“Information is inevitably physical,” IBM Fellow Rolf W. Landauer would often say. During more than four decades at the company, Landauer made an impression as an irascible but kindly colleague — steadfastly honest and possessing little tolerance for academic fads. A pioneering researcher who helped to establish the underlying physics of computing, Landauer is also remembered for his role in shaping the IBM Thomas J. Watson Research Center into a hub of scientific discovery and technological innovation.

From the early 1950s until his death in 1999, Landauer balanced his research with his responsibilities as a manager and research director, while also making a name for himself as a tenacious critic of his field. As for his own work, Landauer’s most well-known discoveries came early in his career. Within his first decade at IBM, he developed a now-famous model for the energy cost of computation and a new theory of electromigration, which helped lay the groundwork for understanding conductivity in nanoelectronic structures.

Through these collective efforts — as his colleague Charles Bennett would later remark in a New York Times obituary — Landauer perhaps “did more than anyone else to establish the physics of information processing as a serious subject for scientific inquiry.”

Landauer was born in Stuttgart, Germany, in 1927 to Jewish parents. His father, an architect and builder, died in 1934, shortly after Hitler rose to power. Four years later, Landauer’s family fled Nazi rule for the United States, settling in New York City. At 16, Landauer graduated from Stuyvesant, one of the city’s most prestigious high schools, before earning an undergraduate degree in physics from Harvard in just two years.

In 1945, shortly after turning 18, Landauer joined the Navy as an electronic technician’s mate. The experience provided him with a practical education in applied science, which helped shape his characteristically pragmatic approach to physics. In 1950, after returning to civilian life, Landauer reached two more milestones. He earned a PhD from Harvard and married Muriel Jussim, his lifelong partner and companion with whom he had three children.

Out of school, Landauer accepted a position at the Lewis Aeronautical Laboratory of the National Advisory Committee for Aeronautics (now NASA) in Cleveland, where he studied the physical properties of metals for speculative aircrafts. He returned to New York in 1952 to research semiconductors as part of a team at the recently established IBM Research Laboratory in Poughkeepsie.

By the 1960s, IBM Research had moved to Yorktown Heights, and Landauer stepped into a managerial role as head of the Physical Sciences Department. In 1961, he made a landmark discovery of his own.

Though less than a decade into his work at the company, Landauer had published 11 papers, exploring topics such as ferroelectrics (materials that exhibit spontaneous electric polarization, which can be reversed by external electric fields) and electromigration, which studies how atoms move in a current of electronic energy. While investigating the latter in 1957, Landauer had put forward the so-called Landauer formula: a scattering theory approach to the analysis and modeling of electromigration, which would later have a fundamental effect on the general understanding of quantum transport in metals and semiconductors, including nanoelectronic structures.

Much of Landauer’s research during this time was guided by his interest in theoretical limits — particularly, the thermodynamic limitations of information processing. This widely held belief posited that computational power would remain inevitably limited by the energy lost during the processing of information. Landauer countered that while an irreducible amount of energy is lost when information is erased — in order to reset the machine for another computation — other operations should, assuming certain advancements in technology, be able to run with more or less no energy loss. Now known as Landauer’s principle, this concept gave scientists, for the first time, an analyzable model for understanding the basic thermodynamics of information processing — and a roadmap for techniques to create more energy-efficient computational devices.

Outside of his own research, Landauer focused his energy on championing the research he believed in. Throughout the early- to mid-1960s, he supported a range of projects on semiconductor lasers and field effect transistors, leading to the discovery of several novel applications in communication technology, among other advances.

In 1966, Landauer became part of a two-man team responsible for managing the entire IBM Research Division. In this role, he advocated strongly for research and investment in MOSFET (metal-oxide-semiconductor field-effect transistor) technologies, which would form the backbone of the next generation of IBM computer memories and microprocessors.

After being named an IBM Fellow in 1969, Landauer stepped back from management positions to refocus on personal research.

Throughout his career, Landauer gained a reputation among his peers as a staunch critic. “He set high standards for his own work and expected (or rather hoped for) the same levels of honesty and taste in others,” wrote Alan Fowler and Bennett, Landauer’s longtime colleagues at IBM, in their 2009 biography of Landauer for the National Academy of Sciences.

He often exhibited an air of intellectual combativeness when challenging the work of others — even while welcoming others to challenge his own findings. For example, when Bennett approached Landauer to posit his own theory — suggesting that loss of energy in computation was not, in fact, inevitable — Landauer met the idea with an open mind, curious to see what Bennett’s theory might mean for the future of computing. “Many people don’t like their scientific ideas overthrown,” Bennett later recalled. “But he was very interested and eager to discover what this all led to.”

Over the course of his career, Landauer became a member of the National Academy of Engineering as well as the National Academy of Sciences. He won several awards for his work, including the American Physical Society’s Oliver E. Buckley Prize, the Stuart Ballantine Medal from the Franklin Institute, and Harvard University’s Centennial Medal.

Landauer died in 1999 of brain cancer. He was remembered by his family as a deeply loyal and dedicated father, husband, brother and son-in-law. His peers recalled him as a truly original thinker who forever changed the way that physicists approached their fields.