18 May 2007
Sonofusion: So, No Fusion Then?
By Rusty Rockets
Where should the responsibility for keeping research institutions honest lay; the internal structures of the institutes themselves, or intervening federal agencies? Anyone with a grasp of the scientific method may argue that an institute practicing good science will already have the checks and balances in place. But a controversy over sonofusion, or bubble fusion, that surfaced at Purdue University seems to indicate otherwise.
In 2002, Rusi Taleyarkhan, the Arden Bement Jr. Professor of Nuclear Engineering at Purdue, and former Distinguished Scientist at Oak Ridge National Laboratory (ORNL), released a paper that detailed a successful sonofusion experiment conducted by Taleyarkhan and a number of his colleagues. In the paper, Taleyarkhan reported results that suggested the possibility of nuclear reactions during the high-energy, luminescent collapse of bubbles in liquid, a process known as cavitation. The sonofusion process works - at least in theory - by bombarding a mixture of acetone and benzene with oscillating sound waves, causing bubbles in the mixture to expand and violently collapse. If neutrons are produced during this process, then this is evidence that a nuclear reaction has taken place; and Taleyarkhan made the claim that this had in fact occurred.
But no other scientist was able to replicate the sonofusion process and Taleyarkhan was accused of what can only be referred to as scientific fraud. The initial challenge to the experimental results came from critics who suggested that the neutrons were in fact mere "leftovers" from an external neutron source used during the experiment. In 2006, however, Taleyarkhan and his team seemed to have answered their critics by redesigning the experiment using natural uranium in the solution, which instead produces bubbles through radioactive decay. This time Taleyarkhan used three independent neutron detectors and one gamma ray detector to seek out any nuclear emissions, with all four detectors producing a positive result. But that wasn't the end of either Taleyarkhan's or Purdue's problems... not by a long shot.
By this stage the damage was already done, and accusations regarding why Purdue allowed the allegedly dodgy research to continue under their auspices began to emerge. Most of the allegations stemmed from interviews with a number of Taleyarkhan's colleagues at Purdue, who revealed some of his experimental methods. In 2006, Nature magazine described aspects of Taleyarkhan's work which included; "claiming he obtained positive results from equipment on which they had seen only negative data, and removing the equipment from their lab altogether."
Equally worrying, graduate student Adam Butt, who appears as a co-author on one of Taleyarkhan's papers, claimed that while he may have been a member of Taleyarkhan's research group, he knew nothing about the experiment he was supposed to have co-authored.
Then, Nature reported that Brian Naranjo, from UCLA, had demonstrated that the nuclear fusion Taleyarkhan claims to have occurred came instead from the radioactive decay of standard lab materials. Following such damning indictments, it's little wonder that the consensus among Taleyarkhan's peers seems to be that the fusion bubble - at least for now - has burst.
But Purdue's troubles didn't end there, as scientists and members of congress began to question the university's reasons for appearing to protect Taleyarkhan from proper scrutiny. In June 2006, Dr. Kenneth S. Suslick, from the University of Illinois at Urbana-Champaign, wrote to Purdue University's associate vice president for research, stating that he believed Taleyarkhan's work might constitute "scientific misconduct." Attempting to emerge from the controversy relatively unscathed, Purdue has since been in damage control, setting up their own internal inquiry; an inquiry that was to eventually clear Taleyarkhan of any misconduct.
But in a recent letter addressed to Purdue President Martin C. Jischke, Brad Miller, chairman of the investigations and oversight subcommittee of the House Committee on Science and Technology, claims that the university was sluggish in its duties to follow proper investigative procedure. Like most other universities, Purdue's system for handling allegations of research misconduct (which federal policy defines as: "fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results") involves a two-step process, which includes an inquiry committee to hammer out the terms of the allegations, followed by a full investigation. Miller says that Purdue was laissez faire about two-step process. As a result, the Congressional subcommittee staff wrote: "It is difficult to understand how the inquiry committee could have then decided that Dr. Taleyarkhan's actions did not constitute research misconduct," which Purdue policy constitutes any "serious deviation" from accepted research practices.
"Purdue is a premier research and educational institution," Miller wrote in conclusion to his letter. "One of its goals is to teach the importance of scientific integrity to its students and future scientists. I sincerely hope that the next inquiry will be conducted in a manner worthy of your great institution." Almost immediately, Purdue officials produced a press release stating that they would be happy to take on Miller's challenge, and fully address the concerns that he raises. One can't help but get the feeling that all parties concerned are in it for the long haul.
Ideally, scientific research should be self-correcting - if it can't be replicated, it probably isn't science - but something seems to be going on in research institutes that warrants the introduction of a research misconduct policy, implemented by federal organizations. Purdue is probably by no means the only institute where questionable practices have surfaced, but the case is an interesting one, as it reveals a level of government-academic interaction not normally seen by the public.
Sure, scientific methodology is by no means as cut-and-dry as popular culture would have us believe - often requiring a combination of imagination, unorthodox methods, and serendipity - but without working toward a certain level of integrity, science just becomes smoke and mirrors. In this respect, federal research misconduct policy is one aspect of government intervention that shouldn't be pooh-poohed. As Purdue spokesperson Joseph Bennett said in a press release; "The Purdue administration cannot answer the ultimate question [of] whether sonofusion works. Only honestly conducted and reported laboratory research and debate in the scientific community can answer that question." You said it!