杏吧原创

San Francisco, California

THE perfect end to a perfect summer鈥檚 day鈥攕ettle down in a comfy
lounger on the porch with a fine white wine, enjoy the satisfying pop of the
cork, savour the heady aroma of cat urine, and sit sipping as the Sun goes down.
. .

Woah. Hang on. Cat urine?

Yes, you read it right. The delicate fragrance of cat pee is one of the key
components in the bouquet of a Sauvignon blanc鈥攋ust as cow urine and horse
manure are crucial to the noses of an old Riesling and a complex Bordeaux. And
bubbling under in the brew are the odours of mouldy socks, rotten eggs and
burning rubber. Such are the delights revealed by scientists trying to
deconstruct the chemistry of the perfect summer wine.

Nose job

Over the past few years, there has been an explosion of research into the
sensory qualities of wine鈥攊ts colour, texture, taste and smell鈥攁nd
chemists have come up with several hundred compounds which are responsible for
the bouquets of our favourite bottles. These have been sniffed out from the
thousands of chemicals in wine by highly sensitive noses, and linked with
certain smells鈥攆lowers or asparagus, for example. Or cat pee.

And by following a bottle of wine from grape to glass, the researchers have
pinpointed the key conditions for making perfect wines. 鈥淲e [want to] enhance
desirable flavours on one hand and avoid off-flavours on the other,鈥� says Peter
Winterhalter, a chemist at Friedrich-Alexander University in Erlangen,
Germany.

So where does the flavour of a wine come from? It鈥檚 not just a matter of
taste, says Ann Noble, a sensory chemist at the University of California at
Davis. Our taste buds pick up four basic sensations鈥攕weet, sour, bitter
and salt. Sweetness in wine comes from any unfermented sugars, sourness from the
acids created during fermentation and present in the original grape, and
bitterness is caused by phenolic compounds, which are mainly found in red wines,
and ethanol, found in all wines. Wines rarely taste salty.

These four tastes alone are not enough to discriminate between the thousands
of wine varieties in the world. The subtle differences that distinguish one wine
from another, says Noble, are due to aromatic compounds wafting up to our
olfactory receptors. In fact, the human nose is so sensitive to these aromas
that scientists are linking people with trained noses up to machines to find out
which chemicals cause which smells. To identify the thousands of compounds in
wine, Gordon Burns, owner of ETS Laboratories in California鈥檚 Napa Valley
separates them using gas chromatography and passes them into a mass spectrometer
to get an idea of their molecular structures. For flavour analysis, sensory
scientists instead pipe the compounds to a panel of human noses.

The problem with using human noses, however, is that most people simply have
hedonistic responses to wine: either they like it or they don鈥檛. 杏吧原创s
trying to isolate aromas in wine want more than that. They need consistent terms
to define what they are studying and so the noses have to be trained.

鈥淔lavour analysis starts with sensory descriptive analysis,鈥� says Thomas
Henick-Kling, a microbiologist from Cornell University in New York. In other
words, people have to describe what is in a wine. 鈥淚n most cases,鈥� says Cornell
flavour chemist Terry Acree, 鈥渄escribing the aroma of a wine is like describing
the colour green to a blind person.鈥�

To help people describe what their noses are sensing, Noble has produced an
鈥渁roma wheel鈥�, a round lexicon consisting of three circles of increasingly
specific descriptive words, all based on real-world smells (see
Diagram).

Sensory scientists ask trainee noses to identify standard reference smells
using the aroma wheel. For example, the chemical dimethyl disulphide should make
trainees think of onion, while dimethyl sulphide should remind them of
asparagus. Those who identify the aromas correctly are then put to work naming
the smells produced by other chemicals.

It鈥檚 a big job: of the 4000 or so chemicals in wine, at least 500 cause
aromas. Acree has collected the responses to more than 300 odour-producing
chemicals on a Web site called FlavorNet, where visitors can discover the
compounds responsible for descriptions from 鈥渇oxy鈥� to 鈥渟weaty鈥濃�攁s well as
鈥渦rine鈥�. But he isn鈥檛 surprised by the large number of smells. 鈥淭he chemistry
and biology that goes on to make a wine covers the whole range of reactions that
can produce aromas,鈥� he says.

To make the perfect wine, then, all you need to do is pick the pleasant
flavours you want, and avoid the nasty ones. Nothing so simple: according to
Acree, it takes a combination of pleasant and unpleasant flavours to make a
full-bodied wine. He thinks there is no such thing as a bad aroma, only one that
is the wrong strength. 鈥淔or example, if a wine has too much methyl mercaptan in
it, you think of burning rubber,鈥� he says. 鈥淏ut at a low level it provides
character, complexity and interest.鈥�

There are a couple of exceptions to Acree鈥檚 rule, however鈥攐ne is a
nasty aroma caused by 2-acetyl-1-pyrroline which smells like mouse urine and
gives one of the longest-lasting aftertastes. Noble describes it as 鈥渓ike
licking out a mouse nest鈥�. Another is the mouldy smell of a spoiled cork (see
鲍苍肠辞谤办别诲鈥�).

Hidden aromas

Instead of identifying 鈥済ood鈥� or 鈥渂ad鈥� aromas, chemists pinpoint combinations
that are unique to specific varieties of wine. 鈥淭here is no one single compound
responsible for, say, Chardonnay flavour,鈥� says Patrick Williams of the
Australian Wine Research Institute in Adelaide. 鈥淚t鈥檚 a combination of many
different volatile compounds.鈥� Finding where these compounds come from could
enable producers to enhance flavours.

Chemists could also boost flavour by unlocking some of the 鈥渉idden鈥� aromas
lurking in wine. Many aromatic compounds are hidden in odourless molecules
called glycosides, discovered in grapes by Williams and colleague Leigh Francis.
But much of this flavour is never released, or is generated slowly over time by
chemical reactions caused by wine鈥檚 acidity.

Francis, working with Williams, has isolated the glycoside compounds that
give 鈥渉oney鈥�, 鈥渢obacco鈥�, 鈥渃hocolate鈥� and 鈥渄ried fig鈥� aromas in Merlot and
Cabernet Sauvignon wines, and is now working on a way of predicting the flavours
bound up as glycoside compounds in grapes before they are made into wine.

鈥淚magine an iceberg floating in water,鈥� says Winterhalter. 鈥淭he tip
represents the free aromatic compounds in wine鈥攖he 80 per cent below the
surface represents flavourless forms of aromatic compounds.鈥� Jane Robichaud,
director of the sensory and research lab at the Beringer Winery in Napa Valley,
says that the work on glycosides explains what happens in wines. 鈥淲hen you
bottle a Riesling or a Gew眉rtztraminer, six months or a year down the line,
it鈥檚 a totally different beast,鈥� she says.

All flavours, even those bound up in glycosides, have to come from somewhere.
And in many cases researchers can work out where a particular chemical comes
from鈥攖he grape, the yeast, or the barrel, for example. So if one or two
aromas were found to make a wine feel fresh and summery, producers could tweak
the wine accordingly.

Flavoursome bacteria

Perhaps the most important step in the wine-making process is fermentation.
Henick-Kling has shown that the strain of yeast used in fermenting the grape
juice strongly influences the flavour of the wine. 鈥淔or example a particular
yeast will make a Riesling taste more floral than another yeast,鈥� he says. His
lab has also found strains of yeast that enhance spicy aromas, such as nutmeg
and cinnamon. 鈥淚f you have five different yeasts,鈥� he says, 鈥測ou have five
different wines.鈥�

It鈥檚 not all pleasant flavours though. The yeast strain Brettanomyces
produces an aroma which has been described as 鈥渨et horse blankets鈥�, 鈥渃ow manure鈥�
and 鈥減lastic bandages鈥�. Henick-Kling鈥檚 graduate student Jonathan Licker found
that the dominant chemical in a 鈥淏retty鈥� wine is 4-ethylphenol. 鈥淎t low levels
it鈥檚 not really a problem,鈥� says Robichaud. 鈥淏ut it depends on how sensitive
somebody is to it as to whether it is adding complexity to the wine or giving
the smell of a sweaty horse blanket,鈥� she adds.

Yeasts are not the only microorganisms influencing the flavours of wines.
Henick-Kling has also studied the flavour-enhancing abilities of bacteria that
carry out a process known as malolactic fermentation (MLF). They convert the
sharply acidic compound malic acid to the softer, creamier lactic acid. Along
with this, the bacteria can give buttery, spicy and nutty flavours, and can
enhance fruity aromas. Thanks to Henick-Kling鈥檚 work, winemakers now know how to
control MLF to get additional desirable flavours.

Before any fermentation gets underway, however, the flavour of a wine can be
influenced by the conditions under which the grapes are grown. Winemakers have
known this for centuries, but now scientists are beginning to understand why.
Malcolm Allen of Charles Sturt University in Wagga Wagga, New South Wales, has
shown that the level of a chemical known as methoxypyrazine in grapes is
influenced by the quality of the soil and the climate. Cold weather, good soil
and lots of rain give high levels of the chemical. Methoxypyrazine is one of the
key aromas in wines such as Cabernet Sauvignon. 鈥淭he perception at low levels is
of chalk dust, old library books even,鈥� says Allen. 鈥淎t higher levels, a bell
pepper aroma dominates.鈥�

Winemakers have another natural tool to control wine flavour. Oak barrels,
which are used for fermenting and ageing, provide a potent source of flavours.
Pascal Chatonnet, a researcher from the University of Victor Segalen in
Bordeaux, France, says the wood contains characteristic chemicals, including
octalactone, eugenol and vanillin, which are responsible for coconut, clove and
vanilla aromas.

To get the desired blend, however, the wine needs to be fermented in the oak
barrels so that the chemicals are properly mixed, says Robichaud. Wine fermented
in a stainless steel tank and then stuck in the oak barrel, on the other hand,
鈥渨ould come out smelling and tasting like a bit of two by four,鈥� she says.

So when it comes to choosing wines, get the advice of the flavour chemists.
But is it safe to invite them to your party? They promise not to bore your other
guests with lists of aromas and chemicals鈥攂ut keep an eye open for the
warning signs. Don鈥檛 let anyone wear a T-shirt emblazoned with the aroma wheel
(available from Ann Noble for $25). And if the conversation turns to the
odours of skunk, wet dog, burnt toast and cat urine, give everyone a beer to
drink instead.

* * *

UNCORKED

PURISTS may turn up their noses, but one wine researcher proposes preserving
delicate wine flavours by using screw tops on wine bottles. Corks can cause a
musty, mouldy odour known as 鈥渃ork taint鈥� which affects the flavour of one
bottle in every case or two.

鈥淭he cork taint is not as repulsive as a rotten egg smell,鈥� says Christian
Butzke of the University of California at Davis. 鈥淚t smells musty, but it is not
that disgusting.鈥� If someone did not recognise the aroma as coming from the
cork, he says, they would simply think they had a bad wine.

The compound causing the mouldy aroma is 2,4,6-trichloroanisole (TCA), which
is one of the five most potent natural aromas to humans. TCA can get into a cork
in a number of ways, including during the washing of corks with chlorine to
bleach them. Butzke wants corks to be treated as a food product, so that
producers would have to ensure that no TCA was present. In the meantime, he
notes that screw-caps or crown caps could do the job, but acknowledges that wine
enthusiasts, himself included, are unlikely to give up the romance of the 鈥減op鈥�
when uncorking a bottle.