A murder occurs in Collin County, Texas. The suspect tries to dispose of the body by burning it on a pyre made of fire wood.
When detectives learn during the course of the investigation that the suspect had brought logs to a party at about the same time the victim came up missing, they collected ten logs from the murder scene fire and four logs from the party fireplace for comparison purposes. If the logs could be somehow shown to match, it might help place the suspect at the crime scene.
“Our initial plan was to compare the tree-rings from the samples,” said Lt. Larry R. Smart, the lead detective on the case for the Collin County Sheriff’s Department. Tree-ring analysis is frequently achieved simply by counting, dating (to the nearest exact year), and measuring (to the nearest 0.001 mm) the tree rings found in specimens.
Smart contacted the Laboratory of Tree-Ring Science at the University of Tennessee directed by Prof. Henri D. Grissino-Mayer to see if the lab could use tree-ring dating to determine whether the two sets of fire wood came from the same tree. Tree-ring analysis, or dendrochronology, has been a reliable forensic tool since playing a major role in solving the 1932 murder of Charles Lindbergh's infant son.
Now You Tree It, Now You Don’t
“ Unfortunately, when I saw the pieces of mesquite (Prosopis glanduloa) I realized that tree-ring dating would not be practical with this type of wood,” Grissino-Mayer said. Mesquite, the most common shrub or small tree of the Desert Southwest, does not form familiar tree-rings patterns due to the erratic nature of tree growth from year to year, which produces annual rings that are not clearly distinguishable. Grissino-Mayer told Smart that tree-ring analysis in this case would not only be challenging, but time-consuming, costly, with no guarantee of results.
While Smart considered the implications of these obstacles, Grissino-Mayer wondered whether some kind of chemical identification technique could identify the wood samples as having come from the same location. He contacted Suzanne Fisher at the Tennessee Valley Authority in Knoxville, who once had conducted dendrochemical analysis for her master’s thesis. Fisher informed Grissino-Mayer of research scientist Madhavi Z. Martin at the Oak Ridge National Laboratory who had already begun initial studies on chemical properties of wood using a technique known as laser-induced breakdown spectroscopy.
“Maybe we could use this technology to obtain enough information from the chemical composition of the wood to perhaps tie the two sets of wood together,” Grissino-Mayer said.
Laser-induced breakdown spectroscopy (LIBS) itself is not new. For the past 15 years or so, for instance, Los Alamos National Laboratory has conducted research and development on elemental analysis using LIBS technology.
The LIBS technique has many advantages compared to conventional analysis methods that make it particularly suited for field based forensic measurements under harsh conditions. For one, LIBS measurements are generally carried out in ambient air at atmospheric pressure, whereas some analytical methodologies require laboratory vacuum chamber processing. The limits of LIBS detection, however, can be improved when analysis is performed with the aide of vacuum processing, Martin indicated.
“The greatest advantage of LIBS, though, is its capability for remote chemical analysis of samples with minimal handling and little or no sample preparation, which minimizes generation of waste to the microgram per pulse of ablated material,” Martin said.
The instrumentation and operation of a LIBS system is simpler than some of the more sensitive techniques, and analysis times on the order of minutes make it more amenable for real-time analysis of chemical processes, according to Martin.
Martin and ORNL colleague Stan Wullschleger originally used LIBS for carbon sequestration applications, such as determining the amount of carbon in soil, a procedure in which they point a flashlight sized laser device at a soil sample in the field or taken from the ground and determine how much carbon the sample contains. Eventually, forensic
investigations may also be achieved just as orderly.
The LIBS process is basically quite simple. A brief, powerful laser beam is fired at a spot on a solid, liquid, gaseous, or aerosol sample. The heat induces a fleck of matter in the target to ionize (vaporize), resulting in the formation of a hot spark, or plasma emission, called the bremsstrahlung process. A small spotting scope mounted near the laser source captures light emitted by the plasma and directs it to a spectral analyzer. Since spectral emissions from ionized, neutral, and molecular species occur sequentially after plasma is formed, by analyzing this light it is possible to deduce the elemental composition of the target substance.
Moreover, the proportionality of spectral line intensity to elemental concentration enables quantitative analysis.
The ability of LIBS to provide rapid multielemental microanalysis of samples with little or no sample preparation has been widely demonstrated. In the last four to five years, LIBS has literally burst on the scientific scene, with widespread applications found from air quality monitoring to archeometallurgy to dentistry to pharmaceuticals. Astronomers have proposed using the technology to determine the composition of the surfaces of other planets, asteroids, moons, and comets. Recently, a bulk explosives application has been identified that may translate into a robust new homeland security and anti-terrorism tool.
LIBS may also become the next forensic Swiss Army Knife.
Forensically, the technique has been used to identify chemical fingerprints in bones and for detecting counterfeit currency. It has also been used to determine whether the hands of a suspected gun user contain traces of gunshot residue.
In gunshot cases, samples are obtained by pressing adhesive tape against the skin of the suspect, then using LIBS to analyze the tape directly. According to a 2003 paper (Applied Optics, Volume 42, October 2003), when the suspect has fired multiple shots, or if the gun has not been cleaned, the gunshot residue provides a spectral signature that is readily apparent, although identification of a person who has fired a single shot from a clean gun is more difficult.
Typically, the detection limit of LIBS for solids is in the low parts-per-million range, although at least one lab (the Industrial Research Institute in Canada) is conducting basic research aimed at lowering the detection limits of LIBS from the parts-per-million to the parts-per-billion range, while also widening its range of applicability.
The Doctor Will Tree You Now
In the Collin County murder case, LIBS essentially produced a chemical fingerprint of the wood from each scene based on heavy metals and other trace elements found in the samples.
“Such elements are not found in abundance in the natural environment and the presence of certain elements is highly dependent on the geological and pedological properties of the immediate environment,” Martin said.
For example, one location may have high concentrations of titanium while another location may not. Because trees uptake these elements from the soil moisture once the elements have become mobilized in solution, trees will fix such elements permanently in the alpha-cellulose, the basic building blocks of wood. LIBS is able to determine which trace elements exist that have been bound in the wood.
Thus, if the chemical fingerprints from logs at one location match the chemical fingerprints of logs from another location, it can correctly be assumed that the logs must have been collected from the same location, or perhaps even from the same tree, she said.
Tree at Last
Martin’s lab sampled eleven logs of the Texas wood using the LIBS technique to assess the presence and amounts of specific elements, namely carbon, magnesium, silicon, aluminum, calcium, manganese, iron, titanium, nitrogen, and sodium.
“Because we did not know what effects the charred wood may have on elemental concentrations, we conducted identical analyses on all eleven logs using both areas of charred wood as well as unburned wood,” Martin said.
Little difference in the chemical spectra between charred logs and unburned logs was found, which she said would be expected because the low temperatures that caused the charred surfaces simply were too low to volatilize and remove the elements.
Martin and Grissino-Mayer found the chemical fingerprint consistent for all the wood samples that were tested, strongly suggesting that the wood logs come from the same type of tree species, and that the logs likely came from the same tree or from a group of trees growing in the same area.
“Because the chemical fingerprints are so similar, we suspect that all eleven logs likely came from the same tree, or even from a few trees that were collected in a very localized area,” she said.
Perhaps anticipating that a mere ‘suspicion’ or ‘suggestion’ of consistency in the wood samples would not satisfy a trial court, Martin and Grissino-Mayer then verified their findings by conducting a correlation analysis using the PROC CORR procedure in the Statistical Analysis System, a process that yielded a near absolute confidence in their LIBS findings.
As the result of the verification process, Martin said they are “99.999 percent confident that the chemical spectra from the eleven logs are identical.”
The defendant in this case was found guilty of capital murder and was given the death penalty. During the trial, several photos of firelogs were shown to jurors.
Douglas Page (firstname.lastname@example.org) is a freelance science and medical writer based in Pine Mountain, California.