Fishy findings sustain, then snuff, stellar career. (Flame Out).

Rare in physics, accusations of scientific fraud have sullied the field for the second time in months.

Investigators have concluded that a young, up-and-coming physicist in fields ranging from molecular electronics to superconductivity repeatedly faked data and committed other types of scientific misconduct.

The misdeeds of Jan Hendrik Schon, 32, of Lucent Technologies' Bell Labs in Murray Hill, N.J., have tainted 25 scientific articles, states a 127-page report issued last week. The report summarizes a 4-month investigation by a five-member committee convened by the company (www.lucent.com/news_events/research review.html). The panel includes scientists at Bell Labs and elsewhere.

The report notes that Schon violated accepted scientific practice by not keeping records of his experiments and by deleting related computer files. Lucent fired Schon shortly after the committee submitted its report, a company official said.

"This is the most extreme case of data manipulation that I've ever heard about in physics," says Arthur P. Ramirez of Los Alamos (N.M.) National Laboratory, where he built a research program inspired by the now-discredited Lucent findings.

Editors at Nature and Science, two leading scientific journals in which many of Schon's findings appeared, say they're in discussion with the coauthors of the studies about retracting the results.

Schon came to Lucent in 1997 from the University of Konstanz in Germany to work as an intern under renowned superconductivity researcher Bertram Batlogg, now at the Swiss Federal Institute of Technology Zurich. The young newcomer quickly became a star of condensed-matter physics as he and his colleagues purportedly carried out a string of remarkable and widely acclaimed feats.

His seeming achievements included single-molecule transistors (SN: 12/8/01, p. 367), plastic transistors in which current could flow remarkably well, and transistors in which certain substances could be transformed by the flick of a switch into superconductors--materials that permit electricity to flow with no resistance.

When those findings were coming out, "I thought this work was the best stuff I'd seen in my scientific career," recalls Paul L. McEuen of Cornell University. "That view was widely held."

However, McEuen, Ramirez, and many other scientists had trouble reproducing the Lucent group's results. "I thought they had a secret they weren't divulging," says Robert C. Dynes, a physics professor and chancellor at the University of California, San Diego. The unexpected nature of that secret is now coming to light.

In written responses to the committee, Schon acknowledges "that I have made various mistakes in my scientific work, which I deeply regret." Nonetheless, he contends, "all the scientific publications that I prepared were based on experimental observations."

Despite having numerous collaborators, Schon almost invariably conducted key measurements and prepared publication figures alone. The investigating committee exonerated all of Schon's coauthors of any misdeeds. However, it questioned whether they--Batlogg, in particular--paid enough attention to what was being presented under their names.

The Schon affair closely follows an uproar this summer at Lawrence Berkeley (Calif.) Laboratory, where physicist Victor Ninov was charged with fabricating data in the subsequently retracted discovery of new heavy elements (SN: 7/20/02, p. 37).