Superconductors play a critical role in the modern-day scientific enterprise, from the magnets used in the CERN Large Hadron Collider to those in medical MRI machines. But these superconductors require very low temperatures to work. The discovery of room-temperature superconductors would open a huge range of applications and new technologies. 

There are no physical laws that rule out such materials, according to a recent perspectives article by a group of scientists including Warren Pickett, Distinguished Professor Emeritus of physics and astronomy at UC Davis. Their article, published March 13 in Proceedings of the National Academy of Sciences, sets out a roadmap to achieve this breakthrough. 

For over three decades, the record superconducting critical temperature at ambient pressure has remained at 133 K, or minus 227 degrees Fahrenheit, for a mercury-based, copper-oxide ceramic known as Hg-1223. Though new classes of higher temperature superconducting materials have been discovered, they remain constrained by the immense pressure required to coax out their superconducting properties. 

In a new study appearing in the same journal issue, researchers from the University of Houston, Argonne National Laboratory and Intellectual Ventures report successfully raising the temperature that Hg-1223 begins to display superconducting behaviors to 151 K, or minus 122 degrees Fahrenheit.

The discovery was achieved through a technique called “pressure quenching” in which the Hg-1223 material was squeezed in a diamond anvil cell at 300,000 times normal air pressure. When the pressure was removed quickly, the material remained in a superconducting state up to 18 degrees higher, for up to two weeks.

The accompanying perspective article advocates for a programmatic approach to superconductivity research over an “Edisonian” approach. 

“Thomas Edison discovered a lot of things just by trying things, by intuition and guessing,” Pickett said. “That needs to be a little more refined.” 

Providing expert input 

For three years, Pickett, who was recently elected to the hall of fame of his alma mater Wichita State University’s Fairmount College of Liberal Arts and Science, has worked as a consultant with Intellectual Ventures’ Enterprise Science Fund. The organization aims to advance experimental technology so it can be produced at the industrial scale. 

“My part is to use my experience and background to comment on and try to help push along what I see as the better direction,” Pickett said. “This whole deep science project is intended to be high risk, high payoff.” 

The programmatic approach that Pickett and his colleagues endorse calls for a crosspollination of knowledge from the conventional and unconventional superconductivity communities and for input from those outside of physics and materials science. It also addresses challenges faced when predicting, engineering and experimenting with superconducting materials. 

The prediction challenge

Pickett and colleagues first outline the “prediction challenge” facing the superconductivity community. They compare the field to the early days of the semiconductor industry, when germanium was used prior to the advent of silicon.

Current superconductors are “useful, but yearning to be supplanted by a better material that can operate at higher temperatures and scale to industrial production,” they write. “While there may not be a ‘silicon’ for superconductivity, this does not mean that we should not search for one, akin to searching for a small needle in a large haystack.” 

To sift through the haystack, the researchers advocate for harnessing leading-edge computational modeling techniques, including machine learning and ab initio modeling, to predict new superconducting materials and their structures. Such methods, combined with thermodynamic screenings, could help prioritize stable candidates for further experimental testing.    

Additionally, theorists and experimentalists should be in constant conversation, creating a bidirectional feedback loop that will help refine both models and experiments. 

For instance, information regarding why a material failed to achieve superconductivity in an experimental setting could help refine the computational model that initially led that material’s selection as a superconducting candidate.  

The authors “imagine a future where materials databases are able to accurately predict properties like [critical temperature] and formation energies, along with viable recipes for making them, maximizing the capabilities” computationally “to minimize time spent toiling in a lab.”

The engineering challenge

A material’s superconductivity, whether intrinsic or engineered, can be enhanced by toying with so-called knobs, which are external ways of manipulation, such as exposing the material to pressure/strain, pulses of light, doping and other techniques that exploit its structure. 

While such techniques are common practice in the superconducting community, Pickett and colleagues highlight areas for improvement and further exploration. 

For example, the researchers note that the “pressure quenching” technique used in the PNAS study requires further research to elucidate how and why it worked. Furthermore, while doping is a well-known technique for chemical manipulation of a material, it can be introduced as a variable in computational models. This would allow theorists to tweak superconducting candidates in computer simulations, showcasing potential effects before suggesting a material be synthesized and experimented with in a lab setting. 

The perspective paper is a rallying cry for the superconductivity community, suggesting directions for further research and exploration. 

“We believe that overcoming this challenge requires a systematic, programmatic approach composed of well-planned paths of investigation spanning computational discovery and design as well as materials synthesis and engineering,” they write.