Alison Davidson is a kind of olfactory Sherlock. A quiet but crisp scientist who runs the forensics lab at the University of Staffordshire, she is working on a form of analysis that in the near future could allow police to gather odour evidence from, say, a coat left at the scene of the crime to build a profile of a suspect: “We’re looking for a smoker here, guv. He probably uses Lynx, he had a curry on the night of the murder and he also has a dog.”
Smell, as every wine lover knows, is a powerful sense with strong links to memory. “That’s what got me started,” says Davidson. “There’s a lot of anecdotal evidence from rape victims who say, ‘I can’t wear those clothes because they still smell of him.’ I thought, ‘It’s just chemicals, what if I can get those chemicals off those garments?’ ”
Davidson wasn’t expecting this to lead her down a wine path, but the techniques used to analyse wine have been crucial to her research.
“I started very simply by looking at perfumes,” she says. “But analytical chemists publish a little more freely about wine than about fragrances, and the approaches are comparable – you’re basically looking at aroma compounds dissolved in alcohol.”
The idea that a wine might have a unique olfactory signature is implicit not only in the prices certain special bottles fetch but also in the classification of wines according to their geographical production and quality level. The possibility of identifying such markers in a scientific way is of great interest to those involved in combatting wine fraud and in safeguarding provenance claims, so there is highly complex work on wine scents under way.
Davidson found herself using gas chromatography – a technique that separates mixtures according to the boiling points of the components – on the Veneto reds valpolicella, amarone della valpollicella and recioto della valpollicella, to try to establish whether a particular producer leaves an identifiable volatile fingerprint on his or her wines.
“Forensic scientists don’t like to use the term fingerprint unless it actually is a fingerprint,” she says. “At the scene of a crime you don’t even get proper fingerprints, you get finger marks.”
Another study provided intelligence on a technique called rapid headspace solid-phase micro-extraction (SPME), which Davidson uses to harvest odour compounds from a piece of evidence.
How does that work? “I’ll show you,” she says, taking the lid off a curved glass vessel and shoving an old cardigan in it.
“Even before we get to putting something on the gas chromatograph where we can look at the patterns, we have to get those chemicals. Sample collection is critical. This is a desiccator. We put the whole garment in there and — see how there’s a bit of headspace at the top? We put the lid on, pop it in the oven for 10 minutes at 110C, and all those volatile chemicals will start preferentially going into the gas phase…”
Preferentially? “Before fats and heavy stuff. Then, after about 10 minutes, they’ll be in the headspace and we can collect them using a fibre made from PDMS, a silicon compound, and put them through gas chromatography.”
Davidson started off looking only at perfumes, to which quite a few police officers said: “You seem to think criminals are well-groomed and smell lovely, and our experience is that they just don’t.”
However, there is apparently one case of a rapist being caught because his victims said that he reeked of a particular aftershave: when he was brought in on another charge, the smell of it triggered the policeman’s memory.
“What I’ve realised as more work goes on with human scent is that I can say: ‘This mixture is very individualising to this particular person.’ It’s made up of endogenous compounds – the things you produce yourself, and some of that is going to be genetic, or related to what you ate – and exogenous compounds, the things we apply to ourselves in beauty and hygiene products. We live in a highly complex chemical world.”
There is a lot of work still to do. Information-sharing by fragrance companies would be a huge boost. “I usually work by pattern-matching, but it would be lovely to be able to identify actual perfumes. At the moment some brands, such as Lynx deodorants, have a component that I recognise.”
There also needs to be more research on human scent — how similar is our smell to that of the clothes we wear? How does it change over time? What is the effect of mood on smell?
Davidson believes she is a couple of years away from developing methods that could be used in the real-life forensics field.
And all of this with more than a little help from techniques initially developed in a wine laboratory.
Source: The Telegraph