What happens when drug testing experts come across a substance they have never seen before? An ANU expert takes us behind the scenes.

Imagine, if you will, a small plastic bag containing a mixture of crystals and powder.

The individual presenting it thinks it “might be ketamine?”, but admits the subjective effects are different to what they’re used to. How do we find out if it’s what they think it is? And what are the consequences if it isn’t?

This is a typical scenario for the people working at CanTEST — Australia’s first and only fixed-site, face-to-face drug checking service, launched in Canberra in August 2022. The public can bring in recreational drugs for analysis, and access information and advice based on the results.

The service is designed to engage young consumers of drugs with the next generation of harm minimisation and it is terribly important to get this engagement right.

The identification of new psychoactive substances presents a major challenge in the context of drug checking.

Chemical fingerprints

Testing provides us with the ‘fingerprint’ of a chemical, which hopefully matches one of the thousands stored in databases available to analysts.

But what happens when a fingerprint does not provide a match? What if a chemical is a ‘cleanskin’ — a chemical unknown?

How we determine the identity of mystery chemicals can be far more convoluted than many might guess.

That brings us back to the original baggy.

White powder in a small clear bag against a 10cm ruler

'CanKet' sample tested at Australia's first fixed pill testing site. Photo: Malcolm McLeod

Identifying chemical X

Patrick Yates is a PhD candidate from the ANU Research School of Chemistry and a barn of a man with a towering IQ to match. He carefully measures a small amount of the product, let’s call it chemical X, and positions it on the Fourier Transform Infrared (FTIR) spectrometer.

The FTIR has become the workhorse of many drug checking programs around the world and was used by Pill Testing Australia at the Groovin’ the Moo music festival in 2018 and 2019. It can usually be made to work quickly and reliably — even at a bush doof — as long as electricity is available. A laser is shone on the substance and the ‘reflection’ (a measure of how the drug shakes and wiggles) is captured and compared to a database of more than 30,000 chemicals.

Yates’ analysis reveals that the contents of our baggy don’t quite add up — and don’t confirm a ketamine match. The FTIR suggests the product might be a ketamine analogue called 2-fluorodeschloroketamine (2-FDCK), but Yates has his doubts. The benefit of having formally trained chemists conducting the analyses at CanTEST is that the intuition of experts is often proven right.

PhD student Cassidy Whitefield, a veteran from the Pill Testing Australia team at Groovin’ the Moo, is the newest member of CanTEST. She turns to an instrument known as an ultra-performance-liquid-chromatography-photodiode array (UPLCPDA), which is humming away in the corner.

Prior to the start of her shift, she ran lab-based standards through it, calibrating the machine to the 10 most common drugs we see, including ketamine. The sample is being asked to run a race against the known standard to compare the identities of these compounds. Through this process, we can not only confirm the identity of products but also measure their quantity. It takes about four minutes.

Something is off

While the sample appears similar to the ketamine standard she ran earlier in the day, Whitefield’s trained eye sees something is off.

The rate at which chemical X ran its race (known as the retention time) was similar to our ketamine standard, but its absorption of UV was different.

Whatever is there is real, quite pure, and neither ketamine nor 2-FDCK, which means it’s time to call the Boss and the Doc.

PhD student Cassidy Whitefield operating a spectrometer at CanTEST. Photo: Tracey Nearmy/ANU

The mystery deepens

The Boss is Associate Professor Malcolm McLeod, also of the ANU Research School of Chemistry. I approached him in 2015 to head up the analytical side of our pill testing project. At the time he was unimpressed. He’s currently in Vienna as a Churchill Fellow, touring the services that have been providing pill testing there for decades.

It is late, he has just had COVID-19 and he is still unimpressed, but also interested. He confirms the spectrum sent to him strongly suggests the presence of a carbonyl group, making it likely to be a ‘ketamine-like’ product, but definitely not ketamine, and probably not 2-FDCK.

The Doc is me — a bit player in this evolving drama, having been involved in the efforts to first identify methoxetamine in the United Kingdom and deschloroketamine in Australia. I provide the medical context for any analysis and in conjunction with my colleagues try to provide a risk assessment for the rapidly evolving illicit drug market. I’m familiar with ketamine, both as an invaluable agent in the emergency and pre-hospital environment, and as part of the illicit arylcyclohexylamine group of drugs emerging on the market.

The punter is still with us, and a three-way conversation is happening across two continents and several time zones. We don’t know what chemical X is and the punter wants an answer. Too bland a consensus from us and someone might get hurt. An overreaction could mean the power of future warnings might be diluted.

How certain is McLeod that we are dealing with a ketamine-like substance? “Pretty certain.” The punter is advised the substance is not ketamine and that the identity of the product could not be ascertained. Our band of peer workers advises extreme caution in using it.

When in doubt, run more tests

It will come as little surprise that this situation was not one our team of analytical chemists could leave unresolved. The chemical X sample was transported back to the den of chemical interrogation that is the ANU Research School of Chemistry and the inquisition began.

The product was first subjected to a method called gas-chromatography-mass spectrometry (GC-MS), meaning the sample was made to run another race and then smashed into pieces to further fingerprint our mystery chemical. The molecular weight was also ascertained. Ketamine has 13 carbons, 16 hydrogens, a chlorine, a nitrogen and an oxygen. McLeod established that a match might be made if ketamine (C13H16ClNO) lost its chlorine and gained a fluorine and an ethyl group in place of a methyl group, resulting in C14H18FNO. He was right.

The GC-MS data correlated closely with a ketamine derivative known as fluorexetamine, but the presence of an isomer — two compounds with the same molecular formula but arranged differently — could not be ruled out.

It was time to bring out the big guns. The nuclear magnetic resonance (NMR) spectrometer is a chemist’s Book of Runes — an instruction booklet that reads like an Ikea manual written in Cyrillic. Answers can be found, but only by those few who can speak the language well. McLeod could see the sample contained an N-ethyl group (like fluorexetamine and 2’-fluoro-2-oxo-PCE) and not an N-methyl group (like ketamine and 2-fluorodeschloroketamine).

So, the team designed a series of multidimensional correlation experiments to test some alternative hypotheses. The chemists were able to divine that there were four hydrogens next to each other around the aromatic ring, meaning that it could not be fluorexetamine. Chemical X could only be a chemical that went by the moniker of 2’-fluoro-2-oxophenylcyclohexylethylamine.

And this had never been seen before.

From chemical X to CanKet

It’s hard to emphasise what a phenomenal piece of work this was. Analytical chemists are not the sort of creatures to blow their own trumpet, which is why I’m writing this piece.

The CanTEST crew (from left) Associate Professor Malcolm McLeod, Patrick Yates and Cassidy Whitefield. Photo: Tracey Nearmy/ANU

We contacted our offsiders at the United Nations Office of Drug Control, the European Monitoring Centre for Drugs and Drug Addiction, as well as several esteemed researchers around the world. None had seen this before.

Our colleagues at the ACT Government Analytical Laboratory wrote to international peers and a global forum of forensic and analytical chemists reviewed their locally acquired data and provided information that supported our findings. We have since found a single further report out of China from a forensically obtained analytical sample, where it was described by another name, 2-(2-fluorophenyl)-2-(ethylamino) cyclohexan1-one (2F-NENDCK).

Our team pushes on. The compound 2’-fluoro2-oxo-phenylcyclohexylethylamine is turning up regularly, as well as other ketamine derivatives. It’s a bit of a mouthful and the unassuming stars of this story have quietly taken to calling it ‘CanKet’, as in Canberra ketamine — a name informally conferred by the analysts, not the consumers.

We are now able to identify CanKet, but in analytical terms this was the chemical equivalent of needing a car to go to work, but being forced to build it yourself from scratch.

The team at CanTEST has identified a product never before seen in Australia and with no clinical information associated with it from elsewhere. We are proud of the bar we’ve set ourselves in maintaining open science and humbled by the opportunity we have been provided by the ACT, a jurisdiction making all the waves in Australia.

CanTEST is a collaboration between Directions Health Services, the Canberra Alliance for Harm Minimisation and Advocacy (CAHMA), Pill Testing Australia and ACT Health, with advice provided by scientists from ANU.

Top image: Zivani/Shutterstock.com

You may also like

Article Card Image

We’ve been testing Canberra’s drugs. Here’s what we found

The first month of pill testing in Canberra has revealed unexpected findings.

Article Card Image

Coke light: drug tests show 40% of ‘cocaine’ had no cocaine

Researchers from ANU have released results from Australia's first government-backed pill testing service, including finding methamphetamine that was actually sugar.

Article Card Image

Can AI produce decent podcasts?

Everybody wants to start a podcast these days – even AI. What does this mean for the future of science communication?

Subscribe to ANU Reporter