- Press Release Distribution

Engineers apply physics-informed machine learning to solar cell production

Lehigh University researchers use Frontera supercomputer to simulate photovoltaic fabrication, train AI to optimize energy production

Engineers apply physics-informed machine learning to solar cell production
( Today, solar energy provides 2% of U.S. power. However, by 2050, renewables are predicted to be the most used energy source (surpassing petroleum and other liquids, natural gas, and coal) and solar will overtake wind as the leading source of renewable power. To reach that point, and to make solar power more affordable, solar technologies still require a number of breakthroughs. One is the ability to more efficiently transform photons of light from the Sun into useable energy.

Organic photovoltaics max out at 15% to 20% efficiency -- substantial, but a limit on solar energy's potential. Lehigh University engineer Ganesh Balasubramanian, like many others, wondered if there were ways to improve the design of solar cells to make them more efficient?

Balasubramanian, an associate professor of Mechanical Engineering and Mechanics, studies the basic physics of the materials at the heart of solar energy conversion -- the organic polymers passing electrons from molecule to molecule so they can be stored and harnessed -- as well as the manufacturing processes that produce commercial solar cells.

Architecture of the OPV bulk-heterojunction structure and the design scope. [Credit: Ganesh Balasubramanian, Joydeep Munshi, Lehigh University]

Using the Frontera supercomputer at the Texas Advanced Computing Center (TACC) -- one of the most powerful on the planet -- Balasubramanian and his graduate student Joydeep Munshi have been running molecular models of organic solar cell production processes, and designing a framework to determine the optimal engineering choices. They described the computational effort and associated findings in the May issue of IEEE Computing in Science and Engineering.

"When engineers make solar cells, they mix two organic molecules in a solvent and evaporate the solvent to create a mixture which helps with the exciton conversion and electron transport," Balasubramanian said. "We mimicked how these cells are created, in particular the bulk heterojunction -- the absorption layer of a solar cell. Basically, we're trying to understand how structure changes correlate with the efficiency of the solar conversion?"

Balasubramanian uses what he calls 'physics-informed machine learning'. His research combines coarse-grained simulation -- using approximate molecular models that represent the organic materials -- and machine learning. Balasubramanian believes the combination helps prevent artificial intelligence from coming up with unrealistic solutions.

"A lot of research uses machine learning on raw data," Balasubramanian said. "But more and more, there's an interest in using physics-educated machine learning. That's where I think lies the most benefit. Machine learning per se is simply mathematics. There's not a lot of real physics involved in it."

Writing in Computational Materials Science in February 2021, Balasubramanian and Munshi along with Wei Chen (Northwestern University), and TeYu Chien (University of Wyoming) described results from a set of virtual experiments on Frontera testing the effects of various design changes. These included altering the proportion of donor and receptor molecules in the bulk heterojunctions, and the temperature and amount of time spent in annealing -- a cooling and hardening process that contributes to the stability of the product.

They harnessed the data to train a class of machine learning algorithms known as support vector machines to identify parameters in the materials and production process that would generate the most energy conversion efficiency, while maintaining structural strength and stability. Coupling these methods together, Balasubramanian's team was able to reduce the time required to reach an optimal process by 40%.

"At the end of the day, molecular dynamics is the physical engine. That's what captures the fundamental physics," he said. "Machine learning looks at numbers and patterns, and evolutionary algorithms facilitate the simulations."

Trade-Offs and Limitations

Like many industrial processes, there are trade-offs involved in tweaking any facet of the manufacturing process. Faster cooling may help increase power efficiency, but it may make the material brittle and prone-to-break, for instance. Balasubramanian and his team employed a multi-objective optimization algorithm that balances the benefits and drawbacks of each change to derive the overall optimal manufacturing process.

Flowchart describing steps in a typical coupled Cuckoo Search-Coarse Grained Molecular Dynamics (CS-CGMD) algorithm. The dashed box represents the augmented machine learned exploration of the regions of interest to supplement ill-performed nests with newer alternatives during each CS optimization generation. [Credit: Ganesh Balasubramanian, Joydeep Munshi, Lehigh University]

"When you try to optimize one particular variable, you are looking at the problem linearly," he said. "But most of these efforts have multi-pronged challenges that you're trying to solve simultaneously. There are trade-offs that you need to make, and synergistic roles that you must capture, to come to the right design."

Balasubramanian's simulations matched experimental results. They determined that the make-up of the heterojunction and the annealing temperature/timing have the largest effects on overall efficiency. They also found what proportion of the materials in the heterojunction is best for efficiency.

"There are certain conditions identified in literature which people claim are the best conditions for efficiency for those select molecules and processing behavior," he said. "Our simulation were able to validate those and show that other possible criteria would not give you the same performance. We were able to realize the truth, but from the virtual world."

With an award of more time on Frontera in 2021-22, Balasubramanian will add further layers to the machine learning system to make it more robust. He plans to add experimental data, as well as other modalities of computer models, such as electronic structure calculations.

"Heterogeneity in the data will improve the results," he said. "We plan to do first principle simulations of materials and then feed that data into the machine learning model, as well as data from coarse-grained simulations."

Balasubramanian believes that current organic photovoltaics may be reaching the limits of their efficiency. "There's a wall that's hard to penetrate and that's the material," he said. "These molecules we've used can only go so far. The next thing to try is to use our framework with other molecules and advanced materials."

His team mined the literature to understand the features that increase solar efficiency and then trained a machine learning model to identify potential new molecules with ideal charge transport behaviors. They published their research in the Journal of Chemical Information and Modeling. Future work on Frontera will use Balasubramanian's framework to explore and computationally test these alternative materials, assuming they can be produced.

"Once established, we can take realistic molecules that are made in the lab and put them in the framework we've created," he said. "If we discover new materials that perform well, it will reduce the cost of solar power generation devices and help Mother Earth."

Balasubramanian's research harnesses the two things that computer simulations are critical for, he says. "One is to understand the science that we cannot study with the tools that we have in the real world. And the other is to expedite the science - streamline what we really have to do, which reduces our cost and time to make things and physically test them."


[Attachments] See images for this press release:
Engineers apply physics-informed machine learning to solar cell production


Understanding gut inflammation may hold clues to mitigating Parkinson's onset

GRAND RAPIDS, Mich. (JUNE 8, 2021) -- Chronic inflammation in the gut may propel processes in the body that give rise to Parkinson's disease, according to a study by scientists at Van Andel Institute and Roche. The study, published in Free Neuropathology, is the latest in a growing list that links the gut and the immune system to Parkinson's. The researchers' findings in an experimental mouse model of gut inflammation track with several large-scale epidemiological studies that show an association between Parkinson's and inflammatory bowel diseases, such as ulcerative colitis and Crohn's disease. Epidemiological evidence from other groups indicates the risk of developing Parkinson's fades in certain people whose ...

Bacteria-sized robots take on microplastics and win by breaking them down

Bacteria-sized robots take on microplastics and win by breaking them down
Small pieces of plastic are everywhere, stretching from urban environments to pristine wilderness. Left to their own devices, it can take hundreds of years for them to degrade completely. Catalysts activated by sunlight could speed up the process, but getting these compounds to interact with microplastics is difficult. In a proof-of-concept study, researchers reporting in ACS Applied Materials & Interfaces developed self-propelled microrobots that can swim, attach to plastics and break them down. While plastic products are omnipresent indoors, plastic waste and broken bits now litter the outdoors, too. The smallest of these ...

Social media use one of four factors related to higher COVID-19 spread rates early on

TORONTO, June 9, 2021 - Researchers from York University and the University of British Columbia have found social media use to be one of the factors related to the spread of COVID-19 within dozens of countries during the early stages of the pandemic. The researchers say this finding resembles other examples of social media misinformation ranging from the initial phase of vaccine rollout to the 2021 Capitol riot in the United States. Countries with high social media use leading to off-line political action prior to the pandemic, as surveyed before the pandemic by V-Dem (a database from the University of Gothenburg), showed ...

Researchers tame silicon to interact with light for next-generation microelectronics

Researchers tame silicon to interact with light for next-generation microelectronics
Skoltech researchers and their colleagues from RAS Institute for Physics of Microstructures, Lobachevsky State University of Nizhny Novgorod, ITMO University, Lomonosov Moscow State University, and A.M. Prokhorov General Physics Institute have found a way to increase photoluminescence in silicon, the notoriously poor emitter and absorber of photons at the heart of all modern electronics. This discovery may pave the way to photonic integrated circuits, boosting their performance. The paper was published in the journal Laser and Photonics Reviews. "Natural selection" in semiconductor technology ...

For early amphibians, a new lifestyle meant a new spine

Vertebrate life began in the water, but around 340-360 million years ago, four-limbed creatures, or tetrapods, made the transition onto land. In the years that followed, some species adapted to terrestrial life, while others turned back to the water and readapted to an aquatic lifestyle. A new study of these early amphibians, published in the journal PLOS ONE and led by Penn paleontologist Aja Carter, suggests that these environmental shifts left an impression--on the shape of the animals' spines. "I'm interested in how the shapes of the vertebrae affect how animals move," she says. "Our findings suggest that, in at least one part of the vertebrae, the shape of the bones ...

Forget wearables: Future washable smart clothes powered by Wi-Fi will monitor your health

Forget wearables: Future washable smart clothes powered by Wi-Fi will monitor your health
WEST LAFAYETTE, Ind. - Purdue University engineers have developed a method to transform existing cloth items into battery-free wearables resistant to laundry. These smart clothes are powered wirelessly through a flexible, silk-based coil sewn on the textile. In the near future, all your clothes will become smart. These smart cloths will outperform conventional passive garments, thanks to their miniaturized electronic circuits and sensors, which will allow you to seamlessly communicate with your phone, computer, car and other machines. This smart clothing will not only make you more productive but also check on your health status and even call for help if you suffer an accident. The reason why this smart clothing ...

Nearly 1 in 5 patients who die from unexplained sudden cardiac death have suspicious gene

As many as 450,000 Americans die every year from a sudden, fatal heart condition, and in slightly more than one in ten cases the cause remains unexplained even after an autopsy. Researchers from the University of Maryland School of Medicine (UMSOM) and their colleagues found that nearly 20 percent of patients with unexplained sudden cardiac death - most of whom were under age 50 - carried rare genetic variants. These variants likely raised their risk of sudden cardiac death. In some cases, their deaths may have been prevented if their doctors had known about their genetic predisposition to heart disease. The study findings were published ...

Scientists identify distinctive deep infrasound rumbles of space launches

WASHINGTON--After their initial blast, space rockets shoot away from the Earth with rumbles in infrasound, soundwaves too low to be heard by human ears that can travel thousands of miles. New research used a system for monitoring nuclear tests to track the infrasound from 1,001 rocket launches. The research identified the distinctive sounds from seven different types of rockets, including the Space Shuttles, Falcon 9 rockets, various Soyuz rockets, the European Space Agency's Ariane 5, Russian Protons and several types of Chinese Long March rockets. In some cases, like the Space Shuttle and the Falcon 9, the researchers were also able to identify the various stages of the rockets' journey. The ...

Smokers needed angioplasty and stenting a decade before non-smokers

Smokers needed their blocked arteries fixed nearly a decade earlier than non-smokers, and patients with obesity underwent these procedures four years earlier than non-obese patients, according to a new statewide study. The research included patients without a history of heart attack who were treated at hospitals across Michigan participating in BMC2, the Blue Cross Blue Shield of Michigan Cardiovascular Consortium. The patients had undergone angioplasty and/or stenting to widen or unblock their coronary arteries and restore blood flow. Almost all of them had at least one traditional ...

New research a 'step change' for diabetes patients

New research a step change for diabetes patients
Millions of people with diabetes are at risk of developing foot ulcers, which often lead to amputations and other health complications. Now, Scientists from the Centre for Biomechanics and Rehabilitation Technologies (CRBT) have developed a new method to reliably detect this risk without the need for complex electronic in-shoe sensors. Dr Panagiotis Chatzistergos, Associate Professor in Orthopaedic Biomechanics, explained: "In the UK alone, 169 people have a toe, foot or limb amputated as a result of diabetes every week, yet importantly up to 80% of these amputations could have been prevented with correct management. "Routine overloading ...


Cat-borne parasite Toxoplasma induces fatally bold behavior in hyena cubs

Nature article: Dieting and its effect on the gut microbiome

NIH scientists describe "Multi-Kingdom Dialogue" between internal, external microbiota

Melatonin in mice: there's more to this hormone than sleep

Wild bees need deadwood in the forest

A triple-system neural model of maladaptive consumption

Milk protein could help boost blueberries' healthfulness

Seeking a treatment for IBS pain in tarantula venom

Addressing inequity in air quality

Roughness of retinal layers, a new Alzheimer's biomarker

Study links sleep apnea in children to increased risk of high blood pressure in teen years

Black patients with cirrhosis more likely to die, less likely to get liver transplant

Researchers outline specific patterns in reading in Russian

These sea anemones have a diverse diet. And they eat ants

Viruses as communication molecules

Spirituality can promote the health of breast cancer survivors

Tiny ancient bird from China shares skull features with Tyrannosaurus rex

University of Minnesota Medical School report details the effects of COVID-19 on adolescent sexual health

Odd smell: flies sniff ammonia in a way new to science

Fracture setting method could replace metal plates, with fewer complications

Sneeze cam reveals best fabric combos for cloth masks (video)

'Lady luck' - Does anthropomorphized luck drive risky financial behavior?

People willing to pay more for coffee that's ethical and eco-friendly, meta-analysis finds

Low-cost imaging technique shows how smartphone batteries could charge in minutes

Pleistocene sediment DNA from Denisova Cave

Quantum birds

Antibody therapy rescues mice from lethal nerve-muscle disease

Life in these star-systems could have spotted Earth

Cutaneous reactions after mRNA COVID-19 vaccines

Skin reactions after COVID-19 vaccination: Rare, uncommonly recur after second dose

[] Engineers apply physics-informed machine learning to solar cell production
Lehigh University researchers use Frontera supercomputer to simulate photovoltaic fabrication, train AI to optimize energy production