"Do I really need to reduce wine separately?"
I was just watching an America's Test Kitchen episode in which they said to reduce a wine/port/red wine vinegar mixture till syrupy, then add chicken stock. I couldn't help wondering why that's any better than adding the chicken stock initially and reducing the whole thing. Water loss is water loss, right? To the same point, recipes are forever distinguishing between simmering sauces slowly versus rapidly reducing them. What's the difference?
—Sent by RobC_
Reduction—the process of simmering a liquid to concentrate its color, aroma, and flavor—may seem at first like an extremely simple process whereby you're just removing water, but in fact, it's a bit more complicated than that, as any chef, distiller, or chemist, for that matter, can tell you.
There are a couple of important processes going on during reduction. The first is yes, water loss. Water molecules are held together in relatively tight formation when in liquid form, but even so, surface molecules have a tendency to get a bit overexcited and jump off into the atmosphere. This happens very slowly at lower temperature (say, a puddle drying up on a warm day), and rather rapidly at hotter temperature (the steam coming off the top of a hot tub, for instance). Eventually, when the water gets hot enough*, it'll escape very energetically indeed as it simmers or boils away.
*Specifically, when the vapor pressure of the water's surface is equal to or greater than atmospheric pressure
But here's the deal: when simmering, water is not the only thing escaping. Ever notice how when you come home to a pot of sauce simmering on the stovetop or perhaps a beautiful pot roast braising in the oven, your entire home smells of it?
Guess what: if those flavorful aromatic compounds are reaching your nose, it means they are leaving the pot.
The smaller and lighter those compounds are, the more likely they are to jump out of the pot with the evaporating water and float off into the air.
Vigorous boiling can exacerbate this effect as it agitates the molecules even harder.
Think of a pot of stock as a ball pit with dozens of kids playing in it. The balls represent small, light water molecules, while the kids (who come in all shapes and sizes) represent various flavorful molecules. Now stick this ball pit on the back of a dump truck driving down a bumpy road.
Drive that truck slowly and it'll minimize bumps. The small plastic balls will bounce a bit and escape at a relatively low rate. Drive slowly enough, and you'll even be able to reduce the number of plastic balls without losing any kids. This is what it's like to reduce a stock very, very slowly.
Speed that truck up a bit and you'll start losing your balls faster, along with a few of the skinnier kids. They fall along the side of the road, never to re-enter the ball pit.
Go faster still and you may end up losing all of the light, skinny kids, leaving you with a few balls, along with all the larger kids. You may have gotten to your destination faster, but it's at the cost of diversity and complexity.
A sauce reduced quickly will be flatter-tasting and less flavorful than a sauce reduced slowly. Try it side by side with two pots of stock or two pots of wine and you will see.
Moral: When reducing any liquid for a sauce, it's best to go low and slow.
With wine, it gets even more complicated, as you've got another factor to contend with: alcohol.
We all know that ethanol has a lower boiling point than water, so you may think that by heating an alcohol and water mix to the boiling point of alcohol, only the alcohol will be escaping. This is not in fact the case, and it's one of the factors that makes distillation complicated. When simmering an alcohol and water mix, the vapor coming off will be a mixture of alcohol and water, even if the liquid hasn't yet reached the boiling point of pure water.
This is because water molecules like to stick to each other, as do alcohol molecules. When you've got 100% water in a pot, the molecules are tightly aligned with each other and stuck tight, making it tough to separate them and cause them to jump off into the atmosphere. Add some alcohol to that mix, and the average distance between water molecules grows—thus their hold on each other is weaker, making it easier for them to evaporate. The same holds for alcohol molecules.
That's why it's nearly impossible to remove the alcohol content from a sauce or stew by simmering it on the stovetop. By the time the alcohol has been removed, most of the water will be evaporated as well.*
*It's for this reason that collecting, condensing, and re-distilling the vapor that comes off an alcohol/water mixture can only get you so far. At a certain point (about 95.6% alcohol and 4.4% water), the alcohol and water form what's known as a positive azeotrope, a mixture whose boiling point is lower than either of its constituents. It boils at 78.2°C, lower than pure alcohol (78.4°C) or pure water (100°C). The vapor coming off of it is also at the same proportion as the liquid itself, meaning that no matter how long you boil it, you'll never alter its alcohol content.**
**This is why if you work in a biology or chemistry lab, it's all right to drink the 95% ethanol on a late night, but stay away from the 98% stuff, which is distilled with benzene to remove those extra few percentage points of water.
Reducing the wine separately, then diluting the resulting reduction, is a far more efficient way of minimizing the overall final alcohol content of the dish than attempting to reduce the alcohol after combining it with the remaining liquids.
For instance, say I start with 500ml of wine with an alcohol content of 10%. I reduce it on the stovetop by half, leaving me with 250ml of reduced wine that now has an alcohol content of, say, 4% (the actual content will vary depending on the exact conditions I performed the reduction under). When I subsequently add 500ml of stock to that mix for a final volume of 750ml, I end up with a final alcohol content of 1.3%.
If, on the other hand, I start by combining 500ml wine with 500ml of stock (resulting in a liquid with 5% alcohol) and reduce it by 33% to get to a final volume of 750ml, I end up with an alcohol content of around 3% (give or take)—more than twice as high as if I had reduced it separately at the start.
Moral: If you want to end up with a reasonable amount of alcohol in your final dish, reduce wine or liquor it separately before adding your stock. (Around 1% alcohol is a good range for most sauces made with booze).
Reducing Vegetable-based Sauces
Reducing fresh vegetable-based sauces, like a tomato sauce, is another kettle of fish. In these sauces, enzymatic action comes into play. Tomatoes and other vegetables naturally contain enzymes that will break down pectin, the carbohydrate glue that holds cell walls together. Cook these raw purees at a relatively low temperature (under 180°F or so), and the enzymes will break down enough molecules into smaller bits that the sauce will thicken up pretty considerably. But the thickening is short lived. As you continue to heat the sauce, molecules continue to break down to a point where they cannot thicken at all, creating a watery sauce.
In his indispensable food science encyclopedia On Food and Cooking, Harold McGee states that for a vegetable-based sauce, rather than starting at a low simmer, "if instead the raw puree is cooked quickly close to the boil," enzymes are deactivated before they get the chance to break down pectin, which in its whole form is a more efficient thickener. "The result is a thicker sauce that requires less subsequent reduction."
Moral: For the best fresh tomato sauces, heat the tomatoes rapidly at the start to deactivate enzymes, then reduce the heat to a bare simmer to concentrate flavor. Note that as canned tomatoes have already been processed with heat, their enzymes have already been deactivated—it's totally fine to cook a canned tomato sauce at a low simmer in order to reduce while minimizing flavor loss.
Got a question for The Food Lab?
Email your questions to [email protected], and please include your Serious Eats user name in your email. All questions will be read, though unfortunately not all can be answered.