By Rachel Fairbank
Illustrations by Jasu Hu
Achieving net-zero emissions is a technological challenge requiring sweeping innovation across multiple aspects of society, such as reducing carbon dioxide emissions from sectors like transportation, agriculture and industry; transitioning to clean energy sources such as wind or solar; increasing the efficiency of renewable energy sources; and devising long-term methods for storing renewable energy.
“The first target that needs to be met for energy transition is electrification,” said Aditya Mohite, associate professor of chemical and biomolecular engineering and the director of the Rice Engineering Initiative for Energy Transition and Sustainability, REINVENTS for short. “If we take care of electrification, then we will take care of about half of the CO2 emissions that are being emitted. The second, harder piece of this is: How can we positively impact industry’s environmental footprint? This is a much harder proposition.”
As the U.S. works toward the goal of net-zero emissions, researchers at the George R. Brown School of Engineering are focused on answering some of the major technological questions of the future. From the Carbon Hub led by Professor Matteo Pasquali to the recently launched REINVENTS, engineers across the school are actively developing solutions. Here are some ways they are making a difference, along with their plans for moving these technologies forward.
Zeroing in on Industrial Emissions
According to the EPA, nearly a quarter of U.S. greenhouse gas emissions come from industry. One major source of industrial emissions is methanol, which is used as a precursor for many chemicals and materials. “Every year, one-quarter of the emissions from the chemical manufacturing industry are due to methanol,” said Haotian Wang, associate professor of chemical and biomolecular engineering. Methanol is used to make some of the key precursor molecules, such as acetic acid, that are, in turn, used in the food industry and to make a wide variety of organic and inorganic materials, including polymers.
In research funded by the U.S. Department of Energy, Wang’s research group works on ways to synthesize acetic acid from carbon monoxide, which can be obtained from carbon dioxide electrochemical conversion, rather than methanol, in order to make the process emissions-free. “We developed a catalyst that can help us to dimerize carbon monoxide molecules,” Wang said. This catalyst, which helps facilitate the reaction, is being combined with a new type of reactor, one that helps reduce the energy requirements of the process.
Part of the synthesis process is bringing the two carbon monoxide molecules together to synthesize acetic acid, which has two carbons. This process, known as carbon-carbon coupling, typically requires a lot of energy and is inefficient, with a low product selectivity. The Wang research group aims to develop new catalysts that facilitate this carbon-carbon coupling process, improving the acetic acid product selectivity. “If this reaction pathway is successful, we could significantly reduce the carbon footprint associated with using methanol during the acetic acid fabrication process,” Wang said.
The second component of this process has been to design a new type of reactor, called a solid electrolyte reactor, which drives the reaction in a solid electrolyte medium, rather than in a liquid electrolyte medium. The advantage of this type of reactor is that once the product has been synthesized, it doesn’t need to be separated from a liquid electrolyte medium that contains ionic impurities. “That separation process is very costly,” Wang said. “These costs can be up to 80% of the overall cost.”
Wang and his collaborators hope to use this type of system to make acetic acid from carbon dioxide, rather than carbon monoxide, and to adapt this system for synthesizing similar products often used in industry, such as ethylene or ethanol.
Adapting AI for Energy Efficiency
Leaders in artificial intelligence have acknowledged that the industry faces an energy crisis and the next wave of generative AI systems will consume several times more power than expected. Anshumali Shrivastava, associate professor of computer science, has founded a company, ThirdAI, which he believes can alleviate the problem.
“AI processes have historically run on larger, less accessible computing hardware,” he said. “With ThirdAI, our tools are able to run on a regular central processing unit, rather than the more powerful graphics processing unit. CPUs handle fewer computations simultaneously, but have more memory than in GPUs. That’s an acceptable trade-off for the purpose of saving energy.”
AI is responsible for carbon emissions from non-renewable electricity and for the consumption of millions of gallons of fresh water. Instead of trying to make AI more efficient by applying enormous amounts of computing power to it, Shrivastava is rethinking its fundamentals in an effort to make it cheaper and more energy-efficient.
“We’ve been using an inefficient process and using even more energy to do it. Our team is working to change that,” Shrivastava said.
Billions of bits of data in an AI model are routinely updated every time there is an error, even though just a few hundred may actually need the update. ThirdAI’s strategy is to process and train AI models by limiting updates to what is relevant. It does that by using hashing, in which data is tagged and stored in memory close to similar kinds of data.
“When we looked at the landscape of deep learning, we saw that much of the technology was from the 1980s and most of the market was using GPUs, investing in expensive hardware and expensive engineers and then waiting for the magic of AI to happen,” Shrivastava said. “Our algorithm eliminates the need for specialized acceleration hardware that wastes so much energy.”
Producing Liquid Fuels Using Sunlight
Another key source of carbon emissions is the transportation sector, where gasoline is still one of the predominant fuel sources for vehicles. The advantage of gasoline is that it can be stored for long periods of time and contains high amounts of energy, which can then be used to power a vehicle. The disadvantage is that oil and gas is a finite resource and emits carbon dioxide. “With gasoline, you can fill your car in two minutes and go 400 miles,” Mohite said. “The transition to replacing that is not going to be straightforward.”
One alternative to gasoline is hydrogen gas. Similar to gasoline, hydrogen gas can be stored for long periods of time and used to power cars, emitting only water rather than carbon dioxide, in the process. However, “most of the hydrogen produced is from cracking methane, but in the process, you emit carbon dioxide,” Mohite said.
One alternative to gasoline is hydrogen gas. Similar to gasoline, hydrogen gas can be stored for long periods of time and used to power cars, emitting only water rather than carbon dioxide, in the process. However, “most of the hydrogen produced is from cracking methane, but in the process, you emit carbon dioxide,” Mohite said.
Hydrogen can also be produced by splitting water into hydrogen and oxygen, but doing so is an energy-intensive process, one that requires more energy than it will generate. If this process can be made more energy-efficient while also using a renewable energy source, such as wind or solar, it can bring hydrogen gas closer to being a viable alternative to gasoline.
In support of this goal, Mohite’s research group is working on more energy-efficient methods for producing hydrogen gas from water using low-cost solar as the energy source, as part of a project funded by the U.S. Department of Energy. “We’ve developed a technology where we use a very low-cost solar cell, which we then couple with catalysts, which allows us to do high-efficiency solar to hydrogen production,” Mohite said. “It’s a platform device that converts sunlight into a liquid fuel.”
A system like this can use the energy from sunlight to convert water into hydrogen fuel, which can then be stored and used in place of gasoline. “The advantage of an integrated system like this is that you can also harvest the heat which is lost. That helps drive the catalysis, making it more efficient,” Mohite said. “You’re harvesting electric power from sunlight, but you are also harvesting the lost heat.”
Reinventing the Energy Landscape
Given the diversity of energy transition work being done by Rice Engineering faculty, a critical component of the research is supporting its move from early stages in the lab toward later stages of technological feasibility.
With support from the Dean of Engineering, Rice Engineering launched a new program in 2023 called the Rice Engineering Initiative for Energy Transition and Sustainability, or REINVENTS. Its goal, aided by an initial commitment of $500,000, is to create interdisciplinary teams to foster emerging technology. This includes the creation of teams in three key areas: energy generation, long-term energy storage and energy-efficient processes and materials.
“We now have groups of faculty members who are formed into teams, based on the various technology portfolios that we have,” said Mohite, the director of REINVENTS. “We think of REINVENTS as an extended research arm for companies across industries.”
The mission of REINVENTS is to support the transition from prototype to fully developed technology, a process that is time- and resource-intensive. “If you want to scale your technology from a technology readiness level of two or three to a technology level of seven, which is where industries want it, how do you do that?” Mohite said. REINVENTS helps bridge that gap, by forming collaborations and offering the necessary infrastructure for testing how these technologies might scale up. “The accelerator is in the works, and now we are progressing toward pilot scale capabilities,” Mohite said.
Research performed by faculty members in the George R. Brown School of Engineering has for years been the seed from which start-up companies have sprung and flourished. Here are three examples:
-
DexMat is a Houston-based climate tech startup that creates high-performance, low-carbon materials with applications in a variety of industries. The company’s co-founders are Matteo Pasquali, A.J. Hartsook Professor of Chemical and Biomolecular Engineering and director of Carbon Hub and Dmitri Tsentalovich, a Rice chemical engineering doctoral alum. Its signature product is Galvorn, patented by Pasquali and made entirely of carbon derived from hydrocarbons, renewable fuels and captured carbon. The company’s goal is to make carbon- and energy-intensive materials like steel, aluminum and copper obsolete. dexmat.com
-
Qilin Li, professor of civil and environmental engineering, is the founding director of SolMem, a company dedicated to developing technologies that can treat water with fewer chemicals and less energy. Its researchers tackle the most challenging water and wastewater problems by developing low-cost, high-efficiency treatment systems that use renewable energy. The technologies provide water to communities and industries at off-grid locations where conventional water sources are unavailable or limited. solmem.com
-
Syzygy Plasmonics is rooted in the work of senior researchers Naomi Halas, founding director of Rice’s Laboratory for Nanophotonics, and Peter Nordlander, professor of electrical and computer engineering and of materials science and nanoengineering. The Houston company focuses on commercializing deep decarbonization of chemical manufacturing processes. Its strategy combines new photocatalyst technology with a novel reactor that uses common, low-cost materials to manufacture hydrogen, ammonia, methanol and other chemicals, rather than relying on thermal energy. The result is a renewable universal platform that uses various feedstocks to enable chemical reactions while reducing feedstock waste and producing fewer emissions. plasmonics.tech
Rice leverages its research expertise and location in the energy capital of the world to advance the clean energy transition. In addition to carbon capture, utilization and storage, and hydrogen generation and storage, the school’s research areas include:
Electrification
Lisa Biswal, William M. McCardell Professor in Chemical Engineering, and Ming Tang, assistant professor of materials science and nanoengineering, explore batteries in relation to electrifying cars. The Biswal group designed a class of organic-inorganic hybrid Si-anode that improves the capacity and cyclability of lithium-ion batteries. The Tang group uses mesoscale modeling and advanced X-ray characterization techniques to analyze how the internal structure of lithium-ion batteries affects battery performance.
Long-term Storage
Daniel Cohan, of civil and environmental engineering, models the integration of renewable resources and storage technologies to complement traditional power plants. Jun Lou, of materials science and nanoengineering, focuses on solid-state battery mechanics to boost capacity and safety, potentially improving battery performance for electronics. Geoff Wehmeyer, of mechanical engineering, builds devices that control heat flows in energy systems to improve efficiency and thermal performance in industrial and electrical applications.
Energy-Efficient Materials
Kai Gong, assistant professor of civil and environmental engineering, leads the Sustainable Infrastructure Materials group, which employs advanced computation, experiments and data-driven modeling to lessen the environmental impact of infrastructure materials. Stavroula Alina Kampouri, assistant professor of chemical and biomolecular engineering, and her research group focus on developing multitasking sponge-like materials capable of capturing waste and converting it into valuable resources using sunlight as the energy source.
Sustainable Computing
Through his research in quantum systems and compilers, Tirthak Patel, assistant professor of computer science, seeks to improve the energy and performance of quantum computers by leveraging quantum mechanical principles of superposition, entanglement and reversibility. This work will enable the efficient execution of large-scale applications in machine learning and scientific computation on quantum computers.