A Trust Betrayed in Physics

It seemed almost too good to be true.

Late last year, scientists at Bell Labs discovered a way to construct a transistor out of a single molecule. The man behind it all was Jan Hendrik Schon, a 32-year-old physicist who seemed destined to win a Nobel Prize one day.

But Schon’s discovery was exposed as a fake Tuesday, and now his case is raising troubling questions about how well the community of physics--whose work is the foundation of key military, industrial and commercial developments--can police itself.

Schon’s dismissal from Bell Labs casts a shadow on what many consider the fastest-moving research area in science today--a field that holds the promise of entirely new approaches to computing and new industries to capitalize on them.

Physics, like all fields of science, is an enterprise that traffics in facts but is sustained by faith in its own integrity.

The Schon case is the second major instance of research misconduct to come to light this year. In July, officials at the Lawrence Berkeley National Laboratory admitted that the stunning discovery of a new element was fabricated.

Some experts say many physicists are too trusting or simply unwilling to look too closely at research practices. The peer-review process meant to guarantee scientific integrity is too much the paper tiger, they say, too easily influenced by a researcher’s reputation.

The extensive pattern of misconduct in the Schon case also highlights the responsibility of other scientists who lend their names as co-workers to the research papers that are published. They share in the discovery and take some credit for it, yet often they may not be able to vouch for all the work that went into an experiment.

“The physics community may be less on guard and more willing to take results on trust,” said Yale University science historian Daniel Kevles, who has written extensively on research misconduct.

Mostly, however, scientists are asking themselves just how Schon thought his fabrications would withstand sustained scrutiny.

“I am absolutely mystified about what made him think he could get away with such a thing,” said Caltech physicist and Vice Provost David Goodstein. “This is high-profile stuff that people would want to build on to make new advances.”

Schon had been recruited to join Bell Labs by Bertram Batlogg, one of the most prominent physicists in the world. Batlogg, then head of Bell Labs’ materials physics research department, was looking for an experimental wizard to study the possibility of building organic transistors--vanishingly small electronic components based not on silicon but carbon.

These organic, molecule-sized components were seen as the next breakthrough technology to replace silicon circuits, which are approaching their physical limits for miniaturization. Chips made of organic materials, like plastics, could lead not only to vastly more powerful computers but also a host of new uses, such as flexible electronic displays.

The problem was trying to make something on such a small scale. Many researchers had tried and failed miserably.

But Schon eventually began publishing breakthrough papers that described a method of transforming organic crystals into electronic components.

At the height of his success barely a year ago, almost everybody at Bell Labs admired Schon’s modest and unpretentious style.

They lauded his deep grasp of condensed-matter physics and solid-state devices. He worked long hours in the lab too, they said, always wiring up new samples of the exotic materials that were the subject of his experiments.

Most of all, they were dazzled by his ability to write such clear, coherent research papers at such a remarkable rate--one paper every eight days last year. He and his colleagues churned out more than 90 papers since 1998.

His findings sent an entire field of physics buzzing. The company that employed him did not shy from trumpeting his success. “This work pushes the miniaturization of electronics to its final frontier,” Federico Capasso, a Bell Labs vice president, said in a news release on one of Schon’s discoveries.

His most startling discovery was the creation of a transistor made of just one molecule.

The device was built by cooking organic molecules in a beaker and pouring them onto a layer of gold atoms. The molecules assembled themselves into a pattern allowing them to function as transistors, the key components of modern electronics. A top layer of gold atoms completed the construction.

This and all his findings were peer-reviewed by panels of other scientists and published in the world’s leading research journals, such as Science and Nature.

Indeed, one of his papers published in Science in February 2000 was one of the most cited physics papers that year, according to the Institute for Scientific Information.

“I was blown out of my chair by his work,” said UCLA chemistry professor James Heath, a leader in the field of nanostructures, or structures on a molecular scale. “I thought it was just amazing. He had a pretty grand run for two years.

“We tried to do experiments inspired by his findings,” Heath said. “Our results were not as inspired as his. Groups tried very hard to reproduce these results and fell on their face trying to do it.”

No one was trying to disprove Schon’s work at first. Rather, they sought to emulate and build on it. A new field of molecular computing was taking shape and everyone wanted to be among its founders.

Lab after lab tried and failed to duplicate the experiments, however, and the buzz about Schon’s brilliance took a disquieting turn starting in the summer of 2001.

Then several researchers noticed that the graphs published in support of several different experiments were suspiciously alike.

By last spring, awe had turned to doubt. The whispers gave voice to formal allegations of fraud.

All universities that conduct publicly funded research are required to have policies and procedures for investigating research fraud. Bell Labs, however, is a corporate enterprise.

There appears to have been no public money involved in Schon’s work, several researchers said, so federal fraud investigators had no reason to become involved.

In May, Bell Labs organized its own committee of outside experts to investigate the problem. Allegations of fraud and misconduct had been leveled at 25 published research papers involving 20 scientists. The questions, however, involved less than a quarter of Schon’s publications over the past five years. Many of his unchallenged papers also claimed significant breakthroughs.

When the five investigators--themselves prominent scientists and academics--started asking questions, they found that those who had spoken most glowingly of the young researcher acknowledged that they actually knew almost nothing about the experiments he had performed.

Indeed, in a field known for its collaborations and collegiality, Schon made almost all his discoveries alone, they learned. Although his papers bore the names of many scientists, none directly participated in the breakthroughs, the committee determined.

Schon had conducted many of his most controversial experiments not at Bell Labs but at the University of Konstanz in Germany, where he was awaiting his entry visa to the United States.

“Working in isolation is a really unusual thing to do in an industrial lab,” said Heath at UCLA.

In interviews with Schon, the investigating committee members found themselves weighing questions of superconductivity in polythiophene and debating fine points of unipolar inverters, load kinks and ring oscillators.

The Bell investigators eventually found that all his electronic data files had been deleted. More tellingly, perhaps, there were no laboratory records, such as signed notebooks, dated sheets of experimental data or logs, like those most scientists are accustomed to maintaining.

All the hundreds of devices that Schon said he had worked on were supposedly damaged during experiments, broken when being moved to his new lab or simply discarded. He readily admitted that he substituted and altered data to make the results look better. Some of the substitutions, he said, were honest mistakes. But he told the panel: “I truly believe that the reported scientific effects are real, exciting and worth working for.”

There were so many such mistakes, the investigators said, that it showed a “deeper problem.”

Schon could not be reached Thursday to review the report’s conclusions.

Several experts said that Schon’s dismissal is proof that the system of scientific checks and balances is working. The process, they said, caught up with him.

But when a scientific truth is revealed to be a lie in disguise, the impact goes well beyond one researcher’s career.

The known facts of a scientific discipline have to be reordered. Promising avenues of research instantly become blind alleys.

“This revelation of misconduct is quite problematic since a number of individuals have sought to direct their research into what appeared to be a promising direction,” said Harvard University chemist Charles Lieber, an authority in the new field of nanotechnology.

“This could have some negative impact on nanoscience and nanotechnology,” Lieber said. “But overall there is too much good work being done to condemn the field for one bad apple.”