Alumni Profile: Sam Melton, A.B. '13, Ph.D. '20

When it comes to renewable energy, consistency is critical. The sun doesn’t always shine, the wind doesn’t always blow, and if a grid relies too much on those sources of energy, it can struggle when they aren’t available.

Geothermal energy offers a stable alternative to solar and wind power. Geothermal plants use the heat generated from radioactive decay in the earth's core, which either naturally generates subterranean pockets of super-heated water or steam, or can be engineered to produce the steam at the plant site. The result is a steady, reliable, consistent output of energy.

But geothermal energy has its own challenge: cost. Plant developers often have to drill many holes into the ground to find a sufficiently large geothermal pocket, and each well adds up.

“There's an upfront expiration cost frequently over $50 million that the geothermal developers need to expend just to evaluate a site and see if it's suitable for generating power,” said Sam Melton, A.B. '13, Ph.D. '20. “This upfront capital cost is very risky for the developers, obviously, but it also poses problems for them getting financing to develop the projects. Because it's so uncertain, the banks are less likely to provide them financing for this exploration.”

Melton wants to bring the cost of exploration down. A former Ph.D. candidate in applied math at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS), Melton is in his first year as co-founder and chief technology officer for New Mantle Technologies. The Massachusetts-based start-up is developing tools that combine machine learning and physics simulation to make it easier to find ideal geothermal plant locations. The company completed its pre-seed funding in February and is currently expanding its workforce with more geologists and computational scientists.

“I wake up every day, and I'm motivated to make progress because I think the technology we're developing could really change the world and could generate a lot more clean power for the U.S.,” Melton said. “We're trying to both make it easier to prospect a new site without drilling holes at all, and then once you've settled on a site, we're hoping to provide more information that reduces the number of test wells that developers need to drill before finalizing their power plant plans.”

A Massachusetts native, Melton studied physics as an undergrad at Harvard. He was interested in a graduate degree, but wanted something different than a lifetime in physics academia. He completed a one-year fellowship at the University of Cambridge, worked for a year at a biotech start-up called Warp Drive Bio, then returned to Harvard for his Ph.D.

“I wanted to do something that would have more of a real world impact, which is something that I've been continuously chasing,” he said. “I talked to faculty like Michael Brenner and my eventual advisor Sharad Ramanathan about SEAS. I liked the idea of SEAS because it felt like I could study a lot of techniques in applied math, and then it was kind of up to me what I was going to do with it.”

Melton’s pivot to geothermal energy came from his next stop after SEAS: MIT. He spent the next four years on a fellowship studying the physics of living and complex systems. While there he met Charles Gertler, who was already researching geothermal plant development for the U.S. Department of Energy. 

“I’d spent a lot of time trying to think of what big problems I would want to spend maybe 10 or 15 years working on,” Melton said. “I got really adamant that I wanted to work on something that would make a difference in climate, so I started thinking about ways I could use the skills I had to make some sort of impact. I worked for a battery manufacturer for a little bit, but most of my thoughts were about doing something a little bigger on my own. I wanted to have more ownership over the process and the product and the company.”

Melton and Gertler eventually co-founded New Mantle this past January. As CTO, Melton splits his time between coding, researching new publications about geothermal energy and plant development, and meeting with potential clients.

“When I was building the original tech base of the company, I was always really pleased to come across a problem I needed to solve, and I’d remember a problem set I did in my first year at SEAS where I solved a very similar problem in a class, or a conversation I had with my advisor on a really similar topic,” Melton said. “There were a lot of methods for physical computing that I took in grad school because it was a requirement, and those kinds of classes have been extremely useful. I also think I have reasonable skills at presenting my findings to investors and clients. I don't remember specifically learning this at SEAS, but I do remember giving a lot of group meetings and talks like that, and all of that practice has definitely helped.”

Alongside technical and communication skills, SEAS also taught Melton an invaluable approach to problem solving. The SEAS philosophy of first characterizing the exact problem you want to address, researching and talking to stakeholders, and then devising clear parameters of what a solution must accomplish has been critical to how he’s built New Mantle Technologies. And as new problems arise as the company continues to expand, he knows that same philosophy will continue to keep him on the right track.

“SEAS sets you up with a huge capacity for solving problems, because they teach you how to solve so many diverse problems with so many diverse methods,” he said. “I think the thing that was most useful, and I remember so fondly about grad school, was conversations with my advisor, Sharad. He’s an incredible scientist, and has such a strong sort of principle that the computations you do should be connected to experiments and should be meaningful. It's better to do a simple calculation that will have an impact rather than a complicated calculation that is maybe not exactly what's happening in the real world. I'm always thinking about lessons Sharad taught me about how to build models, how to think carefully about data, and then combine the two into my analysis.”