The problem had haunted Honda for years: Why were airbags made by Takata exploding in its cars, sending metal fragments flying into the passenger cabin?
By 2010, Honda had recalled more than 800,000 cars to fix the airbags after at least one death. It turned out to be the early stages of what is now one of the largest automotive recalls in American history, involving 12 automakers and at least 19 million vehicles.
Honda assured regulators that the airbags’ design was not to blame, echoing an explanation from Takata, one of the biggest makers of airbags worldwide, that pointed to isolated manufacturing problems.
But behind the scenes, officials from Honda and Takata were exploring another possibility, according to employees of both companies.
They wanted to determine whether the propellant used to inflate the airbags might be at the root of the explosions. Without alerting regulators or the general public, Takata commissioned one of the nation’s most respected pyrotechnics labs, the High Pressure Combustion Laboratory at Pennsylvania State University, to study the compound, ammonium nitrate.
As a condition of the study, the Penn State lab was not permitted to publicly link the research to Takata or Honda, according to accounts that were confirmed by Takata. And when the findings were published in a scientific journal, the lab was forbidden to disclose that Takata had paid for the study. Those omissions violated the journal’s statement of ethics.
And when the study’s conclusion in 2012 cast doubt on the use of ammonium nitrate, suggesting it was sensitive to changes in pressure, Takata disputed the methodology, dismissed the conclusion and waited more than two years before sharing the research with regulators, according to accounts confirmed by Takata.
The role ammonium nitrate might play in the exploding of the airbags is still unknown. And federal officials say Takata and Honda broke no laws in keeping the study quiet because the airbags were not the subject of a formal inquiry during that period. An earlier safety investigation into the airbags had been closed, and the current investigation was not opened until late last year.
But given the long history of problems and the seriousness of the exploding airbags — they are now linked to at least eight deaths and more than 100 injuries — some former regulators and safety advocates sharply criticized the companies’ silence.
“This fits into a pattern that they have been involved in from the beginning — to deny, delay, defer and blame other people,” said Joan Claybrook, a former administrator of the National Highway Traffic Safety Administration and a longtime advocate of auto safety. “Takata has gone a long way to try to suppress information about this disaster with these bad airbags, and this is an important piece of it.”
Today, Takata continues to maintain that ammonium nitrate is safe when properly handled, and it uses the compound in airbags for new cars, as well as in replacement airbags in recalled vehicles. A company spokesman said it was wrong to make too much of the Penn State study, which he described as an “inaccurate and incomplete” depiction of the propellant.
“Takata and Honda pursued the research at Penn State to make sure we weren’t missing something in our analysis,” Jared Levy, a spokesman for Takata, said in written responses to questions from The New York Times. “Takata was doing testing internally, but wanted to seek other experts’ opinions regarding the root cause.”
Honda also defended the delay in submitting the report. “In light of the fact that the Penn State’s findings were preliminary in nature, and presuming that the study would continue, Honda believed it was premature to submit these findings as conclusive,” Chris Martin, a spokesman, said in a written statement.
The lab at Penn State has stood by its research, which was overseen by its founder, Kenneth K. Kuo. The findings were published in 2012 in a journal edited by Mr. Kuo, The International Journal of Energetic Materials and Chemical Propulsion.
Mr. Kuo declined to comment for this article. Another researcher on the study, Eric Boyer, said in a statement that the findings were not meant to be “a universally applicable indictment or confirmation of anything.”
“We hope they contribute to the body of safety knowledge on this issue,” the statement said.
A Penn State spokeswoman said that the university allowed financial sponsors of research to remain anonymous, suggesting it was common practice for major research universities.
The commissioning of the Penn State study, which has not previously been reported, conflicts with public statements made at the time in regulatory filings by Takata and Honda about their internal investigations into the cause of the airbag problems. Both companies minimized the scope of the defect in their interactions with regulators, even as they suspected that a more fundamental problem with the airbags’ design might be at fault, according to the employees.
Apart from the Penn State study, Takata asked another researcher, Vigor Yang, a professor at the Georgia Institute of Technology, for ideas for a new propellant that could replace the ammonium nitrate in its airbag inflaters. Those discussions, in 2010, did not lead to a formal research project, Mr. Yang said.
Eventually, late last year, Honda, under an order by the National Highway Traffic Safety Administration to produce all relevant documents about the defect, turned over the Penn State study to regulators. That came days after Honda had vastly expanded the airbag recall to cover the entire nation and two weeks after lawmakers had sharply questioned a Takata executive at a contentious congressional hearing.
On Thursday, the safety agency will hold a public meeting in Washington to discuss Takata’s defective airbags and the plan to repair them.
The roots of the Penn State study date at least to the spring of 2010, when representatives from Takata and Honda met with researchers at the Penn State lab. The following year, they signed an agreement with the lab.
In 2012, the researchers came back with results that troubled Takata: Rapidly changing pressure inside an airbag’s steel inflater, the study concluded, could trigger a state known as “dynamic burning,” which in turn could lead to excessive internal pressure and potentially cause the inflater to rupture.
In June of that year, Takata told the researchers that the lab’s tests and data did not match the performance of the airbags in the field. Part of the problem, Takata argued, was that the lab had used propellant manufactured to specifications that were different from the propellant in real-world airbags.
If the lab’s findings were true, Takata’s engineers argued, there would have been far more ruptures than the company was aware of, according to a written account that Takata provided to The Times.
The lab made some changes to the final report, Takata said, but stuck with its focus on dynamic burning. A version of the study was published later that year.
Still unhappy with the findings, Takata, with Honda’s support, urged the lab to carry out further testing, under conditions that more closely mimicked those in real inflaters. They agreed to 10 months of follow-up tests.
But the follow-up research ended in a disagreement over what Takata called its slow pace.
“The Penn State report was inaccurate and incomplete. We asked Penn State to conduct more research, but they did not complete it,” said Mr. Levy, the Takata spokesman. “We also provided the report to two outside experts, both of whom said it was flawed.”
The outside experts were also paid by Takata.
One of those experts was Jochen Neutz, deputy head of the energetic systems department at the Fraunhofer Institute for Chemical Technology in Germany. Dr. Neutz, made available to The Times by Takata, said in an interview that he continued to disagree with the Penn State study and that his lab’s own analyses did not support the study’s findings.
Takata also reached out to Dr. Yang, the head of the School of Aerospace Engineering at Georgia Tech, who also said at the time that the study was flawed. The lab’s analyses, he said, were conducted under conditions vastly different from those in the manufacturer’s inflaters. But in an interview, Dr. Yang acknowledged that he could not rule out the possibility that dynamic burning could cause ruptures in airbags.
Ms. Claybrook, the former safety administrator, was skeptical of Takata’s actions.
“This doesn’t pass the taste test,” she said. ”They hired researchers to produce a study. They didn’t like the result, and then they hired more people to look at that study.”
Dr. Yang said Takata also asked him in 2010 for ideas for a new propellant.
“They asked me to look into future propellants. They specifically asked me, ‘What kind of propellant do you see in the future?’” Dr. Yang said. Takata said it was looking for a propellant alternative that was “controllable, low cost and reliable,” he said. “They shared their dream list” of properties, he said.
That discussion did not result in funded research, Dr. Yang said, and he did not put forward any concrete suggestions.
Takata said it was always looking to improve its products. “Takata regularly examines new technologies and consults experts in the field about advancements across all of its products,” Mr. Levy said.
But it was only in May, in connection with a congressional hearing, that Takata publicly acknowledged that its engineers were looking into new types of propellant. At the same time, under pressure from regulators, Takata also admitted that the design of the airbags was defective.
Even today, after millions of cars have been recalled over the last seven years, Takata and automakers are struggling to understand the scope of the problem. Last week, for example, General Motors notified regulators that it would recall about 400 cars from the 2015 model year that have Takata airbags containing ammonium nitrate.
Written by Danielle Ivory and Hiroko Tabuchi of The New York Times
(Source: The New York Times)