Reaching Others University at Buffalo - The State University of New York
Skip to Content
Official UB news and information for the media

Filling Fragments Can Identify Human Remains, Forensic Dentists Show

By Lois Baker

Release Date: October 3, 2006

BUFFALO, N.Y. -- When an explosion, accidental cremation or a fire set deliberately to cover a crime destroys a body, precious little may remain to link it to a life once lived.

Yet even among the ashes, a team of forensic dental researchers at the University at Buffalo has shown that evidence exists that can help identify human remains when all else -- flesh, bones, teeth, DNA -- is lost.

The evidence can be hard to recognize, but it is distinctive.

In a series of experiments reported in the May 2006 issue the Journal of Forensic Science and in an upcoming article in the same journal, the researchers show for the first time that inorganic resins that make up the central matrix of tooth-colored dental fillings can withstand temperatures of 1,800 degrees Fahrenheit, be recovered and named by brand or brand group.

Even when only fragments of resin could be found, the researchers were able to classify the composition of elements in the filling. Comparing those elements and their proportions to the composition of the known filling brands recorded in a deceased's dental chart could, under the best circumstances, help identify the remains unequivocally.

At the very least, the analysis could determine if the filling material was or was not consistent with a person's dental records.

Mary A. Bush, D.D.S., said this new type of evidence could have a major impact on forensic dentistry.

"To date, no one has recognized that many modern restorative resins have unique characteristics that can be distinguished and used for forensic identification," said Bush, assistant professor of restorative dentistry in the School of Dental Medicine at the University at Buffalo and lead author on the studies.

"Nobody has applied the standard analytical methods that we have at UB to survey these materials and determine these properties." 

Peter Bush, director of the instrument center where much of the research analysis was conducted, was a major contributor to the research, along with Raymond Miller, D.D.S., UB clinical assistant professor of oral diagnostic sciences and a forensic dental expert, and Jennifer Prutsman-Pfeiffer, anthropologist and UB doctoral student.

The team's work has yielded unexpected rewards. The FBI has offered to include the information in their database, and the American Society of Forensic Odontology provided a grant to help assemble the data.

"The importance of identifying these properties is, first, to show that it can be done," said Bush, "and second, that it can be done even after extreme events such as mass disasters, plane crashes or explosions," or a murder.

The 1999 trial of Donald Blom, accused of killing Katie Poirier after abducting her from a Minnesota convenience store, demonstrated the usefulness of such forensic evidence. Blom confessed to the crime, but later recanted. The body never was found, but human bone fragments and a single tooth were unearthed in a burn pit on Blom's vacation property. Analysis of the components of the tooth's filling material matched the brand of filling recorded in the victim's dental records. That evidence helped put Blom in prison for life.

Bush and colleagues began their experiments in mid-2005, using UB's specially equipped instrument center, which includes a scanning electron microscopy/energy dispersive X-ray spectroscopy equipment, known as SEM/EDS, and a portable X-ray fluorescence (XRF) unit to conduct material analysis outside the lab.

They had access to cadavers for the second research phase through the UB School of Medicine and Biomedical Sciences' Anatomical Gift Program, to which persons donate their bodies for use in teaching and scientific research.

Initial experiments were carried out with teeth only.

The investigators created disks of 10 different resins used for standard tooth fillings to serve as controls, then filled extracted teeth with the resins and incinerated them in an oven at 900 degrees Centigrade (1,652 degrees Fahrenheit) for 30 minutes.

These conditions were more extreme than in a standard cremation, Bush noted, because teeth normally would be protected by flesh and bone, allowing them to withstand the high temperature for a longer period of time. With no such protection, the extracted teeth fragmented in half-an-hour.

Dental resins consist of an organic matrix surrounding inorganic filler particles. "At these high temperatures, everything organic is destroyed," said Bush. "It was the inorganic material that was recoverable."

After retrieving the resins fragments, the team analyzed their elemental composition using SEM/EDS. In the May 2006 issue of Journal of Forensic Science they reported they were able to identify the concentration and microstructure of the inorganic elements in the fragments and link them to the specific brand or brand group of the material documented in the controls.

"Not only do these materials have various microstructures," said Bush, "they also have unique elemental compositions, which makes it possible to distinguish between brand or brand groups. We showed that the elemental distinction remains even after extreme conditions such as cremation."

To create a true-to-life scenario, the team worked next with cadavers donated to the medical school's Anatomical Gift Program. (Full approval from the university's Human Subject Review Board was obtained for the study.) They removed all existing resin fillings from the teeth of six cadavers and replaced them with a total of 70 fillings representing five different resin brands. The filling brands used were recorded in each cadaver's dental record.

With the new fillings in place, the bodies were put through the standard two-step cremation process: very high heat (1,800 degrees Fahrenheit) for two and a half hours, which destroys all flesh and small bones, then crushed in a grinder and reduced to ashes.

Bush and colleagues were able to find and identify enough of the resins to make a positive identification of each cadaver, using the portable XRF unit to mimic investigations that need to be conducted in the field.

The results of this study will appear in the online version of the Journal of Forensic Science in December 2006 and will be published in the January 2007 print issue.

"Even in the ashes, we were able to retrieve small pieces of resin and distinguish between cadavers," said Bush. "To my knowledge, this is the first time this type of analysis has been done. This study provides hope of identification when little hope may be present.

"If an individual isn't burned to this extreme and the teeth are intact, but the dental X-ray comparison is questionable or teeth are fragmented, this type of analysis can give another level of certainty on which to base an identity," she said.

XRF doesn't provide as much information as the lab-based SEM/EDS equipment, Bush noted, but its speed compensates for lack of precision. The device can identify the chemical spectrum of elements in inorganic material in 6-10 seconds, providing quick on-site screening of suspected material.

The ability to distinguish between resins gives investigators a new tool for use in special circumstances, Bush said.

"Retrieving small amounts of resin as we did in this study would not carry as much weight for identification as a dental chart comparison, but the evidence was indisputable and unequivocal. This evidence would serve as an aid in identification when very little other evidence exists or when added scientific corroboration is needed."

Bush and her co-investigators currently are working with the FBI to construct a database of the most common brands of dental restoration materials and their elemental composition for use in criminal investigations.

"There are more than 50 filling materials on the market today," said Bush. "We have analyzed the 30 most popular resins and 23 historical resins dating back to 1971. There are also many other unique dental materials -- posts, cements, crowns, sealers -- that also will be included in our database. Again, no one else has attempted such a comprehensive survey of their properties."

The database does have limitations: It will be useful only if dentists document all dental restorations, including brand names, in their dental records, noted Bush.

The UB researchers will have a role in bringing that point home to their colleagues and dentists of tomorrow through the new Laboratory for Forensic Odontology Research (LFOR) housed in the UB dental school. Bush will direct the laboratory, with Miller as co-director.

The laboratory will host demonstrations and seminars on forensic dentistry for students, and present lectures and continuing education programs in forensics for practicing dentists.

The knowledge they gain may help deter any future "Donald Bloms."

The University at Buffalo is a premier research-intensive public university, the largest and most comprehensive campus in the State University of New York. The School of Dental Medicine is one of five schools that constitute UB's Academic Health Center.