernzerhof functional58. A plane-wave cut-off energy of 500 eV was used for all calculations. The SnSe NS was modelled using a slab structure with two SnSe monolayers and a vacuum layer thicker than 15 Å in a 4 × 4 × 1 supercell. The Brillouin zone was sampled using a 3 × 3 × 1 Γ-centred Monkhorst–Pack k-point mesh. The lattice constants of the SnSe layer planes were allowed to relax. All adsorbed species were optimized to convergence criteria of less than 10−5 eV for energy and 0.01 eV Å−1 for force. Van der Waals dispersion forces between adsorbates and surfaces were accounted for using Grimme’s D3 dispersion correction method59 with the Becke–Johnson damping function. To investigate Sn vacancy migration, the climbing-image nudged elastic band method was used60. The free energy (G) of a given state was calculated using the following equation61: G = E + ZPE − TS, where E is the calculated electronic energy from DFT computations, ZPE is the zero-point energy and TS is the entropy at a constant temperature of 313 K. Zero-point energy and entropy for the adsorbates were determined through vibrational frequency calculations using density functional perturbation theory. Entropy values for molecules (H2O, O2 and H2O2) were sourced from the Computational Chemistry Comparison and Benchmark Database62. The SnSe(100)–H2O interface was modelled by introducing 70 water molecules and one H2O2 molecule on the SnSe(100) surface in a \(3\sqrt{2\,}\times \,3\sqrt{2\,}\times \,1\) supercell. This set-up allowed the study of both the water–SnSe interface and the H2O2 dissociation process. During dynamic calculations, a 1 × 1 × 1 k-point mesh was used, with the bottom SnSe monolayer fixed and the rest fully relaxed. Simulations were performed using the canonical (NVT) ensemble at 313 K with Nosé–Hoover thermostats for 12 ps and a time step of 1.0 fs. Water concentration profiles were averaged over 4,000 structures from the 8–12 ps AIMD simulation. Constrained AIMD simulations with a slow-growth sampling approach were used to evaluate the kinetic barriers for H2O2 production from OH* desorption63. In this method, transformations from the initial to the final state were evaluated along a chosen reaction coordinate, defined by a suitable collective variable (CV). To simplify the calculation process, we considered the reverse process of H2O2 formation, which is the decomposition of H2O2 into two OH* radicals. For H2O2 decomposition studies, the O–O distance was chosen as the CV. The transformation step size was controlled to be 0.001 Å for each constrained AIMD step. The blue-moon ensemble with the SHAKE algorithm in VASP was used to compute the mean force acting on the CV along the reaction coordinate64. Reaction barriers were obtained by integrating the free-energy gradients to compute the free energy profiles based on thermodynamic integration64. Data availability The data supporting the findings of this study are available within the Article and its Supplementary Information. All other data are available from the corresponding author upon reasonable request. Source data are provided with this paper. References Ciriminna, R. et al. Hydrogen peroxide: a key chemical for today’s sustainable development. ChemSusChem 9, 3374–3381 (2016). Article CAS PubMed Google Scholar Freese, T. et al. An organic perspective on photocatalytic production of hydrogen peroxide. Nat. Catal. 6, 553–558 (2023). Article CAS Google Scholar Zhang, Y. et al. H2O2 generation from O2 and H2O on a near-infrared absorbing porphyrin supramolecular photocatalyst. Nat. Energy 8, 361–371 (2023). Article CAS Google Scholar Disselkamp, R. S. Can aqueous hydrogen peroxide be used as a stand-alone energy source? Int. J. Hydrog. Energ. 35, 1049–1053 (2010). Article CAS Google Scholar Mousavi Shaegh, S. A. et al. A membraneless hydrogen peroxide fuel cell using Prussian Blue as cathode material. Energ. Environ. Sci. 5, 8225–8228 (2012). Article CAS Google Scholar Campos-Martin, J. M. et al. Hydrogen peroxide synthesis: an outlook beyond the anthraquinone process. Angew. Chem. Int. Ed. 45, 6962–6984 (2006). Article CAS Google Scholar Freakley, S. J. et al. Palladium-tin catalysts for the direct synthesis of H2O2 with high selectivity. Science 351, 965–968 (2016). Article CAS PubMed Google Scholar Zhang, J. et al. Photocatalytic phosphine-mediated water activation for radical hydrogenation. Nature 619, 506–513 (2023). Article CAS PubMed PubMed Central Google Scholar Liu, R. et al. Linkage-engineered donor–acceptor covalent organic frameworks for optimal photosynthesis of hydrogen peroxide from water and air. Nat. Catal. 7, 195–206 (2024). Article Google Scholar Teng, Z. et al. Atomically dispersed antimony on carbon nitride for the artificial photosynthesis of hydrogen peroxide. Nat. Catal. 4, 374–384 (2021). Article CAS Google Scholar Tan, H. et al. Photocatalysis of water into hydrogen peroxide over an atomic Ga-N5 site. Nat. Synth. 2, 557–563 (2023). Article CAS Google Scholar Shiraishi, Y. et al. Resorcinol–formaldehyde resins as metal-free semiconductor photocatalysts for solar-to-hydrogen peroxide energy conversion. Nat. Mater. 18, 985–993 (2019). Article CAS PubMed Google Scholar Yan, B. et al. Laser direct overall water splitting for H2 and H2O2 production. Proc. Natl Acad. Sci. USA 121, e2319286121 (2024). Article CAS PubMed PubMed Central Google Scholar Lu, Z. et al. High-efficiency oxygen reduction to hydrogen peroxide catalysed by oxidized carbon materials. Nat. Catal. 1, 156–162 (2018). Article CAS Google Scholar Xia, C. et al. Direct electrosynthesis of pure aqueous H2O2 solutions up to 20% by weight using a solid electrolyte. Science 366, 226–231 (2019). Article CAS PubMed Google Scholar Jung, E. et al. Atomic-level tuning of Co–N–C catalyst for high-performance electrochemical H2O2 production. Nat. Mater. 19, 436–442 (2020). Article CAS PubMed Google Scholar Siahrostami, S. et al. Enabling direct H2O2 production through rational electrocatalyst design. Nat. Mater. 12, 1137–1143 (2013). Article CAS PubMed Google Scholar Zhao, J. et al. Contact-electro-catalysis for direct synthesis of H2O2 under ambient conditions. Angew. Chem. Int. Ed. 62, e202300604 (2023). Article CAS Google Scholar Ran, M. et al. Dynamic defects boost in-situ H2O2 piezocatalysis for water cleanup. Proc. Natl Acad. Sci. USA 121, e2317435121 (2024). Article CAS PubMed PubMed Central Google Scholar Zhao, L.-D. et al. Ultrahigh power factor and thermoelectric performance in hole-doped single-crystal SnSe. Science 351, 141–144 (2016). Article CAS PubMed Google Scholar Li, F. et al. Recent advances in SnSe nanostructures beyond thermoelectricity. Adv. Funct. Mater. 32, 2200516 (2022). Article CAS Google Scholar Wei, Z. et al. Efficient visible-light-driven selective oxygen reduction to hydrogen peroxide by oxygen-enriched graphitic carbon nitride polymers. Energ. Environ. Sci. 11, 2581–2589 (2018). Article CAS Google Scholar Peng, H. et al. Defective ZnIn2S4 nanosheets for visible-light and sacrificial-agent-free H2O2 photosynthesis via O2/H2O redox. J. Am. Chem. Soc. 145, 27757–27766 (2023). Article CAS PubMed Google Scholar Liu, T. et al. Overall photosynthesis of H2O2 by an inorganic semiconductor. Nat. Commun. 13, 1034 (2022). Article CAS PubMed PubMed Central Google Scholar Yue, J.-Y. et al. Thiophene-containing covalent organic frameworks for overall photocatalytic H2O2 synthesis in water and seawater. Angew. Chem. Int. Ed. 62, e202309624 (2023). Article CAS Google Scholar Qin, C. et al. Dual donor–acceptor covalent organic frameworks for hydrogen peroxide photosynthesis. Nat. Commun. 14, 5238 (2023). Article CAS PubMed PubMed Central Google Scholar Zhang, X. et al. Keto-anthraquinone covalent organic framework for H2O2 photosynthesis with oxygen and alkaline water. Nat. Commun. 15, 2649 (2024). Article CAS PubMed PubMed Central Google Scholar Richards, T. et al. A residue-free approach to water disinfection using catalytic in situ generation of reactive oxygen species. Nat. Catal. 4, 575–585 (2021). Article CAS Google Scholar Edwards, J. K. & Hutchings, G. J. Palladium and gold–palladium catalysts for the direct synthesis of hydrogen peroxide. Angew. Chem. Int. Ed. 47, 9192–9198 (2008). Article CAS Google Scholar Li, Q. et al. Shear stress triggers ultrathin-nanosheet carbon nitride assembly for photocatalytic H2O2 production coupled with selective alcohol oxidation. J. Am. Chem. Soc. 145, 20837–20848 (2023). Article CAS PubMed Google Scholar Zhao, L.-D. et al. Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals. Nature 508, 373–377 (2014). Article CAS PubMed Google Scholar Liu, D. et al. Lattice plainification advances highly effective SnSe crystalline thermoelectrics. Science 380, 841–846 (2023). Article CAS PubMed Google Scholar Huang, J. et al. Oxyhydroxide nanosheets with highly efficient electron–hole pair separation for hydrogen evolution. Angew. Chem. Int. Ed. 55, 2137–2141 (2016). Article CAS Google Scholar Sun, Y. et al. Pits confined in ultrathin cerium(IV) oxide for studying catalytic centers in carbon monoxide oxidation. Nat. Commun. 4, 2899 (2013). Article PubMed Google Scholar Duvjir, G. et al. Origin of p-type characteristics in a SnSe single crystal. Appl. Phys. Lett. 110, 262106 (2017). Article Google Scholar Yuan, S. et al. Surfactant-free aqueous synthesis of pure single-crystalline SnSe nanosheet clusters as anode for high energy- and power-density sodium-ion batteries. Adv. Mater. 29, 1602469 (2017). Article Google Scholar Zhong, W. et al. Coupled vacancy pairs in Ni-doped CoSe for improved electrocatalytic hydrogen production through topochemical deintercalation. Angew. Chem. Int. Ed. 59, 22743–22748 (2020). Article CAS Google Scholar Liao, Q. et al. Regulating relative nitrogen locations of diazine functionalized covalent organic frameworks for overall H2O2 photosynthesis. Angew. Chem. Int. Ed. 62, e202310556 (2023). Article CAS Google Scholar Hou, Y. et al. Efficient photosynthesis of hydrogen peroxide by cyano‐containing covalent organic frameworks from water, air and sunlight. Angew. Chem. Int. Ed. 63, e202318562 (2024). Article CAS Google Scholar Derkosch, J. et al. Raman spectroscopic and X-ray diffraction study of Na2Se·9(H, D)2O and comparison between O-H(D)∙∙∙Se and O-H(D)∙∙∙S hydrogen bonds. J. Raman Spectrosc. 17, 75–78 (1986). Article CAS Google Scholar Vijayarangamuthu, K. & Rath, S. Nanoparticle size, oxidation state, and sensing response of tin oxide nanopowders using Raman spectroscopy. J. Alloy. Compd. 610, 706–712 (2014). Article CAS Google Scholar Adams, J. S. et al. Unifying concepts in electro- and thermocatalysis toward hydrogen peroxide production. J. Am. Chem. Soc. 143, 7940–7957 (2021). Article CAS PubMed Google Scholar Dronskowski, R. & Bloechl, P. E. Crystal orbital Hamilton populations (COHP): energy-resolved visualization of chemical bonding in solids based on density-functional calculations. J. Phys. Chem. 97, 8617–8624 (1993). Article CAS Google Scholar Mohsenzadeh, A., Bolton, K. & Richards, T. DFT study of the adsorption and dissociation of water on Ni(111), Ni(110) and Ni(100) surfaces. Surf. Sci. 627, 1–10 (2014). Article CAS Google Scholar Mäkinen, M. & Laasonen, K. Density functional theory study of trends in water dissociation on oxygen-preadsorbed and pure transition metal surfaces. Surf. Sci. 734, 122305 (2023). Article Google Scholar Sraitrova, K. et al. Vacancies in SnSe single crystals in a near-equilibrium state. Phys. Rev. B 99, 035306 (2019). Article CAS Google Scholar Zhang, S. et al. Formation and migration of vacancy defects in GeSe and SnSe. J. Phys. B 54, 035003 (2021). CAS Google Scholar Zhang, P. et al. Effects of four-phonon interaction and vacancy defects on the thermal conductivity of the low-temperature phase of SnSe. Phys. Rev. Appl. 21, 024043 (2024). Article CAS Google Scholar Chen, S. et al. In situ/operando analysis of surface reconstruction of transition metal-based oxygen evolution electrocatalysts. Cell Rep. Phys. Sci. 3, 100729 (2022). Article CAS Google Scholar Wu, Y.-j et al. Evolution of cationic vacancy defects: a motif for surface restructuration of OER precatalyst. Angew. Chem. Int. Ed. 60, 26829–26836 (2021). Article CAS Google Scholar Pang, H. et al. Realizing n-type SnTe thermoelectrics with competitive performance through suppressing Sn vacancies. J. Am. Chem. Soc. 143, 8538–8542 (2021). Article CAS PubMed Google Scholar Zhong, Y. et al. Large scale self-assembly of SnSe nanosheets prepared by the hot-injection method for photodetector and capacitor applications. Mater. Today Energy 12, 418–425 (2019). Article CAS Google Scholar Tomita, O. et al. Partial oxidation of alcohols on visible-light-responsive WO3 photocatalysts loaded with palladium oxide cocatalyst. ACS Catal. 6, 1134–1144 (2016). Article CAS Google Scholar Ravel, B. & Newville, M. ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Rad. 12, 537–541 (2005). Article CAS Google Scholar Rehr, J. J. et al. Parameter-free calculations of X-ray spectra with FEFF9. Phys. Chem. Chem. Phys. 12, 5503–5513 (2010). Article CAS PubMed Google Scholar Berbille, A. et al. Mechanism for generating H2O2 at water–solid interface by contact-electrification. Adv. Mater. 35, 2304387 (2023). Article CAS Google Scholar Kresse, G. & Furthmuller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169 (1996). Article CAS Google Scholar Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996). Article CAS PubMed Google Scholar Grimme, S. et al. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H–Pu. J. Chem. Phys. 132, 154104 (2010). Article PubMed Google Scholar Tao, J. et al. Climbing the density functional ladder: nonempirical meta-generalized gradient approximation designed for molecules and solids. Phys. Rev. Lett. 91, 146401 (2003). Article PubMed Google Scholar Nørskov, J. K. et al. Trends in the exchange current for hydrogen evolution. J. Electrochem. Soc. 152, J23 (2005). Article Google Scholar Johnson, R. D. NIST Computational Chemistry Comparison and Benchmark Database 22nd edn, NIST standard reference database number 101 (NIST, 2022). Jarzynski, C. Nonequilibrium equality for free energy differences. Phys. Rev. Lett. 78, 2690 (1997). Article CAS Google Scholar Sprik, M. & Ciccotti, G. Free energy from constrained molecular dynamics. J. Chem. Phys. 109, 7737–7744 (1998). Article CAS Google Scholar Download references Acknowledgements This work was supported by the National Key Research and Development Program of China (2024YFA1210400), the National Natural Science Foundation of China (grant nos. 22075126, 22209061, 52202242, 52402221 and 52172187; S.L., Y. Wan, Y. Zhang, Y. Zhu and Y.L.), the National Science Fund for Distinguished Young Scholars (grant no. 51925101; L.-D.Z), the Tencent Xplorer Prize (L.-D.Z), the Start-up Fund for Senior Talents in Jiangsu University (grant nos. 5501310030, 21JDG060 and 5501310015; S.L., Y. Wan and Y. Zhang) and the Jiangsu Provincial Dengfeng Program. We thank the Shanghai Synchrotron Radiation Facility of BL11B (https://cstr.cn/31124.02.SSRF.BL11B) for the assistance on XAFS measurements and Renishaw (Shanghai) for in situ Raman spectroscopy support. We also acknowledge the Hefei Advanced Computing Center for supporting the theoretical calculations. Author information Author notes These authors contributed equally: Xinyue Zhang, Yangyang Wan, Yi Wen, Yingcai Zhu. Authors and Affiliations Institute of Quantum and Sustainable Technology, School of Chemistry and Chemical Engineering, Jiangsu University, Zhenjiang, China Xinyue Zhang, Hong Liu, Zhanpeng Zhu, Yuqiao Zhang, Long Zhang, Jianming Zhang & Shun Li Foshan (Southern China) Institute for New Materials, Foshan, China Xinyue Zhang & Yong Liu School of Materials Science and Engineering, Jiangsu University, Zhenjiang, China Yangyang Wan, Jiaxiang Qiu, Zhongti Sun & Xiaohong Yan School of Materials Science and Engineering, Beihang University, Beijing, China Yi Wen, Shulin Bai & Li-Dong Zhao Center for Bioinspired Science and Technology, Hangzhou International Innovation Institute, Beihang University, Hangzhou, China Yi Wen & Li-Dong Zhao Institute of Atomic Manufacturing, International Research Institute for Multidisciplinary Science, Beihang University, Beijing, China Yingcai Zhu Center for High Pressure Science and Technology Advanced Research, Beijing, China Xiang Gao Tianmushan Laboratory, Hangzhou, China Li-Dong Zhao Authors Xinyue Zhang View author publications You can also search for this author inPubMed Google Scholar Yangyang Wan View author publications You can also search for this author inPubMed Google Scholar Yi Wen View author publications You can also search for this author inPubMed Google Scholar Yingcai Zhu View author publications You can also search for this author inPubMed Google Scholar Hong Liu View author publications You can also search for this author inPubMed Google Scholar Jiaxiang Qiu View author publications You can also search for this author inPubMed Google Scholar Zhanpeng Zhu View author publications You can also search for this author inPubMed Google Scholar Zhongti Sun View author publications You can also search for this author inPubMed Google Scholar Xiang Gao View author publications You can also search for this author inPubMed Google Scholar Shulin Bai View author publications You can also search for this author inPubMed Google Scholar Yuqiao Zhang View author publications You can also search for this author inPubMed Google Scholar Long Zhang View author publications You can also search for this author inPubMed Google Scholar Xiaohong Yan View author publications You can also search for this author inPubMed Google Scholar Jianming Zhang View author publications You can also search for this author inPubMed Google Scholar Yong Liu View author publications You can also search for this author inPubMed Google Scholar Shun Li View author publications You can also search for this author inPubMed Google Scholar Li-Dong Zhao View author publications You can also search for this author inPubMed Google Scholar Contributions S.L. and L.-D.Z. conceived the idea and designed the study. X.Z. synthesized the catalysts and conducted catalytic performance tests. Y. Wan, J.Q., S.B. and Z.S. performed theoretical calculations. Y. Wen and X.G. conducted the TEM characterizations. Y. Zhu carried out the XAFS measurements and analysis. X.Z., H.L. and Z.Z. performed materials characterizations. X.Z., Y. Zhang and L.Z. conducted in situ experiments. S.L. and Y. Wan analysed the mechanisms. X.Y., J.Z. and Y.L. provided valuable discussions. S.L., Y. Wan and L.-D.Z. wrote the manuscript. All authors reviewed and commented on the manuscript. Corresponding authors Correspondence to Shun Li or Li-Dong Zhao. Ethics declarations Competing interests The authors declare no competing interests. Peer review Peer review information Nature Catalysis thanks the anonymous reviewers for their contribution to the peer review of this work. Additional information Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Extended data Extended Data Fig. 1 Characterization of Sn vacancy defects. a, Se K-edge k2χ(k) EXAFS oscillation functions for SnSe NSs and bulk crystals. The reduced oscillation amplitudes in the 4 − 10 Å−1 range for SnSe NSs indicate structural disruptions associated with Sn vacancy defects. b, Corresponding Fourier transforms of the EXAFS spectra. c, EPR spectra of Sn1−xSe NSs and SnSe bulk crystals. d, Calculated defect concentration derived from the EPR spectra. e, Raman spectra of Sn1−xSe NSs and SnSe bulk crystals, exhibiting four characteristic peaks at 68.5, 103.1, 127.5 and 149.4 cm–1, correspongding to the Ag(1), B3g, Ag(2) and Ag(3) vibration modes of SnSe, respectively. The Sn1−xSe NSs exhibit broadened and diminished peaks compared to bulk crystals, indicating their ultrathin structure and the presence of defects. f, Calculated B3g/Ag(1) intensity ratios, demonstrating an increase in Sn vacancy defects with higher x values. Source data Extended Data Fig. 2 Thermocatalytic H2O2 production and mechanism analysis. a, Time-dependent H2O2 production over SnSe NSs under different gas environment at 40˚C. Error bars represent the standard deviations (SD) of three replicate tests. b and c, Isotopic labelling experiments using both H218O and 18O2. The labeled product C7H6O218O was identified by LC-MS, characterized by a difference of charge to mass ratio (m/z) of +2. The LC-MS spectra show both oxidation of labeled water (b) and the reduction of 18O2 (c). d and e, In situ EPR spectra of DMPO-•O2− (d) and DMPO-•OH (e) at various temperatures. f, Quantitative analysis of radical concentrations for Sn0.9Se and SnSe NSs. Source data Supplementary information Supplementary Information Supplementary Notes 1–4, Figs. 1–34, Tables 1–8 and References 1–50. Supplementary Video 1 The experimental set-up for hydrogen peroxide production using a flow reactor at room temperature. The catalyst is loaded into a commercial stainless-steel column, effectively shielding the system from any light exposure. The flow reactor operates under room temperature, with water and oxygen introduced into the system. The process relies solely on thermal energy to drive the production of hydrogen peroxide. Supplementary Data 1 The atomic coordinates of the optimized computational models in this study. Source data Source Data Fig. 1 Source data for Fig. 1. Source Data Fig. 2 Source data for Fig. 2. Source Data Fig. 3 Source data for Fig. 3. Source Data Extended Data Fig./Table 1 Source data for Extended Data Fig. 1. Source Data Extended Data Fig./Table 2 Source data for Extended Data Fig. 2. Rights and permissions Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. Reprints and permissions About this article Cite this article Zhang, X., Wan, Y., Wen, Y. et al. SnSe nanosheets with Sn vacancies catalyse H2O2 production from water and oxygen at ambient conditions. Nat Catal (2025). https://doi.org/10.1038/s41929-025-01335-4 Download citation Received: 10 October 2024 Accepted: 07 April 2025 Published: 23 May 2025 DOI: https://doi.org/10.1038/s41929-025-01335-4 Share this article Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative

An illustration of a ‘black widow’ pulsar (right) with its brown dwarf companion star (left).Credit: Mark Garlick/Science Photo Library Astronomers at the world’s largest telescope have found hints of a ‘missing link’ in the evolution of two binary stars that are tightly orbiting one another. It is a long-theorized type of system in which a rapidly spinning neutron star called a pulsar — the remnants of a massive supernova — is engulfed into its larger companion, which it then strips to its core. The neutron star is a rare example of a ‘spider pulsar’ — one that consumes the material around a companion star, or causes it to scatter away by emitting powerful beams of radio waves. The detection, described on 22 May in Science1, sheds light on processes that give rise to the spectacular mergers seen by gravitational-wave observatories — involving two neutron stars, two black holes or one of each. Jin-Lin Han, the leading radio astronomer at the National Astronomical Observatories in Beijing, says that this particular pulsar caught his attention because the radio signals it emitted showed that it orbited a companion about every 3.5 hours — but also disappeared for about one-sixth of that time, suggesting that the companion star was itself hidden by galactic dust. Gigantic Chinese telescope opens to astronomers worldwide Han and his collaborators watched the star repeatedly over four and a half years — a period that included three extended observations covering the full orbit of the system. They used the Five-hundred-meter Aperture Spherical Radio Telescope (FAST) in the Guizhou province of southern China, the most powerful radio dish ever built. The researchers calculated that the companion star must weigh between 1 and 1.6 times the mass of the Sun, but determined that it couldn’t be a common star, like the Sun, because of the system’s tight orbit. The fact that the pulsar gets eclipsed means that the companion couldn’t be too compact, either — ruling out another neutron star, a black hole or the common remnant of an extinguished star, called a white dwarf. The team concluded that the companion must be a helium star — one whose envelope of hydrogen has been stripped away. This points to a system that has undergone a ‘common envelope’ phase: as the companion star aged, it became so puffed up that it engulfed the neutron star. The neutron star then spent a few thousand years inside this envelope, while its radiation slowly blew it into space. Common-envelope phases are thought to cause the orbits of binary systems to tighten to the point at which they begin to emit gravitational waves — which forces them to tighten even further, until they merge. Spider pulsars are named after the female spiders that eat the male after mating, and have names such as black widow or redback, depending on the type of companion star. This is the first one to be found orbiting with a helium star. “Astronomers may have to consult an arachnologist to come up with a suitable spider-inspired name for this system,” says Selma de Mink, a theoretical astrophysicist at the Max Planck Institute for Astrophysics in Garching, Germany.

Wool garments provide a cozy home for the eggs of body lice, which spread the bacterium Borrelia recurrentis.Credit: Kathy deWitt/Alamy A deadly infection that is spread by body lice might have emerged thousands of years ago thanks to a new fashion trend: wool. Ancient genomes of the spiral-shaped bacterium Borrelia recurrentis — which causes a neglected disease called louse-borne relapsing fever (LBRF) — suggest that the pathogen diverged from a less deadly tick-transmitted infection around 5,000 years ago. From Vikings to Beethoven: what your DNA says about your ancient relatives This is not long after humans began domesticating sheep in the Middle East and around the time that wool textiles became common across Eurasia. Wool garments provide a cozy home for the eggs of body lice (Pediculus humanus), says Pooja Swali, a geneticist at University College London (UCL) who co-led the study, published1 on 22 May in Science. Today, LBRF is found mostly in African countries including Ethiopia, Somalia and Sudan and among people fleeing war, poverty and famine there — the disease thrives in overcrowded and disaster conditions. Week-long fever Records from classical Greece and Medieval Europe of days-long, recurring fevers characteristic of LBRF suggest that the disease might have been more common and widespread in the past. Antibiotics can now clear most B. recurrentis infections, but if they are left untreated, 10–40% of cases can be fatal. As part of an effort to study ancient infections, Swali and her colleagues screened a large collection of ancient human genomes from Britain, looking for pathogens. They identified four B. recurrentis infections dating from between 2,300 and 600 years ago. Researchers extracted genomes from human remains in Britain to study ancient infections.Credit: Adelle Bricking Further sequencing of the British strains and comparisons with one other ancient B. recurrentis genome sequence — from medieval Norway — and modern relatives indicated that the species diverged from its closest relative, the tick-borne Borrelia duttonii, between 4,000 and 6,000 years ago. This time period overlaps with increasing use of animal products such as wool and milk by humans across Europe and Asia. “Things that you wouldn’t necessarily think of as creating the opportunity for new pathogens — like clothing technology, like wool — can provide the opportunity for these diseases to emerge,” says Pontus Skoglund, a palaeogeneticist at the Crick Institute in London, who co-led the study with Swali and evolutionary geneticist Lucy van Dorp at UCL. Dropping genes Compared with its tick-borne relatives — which also cause relapsing fevers — B. recurrentis has shed long stretches of DNA and seen numerous genes wither away. The ancient genomes captured some of this genome reduction in action. Ancient DNA traces origin of Black Death It’s not clear why the pathogen has dropped genes, but it is probably related to the shift from being tick-borne to being louse-borne, says Charlotte Houldcroft, a virologist at the University of Cambridge, UK. Humans are the only known hosts of the louse P. humanus, whereas the ticks that carry other Borrelia species occur on multiple animals. “The pathogen is kicking out genetic material it doesn’t need,” Houldcroft says. The idea that sheep domestication and wool use had a key role in the beginnings of LBRF is plausible, says Meriam Guellil, a microbial genomicist at the University of Vienna. But more ancient samples will be needed to establish this. A 2023 preprint2 found another 34 ancient cases of B. recurrentis — in 2.5% of samples tested — including in 5,500-year-old human remains from Scandinavia. Skoglund expects that many more researchers now will look for signs of B. recurrentis infections in archaeological remains, especially in Eastern Europe and the Middle East. “In the next five to ten years, we might have a really good idea when it became widespread.”

Some conferences have been relocated in response to researchers’ concerns about visiting the United States.Credit: skynesher/Getty Several academic and scientific conferences in the United States have been postponed, cancelled or moved elsewhere, as organizers respond to researchers’ growing fears over the country’s immigration crackdown. Organizers of these meetings say that tougher rules around visas and border control — alongside other policies introduced by US President Donald Trump’s administration — are discouraging international scholars from attending events on US soil. In response, they are moving the conferences to countries such as Canada, in a bid to boost attendance. ‘Anxiety is palpable’: detention of researchers at US border spurs travel worries The trend, if it proves to be widespread, could have an effect on US scientists, as well as cities or venues that regularly host conferences. “Conferences are an amazing barometer of international activity,” says Jessica Reinisch, a historian who studies international conferences at Birkbeck University of London. “It’s almost like an external measure of just how engaged in the international world practitioners of science are.” “What is happening now is a reverse moment,” she adds. “It’s a closing down of borders, closing of spaces … a moment of deglobalization.” Unpopular location Conferences help researchers to connect, share new discoveries and shape the priorities of their fields. But events such as the detention and deportation of international scholars are pushing some academic societies and institutes to rethink where they hold their meetings. One of those is the International Society for Research on Aggression (ISRA), which announced last month that it would relocate its 2026 meeting from New Jersey to St. Catharines, Canada, after a survey of its members suggested that many international researchers would not attend a US meeting. “It was clear to us that, if we held a meeting in the US, based on the feedback we received, that we might not get enough people to register,” says Dominic Parrott, a clinical psychologist at Georgia State University in Atlanta, and ISRA’s president-elect. “We wanted to make sure that we were going to get our members and non-members from many different parts of the world, because that makes the best meeting and helps produce the best science.” How Trump 2.0 is reshaping science Organizers of the International Conference on Comparative Cognition have made a similar call. Its 33rd annual conference next year will take place outside the United States for the first time in the society’s history, in Montréal, Canada. “It was really a difficult decision for us,” says Caroline Strang, one of the conference organizers and a psychologist at Western University in London, Canada. “With things as unpredictable as they felt at the time and still feel, it’s a decision we made that allows as many of our attendees to come as possible.” The Northwest Cognition & Memory (NOWCAM) meeting relocated its meeting earlier this month from Western Washington University in Bellingham, to Victoria, Canada. Stephen Lindsay, a psychologist at the University of Victoria, one of the meeting’s organizers, says that most attendees are students in Canada. “I was concerned that many of them would decide not to attend NOWCAM if doing so entailed crossing the border,” he tells Nature. This is a choice he has had to make himself. “I plan to avoid travel to the US until relations improve, even though that means forgoing important annual conferences such as the annual meeting of the Psychonomic Society [in Denver, Colorado, in November], which I have rarely missed over the last 39 years.” Cancelled plans Other US meetings have been postponed, or cancelled altogether, because of similar or related concerns. The International Association of Cognitive Behavioral Therapy has cancelled its conference, originally planned for August 2025 in Nashville, Tennessee, because cuts to federal funding meant it was “no longer financially viable”. The 2026 Cities on Volcanoes conference in Bend, Oregon, has been postponed to 2030 or 2032. The International X-ray Absorption Society cancelled its upcoming 19th conference in Chicago, Illinois, which was scheduled for July this year. “We started to have cancellations by invited speakers,” says Carlo Segre, a physicist at Illinois Institute of Technology in Chicago, and one of the conference organizers. “It is not clear when this conference will be held in the USA once again.” The group will have their meeting in Thailand next year. International PhD students make emergency plans in fear of US immigration raids In February, the Texas Department of State Health Services cancelled its conference in Galveston on immunization, which normally takes place every two years. A spokesperson for the department says that the decision was because of uncertainty about whether officials at the US Centers for Disease Control and Prevention would be able to attend. Among conferences that are due to go ahead in the United States, some have taken steps to improve accessibility for international researchers. The 50th annual meeting of the Society for Social Studies of Science, which will be held in Seattle, Washington, in September, is now offering hybrid attendance, in response to the “unpredictable” situation at the US border. Sustained shift At the moment, there are no data available on how widespread the issue is. Nature asked three large scientific societies that regularly hold conferences in the United States whether they are considering relocating their meetings. The American Chemical Society and American Physical Society said that they had no plans to relocate upcoming conferences. The American Society for Microbiology (ASM) is going ahead with its annual meeting in Los Angeles, California, next month. “We haven’t seen this groundswell of concern that’s led us to say we need to rethink this meeting and moving it outside,” says Christopher DeCesaris, chief financial officer at ASM, based in Washington DC. He adds that, every year, some attendees experience problems with travel arrangements or visas. “I wouldn’t say that’s been such an increase to where it’s something that is a marked difference from what it’s been in the past.” Reinisch says that the broader impacts of cancelled and relocated conferences might not be clear right away, but if the trend is sustained, it could have a cumulative effect. “Conferences require huge infrastructure. They’re business, they generate a lot of money,” she says. A growing reluctance to hold scientific conferences in the United States “will have an impact on the cities that have established themselves as conference hosting cities”. “It’ll be interesting to see to what extent the US responds — or doesn’t,” Reinisch adds. “From the level of individual scholars, researchers, associations, universities and conference hosting cities, it would be considered a loss, or something that should be reversed if at all possible.”

Abstract Recent advancements in the CO2 reduction reaction (CO2RR) target multicarbon chemical production and scalable electrode designs for industrial applications. Here we introduce a zero-gap cell utilizing humidified gas-phase CO2 and circulated alkaline media, achieving a Faradaic efficiency of 66.9% for C3+ products and a current density of −1,100 mA cm−2. In situ spectroscopic analyses revealed formaldehyde as a key intermediate formed on copper oxide/hydroxide interfaces derived from a phosphorus-rich copper catalyst. Unlike conventional pathways based on dimerization of CO intermediates, our study selectively produces liquid-phase multicarbon products because of autonomous local pH variations under a weak alkaline microenvironment, with allyl alcohol as the dominant C3+ product. The high selectivity and efficiency for liquid products provide a substantial advantage for storage and transport, highlighting the scalability and practical feasibility of our approach, which offers a potential economically viable solution for CO2 utilization. This development encourages the adoption of CO2RR technologies in iron–steel and petrochemical industries to mitigate greenhouse gas emissions. Main The urgent need to mitigate greenhouse gas emissions coupled with industrial demand for a broad range of chemicals necessitates innovative production methods1. This is particularly true for C3+ multicarbon products, defined as compounds with three or more carbon atoms, which are essential for various industrial applications. C3+ multicarbon products are traditionally produced through methods such as catalytic cracking, crude oil distillation and oligomerization2,3. However, these traditional methods rely heavily on fossil fuel feedstocks and require high pressures and temperatures, leading to substantial greenhouse gas emissions, environmental pollution and energy consumption4. In contrast, by utilizing CO2, a notorious greenhouse gas, as the primary feedstock, electrochemical CO2 reduction offers the prospect of a more sustainable alternative, especially when powered by renewable electricity sources such as wind or solar power5,6,7,8. This method operates under ambient or near-ambient conditions, reducing energy demands compared with traditional methods. Despite these important advantages, electrochemical methods face challenges such as low Faradaic efficiencies (FEs), slow reaction rates (low current densities) and catalyst stability issues. Moreover, the variety of products formed during electrochemical CO2 reduction necessitates an energy-intensive separation process to achieve industrial purity standards for each hydrocarbon product. Recent research has focused on addressing these challenges to enhance the industrial viability of the CO2 reduction reaction (CO2RR)8,9. For instance, the use of gas diffusion electrodes (GDEs) in electrolytic cells has been explored to establish triple-phase boundaries by incorporating gaseous CO2 inlets and liquid-phase alkaline catholytes into membrane electrode assembly (MEA) cells10,11,12,13. Despite considerable progress—for example, MEA cells have attained a heightened maximum current density (Jmax) of 1.4 A cm−2—obstacles remain, due in particular to the conversion of input CO2 reactant into bicarbonate, resulting in carbonate deposition within the GDEs2,13,14. Additionally, although some systems have demonstrated direct conversion of CO2 into C3+ hydrocarbons, their FEs remain below 20%, with current densities also typically falling below the industrially necessary level of 1 A cm−2 (refs. 15,16,17). To overcome these limitations, we developed an approach that utilizes copper phosphide (CuP2) cathodes paired with non-precious-metal anode catalysts in a zero-gap MEA configuration for the CO2RR. With humidified CO2 as the source for CO2 electrolysis, we achieved successful operation at high current densities of up to −1.1 A cm−2 and a high partial current density of 735.4 mA cm−2 with a FE of 66.9% for C3+ products on a 9-cm2 electrode. Using a larger electrode, we obtained a 30-fold increase in yield rate (μmol cm−2 h−1) compared with previous studies2,10,11,15,16,18,19. Analytical techniques, including time-of-flight secondary ion mass spectrometry (ToF-SIM) and in situ Raman spectroscopy, revealed catalyst reconstruction involving copper oxide/hydroxide formation, demonstrating the potential of this approach to advance CO2 reduction technologies. Results CO2 electroreduction in zero-gap MEA configuration The CO2 supply is a critical factor that markedly influences the electrochemical reduction of CO2 (ref. 20). Supplying humidified CO2 is more effective than using its dry gaseous form21. This is because water vapour in the humidified CO2 facilitates the conversion of CO2 into hydrocarbons by improving mass transport and ensuring an adequate supply of reactants at the electrode surface22. Additionally, humidification can enhance ion conductivity and proton availability within the cell, thereby promoting reaction kinetics23,24. This method reduces the thickness of the CO2 diffusion layers, thereby enabling the application of higher current densities. Utilizing humidified CO2 ensures a consistent and abundant supply, optimizes the conditions for efficient electrochemical reduction, and enhances the current density2,20. To minimize the CO2 diffusion layer on the surface of the electrode for high-current operation, humidified CO2 was supplied to the cathode, as depicted in Fig. 1a. This technique enhances CO2 transfer, leading to increased partial current density and FE even at low cell voltages. Moreover, this method makes it possible to achieve a considerably higher concentration of liquid multicarbon products by providing dissolved CO2 with minimal water content25. Fig. 1: CO2 electrolysis system for multicarbon production. a, Schematic illustration of humidified CO2 supply to CO2-to-C3+ systems, including the MEA with an anion-exchange membrane (AEM) and GDEs. The system uses a mass flow controller (MFC) to regulate the CO2 flow at 500 sccm, maintained at 5 °C in a water bath, with product analysis conducted via headspace GC–MS. The anolyte, 1-M KOH, is circulated using a glow pump. b, Chronopotentiometric cell voltage profile using platinum on carbon (Pt/C) as an anode on a 9-cm2 electrode, showing the voltage response to step changes in applied current density at 20, 40, 60, 80, 100, 120, 140, 180 and 200 mA cm−2, measured at room temperature every 10 min. c, Chronopotentiometric cell voltage profile using nickel–iron (NiFe) as an anode on a 9-cm2 electrode, showing the voltage response to step changes in applied current density at 20, 40, 60, 80, 100, 120, 140, 180 and 200 mA cm−2, measured at room temperature every 10 min. Source data Full size image Initially, a CuP2 catalyst airbrushed onto a gas diffusion layer was used, and various anode catalysts were compared (see Methods for detailed procedures)26. In this study, CO2 reduction experiments were conducted using a zero-gap MEA-type cell (Fig. 1a) with an electrode area of 9 cm2. We used chronopotentiometry with a humidified CO2 supply for CO2 reduction. CO2 electrolysis performance was evaluated by analysing the effluent gases and liquid products at room temperature and atmospheric pressure, specifically by calculating the FE values of the C3+ liquid products at various applied currents. We first assessed the performance of CO2 reduction using the widely recognized Pt/C catalyst and evaluated the performance of the zero-gap MEA-type cell. An experiment was conducted to determine the extent of cell voltage reduction with the inclusion of Pt/C on the anode side (Fig. 1b). We applied current densities (J) of −20 mA cm−2 to −200 mA cm−2 at −20 mA cm−2 intervals, with each stage conducted for over 10 min. The cell voltage remained stable up to −140 mA cm−2; beyond this range, it showed considerable instability with pronounced fluctuations. Given the low stability, high cell potential and the use of precious-metal Pt/C, numerous researchers are actively working on replacing oxygen evolution reaction (OER) catalysts with non-precious-metal alternatives27,28. Hence, we experimented with non-precious-metal NiFe catalysts, which demonstrated competitive performance compared with Pt/C. The cell voltage when using NiFe was substantially lower than that of Pt/C and remained stable even at −200 mA cm−2 (Fig. 1c). When the type of anode was varied, the cell potential was the highest for Pt/C, followed by NiFe and iridium black (11). Nevertheless, NiFe proved to be the most stable anode, maintaining a constant voltage for 2 h even at an increased current density of −200 mA cm−2. During our investigation of the CO2RR, most of the products on our catalysts were liquid phase. Consequently, our analysis and subsequent discussions primarily concentrate on the performance of the CO2RR with respect to these liquid-phase products. Pt/C was used as the anode, and a current density of −100 mA cm−2 was applied for a duration of 2 h. Under these conditions, the FEs for the C2+ product spectrum revealed pronounced selectivity towards allyl alcohol, which comprised 48.7% of the total products (Supplementary Table 1). Within the array of C2+ products, the FEs for C3+ products were even more remarkable, exceeding 50% and reaching 56.0%. When NiFe was used as the anode, the selectivity shifted towards C2 products such as acetaldehyde and ethanol, rather than C3+ products, such as propionaldehyde, propanol, allyl alcohol and butyraldehyde, as shown in Supplementary Table 1. This shift occurred even though the same current density was applied as used with the Pt/C anode. The cell voltage provided by NiFe was insufficient to achieve high selectivity for C3+ products, suggesting that the anode material substantially influences product distribution. To further investigate this phenomenon, additional experiments using a three-electrode system were conducted to discern the effect of anode potential variations on the working electrode (Supplementary Fig. 2a,b). CuP2 was used as the working electrode, with Pt/C and NiFe as counter-electrodes. The results indicated that the choice of counter-electrode material influenced the reduction current at the working electrode. Specifically, when NiFe was used, the system exhibited a lower overpotential for the OER, leading to a lower cell voltage. This change in cell voltage affected the reduction potential at the cathode, thus altering the CO2RR selectivity. These findings underscore that the nature of the anode catalyst could impact the electrochemical environment at the cathode, subsequently affecting the product distribution. All performance evaluations were conducted using a two-electrode system with chronopotentiometry, applying a constant current density and measuring the resulting cell voltage. The cell voltage reflects contributions from both the anode and cathode. The difference in OER potentials between Pt/C and NiFe alters the reduction potential at the cathode. NiFe exhibits a lower overpotential than Pt/C in alkaline media (Supplementary Fig. 2c), meaning the cell voltage required for the OER with Pt/C would be higher. This increased anode potential affects the reduction environment at the cathode, influencing the selectivity of the CO2 reduction products. Thus, the observed variation in product selectivity could be attributed to the different OER potentials of the anode catalysts, altering the electrochemical environment. We also conducted linear sweep voltammetry (LSV) measurements in 0.5-M KHCO3 electrolyte under both argon and CO2 purging conditions, observing lower current densities under CO2 purging (Supplementary Fig. 2d). This suggests that the CO2RR dominate over the HER due to the formation of surface intermediates that inhibit the HER. Despite CuP2’s inherent activity for the HER, as evidenced by the free energy of hydrogen adsorption, the catalyst favours CO2RR in the presence of CO2, with intermediates blocking active sites for the HER and promoting C–C coupling reactions. These results underscore the catalyst’s selectivity towards multicarbon product formation under CO2RR conditions. Reconstruction of CuP2 catalyst under CO2RR conditions To delineate the alterations in the cathode catalyst due to CO2 electrolysis, we used various analytical techniques. A ToF-SIMS analysis of the as-prepared CuP2 catalyst (Fig. 2a,b) revealed that the CuP2 surface was predominantly coated with phosphate species26,29. Even in the unreacted catalyst, the high oxygen affinity of phosphorus allows oxygen to permeate, resulting in a superficial phosphate layer. Fig. 2: Surface states of the CuP2 catalyst before and after the CO2RR. a,c, ToF-SIMS negative-ion depth profile and IL-SEM of a CuP2 cathode sample before the CO2RR (a) and after the CO2RR (c). All ions (Cu−, P−, O2−, PO− and P2−) are indicated on the profile. b,d, ToF-SIM 3D plots of each ion before the CO2RR (b) and after the CO2RR (d). Source data Full size image A complementary line-profiled scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS) analysis, shown in Supplementary Fig. 3a, corroborates the presence of a phosphate layer on the surface. After the CO2RR experiments, a discernible reconstruction of the surface, manifesting as a compositional shift, was observed. The initially thin phosphate layer progressively diminished, and was supplanted by an emerging oxide layer, as shown in Fig. 2d,e by ToF-SIMS analysis. This transition was parallel to the line-profiled SEM-EDS pattern (Supplementary Fig. 3), where oxides became predominant at sites formerly occupied by phosphates. This CO2RR process suggests that the high oxygen affinity of phosphorus catalyses the surface reconstruction of the catalyst. For a more detailed analysis, we used identical-location scanning electron microscopy (IL-SEM) (Fig. 2c,f). This technique makes it possible to observe the same particles before and after the reactions. A marked increase in the oxygen wt% and a corresponding decrease in the phosphorus wt% postreaction, aligning with earlier findings, substantiated the inferred catalyst modifications and confirmed the occurrence of prominent catalytic changes (Supplementary Table 2). X-ray photoelectron spectroscopy (XPS) was performed to investigate the chemical bonding on the catalyst surface (Supplementary Fig. 4). The phosphorus 2p spectrum of the as-prepared CuP2 electrode revealed two distinct doublets. The peak at 134.3 eV corresponds to oxidized phosphorus (phosphate), indicating a hydrated phosphate layer on the surface, consistent with previous findings. After the CO2RR, a hydroxylated phosphate layer formed. The peak at 129.7 eV, lower than red phosphorus at 130.2 eV, suggests charge transfer between copper and phosphorus, with phosphorus being negatively charged. Comparing spectra before and after the CO2RR, the intensity of the phosphorus 2p3/2 peak increased, indicating the persistence of phosphide postreaction. Additionally, the oxygen 1s spectra showed a decrease in the P–O interaction peak intensity and an increase in the Cu–O interaction peak intensity after the CO2RR. These changes highlight notable surface chemistry modifications due to the reaction. Although previous analyses were performed ex situ, leading to concerns about potential surface oxidation postreaction due to air exposure, we investigated surface changes during the CO2RR using in situ Raman spectroscopy. A custom-built cell designed for a three-electrode H-cell configuration (Supplementary Fig. 5) was used for these experiments30. A constant potential of −0.6 VRHE (RHE, reversible hydrogen electrode) was applied for 2 h, and the spectra were sequentially acquired to monitor temporal changes under a CO2-saturated atmosphere. The initial observations at an open-circuit potential indicated the presence of a phosphate layer on the electrode surface, as shown in Fig. 3. Upon application of the potential, we observed a gradual decrease in the intensity of the peaks associated with Cu3(PO4)2 at specific wavenumbers (1,361, 1,595 cm−1), suggesting the depletion of the phosphate species. Concurrently, peaks corresponding to CuO and CuOx/(OH)y (389–394 cm−1) emerged, indicating the formation of copper oxide and hydroxide species31. This transition in the Raman spectra signifies the ongoing reconstruction of the catalyst surface during the CO2RR. To ensure the reliability of our findings and confirm the reproducibility of the observed surface transformations, Raman spectra were collected from multiple locations (≥3) across all samples and under each potential investigated in this study. Fig. 3: In situ Raman spectroscopy results over time. a, Copper oxide region of the reconstructed CuP2 electrode. b, Copper phosphate/phosphide region of the reconstructed CuP2 electrode. The measurements were conducted under a constant potential of −0.6 VRHE, with a total charge of −3.44 C applied for 45 min. Source data Full size image We investigated the electrochemical stability of CuP2 across various pH levels and potential conditions, focusing on the relevant operational range for CO2 reduction reactions—specifically, the stability region for CuP2 within a pH range from neutral to slightly alkaline, which corresponds to the local conditions during the CO2RR as OH− ions are generated, shifting the local pH towards more alkaline conditions. Within this defined stability region, CuP2 remains robust, avoiding considerable degradation or dissolution32. This is corroborated by the formation of stable CuP2 and other protective species, such as phosphate layers, which help mitigate corrosion and maintain the catalyst’s integrity. Based on our results, we observed that CuP2 initially maintains a stable structure under CO2RR conditions but undergoes surface reconstruction as the reaction progresses. It is important to distinguish between reconstruction and degradation. The high oxophilicity of CuP2 leads to strong interactions with oxygen species, resulting in a gradual transformation of the surface to CuxPyOz (copper phosphate) phases, eventually progressing to copper oxide and finally to metallic copper. Despite this reduction to copper, phosphate groups remain on the surface, actively contributing to the catalyst’s performance and aiding in multicarbon product formation. This surface reconstruction aligns with the behaviour of Cu2O-derived copper catalysts, where the formation of active copper metal sites is beneficial for C2+ product selectivity. Furthermore, the stability test conducted at a high current density of 500 mA cm−2 for 60 h demonstrated a consistent cell voltage with minimal fluctuations (Supplementary Fig. 6). This consistent voltage profile indicates that the postreconstruction catalyst maintains a robust structure without signs of marked degradation, as any degradation would have led to noticeable voltage instability. The reconstructed, CuP2-derived catalyst, featuring surface-bound phosphate groups, provides a stable and optimal environment for CO2 reduction, differentiating it from conventional copper catalysts that start in their metallic state. This unique reconstruction pathway enhances the catalyst’s ability to promote multicarbon coupling reactions, highlighting the critical role of phosphate groups in stabilizing the surface and maintaining catalytic activity. Proposed CO2RR pathway for multicarbon product generation Drawing from the detailed analysis of the conversion ratios in Fig. 5e, and reinforcing the mechanisms discussed in our previous work, we highlight a distinct pathway in CO2 electrolysis, leading to multicarbon products that do not primarily rely on CO intermediates and C–C dimerization. Diverging from the traditional focus on CO dimerization, our findings emphasize a formaldehyde-driven condensation mechanism as the initial step in C–C coupling26,29,33. To elucidate the role of formaldehyde in the CO2RR, we carried out additional experiments comparing the reduction of CO, CO2 and formaldehyde under similar conditions. These experiments were performed using a microfluidic GDE cell (Supplementary Fig. 7) at a potential of −0.6 V versus RHE. For the reduction of CO and formaldehyde, we used a phosphate buffer solution saturated with argon to exclude CO2 participation. However, for CO2 reduction, a 0.5-M KHCO3 solution saturated with CO2 was used. As shown in Supplementary Fig. 8, CO reduction did not yield any carbon products, whereas CO2 reduction produced formate, acetaldehyde and butanol, consistent with a previous report26. Formaldehyde reduction led to a slight increase in hydrogen production and a substantial increase in selectivity towards C2+ compounds, specifically acetaldehyde and butanol. Although the product distribution might differ from that in an MEA-type cell, the enhanced selectivity for multicarbon compounds during formaldehyde reduction underscores its crucial role in the C–C coupling process during CO2 reduction. These findings support the hypothesis that formaldehyde is a key intermediate in forming multicarbon products, highlighting its importance in the overall reaction mechanism. To further explore our proposed mechanism, we outline several tentative steps, detailed in Supplementary Fig. 10: surface protonation, CO2 adsorption, rearrangement of CO2 and formation of formate, formate coverage at the CuP2 surface, and formaldehyde formation and aldehyde formation. The literature suggests that the Gibbs free energy (ΔG) values for these steps may be negative, indicating potential thermodynamic favourability, although this is based on preliminary evidence from previous studies29,34,35,36. At this stage, our working hypothesis for the mechanism is tentatively supported by these observations, but this will require additional experimental and computational investigations to confirm its validity. As proposed in Supplementary Fig. 11a, in the initial step for CO2− binding, carbon-binding leads to CO formation, which can further go through CO dimerization to produce multicarbon products such as ethylene, ethanol, 1-propanol, etc. Oxygen binding, however, leads to the formation of formate, which is widely accepted as an end product. In the presence of Lewis acid catalysts, formaldehyde self-condensation can be catalysed to form the C2 compound known as glycolaldehyde. Some non-copper catalysts with Lewis acid sites have been proposed for the self-condensation of formaldehyde as a C2+ formation mechanism. To further investigate the differences in C–C coupling mechanisms and validate the reaction pathways illustrated in Supplementary Fig. 11a, and to demonstrate the related intermediates of formaldehyde condensation shown in Supplementary Fig. 9, we conducted in situ surface-enhanced infrared absorption spectroscopy (SEIRAS) experiments. These experiments aimed to investigate the adsorption behaviour of CO on metallic copper and CuP2 surfaces under applied reduction potentials. As shown in Supplementary Fig. 11c,d, metallic copper exhibited a clear CO adsorption peak (∗COad) under reduction conditions, confirming its role in CO dimerization pathways that lead to multicarbon products. In contrast, CuP2 showed no detectable CO adsorption peaks across the entire range of applied potentials (0.3 VRHE to −1.8 VRHE), indicating a fundamentally different reaction mechanism. These results validate the proposed reaction pathways, where CuP2 avoids the CO-mediated C–C coupling, and instead support alternative routes, such as the formate pathway, for multicarbon product formation. The majority of formate-producing catalysts that require high overpotentials are known to form a curved structure of CO2 during the reaction, which is bound to oxygen atoms on the surface of the catalyst through a proton-coupled electron transfer process. This mechanism is consistent with our previous XPS results, which demonstrate that charge transfer takes place between copper and phosphorus. This suggests that copper has a positive charge and could potentially act as a hydride acceptor in the hydride transfer mechanism (CO2 + H− → HCOO−) producing formate, which is confirmed in Fig. 4a for HCOOad. After HCOOad adsorption, the reaction can proceed through HCOad to form formaldehyde (Fig. 4b). Under mild conditions, it is energetically favourable for two molecules of formaldehyde to undergo self-condensation and form a C–C bond. This reaction has a standard Gibbs free energy of −25 kJ mol−1. During the formaldehyde condensation reaction, OCH3ad (Fig. 4c) appears after the carbon atoms are coupled, and eventually the OC2H5ad (Fig. 4d) intermediate is formed, which is a primary intermediate for acetaldehyde. The onset of the OC2H5ad intermediate starts slightly behind the other intermediates which means it appears after the C–C coupling reaction. Fig. 4: In situ SEIRAS spectrum and proposed reaction pathways of CO2 to each product. a–d, HCOOad (a), HCOad (b), OCH3ad (c) and OC2H5ad (d) intermediates identified by in situ SEIRAS, related to the formaldehyde condensation reaction as a C–C coupling process. e, Proposed reaction pathways to various products, with colour coding for carbon chain length: C1 (purple/red), C2 (blue) C3 (green), C4 (orange). Source data Full size image In this refined pathway (Fig. 4e), the decreased amount of C2 intermediates steers the reaction towards formaldehyde condensation, yielding acetaldehyde37,38,39,40. This compound, together with formaldehyde, undergoes an aldol condensation reaction facilitated by specialized catalysts to form acrolein, a key intermediate of C3 products. This pathway bifurcates with the hydrogenation of acrolein, resulting in the formation of allyl alcohols and propionaldehydes. Subsequent hydrogenation leads to the synthesis of 1-propanol41,42. Moreover, in line with the reaction schemes outlined in our earlier studies, the aldol condensation of two acetaldehyde molecules led to the production of butyraldehyde, a C4 compound. This clearly illustrates that the electrochemical conversion of CO2 to multicarbon chemicals involves a complex interplay of reactions, highlighting formaldehyde condensation as a pivotal step in the formation of higher-order carbon products. By emphasizing the role of formaldehyde in this study, we aimed to refine the understanding of multicarbon product synthesis from CO2 and provide insights that can lead to the development of more efficient catalytic systems for CO2 utilization. High current density electrochemical CO2 reduction performance Following on from the electrolysis system described in Fig. 1a, we conducted a series of experiments under current densities of −100 mA cm−2 to −1.1 A cm−2. We monitored the full-cell voltage under a constant current density for 4 h using non-precious-metal catalysts on both the cathode and anode, as illustrated in Fig. 5a. The cell potential remained stable across all current densities, maintaining −3.4 V at −1.1 A cm−2 without any iR correction, as further detailed in Supplementary Fig. 12. Hydrogen was the predominant gaseous product, accompanied by a spectrum of CO2RR liquid products (Fig. 5b and Supplementary Table 3). Notably, allyl alcohol maintained the highest FE of 51.81% at a current density of −1 A cm−2, underscoring its selectivity in our system. Concurrently, we detected the formation of C1 compounds, such as formates, and C2 compounds, such as acetaldehyde and ethanol. The trend for C3 compound formation, specifically propionaldehyde and propanol, was equally promising, and butyraldehyde confirmed the presence of C4 products. Fig. 5: Electrochemical performance and product analysis of CO2. a, Stability of full-cell voltage over time across a range of current densities from −100 mA cm−2 to −1.1 A cm−2. b, FEs for different products at specified current densities. c, Partial current densities for C3 products (allyl alcohol, 1-propanol and propionaldehyde) plotted against total applied current density. d, Selectivity comparison for C1–4, C2 and C3+ products relative to CO2 reduction at different current densities. e, Conversion rate of CO2 into C1–C4 products at varying current densities. f, Comparison of full-cell voltage under no iR correction conditions across different studies including this work, for various current densities. Cal-Cu, cation-augmenting layer modified copper. g, FE of C3+ products against total current density in comparison to other studies. h, Product yield function against electrode area, demonstrating the scalability of the methodology in this work and other studies (detailed in Supplementary Table 4). The FE values are means, and error bars represent the s.d. from three independent measurements (n = 3). Source data Full size image To gain further insights into the cathodic environment and polarization potential at the CuP2 working electrode under these operating conditions, we designed a modified zero-gap three-electrode cell. This set-up incorporated a reference electrode insertion point to directly measure the cathode polarization potential at various applied current densities. Chronopotentiometric experiments revealed a strong correlation between the cathode polarization potential and CO2RR product selectivity. As shown in Supplementary Table 3 and Supplementary Fig. 13, the cathode polarization potential progressively shifted from −0.82 V versus RHE at −100 mA cm−2 to −1.12 V versus RHE at −1.1 A cm−2. This shift in potential directly influenced product distributions, with lower potentials favouring C1 and C2 products and higher potentials promoting C3 and C4 product formation. A closer analysis of the C3 compounds revealed an escalating partial current density for allyl alcohol, 1-propanol and propionaldehyde, with the current density increasing up to −1.0 A cm−2, as depicted in Fig. 5c. However, a subtle decrease in the quantity of allyl alcohol was observed upon increasing the current density to −1.1 A cm−2. The selectivity trends for the C2 and C3+ products relative to the total CO2 reduction indicated a decline in C2 and an enhancement in C3+ selectivity with increasing current density, as shown in Fig. 5d. This trend suggests a transitional pathway in which C3+ products are derived through a coupling reaction, such as aldol condensation, from C2 intermediates. Figure 5e shows the relationship between the C2, C3 and C4 products, indicating that as the current density increases, discernible shifts occur from C2 to C3 and from C2 to C4. However, the transition from C3 to C4 (Supplementary Fig. 14) did not exhibit a consistent trend across varying current densities, indirectly substantiating that C3 compounds originate from the coupling of C2 and C1, while C4 compounds arise from C2 coupling reactions rather than from C3 and C1. To further investigate the selectivity of CuP2 in formaldehyde reduction and its role as an intermediate in C–C coupling, isotopic labelling experiments were performed using 13C-labelled formaldehyde in an argon-saturated phosphate buffer solution (Supplementary Fig. 15). 13C NMR analysis revealed distinct peaks corresponding to various C1 and C2+ products, confirming the formation of methanol, acetaldehyde, propionaldehyde, ethanol, butyraldehyde and butanol. Notably, methanol was observed, which is attributed to the spontaneous conversion of formaldehyde under reaction conditions. This analysis provides direct evidence for the role of formaldehyde as a crucial intermediate in multicarbon product formation during the reduction process. A comparative analysis of the product distributions obtained from formaldehyde reduction (Supplementary Fig. 16a) and the CO2RR (Supplementary Fig. 16b) experiments highlights pronounced differences driven by the reaction microenvironment and variations in local pH influenced by the cell configuration. In the CO2RR experiment using a MEA at high current density, products such as formate, acetaldehyde, ethanol, propionaldehyde, 1-propanol, allyl alcohol and butyraldehyde were detected. In contrast, the formaldehyde reduction experiment conducted at −0.9 V versus RHE in a flow cell predominantly yielded methanol, acetaldehyde, propionaldehyde, ethanol, butyraldehyde and butanol. The confined reaction zone in the MEA cell, with limited electrolyte presence, leads to rapid proton consumption and substantial changes in local pH near the catalyst surface. This environment favours C–C coupling reactions and facilitates the formation of multicarbon products. Conversely, the flow cell set-up, with a spacer creating a thicker electrolyte layer, results in a different local pH profile that stabilizes formaldehyde and promotes its reduction to methanol. The combined use of headspace gas chromatography–mass spectrometry GC–MS and 13C NMR analysis was crucial in identifying the complete product distribution. Headspace GC–MS (Supplementary Figs. 16a and 17), with its high sensitivity to volatile organic compounds, detected trace-level allyl alcohol, which was not observed in the 13C NMR spectrum due to the latter’s higher detection limit. This highlights the complementary nature of these analytical techniques and underscores the importance of using multiple methods to fully characterize the product profile, especially for trace-level products. The observed differences in product selectivity emphasize the role of formaldehyde as an active intermediate in C–C coupling reactions. During the CO2RR, formaldehyde is probably formed transiently on the catalyst surface but is rapidly consumed, facilitating the formation of higher-order products such as propionaldehyde and butyraldehyde. In the formaldehyde reduction experiment, however, the initial presence of labelled formaldehyde provided direct evidence of its nucleophilic activity, leading to methanol formation and enabling further C–C coupling to produce multicarbon alcohols. This demonstrates the high selectivity of CuP2 for formaldehyde condensation, validating the proposed reaction mechanism and the catalyst’s effectiveness in promoting C–C bond formation. These findings illustrate the critical role of the reaction microenvironment and local pH variations influenced by cell configuration in determining product selectivity. The MEA cell configuration, with its confined reaction environment, promotes rapid C–C coupling and favours the formation of higher-order aldehydes and alcohols. In contrast, the flow cell set-up, with a different local pH profile, stabilizes formaldehyde as a key intermediate, promoting its direct reduction to methanol and subsequent multicarbon synthesis. This underscores the importance of formaldehyde in multicarbon product formation and highlights its pivotal role in C–C coupling pathways during CO2RRs. To address concerns about electro-osmosis at current densities exceeding −1.0 A cm−2, we conducted additional experiments to analyse the anolyte for crossover products postelectrolysis. No products were detected in the anolyte, as detailed in Supplementary Fig. 18, indicating no notable electro-osmosis. Moreover, we monitored cell voltage over time while applying constant currents above −1.0 A cm−2. Figure 5a shows no considerable voltage changes, confirming the absence of oxidation reactions due to crossover products. This stability confirms that our system effectively prevents product loss from electro-osmosis, ensuring reliable CO2 reduction results. To address concerns about potential anode material contamination affecting CO2RR activity, we conducted several analyses and additional experiments. SEM energy-dispersive X-ray analysis on the cathode postreaction showed no deposition of nickel or iron (Supplementary Fig. 19). Additionally, inductively coupled plasma optical emission spectrometry (ICP-OES) analysis of the catholyte in CO2RR experiments detected no nickel or iron ions (Supplementary Table 5). To further validate our findings, we performed a model electrolysis experiment using the same NiFe anode and the same anion-exchange membrane, paired with a Pt/C cathode. We applied constant current densities of 1.1 A cm−2 and 3.0 A cm−2 for 4 h, periodically sampling the catholyte for any potential nickel or iron contamination (Supplementary Table 6). The results of SEM energy-dispersive X-ray and ICP-OES analyses consistently showed no evidence of nickel or iron contamination on the cathode or in the catholyte. These findings confirm that the observed differences in CO2RR activity are attributed to intrinsic catalytic properties and reaction conditions, rather than anode material contamination. This supports our claim that different anode materials result in different cell voltages, influencing the overall CO2RR activity without cross-contamination effects. As delineated in Fig. 5f, our findings, particularly at a high current density of −1.1 A cm−2, showcased the lowest observed cell potential, outperforming even the Pt/C catalysts at equivalent current densities. Furthermore, when juxtaposed with other studies focusing on C3+ product yields10,11,15,16,18 (Fig. 5g), our system exhibited remarkable selectivity for C3+ products, showing an impressive FE of 66.9%. Theoretically, an increased electrode area should proportionally increase the yield of conversion products. However, in practice, upscaling the electrode area can introduce challenges such as increased surface resistance. Nonetheless, our research demonstrates that the strategic enlargement of the electrode area can substantially enhance the yield rate, as evidenced by the production volume exceeding that of other studies by more than 30 times (Fig. 5h). This reinforces the notion that optimizing the electrode microenvironment for CO2 participation can markedly improve the efficiency of CO2RRs. Conclusions Our comprehensive study illuminates a series of pathways for the electrochemical reduction of CO2 to multicarbon compounds, leveraging non-precious-metal catalysts to optimize efficiency and selectivity. The experiments conducted not only demonstrated a stable cell potential across a wide range of current densities but also highlighted the robust selectivity for valuable C3+ compounds, particularly allyl alcohol, which achieved a FE exceeding 50%. These results, situated at the forefront of CO2RR research, mark a step towards the sustainable production of chemicals. In situ Raman spectroscopy revealed the dynamic surface transformations of the catalyst during the CO2RR, offering a deeper understanding of the mechanisms at play. The emergence of copper oxides and hydroxides during the reaction provided clear evidence of the structural evolution of the catalyst, essential for guiding future catalyst designs. Our findings also shed light on the crucial intermediate role of C2 compounds and the importance of aldol condensation reactions in the formation of higher-order alcohols and aldehydes. This insight underlines the complexity of the CO2RR process and its potential to be harnessed for the targeted synthesis of specific products. Additionally, in situ SEIRAS experiments were conducted to observe the intermediate species and support the reaction pathway. The SEIRAS results showed the formation of formate and subsequent formaldehyde on the catalyst surface, consistent with our proposed mechanism and highlighting the importance of these intermediates in the C–C coupling process. Furthermore, our study challenges the traditional constraints associated with electrode area scaling. By strategically enhancing the electrode area, we achieved a remarkable increase in the product yield rate, outperforming existing benchmarks by approximately 30-fold in terms of yield rate (μmol cm−2 h−1), as detailed in Supplementary Table 4, which compares our results with previous studies2,10,11,15,16,18,19.This not only confirms the viability of scaling up the CO2RR but also emphasizes the importance of optimizing the electrode design to maximize reactant participation. This study marks an advancement in the electrochemical conversion of CO2, providing a viable route to high-value multicarbon compounds with high selectivity and efficiency. Our methodology provides guidance for the development of scalable, sustainable and economically feasible CO2RR technologies. The approach achieved a partial current density of 735.4 mA cm−2 and a FE of 66.9% for C3+ products, notably improving yield rates by approximately 30-fold compared with existing benchmarks (Supplementary Table 4), demonstrating a robust step forward in CO2 reduction technologies. Methods Electrode preparation The cathode catalysts were synthesized according to the method described in our previous study26. The catalyst ink was composed of 27 mg of the synthesized catalyst, 27 μl of an ionomer solution (PiperION Dispersion; 5 wt%) and 7,860 mg of isopropyl alcohol. The ink was thoroughly dispersed, and then, using N2 as the carrier gas, was airbrushed onto a 3 cm × 3 cm gas diffusion layer (22BB, Sigracet). For the anode catalyst, nickel nitrate (Ni(NO3)2·6H2O; 99.999%; Sigma-Aldrich) and iron chloride (FeCl2·4H2O; 99.99%; Sigma-Aldrich) were utilized in their as-received forms for the electrodeposition process. A 0.1-M solution of metal precursors was prepared by dissolving these salts in distilled water. The Ni2+ and Fe2+ compositions of the electrolytes were maintained at an optimized ratio of 9:1. The working electrode, consisting of stainless steel fibre paper (Bekipor ST 40BL3, Bekaert), was subjected to a constant current of −2.5 mA cm−2. A carbon cloth was used as the counter-electrode. The deposition time was set to 7,200 s to achieve optimal performance. Surface analysis methods We analysed the catalyst surface by field emission SEM (Verios 5 UC, Thermo Fisher) to observe the morphological characteristics. The distribution of each element in the individual particles was analysed by EDS (Oxford Ultim Max 65, Oxford Instruments) for element line profile and mapping. We utilized ICP-OES (Optima 4300 DV, PerkinElmer) to measure metal ions leached into the electrolyte. XPS (NEXSA, Thermo Fisher Scientific) was used to analyse changes in the electrode before and after the experiments. For IL-SEM, catalyst was drop-cast onto gold transmission electron microscopy grids for use as a working electrode. Raman spectroscopy measurements were conducted using a Horiba Xplora microscope situated within a custom-built PEEK cell, as depicted in Fig. 5b. Observations were made using a 60× water immersion lens (LUMPLFLN60XW). Calibration of the Raman frequency was achieved with a silicon wafer reference (520.6 cm−1). The experimental set-up included a 532-nm laser and an 1,800 lines mm−1 grating monochromator for light dispersion. Signal detection was performed with a 10-s integration time over 10 cumulative scans. Electrochemical measurements utilized a platinum coil as the counter-electrode and featured Ag/AgCl as the reference and catalyst-coated gold substrates as the working electrodes. Electrochemical experiments All electrochemical CO2RR experiments were performed in a zero-gap cell. An anode and cathode with geometric areas of 9 cm2 were placed between two current collectors; CuP2 was used as the cathode catalyst, and Pt/C (1 mg cm−2, 46.7%, TKK) airbrushed on a gas diffusion layer (MGL190, AvCarb) was used as the anode. An anion-exchange membrane (PiperION 20 nm, Versogen) was used after activation in 1-M KOH for 24 h. Gold-coated titanium single serpentine flow fields were used as the cathode and anode. To prevent leakage of the electrolyte and minimize the resistance in the MEA, a rubber gasket (630 μm) was utilized. All the components were assembled within an electrolyser, with a torque wrench being used to secure the screws to a tightness of 60 N·m. The anode was supplied with 1-M KOH electrolyte (at room temperature) at a flow rate of 18 ml min−1, and humidified CO2 (60 °C, with 100% relative humidity) was supplied to the cathode at a flow rate of 500 sccm. A chiller, circulating water at 5 °C, was connected to the outlet of the cathode to liquefy the vaporized products. We applied constant currents of −100 mA cm−2 to −1.1 A cm−2 for 4 h using a potentiostat/galvanostat (PGSTAT30, Autolab), for the CO2 reduction experiments. The liquid-phase products were quantified using headspace GC‒MS (Agilent Technologies). Subsequently, the FE values of each product were calculated. Data availability Data supporting the findings of this study are available from the corresponding authors upon reasonable request. Source data are provided with this paper. References Daehn, K. et al. Innovations to decarbonize materials industries. Nat. Rev. Mater. 7, 275–294 (2021). Article Google Scholar Qi, K. et al. Unlocking direct CO2 electrolysis to C3 products via electrolyte supersaturation. Nat. Catal. 6, 319–331 (2023). Article CAS Google Scholar O’Brien, C. P. et al. Single pass CO2 conversion exceeding 85% in the electrosynthesis of multicarbon products via local CO2 regeneration. ACS Energy Lett. 6, 2952–2959 (2021). Article Google Scholar Alabdullah, M. et al. One-step conversion of crude oil to light olefins using a multi-zone reactor. Nat. Catal. 4, 233–241 (2021). Article CAS Google Scholar Lee, S., Choi, M. & Lee, J. Looking back and looking ahead in electrochemical reduction of CO2. Chem. Rec. 20, 89–101 (2020). Article CAS PubMed Google Scholar Park, G., Hong, S., Choi, M., Lee, S. & Lee, J. Au on highly hydrophobic carbon substrate for improved selective CO production from CO2 in gas-phase electrolytic cell. Catal. Today 355, 340–346 (2020). Article CAS Google Scholar Choi, M. et al. Syngas production for Fischer–Tropsch process via co-electrolytic processes of CO2 reduction and NH3 oxidation. Chem. Eng. J. https://doi.org/10.1016/j.cej.2021.132563 (2022). Creel, E. B. & McCloskey, B. D. Scalable CO2-to-oxygenate production. Nat. Catal. 1, 6–7 (2018). Article CAS Google Scholar Xie, Y. et al. High carbon utilization in CO2 reduction to multi-carbon products in acidic media. Nat. Catal. 5, 564–570 (2022). Article CAS Google Scholar Garcia D. A. et al. CO2 electrolysis to multicarbon products at activities greater than 1 A cm−2. Science https://doi.org/10.1126/science.aay4217 (2020). Huang, J. E. et al. CO2 electrolysis to multicarbon products in strong acid. Science 372, 1074–1078 (2021). Article CAS PubMed Google Scholar Chen, Y. et al. Efficient multicarbon formation in acidic CO2 reduction via tandem electrocatalysis. Nat. Nanotechnol. https://doi.org/10.1038/s41565-023-01543-8 (2023). Article PubMed PubMed Central Google Scholar Zhuang, T.-T. et al. Steering post-C–C coupling selectivity enables high efficiency electroreduction of carbon dioxide to multi-carbon alcohols. Nat. Catal. 1, 421–428 (2018). Article CAS Google Scholar Wen, G. et al. Continuous CO2 electrolysis using a CO2 exsolution-induced flow cell. Nat. Energy 7, 978–988 (2022). Article CAS Google Scholar Gu, Z. et al. Efficient electrocatalytic CO2 reduction to C2+ alcohols at defect-site-rich Cu surface. Joule 5, 429–440 (2021). Article CAS Google Scholar Miao, R. K. et al. Electroosmotic flow steers neutral products and enables concentrated ethanol electroproduction from CO2. Joule 5, 2742–2753 (2021). Article CAS Google Scholar Lee, M. G. et al. Selective synthesis of butane from carbon monoxide using cascade electrolysis and thermocatalysis at ambient conditions. Nat. Catal. 6, 310–318 (2023). Article CAS Google Scholar Xu, Y. et al. Self-cleaning CO2 reduction systems: unsteady electrochemical forcing enables stability. ACS Energy Lett. 6, 809–815 (2021). Article CAS Google Scholar Zhang, R. et al. Synthesis of n-propanol from CO2 electroreduction on bicontinuous Cu2O/Cu nanodomains. Angew. Chem. Int. Ed. 63, e202405733 (2024). Article CAS Google Scholar She, X. et al. Pure-water-fed, electrocatalytic CO2 reduction to ethylene beyond 1,000 h stability at 10 A. Nat. Energy 9, 81–91 (2024). Article CAS Google Scholar Lee, W., Kim, Y. E., Youn, M. H., Jeong, S. K. & Park, K. T. Catholyte-free electrocatalytic CO2 reduction to formate. Angew. Chem. Int. Ed. 57, 6883–6887 (2018). Article CAS Google Scholar Xing, Z., Hu, L., Ripatti, D. S., Hu, X. & Feng, X. Enhancing carbon dioxide gas-diffusion electrolysis by creating a hydrophobic catalyst microenvironment. Nat. Commun. 12, 136 (2021). Article CAS PubMed PubMed Central Google Scholar Vennekötter, J.-B., Scheuermann, T., Sengpiel, R. & Wessling, M. The electrolyte matters: stable systems for high rate electrochemical CO2 reduction. J. CO2 Util. 32, 202–213 (2019). Article Google Scholar Lee, J. et al. Catholyte-free electroreduction of CO2 for sustainable production of CO: concept, process development, techno-economic analysis, and CO2 reduction assessment. Green Chem. 23, 2397–2410 (2021). Article CAS Google Scholar Zhu, P. & Wang, H. High-purity and high-concentration liquid fuels through CO2 electroreduction. Nat. Catal. 4, 943–951 (2021). Article CAS Google Scholar Choi, M., Bong, S., Kim, J. W. & Lee, J. Formation of 1-butanol from CO2 without *CO dimerization on a phosphorus-rich copper cathode. ACS Energy Lett. 6, 2090–2095 (2021). Article CAS Google Scholar Seger, B., Robert, M. & Jiao, F. Best practices for electrochemical reduction of carbon dioxide. Nat. Sustain. 6, 236–238 (2023). Article Google Scholar Weekes, D. M., Salvatore, D. A., Reyes, A., Huang, A. & Berlinguette, C. P. Electrolytic CO2 reduction in a flow cell. Acc. Chem. Res. 51, 910–918 (2018). Article CAS PubMed Google Scholar Calvinho, K. U. D. et al. Selective CO2 reduction to C3 and C4 oxyhydrocarbons on nickel phosphides at overpotentials as low as 10 mV. Energy Environ. Sci. 11, 2550–2559 (2018). Article CAS Google Scholar Kang, S. et al. Durable nickel–iron (Oxy)hydroxide oxygen evolution electrocatalysts through surface functionalization with tetraphenylporphyrin. Angew. Chem. Int. Ed. 61, e202214541 (2022). Article CAS Google Scholar Cheng, C. et al. Amorphous Sn(HPO4)2-derived phosphorus-modified Sn/SnO core/shell catalyst for efficient CO2 electroreduction to formate. J. Energy Chem. 81, 125–131 (2023). Article CAS Google Scholar Bale, C. W. et al. FactSage thermochemical software and databases, 2010–2016. Calphad 54, 35–53 (2016). Article CAS Google Scholar Delmo, E. P. et al. The role of glyoxal as an intermediate in the electrochemical CO2 reduction reaction on copper. J. Phys. Chem. C 127, 4496–4510 (2023). Article CAS Google Scholar David, A. K. et al. Combined computational and experimental investigation on the mechanism of CO2 hydrogenation to methanol with Mn-PNP-pincer catalysts. ACS Catal. 12, 15310–15322 (2022). Article Google Scholar Shyama, C. M. et al. A computational study on ligand assisted vs. ligand participation mechanisms for CO2 hydrogenation: importance of bifunctional ligand based catalysts. Phys. Chem. Chem. Phys. 21, 3932–3941 (2019). Article Google Scholar Bhaskar, M. et al. Toward rational design of 3d transition metal catalysts for CO2 hydrogenation based on insights into hydricity-controlled rate-determining steps. Inorg. Chem. 55, 5438–5444 (2016). Article Google Scholar Tang, M. T., Peng, H.-J., Stenlid, J. H. & Abild-Pedersen, F. Exploring trends on coupling mechanisms toward C3 product formation in CO2R. J. Phys. Chem. C 125, 26437–26447 (2021). Article CAS Google Scholar Cronin, S. P. et al. Direct synthesis of 1-butanol with high Faradaic efficiency from CO2 utilizing cascade catalysis at a Ni-enhanced (Cr2O3)3Ga2O3 electrocatalyst. J. Am. Chem. Soc. 145, 6762–6772 (2023). Article CAS PubMed Google Scholar Li, Z. & Wu, J. Co-electrolysis of CO2 and formaldehyde to ethanol: an inspiring asymmetric C–C coupling pathway. Chem Catal. 2, 2808–2810 (2022). Article CAS Google Scholar Yan, T. et al. Multiscale CO2 electrocatalysis to C2+ products: reaction mechanisms, catalyst design, and device fabrication. Chem. Rev. 123, 10530–10583 (2023). Article CAS PubMed Google Scholar Liu, L., Ye, X. P. & Bozell, J. J. A comparative review of petroleum-based and bio-based acrolein production. ChemSusChem 5, 1162–1180 (2012). Article CAS PubMed Google Scholar Lan, X., Wang, T., Li, X., Huang, N. & Wang, J. Pt@SnOx/SiO2 catalysts with enhanced selectivity to allyl alcohol for acrolein hydrogenation. Catal. Sci. Tech. 6, 7703–7707 (2016). Article CAS Google Scholar Download references Acknowledgements This work was supported by a National Research Foundation of Korea (NRF) grant funded by the Korean government (MSIT) (NRF-2021K1A4A8A01079455). This research was also supported by the NRF, funded by MSIT (RS-2021-NR060081). Author information Authors and Affiliations Department of Environment and Energy Engineering, Gwangju Institute of Science and Technology, Gwangju, Republic of Korea Minjun Choi, Sooan Bae, Yeongin Kim, Youjin Lee, Mokyeon Cho, Sinwoo Kang & Jaeyoung Lee International Future Research Center of Chemical Energy Storage and Conversion Processes (ifRC-CHESS), Gwangju Institute of Science and Technology, Gwangju, Republic of Korea Minjun Choi, Sooan Bae, Yeongin Kim, Youjin Lee, Mokyeon Cho, Sinwoo Kang & Jaeyoung Lee Ertl Center for Electrochemistry and Catalysis, Gwangju Institute of Science and Technology, Gwangju, Republic of Korea Minjun Choi, Sooan Bae, Youjin Lee, Sinwoo Kang & Jaeyoung Lee Authors Minjun Choi View author publications You can also search for this author inPubMed Google Scholar Sooan Bae View author publications You can also search for this author inPubMed Google Scholar Yeongin Kim View author publications You can also search for this author inPubMed Google Scholar Youjin Lee View author publications You can also search for this author inPubMed Google Scholar Mokyeon Cho View author publications You can also search for this author inPubMed Google Scholar Sinwoo Kang View author publications You can also search for this author inPubMed Google Scholar Jaeyoung Lee View author publications You can also search for this author inPubMed Google Scholar Contributions J.L. supervised the project. M.C. conceived the overall idea and structure of the study, designed and performed critical experiments (cathode electrocatalyst fabrication, activity/stability evaluation, in situ Raman and IR spectroscopy, isotope-labelling experiments), interpreted the data regarding mechanistic origins, and wrote the manuscript. S.B. created the experimental design of the zero-gap full-cell configuration with the reference electrode, provided comparable polarization curves between the half-cell and the full cell, and contributed NMR/XPS/ICP-OES analysis. Y.K. provided homemade the NiFe anode and carried out ICP-OES analysis. Y.L. performed near zero-gap cell experiments and carried out headspace analysis of liquid products. M.C. fabricated the CuP2 cathode electrocatalyst. S.K. contributed to the experimental design of the NiFe anode and to in situ Raman spectroscopy. All authors discussed experimental observations throughout the entire submission process. Corresponding author Correspondence to Jaeyoung Lee. Ethics declarations Competing interests The authors declare no competing interests. Peer review Peer review information Nature Catalysis thanks the anonymous reviewers for their contribution to the peer review of this work. Additional information Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary information Supplementary Information Supplementary Methods, Figs. 1–19 and Tables 1–5. Source data Source Data Fig. 1–5 A single file containing all source data, with clearly named tabs for each Figure/Extended Data Figure item. Rights and permissions Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. Reprints and permissions About this article Cite this article Choi, M., Bae, S., Kim, Y. et al. Selective formaldehyde condensation on phosphorus-rich copper catalyst to produce liquid C3+ chemicals in electrocatalytic CO2 reduction. Nat Catal (2025). https://doi.org/10.1038/s41929-025-01341-6 Download citation Received: 10 February 2024 Accepted: 09 April 2025 Published: 22 May 2025 DOI: https://doi.org/10.1038/s41929-025-01341-6 Share this article Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative

Abstract Transformation of economic systems is widely regarded as essential for tackling interacting global crises. Unconventional economic approaches seeking holistic human and planetary well-being have transformative potential, but mainstreaming them is hampered by vested interests and intellectual lock-ins. They are also diffuse and struggle to develop sufficient discursive power to gain more widespread traction in policy. To bring coherence, we undertake a qualitative content analysis of 238 document sources from science and practice. We identify ten ecological, social, political economy and holistic principles cutting across 38 economic approaches. They include: (1) social–ecological embeddedness and holistic well-being; (2) interdisciplinarity and complexity thinking; (3) limits to growth; (4) limited substitutability of natural capital; (5) regenerative design; (6) holistic perspectives of people and values; (7) equity, equality and justice; (8) relationality and social enfranchisement; (9) participation, deliberation and cooperation and (10) post-capitalism and decolonization. We also consider opportunities and barriers to applying these principles in the context of global crises. Our results can help consolidate transformative economic approaches and support future efforts to synthesize conceptual models, methodologies and policy solutions and to validate the identified principles more explicitly within global south contexts. Main The world faces a ‘polycrisis’1 of multiple interacting and compounding crises, including climate change, biodiversity loss, food and energy crises, geopolitical conflicts and ongoing repercussions of the COVID-19 pandemic. These crises severely exacerbate economic, social and health inequalities, greatly set back achievement of the UN Sustainable Development Goals2,3,4 and risk runaway global failures of vital natural and social systems1. They are also deeply rooted in conventional economic institutions and thought5,6,7,8. If the world continues along its present path, it risks, as UN Secretary-General António Guterres put it, “collective suicide”9. Such dire warnings and urgent calls for transformative economic approaches are echoed by thousands of sustainability scientists5, the United Nations3,9, intergovernmental environmental panels6,10 and business leaders11,12. Economics studies the production, consumption, valuation, allocation and exchange of goods and services, including their governance. Since the early twentieth century, neoclassical economics has dominated economic thinking in research, policy, education and public debate, projecting a cohesive narrative with substantial discursive power13. This narrative emphasizes the belief that markets optimally allocate resources without state intervention, embodies a mechanistic worldview of nature separate from people14,15,16 and assumes that individuals follow maximizing behaviour according to fixed rules to optimize outcomes for themselves17. Economic policy then becomes a matter of optimization towards market equilibria. Pursuit of affluence, reflected in gross domestic product (GDP) growth, is explicitly or implicitly the central normative goal18. The limits of these theorems have become increasingly acknowledged in mainstream economics through concepts such as market failures, information asymmetries, motivational diversity and complexity theory. There is debate whether such critiques signify the end of neoclassical dominance or its evolution19,20. Alongside these shifts, there is a growing movement of diverse heterodox, ‘new’ economic perspectives that (1) express holistic views of human and planetary well-being, moving away from the conventional focus on affluence and GDP growth; and (2) embrace the social–economic transformation essential for the long-term sustainability of the biosphere and society5,16,21,22. Transformation here means changing the fundamental attributes and institutions of human systems, including shifts in underlying values, worldviews and paradigms6,23. This movement cuts across economics, management, the broader social and sustainability sciences, policy, business, civil society and grass roots economic institutions such as credit unions, cooperatives and social enterprises. ‘New’ economics does not necessarily refer to ideas being recent, but to a desire to establish a new mainstream. Such a paradigm shift involves fundamental changes to basic principles, which, if present across diverse new economic approaches, could provide the foundation for a coherent new mainstream economic framework. The movement for a transformative new economics will only shape new systemic regimes effectively if it has sufficient discursive power to define influential ideas and frames. This is needed to challenge vested interests that keep decision makers attached to dysfunctional political and economic policy regimes24. Gaining discursive power is also essential in overcoming epistemic lock-ins of economics as a discipline25. For example, editorial control of economic journals is dominated by a small number of western, mostly American institutions26, disincentivizing diversity of thought. The global polycrisis presents not only a pressing need for transformation but also opportunities for mainstreaming unconventional economic discourses. Social–ecological and social–economic systems undergo change in cycles27. Certain phases provide windows of opportunity for new paradigms, including ‘release’ and ‘reorganizing’ processes28. Systemic transformation typically originates with innovations in niches, including new narratives essential for transformation29. In time, these innovations can destabilize established practices and trigger regime transformation, resulting in qualitatively different regimes with new narratives and practices29,30. By exposing key problems of conventional thought, global disruptions can help new thinking to gain traction and innovations to scale up31. However, transformative new economic approaches remain diffuse, collectively lacking sufficient discursive coherence and definition to effectively challenge conventional thinking22,32. To address this, we identified principles (Table 1) that cut across approaches and can act as a boundary object to catalyse more cohesive discourse coalitions for economic transformation. We also investigated to what degree different principles were shared across approaches and which were emphasized most. To achieve this, we analysed document sources from science, policy and practice across 38 approaches (Box 1 and Fig. 1). Following discussion of the ten principles identified, we present a research agenda to support development of discourse coalitions22 between disparate approaches for transforming economics in a time of polycrisis. Fig. 1: The review and analysis process. Schematic showing an overview of sampling, screening and data extraction and synthesis. Full size image Table 1 Ten cross-cutting starting principles that underpin transformative new economic approaches Full size table Box 1 Analytical approach Our approach inductively analysed the content of 102 documents deemed to express the scope of transformative new economic approaches by diverse experts from science and practice, linking qualitative content analysis and expert deliberation (Methods). Qualitative content analysis allows researchers to understand social phenomena and identify emerging patterns, themes and concepts in a subjective but systematic, scientific manner. Our interpretive approach did not define a pre-set number or organizational structure for the principles; these emerged from the data through iterative discussion and analysis. The narratives for the principles were subsequently elaborated based on 238 sources drawn from across transformative new economic approaches, including academic and policy-focused literature (Fig. 1). We also included several ‘hybrid’ approaches put forward in the expert survey, which adhered to some of the principles but did not take an explicitly transformative stance. The narratives associated with the principles are reported as a synthesis of predominant characterizations within sources, with divergent views noted where most salient. Every approach except one (Comanche economics) was characterized by at least three sources and by 8.2 sources on average. Fully referenced results are provided in the Supplementary Information. Results Our analysis identified 38 approaches (Table 2). Six approaches bridged conventional and new economics, embracing several principles but with less emphasis on transformation. We identified ten cross-cutting principles, categorized as holistic, ecological, social and political economy (Table 1). Table 2 Overview of new economics approaches and principles emphasized in the sources analysed Full size table Holistic principles Two ‘holistic’ principles point out that economies are part of societies, which in turn are embedded in wider nature. Principle (1) calls on economists to ‘recognize that economies are embedded within societies and ecosystems and that the basic purpose of economics is to support human and planetary well-being’. Many approaches focus directly on integrating long-term environmental, social and economic value, arguing that economics is fundamentally normative and all economic relations are social–ecological relations. Well-being is conceived as embedded in these relationships, with human and planetary well-being interdependent. An explicit goal is to align economic welfare with planetary well-being, recognizing critical constraints such as social capital (for example, caring and trust essential to healthy communities), Earth’s life-supporting systems, cultural principles that regulate relations between people and nature, intergenerational equity and social–ecological resilience. The second holistic principle, ‘acknowledge complexity and the need for interdisciplinarity in addressing economic problems’, points out the need for economists and policymakers to integrate diverse scientific, humanities and local and indigenous knowledge. Economies are complex, adaptive systems. Conventional neoclassical models reduce real-world complexity to an abstract set of production and consumption measures, which narrows possibilities and can drive problematic policy outcomes, for example, in addressing highly complex problems such as climate change. Many approaches thus advocate complexity-based, nonlinear, integrated social–ecological perspectives, acknowledging interdependencies and dynamics within and between natural and social systems, to underpin more holistic economic, social and environmental policies. Ecological principles Building on the holistic principles of embeddedness and complexity, many approaches are grounded in ecological understanding. Regardless of innovation, economic activity (products and services) always draws on natural resources. Principle (3) encourages economists and policymakers to ‘acknowledge that economies have fundamental biophysical and biochemical limits to growth’. It recognizes that economies are inherently constrained by Earth systems. Bio- and circular economic approaches reframe the relationships between economies and earth systems as metabolic processes: the conversion of matter and energy and generation and recycling of material flows. Because metabolic processes are never completely efficient, growth has fundamental limits. Whereas circular economy perspectives mostly focus on decoupling growth and environmental impacts, steady-state, degrowth, post-growth and many ecological economists consider that decoupling alone is insufficient to address the climate and nature emergencies. They argue that ecological constraints require: world-views, models and policy frameworks at local, national and global scales that explicitly recognize, assess and integrate planetary boundaries and societal metabolism; limiting affluence, particularly in the global north; using holistic, well-being-based indicators to measure progress rather than GDP; and reducing inequality. Principle (4) asks economists and policymakers to ‘recognize that human-derived capital depends on nature’. Conventional economic thinking is primarily concerned with outputs, maintaining that when one form of capital input is diminished, another can replace it. This perspective is strongly challenged. For example, post-Keynesian economists reject direct substitutability of resources in aggregate production functions, whereas circular and ecological economists emphasize inherent connections between natural resource consumption and production through societal metabolic processes. This principle thus strongly advocates nature and resource conservation, including preservation of biodiversity and the climate system and diverse social innovations for environmental stewardship, such as multi-scale networking among businesses, neighbourhoods and cities to manage resources. Principle (5), ‘design economic systems to be circular and regenerative’, reflects the design implications of ecological limits and embeddedness. Circular economy approaches focus on generating value by reducing material and energy use per unit of output and maximizing resource regeneration. Sharing economy thinking takes this further by looking to manufacturers, retailers and cooperatives to act as service providers by supplying the use rather than consumption of products. Doughnut economics adds social boundaries to the circular economy, shifting the emphasis to social–ecological regeneration. Flourishing economics emphasizes long-term socio–ecological benefits generated through collaborations between business and public policy within bioregions. Perspectives such as well-being and living economies highlight the importance of designing community-based living infrastructures that generate security, stability and productivity and decentralize decision making. Many approaches recognize the importance of resilient circular and local economies and supply chains, especially in terms of basic needs such as food and energy, to minimize waste, ‘humanize’ productive activities and improve resource security. Social principles Through its social principles, many approaches explicitly consider the relational and societal implications of economic practices. Principle (6), ‘embed pluralistic models of values and behaviour, based on well-being, dignity, sufficiency and holistic freedom in all economic thinking, decisions and actions’, spells out these social needs and implications. Institutions such as monetary systems, markets, valuation methods and economic education have ‘meta-values’ embedded in their design that determine what values are privileged in decisions. This principle thus recognizes the need to embed more relational and sustainability-aligned values in institutional models, for example, through inclusive community approaches to manage common pool resources and deliberative democratic approaches to social–economic valuation. This principle also invites retirement of the conventional view of homo economicus as a self-interested maximizing agent. Instead, it recognizes the socio–biological reality that humans are diversely motivated and that values and behaviour are grounded in social relationships. It resonates with concepts such as humanistic management and homo integralis, expressing people’s wholeness and environmental embeddedness. Holistic freedom here balances ‘negative’ freedom from constraints, such as free enterprise, with ‘positive’ freedoms to be and do what is intrinsically valuable, such as being educated and participating in community life. This understanding encourages interventions focused on needs and capabilities, emphasizing dignity and sufficiency over unconstrained preference satisfaction. Economics has traditionally focused on growth and efficiency as central goals. This is reflected in many economic and political institutions. Principle (7), ‘consider equity, equality and justice as central questions of economic enquiry’, explicitly challenges this emphasis and its association with ‘trickle down’ theory in economic policy, with global wealth inequality persistently increasing in recent decades33,34, despite decreases in between-country income inequality35. Approaches such as well-being economics and economic democracy point out that not just poverty but also inequality undermines well-being, including for the better off. Equality, equity and justice are strongly emphasized by feminist and indigenous economists, advocating rights for social and cultural groups suppressed by conventional economic systems. Caring and feminist economics approaches challenge the gendered nature of economic relations, exposing assumptions concerning women’s paid and unpaid labour in relation to social reproduction. Sen’s capabilities framework and its extensions constitute an important conceptual lens, redefining progress and development to better integrate social justice. Concepts of environmental justice link environmental impacts and social–economic equality. With growth constrained by Earth systems (Principle (3)), fair distribution is an urgent environmental matter. Ecological economists thus propose a new hierarchy of concerns, where sustainable scale, equitable distribution and social–ecological resilience precede efficiency. Political economy principles The fourth set of principles expresses ways to reshape the political economy to support inclusion and participation. Principle (8), ‘embrace pluralistic social and relational approaches that support social enfranchisement, social needs and the common good’, encourages more central integration of relational worldviews and values such as care, community, love and reciprocity into economic thinking and policy. Proponents generally see important roles for the state and civil society in securing economic and socio–relational priorities and universal access to basic services. Approaches such as foundational, caring and diverse indigenous economics advocate repurposing businesses and financial systems to ensure long-run social and ecological value, overcoming narrow emphases on short-term surplus generation and adapting more relational models of corporate leadership and new structures of accountability towards stakeholders. This also requires more holistic and integrated evaluation and reporting, including metrics that recognize the profound value of unpaid care work and environmental benefits and damages. Conventionally, economics has emphasized modelling and analytical, data-driven methods. Several approaches expand their scope to recognize complex systems and incorporate coupled ecological economic models. Principle (9), ‘embed participation, deliberation and cooperation as core to economic thinking and policy’, advocates integrating analytical approaches with more inclusive social processes in research and policy. It is not possible to fully separate analytical from normative research when considering complex systems. Any choice of technical parameters is ultimately value based, and conflicts between values cannot be fully resolved through optimization but must be democratically deliberated through methodologies such as participatory action research, participatory systems modelling and deliberative valuation. In relation to policy, deliberative economic approaches envisage an active role for citizens participating and cooperating to improve their quality of life and advocate for the rights of workers and others affected by economic policies to have genuine opportunities to participate in decisions. New municipalism and cosmolocalism focus on institutionalizing participative platforms and practices involving collaborations among citizens, municipalities and other levels of governance and through effective cooperative institutions serving local economic development needs. While cooperative perspectives have received some scepticism in conventional economics, institutional and feminist economists have pointed them out as common practice, benefiting sustainability, productivity and equality. Finally, Principle (10), ‘take post-capitalist, decolonized economic perspectives’, underpins diverse models and applications that disrupt conventional relationships between capital and labour, with particular regard for the views of marginalized and previously colonized peoples. Concepts of production conventionally build on labour-capital dichotomies and the concentration of power and capital, reflected in the post-colonial export of western mass consumption lifestyles to the global south. Decolonization and post-capitalism are thus linked in their analyses of ‘unmaking’ colonial and capitalist institutional configurations. Rather than providing a single ideological post-capitalist blueprint, diverse approaches, applications and perspectives on markets and monetary and financial systems are advocated. At the microeconomic scale, sharing economy and post-capitalist approaches envisage economic practices based on new technologies and a reduced need for labour, including new currencies that embed social and ecological values, communal ownership, new forms of cooperatives and online networking spaces to promote non-profit forms of work and address labour mobility, empower disadvantaged individuals and support capabilities. Cosmolocalism envisages collaborations between globally connected citizens and grassroots movements to transform consumption–production regimes through digital innovation, strengthening both local, social–ecologically embedded economies and global citizenship and multilateralism. At the macroscale, degrowth, post-growth, ecological, steady-state and post-Keynesian economics provide new analytical tools in areas such as monetary and physical input–output and system dynamics modelling, whereas post-development theory affirms cultural diversity, aligns new economics with indigenous philosophies, promotes democracy and provides social spaces for conflict resolution and social protocols associated with reciprocity and respect for nature. In summary, these ten principles explicitly shift attention in economic thinking and policy towards holism, heterodoxy, plurality, interdisciplinarity, equity, well-being, participation and aligning economic activities with natural systems. They recognize the context specificity of institutions, values and culture and the need for relational and complexity-based thinking to achieve inclusive and just transformation towards sustainability. Approaches and principles In the sources assessed, the most emphasized principles included relationality and social enfranchisement (n = 25 approaches), holistic perspectives of people and values (n = 24) and holistic well-being and social–ecological embeddedness (n = 23) (Table 1). Transformative approaches expressed three to ten principles each, and hybrid approaches two to four (Table 2). Christian relational perspectives (n = 10), ecological and ecological feminist economics and degrowth (each n = 8) were characterized by most principles. Along with sumak kawsay, buen vivir and foundational and well-being economy, those approaches were also associated with at least one principle in each of the holistic, ecological, social and political economy grouping categories. We identified four groups of approaches in terms of the principles they emphasized. A group cluster of mainly economic perspectives (economic democracy, deliberative economics, new municipalism, new progressivism, post-development, world system theory) emphasize social and political principles, focusing on questions of democracy, participation, cooperation, deliberation and decolonization. A second, larger cluster also mostly consists of economic perspectives (flourishing economics, well-being economy, doughnut economics, degrowth, agrowth, ecological economics, circular economy, steady-state economics, cosmolocalism and post-growth). These approaches have a strong focus on ecological limits and the scale of the economy and often also frame social and political economy questions within the context of earth systems. Buen vivir, enlivenment and Christian humanistic and relational perspectives overlap with this cluster while also linking to the next. A third cluster consists of a wide range of economic and broader societal perspectives (institutional, caring and feminist economics, responsible capitalism, solidarity, foundational and sharing economy, buen vivir, sumak kawsay, ubuntu, kaitiakitanga, enlivenment, Christian humanistic and relational perspectives) that emphasize social and political economic issues such as power, justice and equity and often link these with relationality, quality of life and human dignity. The final, smallest cluster consists of schools of economics and economic perspectives (fair markets and behavioural, complexity, post-Keynesianism and post-capitalist economics) that critically develop mainstream economic knowledge and themes, elaborating on issues such as growth and development, business values, fairness in markets and economies, wealth and distribution and behaviour. Discussion Our synthesis links a large number of transformative new economic approaches across science and practice and systematically distils commonalities (and divergence) in their fundamental principles. No single principle was emphasized by all approaches. Only a minority of approaches cut across all categories of principles. This indicates the salience of the principles for building discourse coalitions where different approaches complement each other. Within the holistic, social and political economy categories, there was at least one principle that was emphasized by a substantial majority of approaches. However, even the most expressed ecological principle, limits to growth, was emphasized by only 45% of the approaches represented in our sample. This does not mean that other approaches ignore sustainability transformation altogether but that the sources for these approaches did not develop a cohesive environmental discourse with clear principles. This is surprising because of the widespread attention to the climate and biodiversity crises and associated risks and the way that these crises point to the need for economic transformation5,6,10,12. Thus, there continue to be important opportunities for diffusion of transformative ecological perspectives into socially oriented approaches. This can lead to mutual enrichment, as demonstrated by, for example, integrations between feminist and ecological economics36. Different approaches also have different epistemic and ontological assumptions, which can generate tensions but also raise opportunities for mutual learning. For example, productive dialogues may be had between indigenous broad societal perspectives and western economic schools and perspectives around understanding complexity37 and decolonization21—or around relationality, which is considered by, for example, behavioural, feminist and deliberative economics, and Christian humanistic and indigenous perspectives but through different concepts, worldviews and knowledge systems38,39,40,41,42. There is also broader opportunity for diffusion of transformative perspectives into hybrid approaches that bridge neoclassical and new economics. For example, the cluster of approaches focusing on critically elaborating mainstream themes form a spectrum from more (for example, post-Keynesian) to less transformative orientations (for example, complexity economics), providing opportunity for debate around the need for transformation. Such debates are important to address the risk of perpetuating fundamental problems and power structures if mainstream economics takes on more sophisticated assumptions and broadens its scope of analysis, without broadening its imagination of what kind of economic systems are possible32. This is also why a paradigmatic shift is needed across the ten principles. Discourse coalitions geared towards mainstreaming their breadth can ensure that current movements towards integration of individual elements (for example, circularity) become a springboard towards broader transformation, rather than a way to evolve neoclassical economics without fundamentally challenging vested interests. Building discourse coalitions is particularly important in a time of polycrisis, not just because of the urgent need for transformation, but also because global economic systems’ consistent failure to respond equitably to crises provides opportunities for change that speak to the principles. For example, many failures recognized by new economists during the COVID-19 pandemic are now acknowledged in public enquiries, such as how neoliberal thinking undermined supply chain resilience43, resisted public health interventions44 and caused pre-existing vulnerabilities resulting from diverse inequalities, which strongly influenced disease and death rates45. The pandemic also heightened recognition of the importance of key workers, the foundational economy, cooperative economic practices and our interdependencies with nature46,47. Furthermore, the large-scale social and economic interventions by governments, rapid changes in behaviour by citizens and adaptations by firms set a precedent for rapid transformative change in response to other crises45. However, crises also generate new barriers. For example, dynamics around inequalities exposed by crises such as the pandemic go two ways, where increased recognition provides a leverage point for change, yet further concentration of wealth further entrenches vested interests48. Overcoming such interests requires discursive power through discourse coalitions, but also disrupting material power relations and reclaiming power through new economic institutions (principles (9)–(10)). Another avenue for mainstreaming the ten principles is their ability to link questions of resource allocation to globally agreed values and norms. These include the UN Declarations of Human Rights and Rights of the Child, the Convention on Biological Diversity’s Ecosystem Approach and the way these are expressed in frameworks such as the Sustainable Development Goals, the Paris Agreement and Kunming–Montreal Global Biodiversity Framework. Normative approaches challenge the conventional economic fact-value dichotomy that reduces social questions to technical problems49. This artificial divide allows policymakers to routinely abuse economic arguments to justify unethical and unsustainable policy. For example, bolstering growth was used to justify pandemic recovery policies weakening social and environmental regulations and strengthening environmentally destructive industries45,48. Crises such as the pandemic and invasion of Ukraine also expose predictable50 policy failures resulting from dogmatic focus on efficiency and growth, such as faltering supply chains and western reliance on authoritarian regimes for fossil fuels. These failures underline the need for transformative new economic approaches that integrate rather than externalize human rights and needs, resilience and environmental limits51. Another example is humanity’s inability to effectively address climate change. Besides the barriers posed by vested interests, this is to an extent driven by simplistic economic modelling approaches that insufficiently acknowledge complexity and the need for interdisciplinarity. A range of more holistic tools integrating new economic principles are available for assessing viable transition paths but are not yet commonly applied in policy analysis or Intergovernmental Panel on Climate Change (IPCC) reports52,53,54. Starting with the 2008 global financial crisis, recent backlashes against globalization may herald the end of neoliberal dominance55. However, it is unclear what structures may replace it—they could be defined by protectionism, nationalism and authoritarianism56 or by new economic initiatives pairing global collaboration with decentralization through regional circular economies and empowered local communities2,45. There are increasing examples where such thinking is becoming more prominent, from community-embedded economic responses during the pandemic to broad well-being indicators for measuring macroeconomic progress, to formal government adoption of doughnut (for example, Amsterdam) and well-being economics (for example, Scotland, New Zealand, Iceland, Wales and Finland). Yet, in academia, whereas some principles are gaining traction through hybrid approaches such as circular and behavioural economics, research on transformative new approaches is still largely absent from top-ranking economic journals, with more prominence in interdisciplinary journals and disciplines such as geography57. Most economics textbooks also continue to present a homogeneous, largely conventional body of knowledge58. The ten principles tie together a range of interdisciplinary, integrative concepts (for example, planetary boundaries, societal metabolism, regeneration, value pluralism, social–ecological embeddedness) that reflects the complexity of human behaviour, societal interactions and human–nature relationships. Such ‘integration by concepts’ is key to a more comprehensive understanding of issues and can underpin more effective and just solutions to global crises59,60. By providing a cohesive narrative grounded in such concepts across new economic approaches, the ten principles can strengthen discursive power through discourse coalitions while respecting ideological pluralism and differences in emphasis, focus, strategy and framing, for example, in relation to capitalism or economic growth32. Global south approaches in particular advocate ontological and epistemological pluralism, challenging the monism of conventional development and aspiring for a pluriverse or ‘world of many worlds’61. Whereas our results demonstrate diffuse adoption of different principles across approaches, competing post-globalization nationalist conservative discourse is also highly diffuse, with potentially more problematic internal contradictions56. However, despite the prominence of global south approaches such as buen vivir and the solidarity economy, transformative new economics research remains heavily concentrated in the global north, risking a bias towards certain issues and frames. For example, the increasing prominence of degrowth frames may resonate less in the global south62, and specific gender-inequality issues in the global south arising in global crises such as the COVID-19 pandemic (for example, women’s livelihood loss, food insecurity, educational setbacks for girls) risk being overshadowed by global north issues (for example, domestic care responsibilities)63. Recognition of non-western worldviews and knowledge systems is a generic scientific challenge64, and new economics is no exception. More research is thus needed to validate and develop new economics principles and narratives more explicitly within global south contexts. Future research could also consider applications of the principles in policy, business and civil society: what can be learned from current applications and institutionalizations and what are the outcomes in terms of equity, sustainability and perceived legitimacy? There are opportunities for further synthesis research in many areas, such as developing more integrated, pluralistic and relational models of value and human behaviour (building particularly on principles (6), (8) and (9)), and cohesive views beyond capitalism and socialism of the relations between capital, labour, markets, the state and communities that connect diverse thinking and practice (principles (1), (5), (8) and (10)). Whereas our review focused on principles, there is also a need for reviewing transformative new economic pathways that address diverse crises and methodologies across approaches, including economic instruments for policy, different analytical methods and boundary methods at the interface of research and policy such as citizen assemblies. Again, the divergent expertise of different new and hybrid economics approaches can strengthen each other, for example, from behavioural economic experimental approaches to post-Keynesian macroeconomic modelling, and recursive methods in complexity economics to deliberative valuation and participatory appraisal in ecological and feminist economics. Finally, more research is needed on understanding barriers and opportunities for mainstreaming the principles in a polycrisis world, which continues to be volatile and uncertain. Future research could consider strategies for overcoming vested interests, breaking through discursive lock-ins, understanding contexts in which transformative new economics approaches are being implemented and connecting niche initiatives into networks that amplify their transformative potential. The importance of such work cannot be overstated, because unless the ideas summarized in the ten principles are rapidly embedded in global and national institutions, humanity is unlikely to overcome the extreme crises it is facing. Methods We conducted an inductive, qualitative content analysis65 to understand the scope of new economic approaches by identifying core principles and systematically synthesizing them across a large number of economic approaches. Figure 1 provides an overview of the sampling, screening and analysis process. The initial set of sources was provided by members of the Leadership Team and Advisory Board for the Global Assessment for a New Economics (GANE) project (http://neweconomics.net). Advisory Board members were affiliated with diverse well-established organizations that embraced or promoted new economics in science, business and policy, such as the Club of Rome, Wellbeing Economy Alliance, World Future Council, World Resources Institute, the Capital Institute, Catalyst 2030, Better Nature, Ethical Markets, the Gross National Happiness Centre and the Green Economy Coalition. Eleven of 24 board members were based in the global south, whereas 17 of 24 members were affiliated with organizations with a global remit. Because of this diversity of backgrounds, we could draw on a mix of sources across research and practice. Board members were (as part of a broader survey on new economics and transformation within a COVID-19 pandemic context) requested to name one or more new economics approaches within their expertise and provide up to five key sources each, describing one or more new economics approaches. GANE leadership team members added further sources for underrepresented approaches. The experts were asked to consider new economics as being broadly associated with a transformative orientation, moving away from an emphasis on GDP growth towards advancing well-being through meeting basic needs, restoring ecosystems and increasing equality. These inclusion criteria drew on three publications that previously reviewed a substantial number of named new economics approaches21,22,66 and Ripple et al.5, an article signed by over 13,000 sustainability scientists that provided a starting point for identifying changes in economics needed for sustainability transformation. One hundred fifty-seven sources were put forward, including journal articles (49%), books (24%), reports (8%), other articles (7%), position statements (6%), web pages (3%), conference proceedings (1%), working papers (1%) and one video (1%). The sources were only included if they clearly described one or more new economics approaches and/or principles. Web pages without a personal or institutional author were also excluded. This left 137 sources (Fig. 1). For practical reasons, 35 of 37 in-scope books were not initially included for detailed analysis but were drawn on for later elaboration of the principles; two books were included because of the limited number of other sources associated with their approach. This left 102 documents that were used for detailed analysis of the new economics approaches and principles. The sources were first screened to extract labels for new economic approaches. The analysis included all distinct named new economics approaches in the main text of these sources, apart from those that were a specific ‘sub-approach’ or concept that could be adopted by multiple new economics approaches (for example, ‘food sovereignty’) or a broad umbrella term or higher-level approach that could embrace multiple new economics approaches (for example, transmodernism). The analysis then followed a hybrid of conventional and directed qualitative content analysis65. The sources were initially inductively analysed across four predefined themes: (1) problems with conventional economic thinking; (2) solutions to address problems; (3) basic principles underpinning solutions; (4) strategies and pathways for change. Verbatim quotes and page numbers were recorded to a spreadsheet. A random pilot sample of 15 sources was coded, with the coding discussed and validated through discussion between five members of the research team before the full dataset was coded. Following initial coding to the four themes, the codes were iteratively consolidated through deliberation among the author team. Here we aimed for comprehensiveness and treated content equally regardless of the number of sources that advocated them to avoid biasing towards more strongly represented approaches. The consolidated codes were then organized together across the four themes in a narrative matrix around the consolidated basic principles, that is, linking principles with problems, solutions and change pathways. Through further deliberation, we then refined the matrix into consolidated narratives surrounding the high-level principles and ordered the principles to form a cohesive overall narrative. There was no pre-set intention to identify a certain number of high-level principles, but we sought to balance between having too many, overly specific principles and having a small number of too wide-ranging principles. The resulting principles were also qualitatively mapped according to broad categories that were emergent from the data (ecological, social, political economy or holistic). To further elaborate and exemplify the narratives for each principle, we reincluded the in-scope book sample and added a further 101 sources through snowballing, expert knowledge and targeted searches to further supplement sources for poorly represented approaches and ensure that each approach was represented by at least three sources. The only exception was Comanche philosophy, for which a single source was submitted in the survey39 and no further relevant sources could be identified. Inclusion of the additional sources generated a final sample (n = 238), including journal articles (60%), books (23%), reports (7%), position statements (3%) and other sources (6%) (Fig. 1). Each approach was then cross checked against the coded principles data to map approaches against principles. In doing so, we interpreted whether the principle was emphasized by one or more of the sources, that is, clearly identifiable as a distinct discursive theme, based on agreement by two of the researchers. This mapping was then validated and updated using the expanded, final sample. Through this process, the approaches were also grouped into emergent categories: (1) schools of economics that reflect particular theoretical and analytical paradigms; (2) economic perspectives that take broader views of how economies should be structured and tend to be less academic and more policy focused than schools; and (3) broad economic philosophies that represent broader philosophical approaches with explicit elements of economic thinking. Characterizing principles and narratives through inductive analysis is inevitably interpretive. Other researchers might have structured the narrative in a different number of principles, labelled and organized them differently and interpreted the relative emphasis of different approaches on particular principles somewhat differently. Surveying different experts would have also led to a different basis for interpretation, and expert selection bias inevitably influenced the relative representation of different approaches and sources selected. We partially mitigated this through increasing the representation of poorly represented approaches as discussed above. Relative representation will also have been influenced by the nature of the approaches: whereas some approaches have built up a vast literature over multiple decades (for example, ecological economics) or have more recently become well known and prolific (for example, degrowth), other approaches originate in a single author (for example, doughnut economics, enlivenment). Rather than addressing such issues through adding arbitrary weights, our focus was on developing discursive synthesis. Consequently, we represented elements of discourse independent of how frequently they were represented. Another important limitation of the expert survey-based approach is that there may be relevant approaches that were not named but may be considered as new economics by other experts. For example, new economic principles can be identified in religious and spiritual traditions besides Christian humanism. Marxist economics was also not put forward, although it has influenced a range of approaches that were included (for example, buen vivir, solidarity economy, new Progressivism, economic democracy). Less prominent approaches may have been unknown to the surveyed experts, considered as sub-approaches or out of scope as economic approaches. However, given satiation in the data and the large number of approaches analysed, it is unlikely that additional approaches would have generated major changes to the principles. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Data availability The dataset generated by the survey research is available via figshare at https://doi.org/10.6084/m9.figshare.28615724 (ref. 67). References Lawrence, M. et al. Global polycrisis: the causal mechanisms of crisis entanglement. Glob. Sustain. https://doi.org/10.1017/sus.2024.1 (2024). Berkhout, E. et al. The Inequality Virus: Bringing Together a World Torn Apart by Coronavirus through a Fair, Just and Sustainable Economy (Oxfam, 2021); https://www.oxfam.org/en/research/inequality-virus Human Development Report 2020: The next Frontier - Human Development and the Anthropocene (UNDP, 2020). Zaliska, O., Oleshchuk, O., Forman, R. & Mossialos, E. Health impacts of the Russian invasion in Ukraine: need for global health action. Lancet 399, 1450–1452 (2022). Article Google Scholar Ripple, W. J., Wolf, C., Newsome, T. M., Barnard, P. & Moomaw, W. R. World scientists’ warning of a climate emergency. BioScience 70, 8–12 (2020). Article Google Scholar IPCC Climate Change 2022: Mitigation of Climate Change (eds Shukla, P. R. et al.) (Cambridge Univ. Press, 2022); https://doi.org/10.1017/9781009157926 Fremstad, A. & Paul, M. Neoliberalism and climate change: how the free-market myth has prevented climate action. Ecol. Econ. 197, 107353 (2022). Article Google Scholar Methodological Assessment of the Diverse Values and Valuation of Nature (eds Balvanera, P. et al.) (IPBES, 2022); https://doi.org/10.5281/zenodo.6522522 Guterres, A. The State of the Planet, Secretary-General’s Address at Columbia University (United Nations, 2020). Brondízio, E. S. et al. (eds) Global Assessment Report on Biodiversity and Ecosystem Services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES, 2019); https://doi.org/10.5281/zenodo.3831673 Vision 2050: Time to Transform (WBCSD, 2021); https://www.wbcsd.org/Overview/About-us/Vision-2050-Time-to-Transform/Resources/Time-to-Transform Nature Risk Rising: Why the Crisis Engulfing Nature Matters for Business and the Economy (WEF and PwC, 2020); https://www.weforum.org/publications/nature-risk-rising-why-the-crisis-engulfing-nature-matters-for-business-and-the-economy/ Finlayson, A. C., Lyson, T. A., Pleasant, A., Schafft, K. A. & Torres, R. J. The ‘invisible hand’: neoclassical economics and the ordering of society. Crit. Sociol. 31, 515–536 (2005). Article Google Scholar Ortiz-Przychodzka, S., Benavides-Frías, C., Raymond, C. M., Díaz-Reviriego, I. & Hanspach, J. Rethinking economic practices and values as assemblages of more-than-human relations. Ecol. Econ. 211, 107866 (2023). Article Google Scholar Söderbaum, P. Neoclassical and institutional approaches to development and the environment. Ecol. Econ. 5, 127–144 (1992). Article Google Scholar Waddock, S. Reframing and transforming economics around life. Sustainability 12, 7553–16 (2020). Article Google Scholar Lawson, T. What is this ‘school’ called neoclassical economics? Camb. J. Econ. 37, 947–983 (2013). Article Google Scholar Jackson, T. Post Growth: Life after Capitalism (John Wiley & Sons, 2021). Colander, D., Holt, R. & Rosser, B. Jr. The changing face of mainstream economics. Rev. Political Econ. 16, 485–499 (2004). Article Google Scholar Neck, R. in The Palgrave Handbook of the History of Human Sciences (ed. McCallum, D.) 1–40 (Springer, 2020). https://doi.org/10.1007/978-981-15-4106-3_5-1 Beling, A. E. et al. Discursive synergies for a ‘great transformation’ towards sustainability: pragmatic contributions to a necessary dialogue between human development, degrowth, and buen vivir. Ecol. Econ. 144, 304–313 (2018). Article Google Scholar Riedy, C. Discourse coalitions for sustainability transformations: common ground and conflict beyond neoliberalism. Curr. Opin. Environ. Sustain. 45, 100–112 (2020). Article Google Scholar Feola, G. Societal transformation in response to global environmental change: a review of emerging concepts. Ambio 44, 376–390 (2015). Article Google Scholar Palley, T. A theory of economic policy lock-in and lock-out via hysteresis: rethinking economists’ approach to economic policy. Economics 11, 1–18 (2017). Article Google Scholar Palley, T. From Financial Crisis to Stagnation: The Destruction of Shared Prosperity and the Role of Economics (Cambridge Univ. Press, 2012). Angus, S. D., Atalay, K., Newton, J. & Ubilava, D. Geographic diversity in economic publishing. J. Econ. Behav. Organ. https://doi.org/10.1016/j.jebo.2021.08.005 (2021). Gunderson, L. H. & Holling, C. S. Panarchy: Understanding Transformations in Human and Natural Systems (Island Press, 2002). Westley, F. et al. A theory of transformative agency in linked social-ecological systems. Ecol. Soc. 18, art27 (2013). Article Google Scholar Simoens, M. C., Fuenfschilling, L. & Leipold, S. Discursive dynamics and lock-ins in socio-technical systems: an overview and a way forward. Sustain Sci. 17, 1841–1853 (2022). Article Google Scholar Geels, F. W. Technological transitions as evolutionary reconfiguration processes: a multi-level perspective and a case-study. Res. Policy 31, 1257–1274 (2002). Article Google Scholar Koch, M. & Buch-Hansen, H. In search of a political economy of the postgrowth era. Globalizations 18, 1219–1229 (2021). Article Google Scholar Buckton, S. J. et al. Reform or transform? A spectrum of stances towards the economic status quo within ‘new economics’ discourses. Glob. Soc. Chall. J. 3, 382–481 (2024). Article Google Scholar Zucman, G. Global wealth inequality. Annu. Rev. Econ. 11, 109–138 (2019). Article Google Scholar Beyond GDP: Measuring What Counts for Economic and Social Performance (OECD, 2018). Hung, H. Recent trends in global economic inequality. Annu. Rev. Socio. 47, 349–367 (2021). Article Google Scholar Perkins, P. E. Feminist ecological economics and sustainability. J. Bioecon. 9, 227–244 (2007). Article Google Scholar Malo Larrea, A., Ambrosi de la Cadena, M., Collado Ruano, J. & Gallardo Fierro, L. Transcending the nature-society dichotomy: a dialogue between the sumak kawsay and the epistemology of complexity. Ecol. Econ. 216, 108044 (2024). Article Google Scholar DellaValle, N. People’s decisions matter: understanding and addressing energy poverty with behavioral economics. Energy Build. 204, 109515 (2019). Article Google Scholar Harris, L. D. & Wasilewski, J. Indigeneity, an alternative worldview: four R’s (relationship, responsibility, reciprocity, redistribution) vs. two P’s (power and profit). Sharing the journey towards conscious evolution. Syst. Res. Behav. Sci. 21, 489–503 (2004). Article Google Scholar Schluter, M. Beyond capitalism: towards a relational economy. Cambridge Pap. 19, 1–4 (2010). Google Scholar Nussbaum, M. Capabilities as fundamental entitlements: Sen and social justice. Feminist Econ. 9, 33–59 (2003). Article Google Scholar Lecture, K. E. & Norgaard, R. B. Deliberative economics. Ecol. Econ. 63, 375–382 (2007). Article Google Scholar Adelodun, B. et al. Understanding the impacts of the COVID-19 pandemic on sustainable agri-food system and agroecosystem decarbonization nexus: a review. J. Cleaner Prod. 318, 128451 (2021). Article CAS Google Scholar Laszlo, C., Cooperrider, D. & Fry, R. Global challenges as opportunity to transform business for good. Sustainability 12, 8053 (2020). Article Google Scholar van Barneveld, K. et al. The COVID-19 pandemic: lessons on building more equal and sustainable societies. Econ. Labour Relat. Rev. 31, 133–157 (2020). Article Google Scholar Flórez, J. J. R., Rodríguez, J. J. J. & Romero, R. Á. S. Economía social como alternativa ante una sociedad post coronavirus. Rev. de Cienc. Soc. 27, 147–162 (2021). Google Scholar Benach, J. We must take advantage of this pandemic to make a radical social change: the coronavirus as a global health, inequality, and eco-social problem. Int. J. Health Serv. 51, 50–54 (2021). Article Google Scholar Spash, C. L. ‘The economy’ as if people mattered: revisiting critiques of economic growth in a time of crisis. Globalizations 18, 1087–1104 (2021). Article Google Scholar Pressman, S. The two dogmas of neoclassical economics. Sci. Soc. 68, 483–493 (2004). Article Google Scholar Stiglitz, J. E. Making Globalization Work (WW Norton & Company, 2007). Jungell-Michelsson, J. & Heikkurinen, P. Sufficiency: a systematic literature review. Ecol. Econ. 195, 107380 (2022). Article Google Scholar Pollitt, H., Mercure, J.-F., Barker, T., Salas, P. & Scrieciu, S. The role of the IPCC in assessing actionable evidence for climate policymaking. npj Clim. Action 3, 11 (2024). Article Google Scholar Hafner, S., Anger-Kraavi, A., Monasterolo, I. & Jones, A. Emergence of new economics energy transition models: a review. Ecol. Econ. 177, 106779 (2020). Article Google Scholar Souffron, C. & Jacques, P. A successful assessment of the economic impacts of ecological transition policies in the EU requires the European Commission to broaden the range of its modelling tools. Preprint at SSRN https://doi.org/10.2139/ssrn.4640677 (2024). Saad-Filho, A. Endgame: from crisis in neoliberalism to crises of neoliberalism. Hum. Geogr. 14, 133–137 (2021). Article Google Scholar ‘National conservatives’ are forging a global front against liberalism. The Economist (15 February 2024). Čajka, A. & Novotný, J. Let us expand this western project by admitting diversity and enhancing rigor: a systematic review of empirical research on alternative economies. Ecol. Econ. 196, 107416 (2022). Article Google Scholar Decker, S., Elsner, W. & Flechtner, S. Advancing Pluralism in Teaching Economics (Routledge, 2019). Spash, C. L. in Towards an Integrated Paradigm in Heterodox Economics: Alternative Approaches to the Current Eco-Social Crises (eds Gerber, J.-F. & Steppacher, R.) 26–46 (Palgrave Macmillan UK, 2012); https://doi.org/10.1057/9780230361850_2 Kapp, K. W. Environmental disruption and social costs: a challenge to economics. Kyklos 23, 833–848 (1970). Article Google Scholar Escobar, A. Designs for the Pluriverse: Radical Interdependence, Autonomy, and the Making of Worlds (Duke Univ. Press, 2017). Tomaselli, M. F., Kozak, R., Gifford, R. & Sheppard, S. R. J. Degrowth or not degrowth: the importance of message frames for characterizing the new economy. Ecol. Econ. 183, 106952 (2021). Article Google Scholar Agarwal, B. Livelihoods in COVID times: gendered perils and new pathways in India. World Dev. 139, 105312 (2021). Article Google Scholar Anderson, C. B. et al. in Methodological Assessment Report on the Diverse Values and Valuation of Nature of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (eds Balvanera, P. et al.) Ch. 2 (IPBES, 2022); https://doi.org/10.5281/zenodo.7154713 Hsieh, H.-F. & Shannon, S. E. Three approaches to qualitative content analysis. Qual. Health Res 15, 1147–1288 (2005). Article Google Scholar Washington, H. & Maloney, M. The need for ecological ethics in a new ecological economics. Ecol. Econ. 169, 106478 (2020). Article Google Scholar Kenter, J. O., Martino, S. & Buckton, S. J. GANE new economics expert survey data. figshare https://doi.org/10.6084/m9.figshare.28615724 (2025). Download references Acknowledgements We thank I. Fazey and I. Kendrick for their support in developing the project. This research was supported by the Leverhulme Trust-funded Leverhulme Centre for Anthropocene Biodiversity (RC-2018-021), the University of York (United Kingdom), Ecologos Research Ltd (United Kingdom) and the United Kingdom Natural Environment Research Council (NE/X002276/1). Author information Authors and Affiliations Aberystwyth Business School, Aberystwyth University, Aberystwyth, UK Jasper O. Kenter Department of Environment and Geography, University of York, York, UK Jasper O. Kenter, Simone Martino, Sam J. Buckton & Adam P. Hejnowicz Ecologos Research Ltd, Aberystwyth, UK Jasper O. Kenter, Simone Martino, Sam J. Buckton, Adam P. Hejnowicz & Jordan O. Lafayette The James Hutton Institute, Aberdeen, UK Simone Martino Carroll School of Management, Boston College, Chestnut Hill, MA, USA Sandra Waddock Global Development Institute, University of Manchester, Manchester, UK Bina Agarwal Institute of Economic Growth, Delhi, India Bina Agarwal Climate Change Policy Group, Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UK Annela Anger-Kraavi School of Economics and Business Administration, University of Tartu, Tartu, Estonia Annela Anger-Kraavi Institute for Global Prosperity, University College London, London, UK Robert Costanza School of Geosciences, University of Edinburgh, Edinburgh, UK Adam P. Hejnowicz Tecnológico de Monterrey, Mexico City, Mexico Peter Jones Flourishing Enterprise Institute, Toronto, Ontario, Canada Peter Jones Lancaster Environment Centre, University of Lancaster, Lancaster, UK Jordan O. Lafayette School of Geography, University of Nottingham, Nottingham, UK Jordan O. Lafayette Partnership for Economic Policy, Nairobi, Kenya Jane Kabubo-Mariara University of Nairobi, Nairobi, Kenya Jane Kabubo-Mariara Department of Social and Political Sciences, College of Business, Arts and Social Sciences, Brunel University London, Uxbridge, UK Nibedita Mukherjee Department of Health Sciences and Leverhulme Centre for Anthropocene Biodiversity, University of York, York, UK Kate E. Pickett Institute for Sustainable Futures, University of Technology Sydney, Sydney, New South Wales, Australia Chris Riedy Bounce Beyond, Boston, MA, USA Steve Waddell Authors Jasper O. Kenter View author publications You can also search for this author inPubMed Google Scholar Simone Martino View author publications You can also search for this author inPubMed Google Scholar Sam J. Buckton View author publications You can also search for this author inPubMed Google Scholar Sandra Waddock View author publications You can also search for this author inPubMed Google Scholar Bina Agarwal View author publications You can also search for this author inPubMed Google Scholar Annela Anger-Kraavi View author publications You can also search for this author inPubMed Google Scholar Robert Costanza View author publications You can also search for this author inPubMed Google Scholar Adam P. Hejnowicz View author publications You can also search for this author inPubMed Google Scholar Peter Jones View author publications You can also search for this author inPubMed Google Scholar Jordan O. Lafayette View author publications You can also search for this author inPubMed Google Scholar Jane Kabubo-Mariara View author publications You can also search for this author inPubMed Google Scholar Nibedita Mukherjee View author publications You can also search for this author inPubMed Google Scholar Kate E. Pickett View author publications You can also search for this author inPubMed Google Scholar Chris Riedy View author publications You can also search for this author inPubMed Google Scholar Steve Waddell View author publications You can also search for this author inPubMed Google Scholar Contributions Conceptualization: J.O.K., S.M., S.J.B., S. Waddock, N.M., B.A., A.A.-K., R.C., P.J., J.K.-M., K.E.P. and S. Waddell. Methodology: J.O.K., S.M., S.J.B. and N.M. Analysis: J.O.K., S.M., S.J.B., S. Waddock, N.M., B.A., A.A.-K., A.P.H., P.J., J.K.-M., J.O.L., K.E.P. and S. Waddell. Writing—original draft: J.O.K., S.M., SJ.B., S. Waddock, B.A., A.A.-K., A.P.H., P.J. and C.R. Review and editing: J.O.K., S.J.B., S.M., S. Waddock, N.M., B.A., A.A.-K., R.C., A.P.H., P.J., J.K.-M., K.E.P. and C.R. Funding acquisition: J.O.K. and K.E.P. Corresponding author Correspondence to Jasper O. Kenter. Ethics declarations Competing interests The authors declare no competing interests. Peer review Peer review information Nature Sustainability thanks Devika Dutt and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Additional information Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Supplementary information Supplementary Information Supplementary Text and Table 1. Reporting Summary Rights and permissions Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Reprints and permissions About this article Cite this article Kenter, J.O., Martino, S., Buckton, S.J. et al. Ten principles for transforming economics in a time of global crises. Nat Sustain (2025). https://doi.org/10.1038/s41893-025-01562-4 Download citation Received: 19 March 2024 Accepted: 09 April 2025 Published: 22 May 2025 DOI: https://doi.org/10.1038/s41893-025-01562-4 Share this article Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative

Asian rice plants inherited a tolerance for cold in just three generations.Credit: Adhi Syailendra/Getty Rice plants can inherit tolerance to cold without changes to their genomes, according to a decade-long study1 carried out by researchers in China. The work, published in Cell today, strengthens the evidence for a form of evolution in which environmental pressures induce heritable changes that do not alter an organism’s DNA. The study conducted experiments that demonstrate, for the first time, the mechanism for these changes — ‘epigenetic’ tweaks to chemical markers on the plant’s DNA that don’t actually tinker with the sequences themselves. “What they’re showing is extremely convincing; I would say that it’s a landmark in the field,” says Leandro Quadrana, a plant geneticist at the French National Centre for Scientific Research in Paris-Saclay. Michael Skinner, who studies epigenetic inheritance at Washington State University in Pullman, says the study adds to the growing body of evidence challenging the prevailing view of evolution that the only way that adaptations emerge is through gradual natural selection of randomly arising DNA mutations. This study shows that the environment isn’t just a passive actor in evolution, but a selective force inducing a targeted change. “This is a very well done, very nice study, and it is easily understood,” says Skinner. Chill time Rice plants originated from tropical regions but can now be grown worldwide, even adapting to colder environments. Shanjie Tang, a plant geneticist at the Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, in Beijing and his colleagues wanted to investigate how the plants acquired this ability to adapt to cold temperatures. They identified an Asian rice (Oryza sativa L.) variety that was particularly susceptible to cold, producing fewer viable seeds when grown in colder environments. Just as the plants were readying to begin reproduction, the researchers placed them in a chamber set to –15 °C for 7 days, before returning them to a natural setting. Seeds were collected from plants that produced the most seeds, and this next generation was subjected to the same cold stress. The researchers noticed that, by the third generation, one variety produced many seeds, despite the cold stress. The plants maintained this ability for the five generations that the researchers continued the experiment. Tang was surprised to see how quickly the plants developed this cold tolerance, a process that seemed to occur at a must faster rate than would have been expected from adaptation through natural selection. Although the finding took four years of growing and planting and many more years of confirming the stability of the changes, says Quadrana. “It’s quite impressive.” To rule out the possibility that this adaptation was due to a difference in the DNA sequence of the cold-tolerant plants, the researchers sequenced the genomes of these plants as well as of those that had not been chilled and had not developed the same tolerance. They found no genetic differences that could have contributed to the plants’ cold tolerance. The biggest challenge, says Tang, was convincing reviewers that there were no genetic changes driving this adaptation, which is why they had to conduct extensive genetic tests to rule that possibility out. The researchers then looked for ‘epigenetic’ markers, small molecules that can regulate gene activity, that differed between the two plant varieties. They found that, compared with the non-cold-tolerant plants, the cold-tolerant ones had fewer chemical tags added to the genome at the start of a specific gene, which they named Acquired Cold Tolerance 1 (ACT1). When they deactivated these chemical tags in the plants grown under normal conditions, they could tolerate cold stress better. And when they edited them back in, they no longer had this ability. “This is demonstrating causality directly. That’s a first,” says Quadrana. The researchers then looked for variations in the number of chemical tags on ACT1 in 131 rice varieties grown across China. Most of the crops found in the coldest, northern region had fewer chemical tags at the ACT1 gene than expected in rice varieties, and most of those from the warmer, coastal southern region had an abundance of these tags. This suggests that this specific feature was probably selected for and facilitated the plants’ migration northwards. Environmental driver Other examples of epigenetic inheritance have been found in animals. But most of the traits have been linked to disease and are not a beneficial adaptation to the environment, as is the case for cold-tolerant rice. For instance, offspring of pregnant rats exposed to fungicides are more susceptible to disease2; and male mice trained to fear a specific chemical scent give rise to offspring that are more sensitive to the same smell3. “Every year, we’ll see more and more papers like this that start to convince the scientific community” that epigenetic inheritance is real, says Skinner.