Scientific Legacy of Nobel Laureate Robert Grubbs Amplified by AI

The Institute pilots new archival methods making one of its largest hybrid collections, donated by the Grubbs family, a rich resource for future researchers.

When Caltech archivist Mariella Soprano began sorting through the extensive professional papers of Caltech professor and Nobel laureate Robert H. Grubbs, who passed away in 2021, she quickly realized her team had a special opportunity. The collection, donated by his wife, Helen, and their children Robert, Brendan, and Kathleen, spanned more than 130 boxes, alongside some 60,000 digital files and 100,000 emails. Its content touched on numerous facets of Bob's prolific career, including interactions with Caltech students, development of patents, and collaborations and consultancy work for companies propelled by his discoveries.

"It feels great to know his legacy will live on, not only through his career but also through his extensive records," Helen says. It is hoped that the archives could prove valuable in continuing research in areas Bob helped pioneer such as "green chemistry," where catalysts are devised that create fewer waste products.

It was the scale and diversity of the donated papers that prompted Soprano, senior collections and special projects archivist, to explore new workflows including the use of AI tools. Helping her lead this effort is Tommy Keswick, digital technologies development librarian.

"Integrating a hybrid collection into a cohesive collection like this is not something that the archives have done yet, and so we wanted to figure out how," says Keswick.

Transforming Archival Practice with AI

The Grubbs papers are one of the first major hybrid collections Caltech has processed, encompassing both physical and digital records.

When the archives team first incorporated AI into their workflow, the goal was simple: to identify duplicate materials. With so much overlap between printed files and digital versions, the technology offered a way to streamline the collection by confirming which records needed to be retained.

As the project developed over the past 18 months, the team discovered broader applications. Today they are testing tools such as named entity recognition, topic modeling, and large language models to generate summaries. One of the most exciting possibilities is the depth and searchability the papers could offer future researchers.

Traditional archival methods create catalogs that briefly describe the contents of a folder or box—but they cannot provide full-text search. According to Keswick, AI enables new ways to catalog content by turning a static inventory into a more dynamic, searchable resource.

"It's a much richer body of information that we can provide to future researchers," says Soprano. "We don't know what questions people might be asking in the future. Archiving all the material in a comprehensive way opens up possibilities for research we can't yet imagine."

The archive also preserves the iterations and false starts behind Bob's breakthroughs, a reminder of the lesson he tried to impart to students: that failure was part of the process. "One thing he'd want to get across is that you won't always succeed at what you set out to do, but if you pay attention along the way, you may find more interesting directions to go in," says son Robert "Barney" Grubbs.

Grubbs and the Caltech Community

For the Grubbs family, donating Bob's papers to the Institute was an obvious choice. "I think he felt very invested in the Caltech community as a whole and viewed his presence there as a big part of his legacy," says Kathleen, known to the family as Katy. The Grubbs children fondly recall growing up around the campus. "Caltech is where we learned to swim, and both Katy and I had our weddings there," adds Brendan.

Bob was originally from Marshall County, Kentucky, and as a proud son of the Bluegrass State he kept a congratulatory letter from then Governor Ernie Fletcher among his papers, which had been sent in recognition of his Nobel Prize win. After completing a postdoctoral fellowship at Stanford University, he taught at Michigan State University. He joined Caltech's faculty in 1978 and remained for more than four decades, holding the Victor and Elizabeth Atkins Professorship of Chemistry. In 2005, he received the Nobel Prize in Chemistry for his pioneering work on olefin metathesis, which led to new classes of advanced plastics and pharmaceuticals. Over his career, Bob was listed as an inventor on more than 200 US patents and co-founded several startups.

He also was known for his collaborative spirit, often working across disciplines and generations to tackle scientific challenges. "He loved his students," Helen says. "He felt very lucky to have such talented young people to work with."

At a memorial gathering in Bob's honor in August 2022, George Grant Hoag Professor of Chemistry Dennis Dougherty stated of his friend and colleague: "Bob's passing creates a huge hole in the Chemistry Division, Caltech, and indeed, the entire world of science." His friends, colleagues, and former students would go on to establish the Bob and Helen Grubbs Fellowship, an endowed fund that supports one graduate student or postdoctoral scholar in chemistry each year.

More recently, the Grubbs family has established the Bob and Helen Grubbs Discretionary Fund for Chemical Catalysis that provides for existing and emerging needs for Caltech's Center for Catalysis and Chemical Synthesis (3CS). The fund will help support 3CS scholars to explore pathbreaking lines of inquiry much like the faculty member for whom the fund is named.

Inspiring Future Research and Discovery

For the archives team, the Grubbs papers represent not just a record of a distinguished career but also a testing ground for new methods. "We're trying to set a model going forward, because we see this playing out across almost everything we collect from now on," Keswick says.

Helen believes Bob would have found the project "thrilling," noting that it reflects his innovative spirit. "Bob was always tinkering, trying to solve problems, great and small," Helen says. "With access to the collection, I hope that students and researchers will be inspired to try new approaches to old problems."