The world needs new materials to solve old problems: climate change, food security, and energy storage. But what if the answer wasn’t a synthetic, lab-grown miracle, but a material we’ve overlooked for millennia—upcycled from waste?

Meet Dr. Francisco Martin-Martinez, a Senior Lecturer in Computational Chemistry at King’s College London and a Research Affiliate at Massachusetts Institute of Technology (MIT). Describing himself as an “impact-driven computational chemist and engineer,” Dr. Martin-Martinez is not just studying biocharBiochar is a carbon-rich material created from biomass decomposition in low-oxygen conditions. It has important applications in environmental remediation, soil improvement, agriculture, carbon sequestration, energy storage, and sustainable materials, promoting efficiency and reducing waste in various contexts while addressing climate change challenges. More; he is leading an interdisciplinary effort to redesign it from the atomic level up. Dr. Martin’s lab operates at the thrilling intersection of computational chemistry, bioinspired materials, and machine learning, fundamentally shifting the paradigm of biochar development from “discovery” to “digital design.” His core mission? “Developing designer biochars with atomic resolution.” His King’s College London lab is moving biochar production from trial-and-error to digital design, using tools like Density Functional Theory (DFT) calculations and molecular dynamics simulations to develop a sort of atomic-scale “CAD platform” for biobased carbon materials. This enables his team to precisely engineer features such as surface topology and heteroatom doping, transforming simple charcoalCharcoal is a black, brittle, and porous material produced by heating wood or other organic substances in a low-oxygen environment. It is primarily used as a fuel source for cooking and heating.   More into high-performance materials. This atomic precision is key for applications from boosting capacity in energy storage devices to creating “good enough” designer biochars for precision agriculture, which he predicts could be one of the most disruptive applications in the next five years. Dr. Martin’s work is central to “BiomassBiomass is a complex biological organic or non-organic solid product derived from living or recently living organism and available naturally. Various types of wastes such as animal manure, waste paper, sludge and many industrial wastes are also treated as biomass because like natural biomass these More Carbon Mining”, redefining waste as a valuable resource to close the loop in the circular economy. I welcome Dr. Francisco Martin-Martinez, a computational architect of atomic-scale simulations, to this session.

Shanthi PrabhaV: Dr. Francisco Martin-Martinez, you describe yourself as an “impact-driven computational chemist and engineer.” If you had to boil down your lab’s core mission in biochar research into a single, memorable phrase, what would it be? (The simpler, the better!)

Dr. Francisco Martin-Martinez:  I will try to be “memorable”: Developing designer biochars with atomic resolution.

SP: Your work sits at the intersection of computational chemistry, materials science, and sustainable applications. Which of those three fields do you think is currently the “bottleneck” in getting high-performance, engineered biochar materials to market, and why?

FW: While innovation in all three areas is crucial, I believe the primary bottleneck right now is in process engineering—specifically, the challenge of achieving cost-effective production at scale. From a computational standpoint, we are becoming increasingly proficient in designing biochar on a computer. Using atomistic models and now AI, we can design biochar with an ideal pore structure for water retention or specific surface chemistry for capturing contaminants, and we are starting to understand the fundamental mechanisms occurring at the nanoscale. However, this virtual precision needs to be translated into real biochar that is produced with the same level of accuracy. How do we translate a precise digital model into a consistent physical product when our input feedstockFeedstock refers to the raw organic material used to produce biochar. This can include a wide range of materials, such as wood chips, agricultural residues, and animal manure. More is inherently variable? This remains a significant challenge for materials science and process engineering.

From a sustainability point of view, there are challenges at the system level, such as the energy cost of producing biochar or the need for scalable supply chains. PyrolysisPyrolysis is a thermochemical process that converts waste biomass into bio-char, bio-oil, and pyro-gas. It offers significant advantages in waste valorization, turning low-value materials into economically valuable resources. Its versatility allows for tailored products based on operational conditions, presenting itself as a cost-effective and efficient More is energy-intensive. The cost of producing a highly engineered biochar often outweighs the price farmers or industries are willing to pay for it. Valorizing any side products from biochar production, such as bio-oils, could help. In fact, a new PhD student in my group, Rebecca Driver, is working on designer bio-oils for asphalt applications, which also contributes to the valorization of biochar production. Additionally, establishing the logistics to collect diverse biomass feedstocks and distribute the final biochar product is a massive systems-level challenge that needs to be addressed.

SP: Many are familiar with the traditional production methods of biochar(pyrolysis). Can you walk us through how your group’s use of Density Functional Theory (DFT) and molecular dynamics simulations changes the game? How does the “digital design” of biochar actually work?

 FM: Traditional biochar production is based on experience and trial-and-error. You heat the material, work with it, and learn what works over time. Our computational approach is akin to being an engineer at the molecular scale, utilizing sophisticated atomic-scale simulation programs. We design and simulate the material digitally, hoping to inform the production process. Dr Anna Bachs-Herrera, who recently graduated from my group has led most of the initial work in this area. It shifts the entire paradigm from discovery to design. In the specific case of DFT and molecular dynamics simulations, we analyze the properties of biochar by modeling exact surface features and chemically active sites, thereby simulating how well it will retain water or adsorb a specific nutrient. With molecular dynamics simulations, we can conduct hundreds of these digital experiments, varying the temperature, pressure, or chemical properties. There are still some challenges to be addressed, such as developing more reliable and accurate methods to describe the interactions between atoms in biochar. A new PhD student in my group, Xiaoqi Zhu, is focusing on this. We collaborate with Dr. Carla de Tomas at King’s College London, a leader in worldwide research on simulating porous carbon materials using large-scale molecular dynamics simulations. We are also starting to work with Dr. Meredith Barr at London South Bank University, an expert in biochar production. It is exciting to learn from them in this highly collaborative research.

SP: Your profile mentions specifically tuning biochar’s properties through surface topology and heteroatom doping (like nitrogen and oxygen). Why are these features—the atomic and molecular-scale features—so critical for real-world performance, say, in an energy storage device?

 FM: A high-performance carbon electrode requires a vast surface area (controlled by the surface topology) to provide storage space, and strategically placed heteroatoms (such as nitrogen and oxygen doping) to make that space chemically active and efficient. Without precise control at the atomic scale, you are just left with simple charcoal instead of a high-performance electrode. The surface topology is the physical landscape of pores, bumps, and channels on the biochar’s surface. A high surface area, created by a network of pores with varying sizes and shapes, is akin to having a massive grid. It makes a vast number of physical “parking spots” where electrolyte ions can be stored. The more accessible surface area you have, the more ions you can pack in, which translates, together with other factors, to a higher energy capacity. When we intentionally introduce nitrogen or oxygen atoms into the carbon lattice, we create chemically active sites. These sites do more than passively hold ions; they actively attract them and can undergo fast, reversible electrochemical reactions. They also enhance biochar’s wettability. These active sites enable the device to not only store more energy but also charge and discharge much faster, thereby boosting its power density. We collaborate with the team of Prof. Antoni Forner-Cuenca at TU Eindhoven, who is the expert in energy storage, from whom I have learned everything I know about this application of biochar. We also work with Dr. Nieves Lopez Salas at the University of Paderborn, whose knowledge on the interaction of water molecules inside carbon pores is endless.

SP: You’re now moving into Machine Learning to develop better interatomic potentials and predictive models. What’s one unexpected or surprising insight you’ve already gained about biochar’s behavior using an AI model that you might have missed with traditional simulations?

FM: With traditional DFT calculations, we were limited by computational cost. We would focus on a specific active site, such as a nitrogen dopant, in a relatively clean and idealized environment. We understand that environment very well, but we are essentially studying it in isolation. With machine learning models, we aim to navigate through thousands of structural variations and predict more complex features at a lower computational cost. If we adopt some forward-thinking approaches, we could envision a future in which we ask AI more sophisticated questions, such as, ‘What are the most impactful atomic patterns for enhancing catalytic activity?’ This allows us to move beyond simple doping and start thinking about how to engineer these specific, high-performing local environments. It would completely reframe our future experimental and computational work. Machine learning is also transforming the way we conduct molecular dynamics simulations. These simulations rely on interatomic potentials, a set of mathematical functions that describe how atoms interact. Machine Learning Interatomic Potentials are expected to achieve the accuracy of quantum mechanics while running at the speed of classical potentials. Our work with Dr. Carla de Tomas focuses on applying these machine learning interatomic potentials to disordered carbon materials.

SP: Let’s talk about applications. You focus on energy storage and soil remediation. Are the material properties you tune for a high-performance carbon electrocatalyst similar to or completely different from those needed for an effective soil amendmentA soil amendment is any material added to the soil to enhance its physical or chemical properties, improving its suitability for plant growth. Biochar is considered a soil amendment as it can improve soil structure, water retention, nutrient availability, and microbial activity.   More?

FM: Well, while the ideal material properties for a high-performance electrocatalyst and an effective soil amendment are vastly different, the underlying computational physics and chemistry often overlap. For instance, a core challenge in both agriculture and the development of electrodes for energy devices, such as redox flow batteries, is accurately simulating the behavior of electrolytes in water. Whether we are modeling how nutrient ions are adsorbed and transported from water into the pores of biochar in soil, or how redox-active ions move and react at an electrode surface, we are fundamentally studying ion-solvent-surface interactions. This means we can often use very similar multiscale models and molecular dynamics simulations, allowing the fundamental insights we gain in one area to accelerate our progress in the other, even if the final ‘designer’ materials are chemically distinct.

SP: Your latest publication discusses “Biomass carbon mining to develop nature-inspired computational materials for a circular economy”. Can you explain the concept of “biomass carbon mining” in simple terms and its role in an actual circular economy?

FM: When we published that perspective article in iScience, we wanted to emphasize “Biomass carbon mining” as the idea of treating waste biomass, such as agricultural residues or forestry scraps, not as trash, but as a valuable above-ground mine for carbon. Just as you would mine the earth for carbon, gold or iron, we “mine” this waste for its fundamental carbon building blocks. It is deeply connected to the concept of biorefinery. Its role in a circular economy is to close the carbon loop fundamentally. Instead of letting that plant-captured carbon return to the atmosphere as CO₂ when the biomass rots or is burned, we “mine” it and lock it into a stable, valuable form. This process transforms a linear path (grow, waste, emission) into a circular one. It’s about seeing waste as a resource and using technology to upcycle it into the building blocks for a more sustainable economy.

SP: You’ve advised companies like 2050 Materials. When a start-up approaches you, what is the single most crucial piece of advice you give them about translating cutting-edge research into a viable, large-scale commercial product?

FM:  I would not describe myself as an expert on advising start-ups, although I have held this role with four or five companies in my career. I suppose a single important piece of advice is to focus on the customer’s problem, and ideally, in your initial technical solution. In the world of research, we are trained to focus on our solution, the molecule, the algorithm, the high-performance material. But a successful product is not defined by its technical elegance, but by its ability to solve a real-world problem for a customer in a way that is reliable, scalable, and affordable. We should not try to commercialize our ‘perfect’ lab prototype, but rather the one that we can scale economically. Identify the ‘good enough’ threshold that solves the problem. Once you have a product that people are willing to pay for, you have a business. You can then use the revenue to improve towards perfection. But if you start by chasing perfection, you will likely run out of money before you ever get to market.

SP: Biochar is a hot topic, but the public conversation sometimes focuses more on carbon sequestration than engineered performance. In the next five years, which application do you predict will be the most disruptive due to advances in computational design: energy or agriculture?

FM: While the high-tech energy applications are fascinating and one of the hottest topics nowadays, I think agriculture can see the most significant and disruptive impact. One reason is the lower barrier to commercial entry. Creating a biochar that measurably improves crop yield by retaining more water and nutrients is a far less demanding engineering challenge than creating a carbon electrode that can compete with state-of-the-art energy storage materials in a commercial device. In agriculture, a “good enough” designer biochar can provide an immediate and profitable return on investment for a farmer. There are companies like HyveGeo that are making remarkable advances in the field. For energy storage, the performance standards are brutally high. An electrode for a supercapacitor or battery must have exceptional conductivity, purity, and stability over thousands of cycles.

SP: Moving beyond the lab, how do you see the future of site-specific precision agriculture being enabled by combining your engineered biochar with IoT and sensor technology? How will a farmer know exactly which biochar to apply and where?

FM: It is not easy to know, as it is very challenging. Maybe in the future, farming will be like personalized medicine for soil. Smart sensors in plants, drones, and tractors could create detailed maps of fields and plants, showing exactly which spots are dry or lacking nutrients. AI could analyze this data and prescribe designer biochar for each specific problem area. This eliminates all the guesswork from soil management, enabling farmers to use fewer resources while growing healthier, more abundant crops. The work of Prof. Benedetto Marelli at MIT on precision agriculture is definitely worth exploring, as it is poised to revolutionize the field (no pun intended) in the coming years. 

SP: Finally, for our readers who want to follow the exciting work being done at the Martin-Martinez Lab and your collaborations with King’s College London and MIT, where are the best places they can find your work and profile? (Please share a website, LinkedIn, or the best place to contact you.)

FM: https://www.linkedin.com/in/franmartinmartinez/

https://www.kcl.ac.uk/people/francisco-j-martin-martinez

https://www.martinmartinezlab.com