Safe to cook with alcohol

In general, we can assume that it will never boil off to zero. By a combination of reduction and dilution we can get to undetectable levels, less than in ripe fruit and some breads. The "zero BAC" requirement has to be able to handle people eating normal foods, so tends to have some margin, just not enough to have a drink.

To do this you'd simmer the wine on its own until significantly reduced, top up with water to the original volume, then use. The amount of wine per portion in many dishes is low anyway - not in everything of course, but there's often more stock than wine, further diluting it.

If this doesn't go far enough, de-alcoholised wines vary with jurisdiction but are often 0.5% or even 0.05%, the latter being less than some fruit juices, and less than vinegar (link to a recent answer of mine, where I discuss a relevant scientific paper). The alcohol in those has been removed by a more effective process than simple simmering.

You are quite safe cooking with alcohol as long as you are not using ridiculous amounts and it goes in before you cook the food (simmer/boil/flambe etc.) and not after. The ethanol component of alcohol evaporates at 78°C (173°F), while simmering happens between 85°C and 96°C(ish) so the ethanol will have boiled off.

Edit for detail - The physics of cooking (and the rest of the universe) mean that the temperature of a mixture will not significantly exceed the state change temperature of one of the components of that mixture while sufficient quantities of that component exist in a state that can absorb the excess available energy. This is because the energy required to cause a state change is significantly higher than the energy required to raise temperature so all liquid alcohol at 78°C will become alcohol steam at 78°C before any water content can become 79°C.

This is just a complicated and long winded way of saying that if you have temperature of 85°C or greater in an unpressurized and ventilated environment, then you can be 100% certain that you have no liquid alcohol in that environment. The extra few degrees are to allow for localized variations in temperature and pressure. A thick gloppy tomato sauce is going to have higher pressures which increase the state change temperature then a thin white wine reduction. If your environment is partially sealed, then there is a risk that your steam may contain alcohol molecules which can condense back into your food (affecting its flavor). This is why it is normal after adding alcohol to return to a boil on the stove before putting a lid on and simmering gently. This means the steam in the pot is clear of alcohol.

Would a chemist be able to find alcohol molecules in your food? Probably yes, if they were careful. Would a chemist or police officer be able to find alcohol molecules in the blood of someone who ate your food? No, not at levels above normal background in the human body.

As a general rule, the answer by Chris H is accurate. Most recipes that involve wine only use a small amount and typically involve cooking in a way that causes most of the alcohol to evaporate (or combust if it’s a flambé), and in general they will have no more impact on BAC than a cup of fruit juice would. In cases where it really matters you can indeed get de-alcoholised wines with very low %ABV specifically for cooking. The same (other than the de-alcoholised aspect) is usually also true of many recipes that involve other forms of alcohol (or, in some cases, you can find alcohol free versions of the same recipes).

The particular recipe you linked, however, is not such a recipe. There are two reasons that it will be an issue:

• The stock is more wine than water by a ratio of 2:1 (the recipe calls for one cup of water for the stock but uses two cups of wine for the stock). This by itself is significantly more wine than is normally found in most recipes that call for wine (most have a ratio less than 1:5).
• The way this is cooked (covered by first a layer of foil and then a lid) is intended to retain moisture. It will be just as effective at retaining the alcohol content as well, which means the alcohol will not ‘cook-off’ as is normally the case with recipes that call for wine.

This means that the stock, after cooking and only factoring in liquid components, will probably still have a %ABV of about 2/3 of that of the wine, probably somewhere around 7-9%, which is actually higher than many beers. Given this, if you really want to make this recipe, you will need to find a de-alcoholised wine that is equivalent to a Chianti (given that the recipe does not list a particular variety of Chianti, it’s probably intended to use a Chianti Classico).