With luminosities exceeding entire galaxies (i.e., billions of times brighter than our Sun) for brief periods of time (months to years), accretion events enable astrophysicists to study supermassive black holes (SMBHs) from cosmological distances, providing a window into the central regions of otherwise-quiescent — or dormant — galaxies. By probing these “strong-gravity” events, where Einstein’s general theory of relativity is critical for determining how matter behaves, TDEs yield information about one of the most extreme environments in the universe: the event horizon — the point of no return — of a black hole.

TDEs are usually “once-and-done” because the extreme gravitational field of the SMBH destroys the star, meaning that the SMBH fades back into darkness following the accretion flare. In some instances, however, the high-density core of the star can survive the gravitational interaction with the SMBH, allowing it to orbit the black hole more than once. Researchers call this a repeating partial TDE.

A team of physicists, including lead author Thomas Wevers, Fellow of the European Southern Observatory, and co-authors Eric Coughlin, assistant professor of physics at Syracuse University, and Dheeraj R. “DJ” Pasham, research scientist at MIT’s Kavli Institute for Astrophysics and Space Research, have proposed a model for a repeating partial TDE. Their findings, published in Astrophysical Journal Letters, describe the capture of the star by a SMBH, the stripping of the material each time the star comes close to the black hole, and the delay between when the material is stripped and when it feeds the black hole again. The team’s work is the first to develop and use a detailed model of a repeating partial TDE to explain the observations, make predictions about the orbital properties of a star in a distant galaxy, and understand the partial tidal disruption process.

The team is studying a TDE known as AT2018fyk (AT stands for “Astrophysical Transient”). The star was captured by a SMBH through an exchange process known as “Hills capture,” where the star was originally part of a binary system (two stars that orbit one another under their mutual gravitational attraction) that was ripped apart by the gravitational field of the black hole. The other (non-captured) star was ejected from the center of the galaxy at speeds comparable to ~ 1000 km/s, which is known as a hypervelocity star.

Once bound to the SMBH, the star powering the emission from AT2018fyk has been repeatedly stripped of its outer envelope each time it passes through its point of closest approach with the black hole. The stripped outer layers of the star form the bright accretion disk, which researchers can study using X-Ray and Ultraviolet /Optical telescopes that observe light from distant galaxies.

According to Wevers, having the opportunity to study a partial TDE gives unprecedented insight into the existence of supermassive black holes and the orbital dynamics of stars in the centers of galaxies.

“Until now, the assumption has been that when we see the aftermath of a close encounter between a star and a supermassive black hole, the outcome will be fatal for the star, that is, the star is completely destroyed,” he says. “But contrary to all other TDEs we know of, when we pointed our telescopes to the same location again several years later, we found that it had re-brightened again. This led us to propose that rather than being fatal, part of the star survived the initial encounter and returned to the same location to be stripped of material once more, explaining the re-brightening phase.”

First detected in 2018, AT2018fyk was initially perceived as an ordinary TDE. For approximately 600 days the source stayed bright in the X-ray, but then abruptly went dark and was undetectable — a result of the stellar remnant core returning to a black hole, explains MIT physicist Dheeraj R. Pasham.

“When the core returns to the black hole it essentially steals all the gas away from the black hole via gravity and as a result there is no matter to accrete and hence the system goes dark,” Pasham says.

It wasn’t immediately clear what caused the precipitous decline in the luminosity of AT2018fyk, because TDEs normally decay smoothly and gradually — not abruptly — in their emission. But around 600 days after the drop, the source was again found to be X-ray bright. This led the researchers to propose that the star survived its close encounter with the SMBH the first time and was in orbit about the black hole.

Using detailed modeling, the team’s findings suggest that the orbital period of the star about the black hole is roughly 1,200 days, and it takes approximately 600 days for the material that is shed from the star to return to the black hole and start accreting. Their model also constrained the size of the captured star, which they believe was about the size of the sun. As for the original binary, the team believes the two stars were extremely close to one another before being ripped apart by the black hole, likely orbiting each other every few days.

So how could a star survive its brush with death? It all comes down to a matter of proximity and trajectory. If the star collided head-on with the black hole and passed the event horizon — the threshold where the speed needed to escape the black hole surpasses the speed of light — the star would be consumed by the black hole. If the star passed very close to the black hole and crossed the so-called “tidal radius” — where the tidal force of the hole is stronger than the gravitational force that keeps the star together — it would be destroyed. In the model they have proposed, the star’s orbit reaches a point of closest approach that is just outside of the tidal radius, but doesn’t cross it completely: some of the material at the stellar surface is stripped by the black hole, but the material at its center remains intact.

How, or if, the process of the star orbiting the SMBH can occur over many repeated passages is a theoretical question that the team plans to investigate with future simulations. Syracuse physicist Eric Coughlin explains that they estimate between 1 to 10% of the mass of the star is lost each time it passes the black hole, with the large range due to uncertainty in modeling the emission from the TDE.

“If the mass loss is only at the 1% level, then we expect the star to survive for many more encounters, whereas if it is closer to 10%, the star may have already been destroyed,” notes Coughlin.

The team will keep their eyes to the sky in the coming years to test their predictions. Based on their model, they forecast that the source will abruptly disappear around March 2023 and brighten again when the freshly stripped material accretes onto the black hole in 2025.

The team says their study offers a new way forward for tracking and monitoring follow-up sources that have been detected in the past. The work also suggests a new paradigm for the origin of repeating flares from the centers of external galaxies.

“In the future, it is likely that more systems will be checked for late-time flares, especially now that this project puts forth a theoretical picture of the capture of the star through a dynamical exchange process and the ensuing repeated partial tidal disruption,” says Coughlin. “We’re hopeful this model can be used to infer the properties of distant supermassive black holes and gain an understanding of their “demographics,” being the number of black holes within a given mass range, which is otherwise difficult to achieve directly.”

The team says the model also makes several testable predictions about the tidal disruption process, and with more observations of systems like AT2018fyk, it should give insight into the physics of partial tidal disruption events and the extreme environments around supermassive black holes.

“This study outlines methodology to potentially predict the next snack times of supermassive black holes in external galaxies,” says Pasham. “If you think about it, it is pretty remarkable that we on Earth can align our telescopes to black holes millions of light years away to understand how they feed and grow.”

Additional co-authors include: M. Guolo, Department of Physics and Astronomy, Johns Hopkins University; Y. Sun, University of Arizona; S. Wen, Department of Astrophysics/IMAPP, Radboud University ; P.G. Jonker, Department of Astrophysics/IMAPP, Radboud University and SRON, Netherlands Institute for Space Research ; A. Zabludoff, University of Arizona; A. Malyali, R. Arcodia, Z. Liu, A. Merloni, A. Rau and I. Grotova, Max-Planck-Institut fu ?r extraterrestrische Physik , Germany; P. Short, Institute for Astronomy, University of Edinburgh; and Z. Cao, Department of Astrophysics/IMAPP, Radboud University

Video: https://youtu.be/_TRtPDbaQ2k

In the study, researchers surveyed 644 elementary school children about how motivated they were to act in ways that would help the environment — such as by using a reusable water bottle at home or refusing to use plastic straws in restaurants — before and after participating in an environmental education program.

The program, developed by the Duke University Marine Lab, focuses on trash in oceans and other waterways, and includes lessons on how long different types of trash persist in waterways, a trash clean-up, and hands-on investigations of challenges related to marine debris. After the program, students scored higher on average on the survey gauging their motivation to act for the environment. Bilingual or multilingual students saw bigger gains on average compared to students who spoke English primarily at home — a finding researchers say is promising, and needs to be investigated further.

“What we saw was that in aggregate, the programs seemed to encourage environmentally friendly actions among everybody, but when we dug down, most of the program’s effect was explained by the response from linguistically diverse children,” said study co-author Kathryn Stevenson, associate professor of parks, recreation and tourism management at NC State. “This is encouraging, as linguistically diverse children are making up more and more of the U.S. population, and we want our programs to resonate with everyone. It also highlights how young people with different backgrounds can make important contributions. It also makes us wonder: Are students bringing these lessons home?”

The study is part of a research series looking at how environmental education can impact children, their families and their communities. In a previous study, researchers found parents’ climate change concern increases after their children are educated. In another study, they found that local leaders’ as well as voters’ views shifted after watching children’s presentations on an environmental issue.

“We’ve been interested in the mechanisms of inter-generational learning,” Stevenson said. “We saw this program can impact all children involved, but this suggests it might work differently for children who speak more than one language. For kids who act as translators for their family, they might be even more practiced at translating on many levels — linguistically or culturally — and we want to know how that might impact inter-generational learning about the environment.”

The newly discovered structures were formed by a process called self-assembly, in which a material’s molecules organize themselves into unique patterns. Scientists at Brookhaven’s Center for Functional Nanomaterials (CFN) are experts at directing the self-assembly process, creating templates for materials to form desirable arrangements for applications in microelectronics, catalysis, and more. Their discovery of the nanoscale ladder and other new structures further widens the scope of self-assembly’s applications.

“Self-assembly can be used as a technique for nanopatterning, which is a driver for advances in microelectronics and computer hardware,” said CFN scientist and co-author Gregory Doerk. “These technologies are always pushing for higher resolution using smaller nanopatterns. You can get really small and tightly controlled features from self-assembling materials, but they do not necessarily obey the kind of rules that we lay out for circuits, for example. By directing self-assembly using a template, we can form patterns that are more useful.”

Staff scientists at CFN, which is a DOE Office of Science User Facility, aim to build a library of self-assembled nanopattern types to broaden their applications. In previous studies, they demonstrated that new types of patterns are made possible by blending two self-assembling materials together.

“The fact that we can now create a ladder structure, which no one has ever dreamed of before, is amazing,” said CFN group leader and co-author Kevin Yager. “Traditional self-assembly can only form relatively simple structures like cylinders, sheets, and spheres. But by blending two materials together and using just the right chemical grating, we’ve found that entirely new structures are possible.”

Blending self-assembling materials together has enabled CFN scientists to uncover unique structures, but it has also created new challenges. With many more parameters to control in the self-assembly process, finding the right combination of parameters to create new and useful structures is a battle against time. To accelerate their research, CFN scientists leveraged a new AI capability: autonomous experimentation.

In collaboration with the Center for Advanced Mathematics for Energy Research Applications (CAMERA) at DOE’s Lawrence Berkeley National Laboratory, Brookhaven scientists at CFN and the National Synchrotron Light Source II (NSLS-II), another DOE Office of Science User Facility at Brookhaven Lab, have been developing an AI framework that can autonomously define and perform all the steps of an experiment. CAMERA’s gpCAM algorithm drives the framework’s autonomous decision-making. The latest research is the team’s first successful demonstration of the algorithm’s ability to discover new materials.

“gpCAM is a flexible algorithm and software for autonomous experimentation,” said Berkeley Lab scientist and co-author Marcus Noack. “It was used particularly ingeniously in this study to autonomously explore different features of the model.”

“With help from our colleagues at Berkeley Lab, we had this software and methodology ready to go, and now we’ve successfully used it to discover new materials,” Yager said. “We’ve now learned enough about autonomous science that we can take a materials problem and convert it into an autonomous problem pretty easily.”

To accelerate materials discovery using their new algorithm, the team first developed a complex sample with a spectrum of properties for analysis. Researchers fabricated the sample using the CFN nanofabrication facility and carried out the self-assembly in the CFN material synthesis facility.

“An old school way of doing material science is to synthesize a sample, measure it, learn from it, and then go back and make a different sample and keep iterating that process,” Yager said. “Instead, we made a sample that has a gradient of every parameter we’re interested in. That single sample is thus a vast collection of many distinct material structures.”

Then, the team brought the sample to NSLS-II, which generates ultrabright x-rays for studying the structure of materials. CFN operates three experimental stations in partnership with NSLS-II, one of which was used in this study, the Soft Matter Interfaces (SMI) beamline.

“One of the SMI beamline’s strengths is its ability to focus the x-ray beam on the sample down to microns,” said NSLS-II scientist and co-author Masa Fukuto. “By analyzing how these microbeam x-rays get scattered by the material, we learn about the material’s local structure at the illuminated spot. Measurements at many different spots can then reveal how the local structure varies across the gradient sample. In this work, we let the AI algorithm pick, on the fly, which spot to measure next to maximize the value of each measurement.”

As the sample was measured at the SMI beamline, the algorithm, without human intervention, created of model of the material’s numerous and diverse set of structures. The model updated itself with each subsequent x-ray measurement, making every measurement more insightful and accurate.

In a matter of hours, the algorithm had identified three key areas in the complex sample for the CFN researchers to study more closely. They used the CFN electron microscopy facility to image those key areas in exquisite detail, uncovering the rails and rungs of a nanoscale ladder, among other novel features.

From start to finish, the experiment ran about six hours. The researchers estimate they would have needed about a month to make this discovery using traditional methods.

“Autonomous methods can tremendously accelerate discovery,” Yager said. “It’s essentially ‘tightening’ the usual discovery loop of science, so that we cycle between hypotheses and measurements more quickly. Beyond just speed, however, autonomous methods increase the scope of what we can study, meaning we can tackle more challenging science problems.”

“Moving forward, we want to investigate the complex interplay among multiple parameters. We conducted simulations using the CFN computer cluster that verified our experimental results, but they also suggested how other parameters, such as film thickness, can also play an important role,” Doerk said.

The team is actively applying their autonomous research method to even more challenging material discovery problems in self-assembly, as well as other classes of materials. Autonomous discovery methods are adaptable and can be applied to nearly any research problem.

“We are now deploying these methods to the broad community of users who come to CFN and NSLS-II to conduct experiments,” Yager said. “Anyone can work with us to accelerate the exploration of their materials research. We foresee this empowering a host of new discoveries in the coming years, including in national priority areas like clean energy and microelectronics.”

This research was supported by the DOE Office of Science.

A multidisciplinary team of investigators at Vanderbilt University Medical Center has discovered that blocking an iron uptake receptor reduces disease pathology and promotes the activity of anti-inflammatory regulatory T cells in a mouse model of SLE. The findings were published Jan. 13 in the journal Science Immunology.

Lupus, including SLE, occurs when the immune system attacks a person’s own healthy tissues, causing pain, inflammation and tissue damage. Lupus most commonly affects skin, joints, brain, lungs, kidneys and blood vessels. About 1.5 million Americans and 5 million people worldwide have a form of lupus, according to the Lupus Foundation of America.

Treatments for lupus aim to control symptoms, reduce immune system attack of tissues, and protect organs from damage. Only one targeted biologic agent has been approved for treating SLE, belimumab in 2011.

“It has been a real challenge to come up with new therapies for lupus,” said Jeffrey Rathmell, PhD, professor of Pathology, Microbiology and Immunology and Cornelius Vanderbilt Chair in Immunobiology. “The patient population and the disease are heterogeneous, which makes it difficult to design and conduct clinical trials.”

Rathmell’s group has had a long-standing interest in lupus as part of a broader effort to understand mechanisms of autoimmunity.

When postdoctoral fellow Kelsey Voss, PhD, began studying T cell metabolism in lupus, she noticed that iron appeared to be a “common denominator in many of the problems in T cells,” she said. She was also intrigued by the finding that T cells from patients with lupus have high iron levels, even though patients are often anemic.

“It was not clear why the T cells were high in iron, or what that meant,” said Voss, first author of the Science Immunology paper.

To explore T cell iron metabolism in lupus, Voss and Rathmell drew on the expertise of other investigators at VUMC:

• Eric Skaar, PhD, and his team are experienced in the study of iron and other metals;

• Amy Major, PhD, and her group provided a mouse model of SLE; and

• Michelle Ormseth, MD, MSCI, and her team recruited patients with SLE to provide blood samples.

First, Voss used a CRISPR genome editing screen to evaluate iron-handling genes in T cells. She identified the transferrin receptor, which imports iron into cells, as critical for inflammatory T cells and inhibitory for anti-inflammatory regulatory T cells.

The researchers found that the transferrin receptor was more highly expressed on T cells from SLE-prone mice and T cells from patients with SLE, which caused the cells to accumulate too much iron.

“We see a lot of complications coming from that — the mitochondria don’t function properly, and other signaling pathways are altered,” Voss said.

An antibody that blocks the transferrin receptor reduced intracellular iron levels, inhibited inflammatory T cell activity, and enhanced regulatory T cell activity. Treatment of SLE-prone mice with the antibody reduced kidney and liver pathology and increased production of the anti-inflammatory factor, IL-10.

“It was really surprising and exciting to find different effects of the transferrin receptor in different types of T cells,” Voss said. “If you’re trying to target an autoimmune disease by affecting T cell function, you want to inhibit inflammatory T cells but not harm regulatory T cells. That’s exactly what targeting the transferrin receptor did.”

In T cells from patients with lupus, expression of the transferrin receptor correlated with disease severity, and blocking the receptor in vitro enhanced production of IL-10.

The researchers are interested in developing transferrin receptor antibodies that bind specifically to T cells, to avoid any potential off-target effects (the transferrin receptor mediates iron uptake in many cell types). They are also interested in studying the details of their unexpected discovery that blocking the transferrin receptor enhances regulatory T cell activity.

Skaar is the Ernest W. Goodpasture Professor of Pathology and director of the Vanderbilt Institute for Infection, Immunology, and Inflammation. Major, associate professor of Medicine, and Ormseth, assistant professor of Medicine, are faculty members in the Division of Rheumatology and Immunology. Rathmell is the director of the Vanderbilt Center for Immunobiology.

Other authors of the study include Allison Sewell, Evan Krystofiak, PhD, Katherine Gibson-Corley, DVM, PhD, Arissa Young, MD, Jacob Basham, MD, Ayaka Sugiura, PhD, Emily Arner, PhD, William Beavers, PhD, Dillon Kunkle, PhD, Megan Dickson, Gabriel Needle, and W. Kimryn Rathmell, MD, PhD.

The research was supported by the National Institutes of Health (grants DK105550, AI153167, DK101003, AI150701, CA253718) and the Lupus Research Alliance William Paul Distinguished Innovator Award to Jeffrey Rathmell.

“Scientific research on herd and flock behaviour is usually based on field observations. In such cases, it is usually difficult to reliably record the internal states of the herd animals,” Kroy said. As a result, the interpretation of observations frequently relies on plausible assumptions as to which individual behavioural rules are necessary for the complex collective groups under observation. Researchers at Leipzig University therefore developed an experimental model system of microswimmers that elicits properties of natural swarm intelligence and provides complete control over the individuals’ internal states, strategies, and transformation of signal perception into a navigational reaction.

Thanks to a sophisticated laser heating system (see image), the colloidal swimmers, which are visible only under the microscope, can actively self-propel in a water container by a kind of “thermophoretic self-propulsion” while their travel is permanently disturbed in a random manner by Brownian motion. “Apart from Brownian random motion, which is ubiquitous in microphysics, the experimental set-up provides complete control over the physical parameters and navigation rules of the individual colloidal swimmers and allows long-term observations of swarms of variable sizes,” Cichos said.

According to Cichos, when just a very simple and generic navigation rule is followed identically by all of the swimmers, a surprisingly complex swarm behaviour results. For example, if the swimmers are aiming at the same fixed point, instead of them gathering at the same place a kind of carousel can form. Similar to satellites or atomic electrons, the swimmers then orbit their attractive centre on circular paths of varying heights. The only “intelligent” behavioural rule required for this is that the self-propulsion responds to environmental perception with a certain time delay, which usually occurs in natural swarm phenomena from mosquito dances to road traffic anyway. It turns out that such a “delayed” effect alone is sufficient to form complex dynamic patterns such as the carousel described above. “Physically speaking, each individual swimmer can spontaneously break the radial symmetry of the system and go into circular motion if the product of the delayed time and swimming speed is large enough,” Kroy said. In contrast, the orbits of larger swarms and their synchronisation and stabilisation depend on additional details such as the steric, phoretic and hydrodynamic interactions between the individual swimmers.

Since all signal-response interactions in the living world occur in a time-delayed manner, these findings should also further the understanding of dynamic pattern formation in natural swarm ensembles. The researchers deliberately chose primitive and uniform navigation rules for their experiment. This allowed them to develop a stringent mathematical description of the observed phenomena. In the analysis of the delayed stochastic differential equations used for this purpose, the delay-induced effective synchronisation of the swimmers with their own past turned out to be the key mechanism for the spontaneous circular motion. To a large extent, the theory allows us to mathematically predict the experimental observations. “All in all, we have succeeded in creating a laboratory for swarms of Brownian microswimmers. This can serve as a building block for future systematic studies of increasingly complex and possibly still unknown swarm behaviour, and it may also explain why puppies often circle their food bowl when they are being fed,” Cichos said.

The study, published in Frontiers in Communication, builds on existing research on how the brain processes language.

Vicky Lai, a UArizona assistant professor of psychology and cognitive science, worked with collaborators in the Netherlands to explore how people’s brains react to language when they are in a happy mood versus a negative mood.

“Mood and language seem to be supported by different brain networks. But we have one brain, and the two are processed in the same brain, so there is a lot of interaction going on,” Lai said. “We show that when people are in a negative mood, they are more careful and analytical. They scrutinize what’s actually stated in a text, and they don’t just fall back on their default world knowledge.”

Lai and her study co-authors set out to manipulate study participants’ moods by showing them clips from a sad movie — “Sophie’s Choice” — or a funny television show — “Friends.” A computerized survey was used to evaluate participants’ moods before and after watching the clips. While the funny clips did not impact participants’ moods, the sad clips succeeded in putting participants in a more negative mood, the researchers found.

The participants then listened to a series of emotionally neutral audio recordings of four-sentence stories that each contained a “critical sentence” that either supported or violated default, or familiar, word knowledge. That sentence was displayed one word at a time on a computer screen, while participants’ brain waves were monitored by EEG, a test that measures brain waves.

For example, the researchers presented study participants with a story about driving at night that ended with the critical sentence “With the lights on, you can see more.” In a separate story about stargazing, the same critical sentence was altered to read “With the lights on, you can see less.” Although that statement is accurate in the context of stargazing, the idea that turning on the lights would cause a person to see less is a much less familiar concept that defies default knowledge.

The researchers also presented versions of the stories in which the critical sentences were swapped so that they did not fit the context of the story. For example, the story about driving at night would include the sentence “With the lights on, you can see less.”

They then looked at how the brain reacted to the inconsistencies, depending on mood.

They found that when participants were in a negative mood, based on their survey responses, they showed a type of brain activity closely associated with re-analysis.

“We show that mood matters, and perhaps when we do some tasks we should pay attention to our mood,” Lai said. “If we’re in a bad mood, maybe we should do things that are more detail-oriented, such as proofreading.”

Study participants completed the experiment twice — once in the negative mood condition and once in the happy mood condition. Each trial took place one week apart, with the same stories presented each time.

“These are the same stories, but in different moods, the brain sees them differently, with the sad mood being the more analytical mood,” Lai said.

The study was conducted in the Netherlands; participants were native Dutch speakers, and the study was conducted in Dutch. But Lai believes their findings translate across languages and cultures.

By design, the study participants were all women, because Lai and her colleagues wanted to align their study with existing literature that was limited to female participants. Lai said future studies should include more diverse gender representation.

In the meantime, Lai and her colleagues say mood may affect us in more ways than we previously realized.

Researcher Jos van Berkum of the Netherlands’ Utrecht University, co-authored the study with Lai and Peter Hagoort of the Max Planck Institute for Psycholinguistics in the Netherlands.

“When thinking about how mood affects them, many people just consider things like being grumpy, eating more ice cream, or — at best — interpreting somebody else’s talk in a biased way,” van Berkum said. “But there’s much more going on, also in unexpected corners of our minds. That’s really interesting. Imagine your laptop being more or less precise as a function of its battery level — that’s unthinkable. But in human information processing, and presumably also in (information processing) of related species, something like that seems to be going on.”

The advent of spatial transcriptomics technologies has allowed researchers to observe gene expression in their spatial context across entire tissue samples. But new computational methods are needed to make sense of this data and help identify and understand these gene expression patterns.

A research team led by Jian Ma, the Ray and Stephanie Lane Professor of Computational Biology in Carnegie Mellon University’s School of Computer Science, has developed a machine learning tool to fill this gap. Their paper on the method, called SPICEMIX, appeared as the cover story in the most recent issue of Nature Genetics.

SPICEMIX helps researchers untangle the role different spatial patterns play in the overall gene expression of cells in complex tissues like the brain. It does so by representing each pattern with spatial metagenes — groups of genes that may be connected to a specific biological process and can display smooth or sporadic patterns across tissue.

The team, which included Ma; Benjamin Chidester, a project scientist in the Computational Biology Department; and Ph.D. students Tianming Zhou and Shahul Alam, used SPICEMIX to analyze spatial transcriptomics data from brain regions in mice and humans. They leveraged the unique capabilities of SPICEMIX to uncover the landscape of the brain’s cell types and spatial patterns.

“We were inspired by cooking when we chose the name,” Chidester said. “You can make all sorts of different flavors with the same set of spices. Cells may work in a similar way. They may use a common set of biological processes, but the specific combination they use gives them their unique identity.”

When applied to brain tissues, SPICEMIX identified spatial patterns of cell types in the brain more accurately than other methods. It also uncovered new expression patterns of brain cell types through the learned spatial metagenes.

“These findings may help us paint a more complete picture of the complexity of brain cell types,” Zhou said.

The number of studies using spatial transcriptomics technologies is growing rapidly, and SPICEMIX can help researchers make the most of this high-volume, high-dimensional data.

“Our method has the potential to advance spatial transcriptomics research and contribute to a deeper understanding of both basic biology and disease progression in complex tissues,” Ma said.

“The advantage of a coal power plant is it’s very steady,” said Nian Liu, an assistant professor at the Georgia Institute of Technology. “If the power source fluctuates like it does with clean energy, it makes it more difficult to manage, so how can we use an energy storage device or system to smooth out these fluctuations?”

Flow batteries offer a solution. Electrolytes flow through electrochemical cells from storage tanks in this rechargeable battery. The existing flow battery technologies cost more than $200/kilowatt hour and are too expensive for practical application, but Liu’s lab in the School of Chemical and Biomolecular Engineering (ChBE) developed a more compact flow battery cell configuration that reduces the size of the cell by 75%, and correspondingly reduces the size and cost of the entire flow battery. The work could revolutionize how everything from major commercial buildings to residential homes are powered.

The all-Georgia Tech research team published their findings in the paper, “A Sub-Millimeter Bundled Microtubular Flow Battery Cell With Ultra-high Volumetric Power Density,” in Proceedings of the National Academy of Sciences.

Finding the Flow

Flow batteries get their name from the flow cell where electron exchange happens. Their conventional design, the planar cell, requires bulky flow distributors and gaskets, increasing size and cost but decreasing overall performance. The cell itself is also expensive. To reduce footprint and cost, the researchers focused on improving the flow cell’s volumetric power density (W/L-of-cell).

They turned to a configuration commonly used in chemical separation — sub-millimeter, bundled microtubular (SBMT) membrane — made of a fiber-shaped filter membrane known as a hollow fiber. This innovation has a space-saving design that can mitigate pressure across the membranes that ions pass through without needing additional support infrastructure.

“We were interested in the effect of the battery separator geometry on the performance of flow batteries,” said Ryan Lively, a professor in ChBE. “We were aware of the advantages that hollow fibers imparted on separation membranes and set out to realize those same advantages in the battery field.”

Applying this concept, the researchers developed an SMBT that reduces membrane-to-membrane distance by almost 100 times. The microtubular membrane in the design works as an electrolyte distributor at the same time without the need for large supporting materials. The bundled microtubes create a shorter distance between electrodes and membranes, increasing the volumetric power density. This bundling design is the key discovery for maximizing flow batteries’ potential.

Powering the Battery

To validate their new battery configuration, the researchers used four different chemistries: vanadium, zinc-bromide, quinone-bromide, and zinc-iodide. Although all chemistries are functional, two were most promising. Vanadium was the most mature chemistry, but also less accessible, and the reduced form of it is unstable in air. They found zinc iodide was the most energy-dense option, making it the most effective for residential units. Zinc-iodide offered many advantages even compared to lithium: It has less of a supply chain issue and also can be turned into zinc oxide and dissolve in acid, making it much easier to recycle.

This electrochemical solution for this unique shape of the flow battery proved more powerful than conventional planar cells.

“The superior performance of the SMBT was also demonstrated by finite element analysis,” said Xing Xie, an assistant professor in the School of Civil and Environmental Engineering. “This simulation method will also be applied in our future study for cell performance optimization and scaling up.”

With zinc-iodide chemistry, the battery could run for more than 220 hours, or to > 2,500 cycles at off-peak conditions. It could also potentially reduce the cost from $800 to less than $200 per kilowatt hour by using recycled electrolyte.

Building the Future of Energy

The researchers are already working on commercialization, focusing on developing batteries with different chemistries like vanadium and scaling up their size. Scaling will require coming up with an automated process to manufacture a hollow fiber module, which now is done manually, fiber by fiber. They eventually hope to deploy the battery in Georgia Tech’s 1.4-megawatt microgrid in Tech Square, a project that tests microgrid integration into the power grid and offers living laboratory for professors and students.

The SBMT cells could also be applied to different energy storage systems like electrolysis and fuel cells. The technology could even be strengthened with advanced materials and different chemistry in various applications.

“This innovation is very application driven,” Liu said. “We have the need to reach carbon neutrality by increasing the percentage of renewable energy in our energy generation, and right now, it’s less than 15% in the U.S. Our research could change this.”

Now, city planners and government health officials can obtain a more precise accounting about the well-being of urban dwellers and the air they breathe, from new research published December 2022 in the journal Transportation Research Part D.

“Infrastructure determines our living environment, our exposure,” said senior author Oliver Gao, the Howard Simpson Professor of Civil and Environmental Engineering in the College of Engineering at Cornell University. “Air pollution impact due to transportation — put out as exhaust from the cars and trucks that drive on our streets — is very complicated. Our infrastructure, transportation and energy policies are going to impact air pollution and hence public health.”

Previous methods to gauge air pollution were cumbersome and reliant on extraordinary amounts of data points. “Older models to calculate particulate matter were computationally and mechanically consuming and complex,” said Gao, a faculty fellow at the Cornell Atkinson Center for Sustainability. “But if you develop an easily accessible data model, with the help of artificial intelligence filling in some of the blanks, you can have an accurate model at a local scale.”

Lead author Salil Desai and visiting scientist Mohammad Tayarani, together with Gao, published “Developing Machine Learning Models for Hyperlocal Traffic Related Particulate Matter Concentration Mapping,” to offer a leaner, less data-intensive method for making accurate models.

Ambient air pollution is a leading cause of premature death around the world. Globally, more than 4.2 million annual fatalities — in the form of cardiovascular disease, ischemic heart disease, stroke and lung cancer — were attributed to air pollution in 2015, according to a Lancet study cited in the Cornell research.

In this work, the group developed four machine learning models for traffic-related particulate matter concentrations in data gathered in New York City’s five boroughs, which have a combined population of 8.2 million people and a daily-vehicle miles traveled of 55 million miles.

The equations use few inputs such as traffic data, topology and meteorology in an AI algorithm to learn simulations for a wide range of traffic-related, air-pollution concentration scenarios.

Their best performing model was the Convolutional Long Short-term Memory, or ConvLSTM, which trained the algorithm to predict many spatially correlated observations.

“Our data-driven approach — mainly based on vehicle emission data — requires considerably fewer modeling steps,” Desai said. Instead of focusing on stationary locations, the method provides a high-resolution estimation of the city street pollution surface. Higher resolution can help transportation and epidemiology studies assess health, environmental justice and air quality impacts.

Funding for this research came from the U.S. Department of Transportation’s University Transportation Centers Program and Cornell Atkinson.