Not quite gas and not quite liquid, plasmas have a reputation for being tricky to characterize and predict. That makes it challenging for scientists to build accurate computer models to simulate them, even in an age of unparalleled processing power.  

For example, certain computer models work easily and efficiently by treating plasma as a fluid, but they make a lot of assumptions about how the plasmas will behave. On the opposite end of the spectrum are models that treat plasmas as kinetic particles. They’re more accurate but are costly and need far greater amounts of computing resources.

Iman Datta thought there should be a best-of-both-worlds solution.

“I wanted to see what would happen if I used a fluid model in general regions of the plasma,” says Datta, a research scientist at Zap Energy. “But in areas such as around the walls of a fusion device, I wanted to try the more intensive kinetic modeling.”

“What he did is still quite cutting-edge,” adds Eric Meier, Zap’s head of theory & modeling. “We really hadn’t tapped into the idea of bringing together fluid and kinetic modeling in that way.”

After publishing his PhD work earlier this year, Datta is applying the methods to develop computational models that match and optimize the real-life conditions measured within Zap’s FuZE-Q and FuZE devices. The relative simplicity of the company’s technology means Datta can suggest changes that can often be quickly implemented and tested, speeding the iteration and feedback cycle between theory and observation.

“The simplest model that can succeed and solve your problem is the one you want,” he says.  

High-fidelity simulations

Computational modeling has long been used as a predictive tool. Few industries today remain untouched by computer processing power to predict weather, create safer high-rise buildings, simulate flying conditions, and more.

But modeling also has the potential to optimize designs of complex devices like fusion systems.

In his work, Datta compares computer models with real-world results to ensure they closely match, a process known as validation. If there’s high fidelity – a reproduction faithful to actual conditions – modelers can run any number of computational experiments to tweak, enhance, and test how a plasma is expected to behave in Zap’s fusion devices.

“We can say, ‘All the knobs need to be turned to these settings,’ as a way to improve plasma performance,” Datta explains.

Interest in the stars leads to fusion

Exploring the night sky in his childhood home of Newfoundland stirred the scientific curiosity that drives Datta today. From an early age, his father took him to Signal Hill in St. John’s, on a far-flung corner of North America that was also the receiving site of the first wireless transatlantic communication. In awe of the universe, Datta's ambition to learn how it all worked ultimately led him to plasma physics.

His first exposure to fusion was learning about astronomy through the Texas Science Olympiad program following a move to the Houston area when he was 11. By his senior year at Cypress Ridge High School, he won the Texas state championship.  

Datta studied mechanical engineering in his undergraduate work and got interested in fluid dynamics – what became the basis of a fascination with plasma. After jobs as both an oil refinery engineer in Texas and at NASA Ames Research Center in California, he followed an invitation from Zap Chief Scientist and Co-Founder Uri Shumlak to pursue a PhD in plasma physics at the University of Washington.

Meier recalled Datta taking time to teach him the ways of WARPXM, a computer code developed at the University of Washington for computational plasma dynamics.

“He has a mind that is very methodical and thorough,” Meier said of Datta. “When he’s tackling a problem, he’s very good at starting at the roots and putting the pieces of the puzzle together holistically.”