Forensic researchers use mass spectrometry to identify smuggled wood

Some years ago, David Gehl made an unusual trip to Siberia. “I went to lots of sawmills and forests and came back from Russia and China with, like, 20 kg of wood boards on my back,” he recalls.

Gehl was no ordinary tourist. He was in the region as part of an undercover probe by a nonprofit group called the Environmental Investigation Agency (EIA). He was following up on suspicions that oak imported into the US from China had come from forests in eastern Russia where logging was prohibited. The forests are the world’s last habitats for wild Siberian tigers and Amur leopards.

With the evidence that Gehl gathered and other information from a report the EIA compiled, federal agents obtained a search warrant and raided the headquarters of Lumber Liquidators (now called LL Flooring). The Department of Justice brought charges against the company for trading illegally logged wood. When Lumber Liquidators pleaded guilty in 2015, it was the first US felony conviction against an importer of illegally harvested wood, and the company paid $13.2 million in fines and penalties.

Proving this case required undercover sleuthing and scrutiny of trade reports. It might have been easier if federal agents could identify protected wood shipments at ports of entry. Since around the time when Gehl was making his undercover observations, chemists at the US Fish and Wildlife Service (USFWS) have been developing a method to identify trafficked wood on the spot using chemical fingerprinting. Ed Espinoza and his colleagues at the USFWS have been instrumental in developing this technique for wood identification and sharing it with law enforcement labs around the world. The technique is called direct analysis in real time mass spectrometry, or DART-MS for short. Today, the suite of molecules it measures can be used to rapidly identify the species, and sometimes even the region, a piece of wood came from.

In the meantime, governments around the world have become more aware of the scope of international wood trafficking, which may be worth as much as $150 billion a year according to the nonprofit Global Financial Integrity. US laws hold importers responsible for verifying that wood they bring into the country is harvested legally. International treaties protect certain species from being traded and cap trading on others. Enforcing those laws, however, can be a challenge because of the massive scale of the trade: hundreds of millions of cubic meters of logs and lumber, enough to fill the Great Pyramid of Giza 100 times over, pass through ports each year.

“It’s very, very difficult for anyone, even a well-meaning company, to realistically track their wood back” to its forest of origin, says Gehl, who continues to trace the origins of illegal lumber for the EIA. Wood often passes through several countries before reaching its end user, and it’s difficult to identify a protected species by eyeballing it. DART-MS, Espinoza and his colleagues say, could enable species identification at the border. But before it can crack lumber-smuggling cases at ports of entry, scientists will need to overcome a number of challenges.

Rapid collection, complicated data

When the USFWS suspects a crime involving wildlife has happened, it sends the evidence to the Clark R. Bavin National Fish and Wildlife Forensics Lab, in Ashland, Oregon. To support a criminal case, researchers at the lab find out as much as they can about the specimen in question, including its species and what region or population it might have come from. Sometimes these questions are answered by scientists in the lab’s morphology and genetics divisions. Problems that they cannot solve come to the chemistry team. Espinoza jokes, “We get the leftovers—what everybody else does not want.”

Wood has long been one of those leftovers, and the USFWS chemists employ a variety of chemical strategies to pin down a sample’s identity and geography. But Espinoza and his colleagues favor DART-MS.

DART-MS has gained popularity in the forensic science community, which uses it in criminal investigations, such as identifying illegal drugs. But analytical chemists tend to regard it with some suspicion. The information it gives is not as tidy as data from other mass spectrometry techniques because it skips the common step of using chromatography to separate analytes before they enter the spectrometer. This additional step adds time to the analysis but reduces the number of ions entering the mass spectrometer at any moment, making each one easier to identify. But according to Rabi Musah, a forensic scientist at the University at Albany, DART-MS’s peculiarities make it a strong fit for forensic analysis. Because the method is fast, is comparatively inexpensive, and usually requires no sample preparation, it lends itself to a field that prizes ease of use and rapid turnaround.

Mass spectrometry gets information about the molecules in a sample by ionizing them and then measuring their mass-to-charge ratio when they hit a detector. While typical mass spectrometers do this within a vacuum, DART-MS ionizes samples in the open air in a process known as ambient ionization. A blue barrel capped at one end with a steel nozzle heats helium to about 400 °C and blows it across the sample held in an open-air gap and into the mass spectrometer. The helium stream ionizes molecules blown off the sample and directs them into the mass spectrometer for analysis.

In a forensic lab outside Washington, DC, at the National Institute of Standards and Technology (NIST), chemist Edward Sisco, who collaborates with Espinoza, is firing up a DART-MS system.

Normally Sisco studies illegal drugs, but he gamely tucks the narrowest edge of a shim of wood from the hardware store into the gap on the instrument as a demonstration. While helium blows over the sample, he glances often at the real-time monitor to make sure he hasn’t broken the gas stream. It takes just a moment to acquire a spectrum.

An increasing number of forensic labs have adopted DART-MS for drug identification since the technique was published in 2005 (Anal. Chem., DOI: 10.1021/ac050162j). But its adoption has not been automatic or universal. “It takes people time to buy in,” says Sisco, who keeps above his desk a sign with the phrase “But we’ve always done it this way” crossed out in red. For one thing, identifying compounds can require some fiddling. Two molecules with the same chemical formula but very different structures and properties—such as cocaine and the antinausea drug scopolamine—can initially look the same and require further analysis to differentiate. The more complex the mixture, the more difficult this problem becomes because everything leaving the surface of the sample enters the mass spectrometer at once.

Many chemists hesitate to analyze “a spectrum with God knows how many things in it,” Sisco says. Part of his work at NIST has focused on making that analysis easier.

Species analysis with DART-MS

Wood is more complex than most materials, and that makes it even harder to analyze than Sisco’s typical mixtures of drugs and adulterants. To simplify DART-MS data analysis, Sisco and NIST colleague Arun Moorthy recently created a software program (J. Am. Soc. Mass Spectrom. 2021, DOI: 10.1021/jasms.1c00097). Moorthy says the easier approach has led several state labs to use the technology more regularly for drug analysis. It also attracted Espinoza’s attention, and led to a collaboration between the NIST and USFWS scientists.

“I am first a strict chemist, and I know what all chemists say about ambient ionization: ‘Ah, it’s a bunch of baloney,’ ” Espinoza says. But in his opinion, the complex spectra aren’t useless­—just unfamiliar to most chemists.

When Espinoza and USFWS colleagues started analyzing wood with DART-MS around 2012, they approached it like a drug identification problem (Rapid Commun. Mass Spectrom. 2012, DOI: 10.1002/rcm.6388). They first used the method to study a pungent protected species called agarwood, which is sold for incense at around $2 million a kilogram. Initially they scanned agarwood spectra for unique metabolites—organic compounds that only agarwood can produce—as their presence would be a sure sign that a product was agarwood. But not all species produce unique signature metabolites. To identify more species, they’ve since expanded their analysis to consider a more complex chemical fingerprint from each piece of wood.

All wood is mostly composed of lignins and cellulose, macromolecules that give the material its structure and sturdiness. But they—and other ubiquitous metabolites, like sugars and amino acids—don’t tell chemists much about the material’s identity. For that, researchers use nonstructural metabolites called extractives, which can include phenolic compounds, terpenoids, and waxes.

Sisco’s analysis at NIST generates a list of the 22 most abundant protonated ions in the wood sample. Sisco sends that ion list to Espinoza to compare with his database of wood samples. Can Espinoza tell what type of tree was felled to become part of a home improvement project? Some of the peaks, he reports, are strongly suggestive of species in the Magnoliaceae family such as poplar. But it’s impossible to say for sure because NIST’s software picks out only the most abundant ions instead of saving a full spectrum.

“The number of compounds that are in a wood sample is just far, far greater than what we have in our drug mixtures,” Moorthy says. The scientists usually identify a single illegal molecule in a mixture by comparing its spectrum with a few signature peaks in a pure sample. But analyzing a wood species’ fingerprint involves looking at its whole DART-MS spectrum, which may include peaks that have yet to be matched to specific molecules.

The species-matching process that the USFWS uses begins, like a drug-hunting lab’s protocol, by searching a spectrum for signature ions that match one or more reference species in the database. Then the researchers compare the unknown sample’s full molecular fingerprint to species that might match. Sometimes the process yields a clear species. Sometimes, with enough in the database, the researchers can even differentiate between geographically distinct populations of the same species. Other times, the closest they can get is a genus or family.

Complicating matters further, researchers must populate that database with many spectra of individual samples to capture chemical diversity within a species. Each red oak is genetically distinct—and that individuality can be reflected in subtle differences in the level of each small molecule it produces. To make sense of how the differences between single trees can still generalize to trends within a species, researchers have “had to dip into biochemistry or population genetics,” Espinoza says.

The chemistry problems are what have kept Espinoza fascinated by wood forensics. When he joined the USFWS forensic lab decades ago, he thought of it as “a bunny-and-tree-hugger lab in Oregon,” he says. “I thought, ‘Oh, that sounds interesting. I’ll probably get a couple of papers and go back to academia.’ ” He has worked there since 1989.

The global context

The USFWS has a strategic appetite to solve the difficult problem of wood identification. “Forests are disappearing at a rapid rate, and there’s not good monitoring of illegal transactions, so there’s a scramble to create the technologies to allow detection of illegal timber,” says Mark Roberts, founder of an environmental law consultancy called Eco Policy Advisors, which focuses on international wood crimes.

International law enforcement organizations are increasingly noticing that the groups smuggling timber and lumber are the same as those committing other trafficking crimes. A 2019 World Bank Group report says that the financial scale and global scope of the wildlife smuggling and drug trafficking industries are nearly the same—but the risk of legal consequence for perpetrators of wildlife crimes is much lower. Law enforcement has historically invested much less to halt illicit wood trafficking than to combat the smuggling of humans, weapons, or drugs, the report notes. But activists say that wood smuggling’s profits often fuel other types of criminal activities.

Some of the clearest-cut wood smuggling crimes deal in precious wood species protected by the Convention on International Trade in Endangered Species of Wild Fauna and Flora, or CITES, the same treaty that prohibits trade in elephant ivory and pangolin scales. The past two triennial CITES conferences have paid increasing attention to intercepting and prosecuting wood smugglers.

The USFWS has worked with numerous botanical gardens, xylaria, and government labs around the world to collect mass spectra from trees of known species and geographic origin. It isn’t alone in this effort; another initiative, World Forest ID, is also collecting mass spectra and other forensic data on known trees.

While DART-MS-based metabolomics has become a leading technique internationally for tree identification, other techniques are available. Several US ports of entry have installed sophisticated imaging systems that capture magnified multispectral photos that wood anatomists can later investigate. Other labs conduct genetic analysis. Although it is hard to extract a genome from hardwood, researchers can retrieve enough smaller genetic markers to make a species identification and in one case even matched confiscated wood to individual illegally cut stumps in a US national forest. Stable-isotope analysis can help determine the region wood has come from, which aids in certain smuggling cases, and at least one lab has published several studies on using near-infrared spectroscopy to identify certain tree species.

DART-MS seems to have the most momentum among these techniques. “Because it’s so cheap and quite effective, it just has a lot of promise,” the EIA’s Gehl says.

The USFWS is a world leader in timber identification. A February 2022 CITES task force report on tackling the illicit lumber trade mentioned the agency’s forensic lab specifically. At the US Forest Service’s Wood Identification and Screening Center, which shares lab space with the USFWS forensic team, it’s the only technique in use. “We are all about DART,” Forest Service chemist Erin McClure-Price jokes.

The federal scientists from several agencies who work on wood identification in Oregon have trained scientists from Vietnam, Peru, Gabon, and other countries on using DART-MS, but they think—and evidence from DART’s rapid expansion in drug-testing labs seems to support—that easier analytical workflows will help the technology spread even farther. Meanwhile, the team is also trying to bring the technology to more locations within the US by making its laboratory mobile.

Taking it on the road

Parked outside the joint lab in Oregon is a gooseneck trailer hitched to a USFWS pickup truck. The trailer, which Espinoza and his team have dubbed the Woodshed, was officially commissioned in November but has yet to be deployed. It’s equipped as a mobile lab that can take his DART-MS operation for wood identification to ports of entry for rapid screening.

Agents can send just a sliver of wood for analysis to the labs in Oregon, but until it is processed, the whole shipment must be seized and held, adding enormous costs in storage and handling. In the US, there are limits on how long the government can hold seized shipments. Cases have involved up to 144 containers at a time, each holding roughly $2 million worth of logs.

That’s why the forensic lab decided to invest in taking a mass spectrometer on the road. Espinoza says that screening between 200 and 400 samples a day on-site at a port would be a reasonable goal. “We can go to places to respond to shiploads of timber and very quickly say, ‘Let it go, let it go, let it go,’ ” but quickly identify shiploads that should be stopped for more screening, he says.

Plenty of mass spectrometers are portable, and chemists pursuing drug identification or chemical weapon detection have contemplated taking a DART-MS on the road in the past. Retired chemist H. Dupont Durst, a coauthor on the original DART-MS paper, spent years in the 1990s retrofitting gas chromatography/mass spectrometry systems to fit onto tractor trailers that could visit chemical weapons depots to verify that weapons had been neutralized. But because no chemical separation precedes the mass spectrometry stage in DART-MS, the technique demands a higher-​resolution spectrometer, necessitating a longer ion path that could be more sensitive to misalignment if jostled. After DART-MS was invented, Durst tried to interest the US military in equipping it to travel but never secured the funding.

“Dupont was so excited about this that he even called me at home at dinnertime the other night when he heard that Ed finally got the mobile lab working,” DART-MS coinventor Robert “Chip” Cody of the analytical instrument company Jeol recalls.

Espinoza considered putting the mass spectrometer in a recreational vehicle but found that option too rattly. To guarantee a softer ride, he upgraded to a trailer designed to move thoroughbred racehorses. It came with air suspension, and Espinzoa added a vibration-dampening mat for good measure.

During an interview with C&EN in November, Espinoza expressed a little concern that despite the modifications for stability, the mass spectrometer might simply be too sensitive to survive a highway trip. But after a test drive early in January, he jubilantly reports by email, “Just got back with the Mobile Lab intact!! . . . It sure looks tiny next to the big trucks, but it feels huge when you are driving it.”

One thing the researchers are still not sure about is whether differences in environment will affect the spectra they collect. Water in the atmosphere, for example, plays an important role in the DART ionization process. Sisco says, “If you go to a port of entry like the New Jersey port authority or LAX or whatever, you’re going to have very different environmental conditions. . . . What is that going to do to the spectra?”

It’s possible that the Woodshed’s first major mission will focus on public engagement instead of law enforcement: Espinoza has been asked to bring it to Washington, DC, in March to participate in celebrations of the 50th anniversary of the Endangered Species Act. “Endangered species,” for many in the US, may connote whooping cranes, pronghorn antelope, and salmon rather than rosewood, mahogany, and ebony. But if Espinoza has his way, by the 100th anniversary of the Endangered Species Act, all these species will still be around.