Stefan Franzen is an intense man. He rattles off ideas and facts in machine-gun manner, in person or in early morning emails.
A classical pianist, he plays Bach to relax. He spent three years in Kenya as a Peace Corps volunteer teaching science. He learned two languages there and became so proficient in Swahili that he taught the language at his school. He speaks 10 languages, including Polish and Chinese, which he learned during annual travels to research projects he has established.
His colleagues at N.C. State say hes driven to be a researcher and a mentor.
And he retains the idealism that led him to the Peace Corps. Scientists, he says, must correct the record when they discover error. And professors have a duty to train the next generation of students to become rigorous, ethical scientists.
In 2005 and 2006, a graduate student and a postdoctorate fellow had developed strong evidence that a $1 million grant that Franzen shared with two colleagues was built on a false premise. The proposal, building on a prominent Science journal article, had said the other two scientists had used RNA, the bodys messenger service for genetic information, to create tiny crystals of palladium, an element. The concept promised advances such as extracting hydrogen from water to create an endless source of clean energy.
Franzen had been asked to join the grant proposal. His colleagues Bruce Eaton and Dan Feldheim, who produced the original research, were unwilling to correct the record. But Franzen believed strongly that their published claims to have produced ground-breaking nanoparticles were inaccurate.
The standoff raised significant questions for N.C. State. Verified research misconduct would embarrass the university, could force it to refund grants and would jeopardize future funding.
The feud also brought tension to one of the Triangles most potent economic engines: scientific research.
The Triangle is home to growing alliances among businesses, academics and venture capitalists, one of the reasons President Barack Obama was here last week to announce a major research initiative at N.C. State. Dozens of research-based companies in and around Research Triangle Park thirst for the discoveries and will pay handsomely. But what are they getting for their licensing fees?
For years, this fight stayed below the surface. Since 2006, Franzen has worked to expose it, first within the university and now to the general public. The scientists whose work is in question Eaton, Feldheim and Lina Gugliotti are silent.
An angry exchange
In October 2006, Franzen resigned from the project, writing a three-page, single-spaced letter to Feldheim and Eaton. Every finding was wrong, he wrote: The hexagonal crystals arent palladium; RNA isnt necessary for the hexagons; and the hexagonal crystals arent even crystals. Evolutionary Chemistry, Eatons trademarked phrase, was not panning out.
These findings were completely unexpected when he embarked on the ambitious project, Franzen wrote.
None of the people who have put effort into this can understand why you, as our collaborators, would simply ignore data, Franzen wrote. But you have ignored it.
Franzen was blunt: It would be unethical for a scientist to advise students and postdocs to continue work that should be retracted. Franzen urged them to retract the Science article and two others.
Feldheim responded the following day.
To say you have been ignored or that Bruce and I have not taken this seriously is incorrect, Feldheim wrote. However, I feel that a correction or quick retraction of all of our papers without looking into this deeply to understand what is going on would not be very smart.
The following day, Feldheim provided images from an electron microscope long requested by Franzen. I would normally find it difficult to hand over my data to someone who has called me a cheat and a liar, but what the hell, you have seen these at least twice before, Feldheim wrote in an email.
The images that Feldheim used to support the articles included two slides that he said led to the statement about the formation of palladium crystals. The first was from Gugliottis original research. The second was a composite image.
So my students may only report what they think I want to see, Feldheim wrote, I may have misinterpreted the diffraction, or maybe these complexes are just weird. But I am not a liar or a cheat.
Franzen was not satisfied. The first image was among those that were not indexed, meaning that the microscope had not been calibrated properly. And Feldheim would not provide him with the original data involving the second image, which Franzen acknowledged could be palladium.
The matter came to a head at a meeting on Oct. 17, 2006. Franzen handed out a short manuscript that would correct errors in the 2004 Science article.
Feldheim reacted angrily and responded with a legal threat, Franzen said. Feldheim claimed the correction relied on a chemical formula made in his laboratory. The formula, closely related to the RNA secret sauce used in the experiment that led to the article, was the property of SomaLogic, the company Eaton had joined in Colorado.
Feldheim said SomaLogic would sue N.C. State if the correction were submitted, documents show.
Feldheim did not discuss the scientific issues at the meeting, Franzen said. Instead, he accused Franzen of misusing grant money. Feldheim also said a scientist at the Lawrence Livermore National Laboratory had independently verified the work. He wouldnt identify the scientist.
Soon after the meeting, Franzen formally accused Feldheim and Eaton of academic misconduct and asked N.C. State to conduct an investigation. The university appointed a panel of five professors to look into the claim.
Calling in the lawyers
The dispute soon spilled beyond the pages of scientific journals into the offices of lawyers employed by N.C. State.
Franzen had to get permission from the universitys legal counsel before proceeding with the correction to Science. Letters flew between lawyers for the university and for Eaton and SomaLogic. (Asked by The News & Observer for the correspondence, N.C. State refused or provided letters with the text completely blacked out, saying the letters were part of the personnel files of Eaton and Feldheim.)
After Franzen sent his correction to Science, Feldheim and Eaton submitted a rebuttal. Their argument against Franzen was mostly scientific, yet sprinkled with adjectives typically not found in science journals: misguided, ill-informed, technically imprecise, erroneous, brazen and illogical.
After having two referee scientists examine the exchange, Science chose not to run the correction. According to one referee, much of this seems to be a personal attack on Feldheim and Eaton. One said that Franzens approach was unconscionable and arrogant, while the other said that the proposed correction submitted by Franzen makes these authors look like amateurs.
Franzens submission covered his attempts to duplicate the experiments. He did not reveal that the images from Gugliottis research were not indexed. Unindexed images are useless random patterns.
Franzen rewrote the correction into a full article published in December 2007 in perhaps the most prestigious journal for chemists, the Journal of the American Chemical Society.
The article drew a scathing response from Feldheim, who had left NCSU in summer 2007 to join Eaton at the University of Colorado. Feldheim repeatedly labeled Franzens work as false (four times), deceptive (six), misleading (three), fictitious (twice) and demanded that the journal retract the article. It did not.
His top complaint was legal, not scientific: Franzen and his colleagues had analyzed a material that did not belong to them.
Eaton soon followed with a similar legal threat. A Colorado lawyer accused N.C. State of infringing on patents owned by Eaton and SomaLogic and demanded that the research immediately stop. The lawyer accused Franzen of using the patented compounds to attack the integrity of Eaton and his published research.
On June 30, 2008, the university investigative committee submitted its report on the 2004 Science paper to the NCSU administration. The investigation had a three-part test:
• Was there falsification? Yes. The paper had falsely asserted that the hexagonal particles were crystalline palladium, the committee ruled.
• Was the conduct a significant departure from accepted practices? Yes. The scientists had failed to identify the exact makeup of the particles through indexing, the committee said. The report quoted Donovan Leonards reaction upon first seeing the images: My jaw hit the floor at this point.
• Was the falsification knowing or intentional? No. The committee found that Lina Gugliotti was a novice in the use of a key electron microscope. The committee said the brunt of the responsibility fell on Feldheim, who should have been aware of the importance of proper indexing and failed to thoroughly consult with experts.
By a 4-1 vote, the committee said Feldheims behavior did not rise to the level of recklessness.
The university forwarded the investigative report to the National Science Foundation and urged Feldheim and Eaton to correct their errors in the Science article. The NSF launched an investigation.
Franzen: not enough
Franzen was dissatisfied with the N.C. State investigation. He was interviewed, and he asked to present evidence but was not allowed to point the committee to where he believed they would find evidence of fraud. He asked the university to let him look at all material it gathered. In particular, he wanted to see the lab notebooks of Gugliotti, who conducted the experiments for the Science paper.
He also turned to Eaton and Feldheims employer. On Aug. 6, 2008, he sent the University of Colorado a seven-page letter outlining what he labeled as four years of misconduct.
The entire claim that RNA can produce metals is a fraud, he wrote, asserting that Feldheim and Eatons litigious behavior betrayed financial motives, not a search for scientific truth.
The falsified solubility and composition are only the manifestations of a fraud that was perpetrated with a profit motive, he wrote.
In December of that year, the Office of Research Integrity at the University of Colorado disagreed: Given the thorough evaluation completed by the North Carolina State Investigation Committee and its exoneration of Professors Eaton and Feldman (sic) of any and all charges, the Inquiry Subcommittee unanimously concluded the two allegations were unfounded.
Similarly, Franzen tried to fix the errors in Science and other journals. The responses ranged from sympathetic to curt, but none agreed to correct the record.
Franzen then tried to get N.C. States help. Terri Lomax, vice chancellor of research integrity, drafted a letter asking Science and the other journals to retract the articles or demand that the authors correct them.
NCSUs lawyers forwarded the letter to Feldheim and Eaton for comment. NCSU never sent that letter to the journals; Franzen said they got cold feet after Eatons lawyer threatened to sue. NCSU declined to provide letters from Eatons lawyers, saying they were part of Eatons personnel file.
In October 2009, the university sent a shorter letter to the journals pointing to two errors in the articles. Lomax said the process that led to the changes in the letters was routine.
Franzen made a formal request under the states public records law to see Gugliottis lab notebooks; university lawyers turned him down.
In fall 2009, he spent $1,334 of his own money to hire Mike Tadych, a Raleigh lawyer who specializes in public records law and who has represented The News & Observer. In 2010, the university relented and allowed Franzen into the room where the investigation records were locked away.
Franzen found the lab notebooks, which track experiments and results. As he turned the pages, he recognized that Gugliotti kept a thorough and well-organized record.
I found an open-and-shut case of research fraud, Franzen said.
The aqueous solution mentioned in the Science article? The experiments routinely used 50 percent solvent. The experiments only produced the hexagonal crystals when there was a high level of solvent, typically 50 percent or more. It was the solvent creating the hexagonal crystals, not the RNA.
On Page 43 of notebook 3, Franzen found what he called a smoking gun.
Gugliotti had pasted four images of hexagonal crystals, ragged around the edges. The particles were degrading at room temperature. The same degradation was present in other samples, she noted.
Palladium has a melting point of 2,831 degrees Fahrenheit. The metal, like gold or platinum, does not flake and fall apart at room temperature.
The page was dated Nov. 7, 2002, 19 months before the publication of the Science article.
Franzen said he does not blame Gugliotti. He said Feldheim and Eaton had the responsibility to know that the degrading crystals were proof that the hexagons were not palladium before the article was published in Science.
A secret meeting
Franzen shared his findings with his chemistry department colleagues, who were concerned at what they saw.
In spring 2011, the senior faculty met with Matt Ronning, a research integrity officer. Ronning had been Franzens contact person for the misconduct complaint, stuck among Franzen, university lawyers and the research integrity office. Franzen gave this account of the meeting:
Ronning began by saying the meeting was secret.
Ronning said he had consulted with administrators at the highest level of the university. They were prepared to revoke Gugliottis doctoral degree and that of another researcher if that would satisfy the department and end the matter. Faculty members objected. They said Eaton and Feldheim should be held responsible, not their students.
Ronning said the process was over once the university finished the investigation in 2008, according to Franzens account. Some professors requested a copy of the investigation; Ronning refused.
The chemistry department formally requested a copy of the investigation from Chancellor Randy Woodson. In a letter to the department chair, Woodson declined to provide it, saying a National Science Foundation investigation was ongoing.
Ronning and Woodson declined to discuss the case with The N&O.
The NSF investigation into research fraud is still unfinished after more than five years. Franzen and colleagues say it is the longest-running investigation in its history.
In the heated meeting of October 2006, Feldheim had referred to an ace in the hole; he said their work had been replicated by a Lawrence Livermore scientist. That scientist was James DeYoreo, who published a 2008 paper with Feldheim and Eaton. At Franzens urging, DeYoreo later published a correction: The solution was 50 percent solvent, not 5 percent, and authors werent sure the so-called palladium crystals were metallic.
In 2013, DeYoreo and Franzen published a paper in Particle, a science journal, reporting that RNA has no apparent role in forming the hexagonal crystals from the palladium compound. It was Franzens ninth paper debunking the notion that RNA could create metallic crystals.
The 10th was published in August in the Journal of Materials Chemistry, whose editors asked Feldheim and Eaton to comment before publication. They declined.