Molecular Gastronomy: Exploring the Science of Flavor (37 page)

covered a molecule in Sauvignon wines having an agreeable odor that belongs

to the thiol family (characterized by a group composed of a sulfur atom and

an –sh hydrogen atom that is directly attached to a carbon atom). This raised

the possibility that other sulfur compounds might contribute to the aroma of

Bordeaux wines. Further investigation by Takatoshi Tominaga, Valérie Lavigne-

248 |

Cruege, and Patricia Bouchilloux revealed the presence of sulfur compounds

in very weak concentrations.

Sulfur and Yeast

Many known sulfur molecules are aromatically very active thiols. Where-

as the perception threshold for alcohols (molecules with an –oh hydroxyl

group) is on the order of a milligram or a microgram per liter in wine, the

threshold for thiols is roughly one-thousandth as much—hence the interest

in developing a method of analysis capable of detecting these trace aromas,

agreeable and noxious ones alike.

The studies by Tominaga and his colleagues showed that many volatile

sulfur compounds responsible for aromatic defects in wines have their ori-

gin in the action or metabolism of yeast, which during fermentation trans-

forms the sulfur amino acids of the grape and the sulfur dioxide added as

a preservative. This finding suggested a way to correct the known defects

of sulfited wines. But whereas it is simple to lower the concentration of

hydrogen sulfide, either by reducing the share of sulfur dioxide or by rack-

ing the wine in order to aerate it, other molecules such as ethanethiol and

methanethiol, as well as a similar molecule known as methionol, are more

difficult to control. The chemists observed that the concentration of these

compounds depends directly on the turbidity of the grape juice before fer-

mentation. This cloudiness, which brings the juice into contact with the

yeast, has to be reduced.

The Misdeeds of Copper

Investigation of the good side of sulfur molecules has yielded its own har-

vest of results. Certain thiols were known to share the characteristic odor of

plants (broom, boxwood), fruits (cassis, grapefruit, passion fruit, guava, papa-

ya), and even foods such as roasted meat and coffee. It happens that the great

variety of aromatic nuances found in Sauvignon wines are lost when copper is

added. Because copper bonds chemically with thiols and blocks their aromatic

action, the Bordeaux chemists suspected that thiols were also responsible for

the odors found in Sauvignon.

Sulfur and Wine
| 249

The first thiol was identified in a red Sauvignon wine, in 1993, and had an

odor of boxwood and broom at a concentration of 40 nanograms per liter. The

chemists succeeded next in finding it in boxwood and broom as well. They

then found other molecules in white wines that are also present in fruits, nota-

bly 3-mercaptohexyl acetate, which has a dominant hint of boxwood and recalls

the odors of grapefruit zest and passion fruit (in which it was subsequently de-

tected) and another thiol that, depending on its chemical environments, gives

off the same fragrances.

And in red wines? Here again the chemists proceeded on the basis of a

sensory observation, namely that the complexity and intensity of fruit and meat

aromas in young red Merlots and Cabernet Sauvignons decrease when these

wines contain a small amount of copper. Several thiols found in Sauvignon

wines recently have been identified in young wines made from Merlot, Cab-

ernet Sauvignon, and Cabernet Franc grapes, in which they contribute notes

of cooked fruit, cassis, meat, and coffee, at concentrations as low as 0.1 nano-

grams per liter.

Oenologists are now seeking to use these discoveries to improve wine-mak-

ing methods. In particular, they know that aging white wines on their lees,

where the sulfur molecules of the yeast are found, progressively increases the

concentration of certain aromatic thiols.

250 | investigations a nd mod el s

75

Wine Glasses

The same well-calibrated glass is best for both white and red wines.

d i s c i p l e s o f t h e g r a p e a r e s e l d o m without a glass. But which one?

Generations of gourmets have debated the optimum form and size of a wine

glass. In France, the Institut National des Appellations d’Origine (inao) has

mandated use of a glass recognized by the International Standards Organiza-

tion (iso) whose bowl is about twice as high as it is wide. Is it really the best?

Because German tasters recommended a rounder glass, Ulrich Fischer, of the

Oenology Department at the University of Neustadt, and his colleague Britta

Loewe-Stanienda decided to explore the intensity of perceptions of aromas in

glasses of various shapes and sizes.

Standards organizations such as the iso and inao have stepped in to regu-

late glassware without having first determined the exact effects of form and

shape on perception. It is for reasons of habit, not science, that red wine is

served in glasses that are more voluminous than those used for white wine,

and mellow wines are served in glasses whose opening is larger than those

meant for dry wines.

Calculated Evaporation

Before studying the shape of glasses, the Neustadt researchers sought to

determine the physicochemical effects of pouring wine into a glass on the per-

ception of aromas. As a matter of physics we know that the vapor pressure of

| 251

a molecule dissolved in a liquid depends on the solvent, the solution, and the

temperature. We also know that molecules that have passed into the gaseous

phase in a glass diffuse into the surrounding air at a rate that depends on the

diameter of the rim. On smelling the air in the bowl one inhales these volatile

gaseous phase molecules. But many connoisseurs inhale several times in rapid

succession. Does another wave of odorant molecules have enough time to pass

into the gaseous phase between inhalations? To find out, the researchers used

gas chromatography to analyze the air that lies above the wine at different

heights and at different times after the wine has been poured into a glass.

Some forty compounds were studied at different temperatures.

It was known, too, that the rate of release of the odorant molecules depends

on their chemical constitution. Other differences appeared as well: The release

rate of esters varied much less, as a function of temperature, than the rate

of release of alcohols and volatile phenols. This phenomenon explains why

the esters in wines that are consumed when they are cool are concentrated

more rapidly than the alcohol, stimulating the perception of fruitiness. Does it

also tell us why white wines are drunk at cooler temperatures than red wines?

Fruity esters, which often dominate the bouquet of white wines, rapidly pres-

ent themselves to the nose at low temperatures in sufficient quantities to be

perceived, whereas red wines must be served at warmer temperatures for the

effects of their volatile phenols to be registered.

Measurements indicated that gourmets are well advised not to quaff their

wine at intervals of less than fifteen seconds if they want to have comparable

sensations. Their impressions will be strongest if they sniff the air nearest to

the rim of the glass.

Tests of Glasses

Having specified the conditions for tasting in detail, the Neustadt research-

ers tested ten glasses with different maximum diameters, heights, and diam-

eters at the rim. Tasters were instructed to note the intensity of some ten notes

(buttery, floral, red fruits, and so on) for the same wine in each glass.

These experiments yielded a great deal of information. With white wines,

for example, a narrow bowl brought out the reduced aroma (a faintly vege-

tal odor suggestive of cabbage) and sulfur character of the wines more than

the calix of the iso glass. Glasses with a large bowl of small or medium height

252 | investigations a nd mod el s

produced weaker sensations than ones with narrow bowls, whatever their

height. Although increasing the volume of the bowl did not reduce the inten-

sity of perceptions, glasses with a larger bowl and rim (which are reputed to be

particularly well suited for the tasting of red wines) are not superior to the iso

glass, which has a relatively narrow bowl.

Three discoveries in particular upset conventional oenological wisdom.

First, the intensity of all perceptions was found to change with the type of glass.

Second, increasing the height of the bowl and the ratio between the diameter

of the rim and the maximum diameter of the bowl intensified all perceptions.

Finally, and contrary to what many manufacturers have claimed, the glass that

gave the best results in white wine tastings—the iso glass—was also the best

for red wines. A dogma is slain.

Wine Glasses
| 253

76

Wine and Temperature

Whether one wants to chill champagne or bring wine up to room tempera-

ture, it pays to be patient.

e v e r y o n e k n o w s t h a t c h a m p a g n e is best drunk chilled. But how

long does the bottle need to be left in the refrigerator? Does it matter whether

a child comes along and opens the door of the refrigerator in the meantime?

Or say you want to bring wine up from the cellar in order to warm it to room

temperature before a meal. How far in advance should you do this? The

use of a thermocouple, a device that precisely measures temperatures, gives

gourmets useful orders of magnitude.

Let’s start with the bottle of champagne. In a simple experiment, the

probe of the thermocouple was inserted and the bottle placed in the door

of a refrigerator at thermal equilibrium next to bottles that had been there

for more than two days and that recorded an internal temperature of 11°c

(52°c). The initial temperature of the bottle of champagne was 25°c (77°f).

Measurements taken every ten minutes showed a slow cooling: After 30

minutes the temperature of the champagne was still above 20°c (68°f), and

it took three hours to reach 15°c (59°f). Only after six hours did it fall to

12°c (54°f), proof that glass is a poor conductor of heat (which propagates

in it solely by conduction, whereas in water heat is distributed by convec-

tion as well).

254 |

Untimely Openings

Suppose you have taken this strong thermal inertia into account, but then

a child comes along and opens the door of the refrigerator. Is there a risk that

the champagne you have gone to the trouble of chilling all this time will now

be warmed up?

This time the temperature in the refrigerator was initially 8°c (46°f). Out-

side the temperature was 20°c (68°f). When the door is left open for only

seven seconds, the temperature rises to 11°c (52°f); after a few minutes it goes

back down to 8°c (46°f). When the door is left open for twenty seconds, the

temperature rises to 18°c (64°f), then drops rapidly at first and slowly after-

ward. The air inside the refrigerator heats up very quickly, then, because the

cool, dense air inside flows out from the bottom and is replaced by air entering

from outside. But this heating up involves only the air itself, not the bottles,

whose calorific capacity—and thermal inertia—is substantial. The temperature

of the bottles themselves rises hardly at all as a result of a momentary opening

of the door, and after a few minutes they cool back down with the rapid drop

in the air temperature.

Up from the Cellar

The opposite question, how long it takes to raise the temperature of a wine,

has also been neglected. Say you are looking forward to dinner at home with

friends and you want to bring up a good bottle from the cellar. Suppose, too,

that your guests happen to know that the wine should be drunk at a tempera-

ture of 18°c (64°f), but the temperature in your cellar is 12°c (54°f). How far

ahead of time should you bring the bottle upstairs?

Naturally one could work out the answer using the laws of physics. But let’s

conduct a little experiment and measure the actual change in temperature over

time. Bring up a bottle of wine whose initial temperature is 9°c (48°f). If the

temperature in the dining room is 24°c (75°f), we will find that it takes about

half an hour for it to reach a temperature of 12°c (54°f), three-quarters of an

hour to reach 14°c (57°f), an hour and a quarter to reach 16°c (61°f), an hour

and three-quarters to reach 18°c (64°f), and more than three hours to reach

20°c (68°f).

Wine and Temperature
| 255

The long period of time needed to bring a bottle up to room temperature

again results from its low thermal conductivity, especially when the difference

in temperature between the room and the bottle is small. The answer to our

earlier question, then, assuming the same ambient temperature, is that the