TL;DR: A September research roundup published by Ars Technica on Oct. 2, 2026 collected six studies that received little attention, including a possible dark-matter signal in South Dakota, entanglement between Z bosons at the LHC, a yeast-based Mars building material, and evidence for betel-nut sucking in Sulawesi.

A month of missed science, in one package

According to Ars Technica, there is never enough time to cover every interesting study, so its staff publishes a monthly selection of work that nearly slipped through the cracks. The September installment, written by Jennifer Ouellette and posted Oct. 2, 2026, brought together particle physics, off-world construction, archaeology of psychoactive-plant use, and history of the telescope.

The through-line is measurement under difficult conditions: a single unusual event in a quiet underground detector, short-lived particles tracked inside ATLAS, a printing process that uses cold and near-vacuum as a curing step, and worn teeth that preserved a behavior seeds alone could not. According to Ars Technica, the September list includes a tantalizing hint of dark matter at the LUX-ZEPLIN detector, entangled Z bosons at the Large Hadron Collider, possible use of psychoactive substances in the Late Pleistocene, and an intriguing technical description of a telescope attributed to Galileo.

A single event deep underground

What LUX-ZEPLIN reported

Physicists believe dark matter makes up roughly 85 percent of matter in the universe, with weakly interacting massive particles, called WIMPs, described as the strongest but not only candidate, according to Ars Technica. Direct detection of WIMPs or other hypothetical dark-matter particles has eluded physicists for decades.

The possible glimpse came from the LUX-ZEPLIN detector buried in a gold mine in South Dakota. According to Ars Technica, the results were presented in a scientific talk at the 2026 TeV Particle Astrophysics conference in Japan, with a preprint available on the arXiv and submitted to Physical Review Letters for peer review.

Reporting by WIRED.jp identified the presentation as work of the LUX-ZEPLIN collaboration, an international project involving 250 scientists and engineers from 39 institutions, operated by Lawrence Berkeley National Laboratory at the Sanford Underground Research Facility about 1.6 km underground in South Dakota. That report described a detector holding 10 tons of liquified high-purity xenon to look for flashes characteristic of a WIMP striking an atomic nucleus.

Why the team aired a caveat-heavy result

According to Ars Technica, the announcement came with strong caveats: the signal is only a hint, well below the threshold needed to confirm discovery, and might be a statistical fluke. Past anomalies faded as more data arrived, but the LUX-ZEPLIN team could not account for this event and decided to report it to the broader physics community.

WIRED.jp quoted Brown University professor and LZ spokesperson Rick Gaitskell saying the collaboration did not claim to have observed dark matter, describing the item as one event and saying the team saw something interesting worth sharing to seek views from the science community. The same report quoted Berkeley physicist and LZ institutional board chair Aaron Manalaysay saying outlier events are not rare, that most turn out to be background on closer inspection, and that the team continued to worry about an overlooked rare background mechanism while finding the possibility of an early dark-matter sign exciting.

WIRED.jp reported that the analysis covered 220 days of data collected from March 2023 to April 2024 and extended the search to a broader class of interactions that could deposit larger energy. One particle interaction appeared in a region where dark matter was expected and competing background was extremely low.

The statistical indicator reported by WIRED.jp was 2.6 sigma, meaning about a 0.5% chance such a prominent event would occur by accident if only known background were present, still far from the 5-sigma standard used in physics for a discovery. Researchers checked timing just after calibration, unusual edge-related false signals, and recoils from neutrons and atmospheric neutrinos, and none explained the event, according to WIRED.jp.

An energy that did not fit the simple picture

According to Ars Technica, if the case proves to be a WIMP striking a xenon nucleus, it is not what physicists would predict. The event showed higher nuclear-recoil energy than expected, which implied the detector should also have recorded several lower-energy WIMPs; it did not.

That mismatch set theorists to alternative models. According to Ars Technica, possibilities raised include an interaction rate that rises with collision energy, WIMPs with internal structure like an atom that interact only when hit hard enough, or coupling to ordinary matter that depends on momentum.

Independent theory papers posted on arXiv after the talk illustrate that scramble. One analysis described the LUX-ZEPLIN candidate as a nuclear recoil at 248 keV, with statistical and systematic uncertainties of 23 keV each, observed during 220 days between March 27, 2023 and April 1, 2024. That measured energy, far above 200 keV, made simple elastic WIMP scattering unlikely and pushed attention toward inelastic or momentum-dependent pictures, including hyperfine excitation of composite dark matter, effects of the Large Magellanic Cloud on the high-speed tail, and revised halo models informed by Gaia.

What comes next for direct detection

According to Ars Technica, the LUX-ZEPLIN team is still collecting data, as is XENONnT at the Gran Sasso National Laboratory in Italy, while China's PandaX detector is being built. WIRED.jp reported that LZ already holds one of the world's largest dark-matter search datasets and plans continued operation, so accumulation of data may show whether the hint gains significance or fades as fluctuation.

Entanglement at collider energy

From top quarks to Z bosons

Cornell University physicist N. David Mermin once described quantum entanglement as the closest thing to magic, because disturbance in one part of the universe can instantly affect a distant part, in a way that appears to bypass the cosmic speed-of-light limit, according to Ars Technica. Albert Einstein called it spooky action at a distance. Today entanglement underpins quantum computing and next-generation sensors, among other applications.

Maintaining entanglement is usually delicate because interference can cause rapid decoherence. Despite that, physicists found evidence in 2023 of entanglement between pairs of top quarks produced in the Large Hadron Collider's ATLAS experiment, raising the possibility that large colliders could serve as tools to study the effect, according to Ars Technica. ATLAS physicists have now observed entanglement in pairs of Z bosons, according to a paper published in Physical Review Letters.

An ATLAS briefing dated June 5, 2026 described the same line of work as using the Higgs boson as a testbed for complex quantum phenomena at high energy, delivering precise Higgs-boson property measurements and providing the first strong evidence of quantum entanglement between two massive vector bosons.

How ATLAS read spins it cannot see directly

Z bosons are produced when a Higgs boson decays and are extremely short-lived, decaying into pairs of electrons and muons. The ATLAS detector tracks those decay patterns precisely. Physicists used that data to reconstruct emission angles of electrons and muons to infer spins of the original Z bosons, according to Ars Technica. The authors claimed the result as the highest-energy example of entanglement observed to date.

Coverage in Techzle on Oct. 1, 2026 described the finding as the first demonstration of entanglement in massive, short-lived Z bosons using ATLAS at the LHC, with entangled particles created even as Higgs bosons decay in the accelerator. That account said the result sheds new light on how widespread the quantum phenomenon is.

A Japanese-language analysis in Fastwards, dated Sept. 27, 2026, said the paper appeared in Physical Review Letters on Sept. 11, 2026, reported rejection of a separable non-entangled state at 4.7 sigma, and treated the Z boson as a spin-1 qutrit with polarization states +1, 0 and -1. That analysis noted the 5-sigma benchmark used for discovery in particle physics was not reached, but described the measurement as the first for a particle pair with qutrit structure. It also noted the Z boson lifetime near 10^-25 seconds, meaning the work is a test of quantum mechanics at high energy rather than a computing component.

A living brick that likes Martian weather

Sand, gelatin and sticky yeast

Dreams of colonizing Mars face obstacles including housing, because transporting anything from Earth to Mars is costly and building materials are especially so, according to Ars Technica. Hong Kong engineers devised a recipe combining Martian rocks with terrestrial gelatin and yeast, which hardens into a durable living building material on par with cheap concrete, according to a paper published in Chem Circularity.

Co-author Jishen Qiu of The Hong Kong University of Science and Technology drew inspiration from hardening of freeze-dried fruits, noting the Martian environment has extremely low temperatures and pressures akin to freeze-drying, according to Ars Technica. The team tested mixtures of sand and glue made from gelatin and yeast, using a special yeast coated with sticky proteins similar to those mussels use to adhere to rocky surfaces. Gelatin provided scaffolding for cells to grow, and sand provided structure and mass. The mixture was tested in simulated Martian conditions by 3D-printing small domes the size of wine corks.

Reporting in Earth.com identified Qiu as a civil engineer at the university who led the work and quoted his inspiration from freeze-dried fruits that become harder. The Martian surface does much the same to anything wet, pulling water out as vapor. The account said the recipe is sand, gelatin and yeast engineered to grow its own glue, with cold and near-vacuum doing work a furnace would otherwise do.

Printing at minus 30 degrees

According to Ars Technica, the mixture freeze-dries after exiting the nozzle, leaving a light, porous foam with compressive strength of 10 to 12 megapascals. The authors said that is strong enough for a one- or two-story structure on Earth, where gravity is three times stronger than on Mars.

Earth.com reported that the team stirred binder into fine sand as a stand-in for building sand on Mars and cast sugar-cube-sized blocks. The blocks froze, then spent two days at minus 55 degrees Celsius under about one ten-thousandth of Earth's air pressure. At that pressure ice goes directly to vapor without melting, leaving open pores, so the finished block has more in common with stiff foam than poured slab. Gelatin sets into a scaffold around sand grains as water leaves, with yeast cells caught inside its walls.

Because gelatin firms as it cools, the extrusion window is narrow. Earth.com reported the team placed that window between 25 and 40 degrees Celsius and extruded at body temperature, 37 degrees Celsius. Inside a chamber at minus 30 degrees Celsius and roughly 1 percent of Earth's air pressure, they printed a test structure 45 millimeters tall. Milder conditions performed worse: at 20 degrees Celsius and 5 percent of Earth's pressure, water in the fresh mix boiled on exiting the nozzle.

A summary in Good News Brief, citing New Atlas, gave the same test point of minus 30 degrees Celsius and 0.01 atmosphere, with printed samples 45 mm tall and 30 mm wide. A Portuguese-language report in Antihype likewise described printing at minus 30 degrees Celsius and 0.01 atmosphere, with the warm mass leaving the nozzle at 37 degrees Celsius and freezing in seconds.

Strength, bending and recycling

Crushed in a press, cured blocks failed at about 12 megapascals, with at least six samples tested for every mix, according to Earth.com. Bending reversed the comparison with lightweight concrete of the same density: blocks reached roughly 6 megapascals in three-point bending, up to twice what the matching concrete manages, and absorbed about as much energy before cracking as a steel-fiber-reinforced version. That bending behavior matters where surface winds drive sand into structures, according to the same account.

Qiu said the strength would support one- or two-story buildings on Earth and probably multistory construction in lower Martian gravity, according to Earth.com and Good News Brief. The Antihype report quoted Qiu saying the material is strong enough for a one- or two-story building on Earth, whose gravity is three times that of Mars, so it could probably raise a multistory building on Mars.

Most plans call for heating loose rock past 1,000 degrees Celsius until grains fuse. Qiu's team estimated its own route at 0.02 to 0.12 gigajoules per cubic meter of finished material, tens of times lower, according to Earth.com and Antihype. That estimate covers processing alone, not yeast cultivation, pumps or water recovery. The same reporting said the route would use about 20 kilograms of ingredients flown from Earth per cubic meter, and that a solar array that needs days to fuse a cubic meter by heat could produce that much of the living material in under an hour.

Broken pieces can return to the loop. The team crushed cured blocks, added water and warmed the slurry to 45 degrees Celsius, dissolving gelatin into a paste like the original mix, according to Earth.com. Recast and freeze-dried, it became the next batch. Four generations on, blocks still tested at 11.8 megapascals, with yeast inside still alive.

Biology had been tried on loose planetary grit before. A 2020 experiment used chitosan from shellfish shells, and a 2022 test used bacteria to cement grains with calcium carbonate, according to Earth.com. Hardening any of it in Martian cold and near-vacuum had gone largely unverified before the Hong Kong group measured that step.

The work appeared Sept. 10 in Chem Circularity, with Ning Liu heading the author list and co-authors from The Hong Kong Polytechnic University, according to Antihype. All of it happened on Earth in a chamber with inert sand and cork-sized pieces; radiation, real Martian soil and air sealing remain untested.

Chewing, sucking and 25,000-year-old teeth

Grooves that did not match known wear

According to Ars Technica, the roundup also covered possible use of psychoactive substances in the Late Pleistocene. The underlying study, published in Science Advances and led by Griffith University, examined skeletal remains of two adult hunter-foragers excavated in South Sulawesi, Indonesia.

A Daily Star account by the study authors said one individual dated to between 25,000 and 16,000 years ago and the other to between 7,600 and 6,300 years ago. Although they lived at least 10,000 years apart, both showed deep, rounded grooves in some teeth. The grooves were unlike previously documented prehistoric dental wear and pointed to an unusual activity repeated over a long period.

The team concluded the foragers had regularly held a hard, roughly spherical object between teeth while sucking for extended periods, with abrasion wearing deep grooves. In one individual wear was so extreme that the pulp chamber, the living core of the tooth, became exposed in several places.

The grooves recalled clay-pipe facets, rounded notches worn into teeth of habitual smokers who clenched abrasive clay-pipe stems, according to the same account. That parallel led the team to consider betel nut. The hard, rounded seeds of the Areca catechu palm generally matched groove size and shape, and habitual use could explain persistence in a damaging practice.

A simpler habit than modern quid

Modern users mix an areca seed with slaked lime, typically made by pulverizing burnt shells, limestone or coral, to form a quid chewed for psychoactive effects, according to the Daily Star account. Slaked lime triggers a chemical reaction that lets psychoactive compounds absorb more rapidly, turns saliva red, and often permanently stains teeth of long-term users.

But chewing betel nut does not produce the distinctive grooves, and the Sulawesi teeth were unstained. The team reasoned that before modern quid chewing developed, foragers simply sucked intact nuts.

To test whether sucking alone could matter, researchers soaked betel nuts in artificial saliva and tested extracts on frog egg cells engineered to produce human brain receptors. Prolonged exposure released arecoline, the principal psychoactive alkaloid in the seeds, so sucking alone was sufficient to liberate brain-altering compounds. Biomolecular analysis also found arecoline in the ancient foragers' teeth, confirming ingestion, according to the same account.

Techno-Science, reporting Sept. 28, 2026, gave the same dates for the two individuals and the same interpretation: regular holding and sucking of whole areca nuts, with arecoline detected in dental tissues and laboratory release demonstrated in artificial saliva at levels sufficient to act on proteins involved in nervous activity.

Why a painful habit may have persisted

Betel nut is an addictive stimulant that induces mild euphoria and heightened alertness, among other effects, with high rates of mouth cancer among habitual users, according to the Daily Star account. After alcohol, caffeine and nicotine, it is the world's fourth most widely consumed psychoactive substance. Written accounts in India and China go back at least 2,000 years, while the earliest prior archaeological evidence dated to around 3,500 years ago in chemical compounds on teeth of a person in Thailand.

The new study suggests use originated much earlier, perhaps 25,000 years ago, and that modern chewing evolved from an older, simpler form that also carried health consequences. One reason may have been pain relief: both individuals had chronic dental problems, and arecoline has pain-killing properties. The account described a cruel irony in which sucking hard, abrasive seeds wore teeth until pulp was exposed, which would have been painful, hampered eating and raised chronic infection risk, while pain-dulling properties may have compelled continuation.

A Galileo-attributed telescope description

According to Ars Technica, the September selection closed with an intriguing technical description of a telescope attributed to Galileo. The roundup framed it as a historical strand alongside the laboratory and field studies, a reminder that instrument history can also hide in plain sight until a manuscript detail gets fresh attention.

That entry did not carry the same kind of numerical result as the physics and engineering papers. Its interest lies in attribution and technique: how a lens arrangement was described, who was credited with it, and what that implies for the early telescope story that runs from Dutch spyglasses to Galileo's astronomical use.

Why this kind of roundup matters for U.S. readers

U.S. facilities and collaborations sit at the center of two of the biggest items. The dark-matter hint came from a detector in South Dakota operated by a U.S. national laboratory, with analysis aimed at Physical Review Letters. The entanglement result came from ATLAS at CERN, a Large Hadron Collider experiment with broad U.S. participation, also reported in Physical Review Letters.

The Mars-material study speaks to in-situ resource thinking that NASA and U.S. universities follow closely: use local mass, bring only light ingredients, cure without a furnace, and recycle. The betel-nut study speaks to methods U.S. labs use every day, from dental microwear to receptor assays, applied to a 25,000-year question. The Galileo manuscript note speaks to archives and instrument collections that U.S. historians consult.

None of the findings arrives as a closed case. The dark-matter signal is a hint that could vanish. The Z-boson entanglement claim sits below the strictest discovery line. The Mars domes are wine-cork scale. The ancient-drug case rests on two individuals. The telescope description turns on attribution. Taken together, they show how science often advances in small, checkable increments before any consensus forms.

Sources referenced