# Calculating time to reduce alcohol in wine using heating method

JT Smith
I don't believe you will be able to determine the %ABV from that refractometer, at least not as easily as I suspect you are thinking it will be. If you are a winemaker you can measure the sugar content of the juice from the grapes, usually in units of Brix. From that one can infer a potential alcohol content of the finished wine. It is also possible to do a calculation to determine the alcohol content by using measurements of the juice and wine. This is because the sugar content drops in concert with an increase in ethanol. But if you just measure a random wine with a refractometer there's no way to know what the value means. Wine is not a simple ethanol-water mixture.

JT Smith
We have a number of bottles of wine that have gone off the cliff, so to speak. Usually still drinkable but some years past their prime. I use them in spaghetti sauce. I pulled one of those out and measured 250g into a large sauce pan. After 25 minutes at 60-90°C (I wasn't being especially careful) it had reduced to 108g.

I have an inexpensive optical refractometer that I purchased for $25 from amazon to make crude measurements of my espresso coffee concentration. I have tested it against a more accurate method and found it's good to about ±0.2% Brix, if I'm careful with it's use. I know how Brix relates to refractive index and I looked up some numbers for the refractive index of ethanol-water mixtures. So I figured I could use this tool to measure alcohol -- assuming it's alcohol in water alone. Starting out, the wine measured 8.7% Brix which is an RI of about 1.3485. That in turn translates into an ABV of something in the vicinity of 30%. The label says this wine is 14.4% alcohol by volume. So obviously the other components of the wine are significant. The reduced wine also measured 8.7% Brix. Clearly a coincidence. A decrease in alcohol should drive the RI (and apparent Brix) down and the increase in dissolved solids concentration should drive it up. They balanced! I then added water to the 108g of reduced wine so that it again weighed 250g. That mixture measured 3.5% Brix. That corresponds to an RI of about 1.3393. If this were a pure ethanol-water mixture it would be about 12.2% ABV. I was thinking that one might possibly infer the new ABV from the two RI readings and the ABV listed on the wine bottle. But I'm not sure how exactly. Probably I'm wasting time and old wine. It made the kitchen smell nicely though! EDIT: By the way, the old wine actually wasn't too bad. It wasn't astringent or sour and it had retained most of its character. In contrast the cooked down reconstituted wine was quite sour and otherwise pretty bland. I'm not sure what that means in terms of the chemistry, and maybe it would still be okay in certain cooked dishes. But I sure wouldn't want to drink it. Last edited: ArtZ I don't believe you will be able to determine the %ABV from that refractometer, at least not as easily as I suspect you are thinking it will be. If you are a winemaker you can measure the sugar content of the juice from the grapes, usually in units of Brix. From that one can infer a potential alcohol content of the finished wine. It is also possible to do a calculation to determine the alcohol content by using measurements of the juice and wine. This is because the sugar content drops in concert with an increase in ethanol. But if you just measure a random wine with a refractometer there's no way to know what the value means. Wine is not a simple ethanol-water mixture. You, you may very well right about this. The digital refractometer I purchased is intended for wine and beer making. And yes, it measures potential alcohol not absolute alcohol, ABV. The potential alcohol may not be indicative the actual ABV as you point out. ArtZ We have a number of bottles of wine that have gone off the cliff, so to speak. Usually still drinkable but some years past their prime. I use them in spaghetti sauce. I pulled one of those out and measured 250g into a large sauce pan. After 25 minutes at 60-90°C (I wasn't being especially careful) it had reduced to 108g. I have an inexpensive optical refractometer that I purchased for$25 from amazon to make crude measurements of my espresso coffee concentration. I have tested it against a more accurate method and found it's good to about ±0.2% Brix, if I'm careful with it's use. I know how Brix relates to refractive index and I looked up some numbers for the refractive index of ethanol-water mixtures. So I figured I could use this tool to measure alcohol -- assuming it's alcohol in water alone.

Starting out, the wine measured 8.7% Brix which is an RI of about 1.3485. That in turn translates into an ABV of something in the vicinity of 30%. The label says this wine is 14.4% alcohol by volume. So obviously the other components of the wine are significant.

The reduced wine also measured 8.7% Brix. Clearly a coincidence. A decrease in alcohol should drive the RI (and apparent Brix) down and the increase in dissolved solids concentration should drive it up. They balanced!

I then added water to the 108g of reduced wine so that it again weighed 250g. That mixture measured 3.5% Brix. That corresponds to an RI of about 1.3393. If this were a pure ethanol-water mixture it would be about 12.2% ABV.

I was thinking that one might possibly infer the new ABV from the two RI readings and the ABV listed on the wine bottle. But I'm not sure how exactly. Probably I'm wasting time and old wine. It made the kitchen smell nicely though!

EDIT: By the way, the old wine actually wasn't too bad. It wasn't astringent or sour and it had retained most of its character. In contrast the cooked down reconstituted wine was quite sour and otherwise pretty bland. I'm not sure what that means in terms of the chemistry, and maybe it would still be okay in certain cooked dishes. But I sure wouldn't want to drink it.
Some good data points for me moving forward. That's a huge discrepancy between the wine label and your refractometer measurement. Since my digital refractometer is also measuring RI, likely it won't any better in accuracy just easier to get answer with the digital display. Maybe the Brix to PA algorithm is smarter?

As a baseline, the first measurement I was going to make was to check the refractometer result against the 15% ABV stated on the bottle. So in a retry of the cal procedure, I confirmed that the unit is defective. I will return the unit on Monday and purchase a new one.

If I get agreement ( stated ABV vs. measured PA) for this initial, I'll proceed with 250mL samples using my sous vide bath. (0.1C accuracy)

Otherwise, I'll need to rethink my strategy. I did purchase a beer/winemaking hydrometer and haven't yet returned it. The downside to using this device is that it requires a 250mL sample volume. and that would be disruptive to the experiment to say the least. There are digital hydrometers but they are pricey.

Thanks for trying this experiment out; it's a big help.

ArtZ
Some good data points for me moving forward. That's a huge discrepancy between the wine label and your refractometer measurement. Since my digital refractometer is also measuring RI, likely it won't any better in accuracy just easier to get answer with the digital display. Maybe the Brix to PA algorithm is smarter?

As a baseline, the first measurement I was going to make was to check the refractometer result against the 15% ABV stated on the bottle. So in a retry of the cal procedure, I confirmed that the unit is defective. I will return the unit on Monday and purchase a new one.

If I get agreement ( stated ABV vs. measured PA) for this initial, I'll proceed with 250mL samples using my sous vide bath. (0.1C accuracy)

Otherwise, I'll need to rethink my strategy. I did purchase a beer/winemaking hydrometer and haven't yet returned it. The downside to using this device is that it requires a 250mL sample volume. and that would be disruptive to the experiment to say the least. There are digital hydrometers but they are pricey.

Thanks for trying this experiment out; it's a big help.
This site has some very small hydrometers but the site is not in English (Dutch?):
https://nl.dreamstime.com/de-hydrometer-van-alcoholmeter-een-hand-image136343314

JT Smith
Some good data points for me moving forward. That's a huge discrepancy between the wine label and your refractometer measurement. Since my digital refractometer is also measuring RI, likely it won't any better in accuracy just easier to get answer with the digital display. Maybe the Brix to PA algorithm is smarter?

Of course there's a discrepancy. Wine is not only ethanol and water.

Your refractometer "PA" will simply be the same measurement but in different units. The Brix value refers to the sucrose content by weight. Ten grams of sucrose in 100 grams of solution is 10% Brix. Grapes are actually mostly fructose and glucose but the difference in RI is not large and your refractometer may even be applying a correction. The PA reading is just another way of expressing the sugar concentration. You could probably google a formula for doing it yourself.

Each molecule of fructose or glucose can be converted into two molecules of ethanol during fermentation. So it's possible to estimate the eventual ABV by knowing the sugar concentration. Wines, even dry wines, have residual sugar, as well as other "stuff". So it's just an estimate, a "potential". The way alcohol is traditionally calculated is by measuring before and after fermentation. Even then there are some assumptions being made.

Using a hydrometer really isn't better. They are awkward because they require large samples, temperature control is harder as a result, and the good ones are glass and quite fragile. And in the end you have the same issue: how do you measure ethanol in a solution that contains other components?

One way would be with a GC. That would be the tool of choice if you had one in your kitchen. I wonder if one would fit next to your sous vide?

JT Smith
I looked at the numbers I came up with yesterday and realized I had been reading the wrong column in a table. When I did it again I got this:

8.7% Brix - original wine measurement
5.2% Brix - what a 14.2% w/w ethanol-water solution would measure

Subtract those two and you get 3.5% Brix, which is what I measured in my reduced/reconstituted wine. A naive assessment of that would be that 3.5% Brix is the residual "sugar" of the wine and all of the ethanol was gone. I'm pretty sure that isn't true so either I botched the experiment somehow or the hopeful assumption that a dealcoholized wine would refract light as if it were a sucrose solution is pretty far off.

Maybe a hydrometer would work better. But I still think using about 1/3 as much as wine in your food is the best solution.

ArtZ
Of course there's a discrepancy. Wine is not only ethanol and water.

Your refractometer "PA" will simply be the same measurement but in different units. The Brix value refers to the sucrose content by weight. Ten grams of sucrose in 100 grams of solution is 10% Brix. Grapes are actually mostly fructose and glucose but the difference in RI is not large and your refractometer may even be applying a correction. The PA reading is just another way of expressing the sugar concentration. You could probably google a formula for doing it yourself.

Each molecule of fructose or glucose can be converted into two molecules of ethanol during fermentation. So it's possible to estimate the eventual ABV by knowing the sugar concentration. Wines, even dry wines, have residual sugar, as well as other "stuff". So it's just an estimate, a "potential". The way alcohol is traditionally calculated is by measuring before and after fermentation. Even then there are some assumptions being made.

Using a hydrometer really isn't better. They are awkward because they require large samples, temperature control is harder as a result, and the good ones are glass and quite fragile. And in the end you have the same issue: how do you measure ethanol in a solution that contains other components?

One way would be with a GC. That would be the tool of choice if you had one in your kitchen. I wonder if one would fit next to your sous vide?
I ordered a replacement refractometer yesterday. And yes, my fancy unit with uP control and digital display, is still measuring residual sugar based on the Brix measurement. The fancy accoutrements may be just be putting lipstick on a pig. I still want to get a baseline with the digital refractometer at full 15% ABV and then adding water to the wine sample to dilute it to some low ABV, 3-4% and perform a retest. This is just a sensibility test.

ArtZ
Of course there's a discrepancy. Wine is not only ethanol and water.

Your refractometer "PA" will simply be the same measurement but in different units. The Brix value refers to the sucrose content by weight. Ten grams of sucrose in 100 grams of solution is 10% Brix. Grapes are actually mostly fructose and glucose but the difference in RI is not large and your refractometer may even be applying a correction. The PA reading is just another way of expressing the sugar concentration. You could probably google a formula for doing it yourself.

Each molecule of fructose or glucose can be converted into two molecules of ethanol during fermentation. So it's possible to estimate the eventual ABV by knowing the sugar concentration. Wines, even dry wines, have residual sugar, as well as other "stuff". So it's just an estimate, a "potential". The way alcohol is traditionally calculated is by measuring before and after fermentation. Even then there are some assumptions being made.

Using a hydrometer really isn't better. They are awkward because they require large samples, temperature control is harder as a result, and the good ones are glass and quite fragile. And in the end you have the same issue: how do you measure ethanol in a solution that contains other components?

One way would be with a GC. That would be the tool of choice if you had one in your kitchen. I wonder if one would fit next to your sous vide?
I ordered a replacement refractometer yesterday. And yes, my fancy unit with uP control and digital display, is still measuring residual sugar based on the Brix measurement. The fancy accoutrements may be just be putting lipstick on a pig.
I looked at the numbers I came up with yesterday and realized I had been reading the wrong column in a table. When I did it again I got this:

8.7% Brix - original wine measurement
5.2% Brix - what a 14.2% w/w ethanol-water solution would measure

Subtract those two and you get 3.5% Brix, which is what I measured in my reduced/reconstituted wine. A naive assessment of that would be that 3.5% Brix is the residual "sugar" of the wine and all of the ethanol was gone. I'm pretty sure that isn't true so either I botched the experiment somehow or the hopeful assumption that a dealcoholized wine would refract light as if it were a sucrose solution is pretty far off.

Maybe a hydrometer would work better. But I still think using about 1/3 as much as wine in your food is the best solution.
Unless I am missing the point, it sounds like your recheck was revealing that there's some hope for the refractometer. BTW, what is a GC?

JT Smith
Unless I am missing the point, it sounds like your recheck was revealing that there's some hope for the refractometer. BTW, what is a GC?

GC refers to gas chromatography. Probably more than you are willing to spend. :-)

Being human we can always hope... even if it's impossible. I'm pessimistic about using a refractometer because we don't have a formula or table for the RI of a dealcoholized wine, never mind your particular wine. Somebody smarter and more experienced could probably tell you if there were reason to hope.

I wonder if it might work better to measure the density with a hydrometer. You could heat a small sample of the wine in an oven until it is fully dry and weigh the residue. You'd need a pretty good scale. Alternatively you could heat some wine until it is so reduced that you can feel confident that very little ethanol remains. Then reconstitute that and measure the density. Either of those approaches will tell you how much dissolved material is in the wine, besides ethanol. Then I think you could subtract that off of hydrometer measurements of wine with ethanol. I think that might work. You could check it against the unadulterated wine to see if it really works out to 15%. I could just be dreaming. I'm not a scientist, I just play one in the kitchen sometimes.

ArtZ
GC refers to gas chromatography. Probably more than you are willing to spend. :-)

Being human we can always hope... even if it's impossible. I'm pessimistic about using a refractometer because we don't have a formula or table for the RI of a dealcoholized wine, never mind your particular wine. Somebody smarter and more experienced could probably tell you if there were reason to hope.

I wonder if it might work better to measure the density with a hydrometer. You could heat a small sample of the wine in an oven until it is fully dry and weigh the residue. You'd need a pretty good scale. Alternatively you could heat some wine until it is so reduced that you can feel confident that very little ethanol remains. Then reconstitute that and measure the density. Either of those approaches will tell you how much dissolved material is in the wine, besides ethanol. Then I think you could subtract that off of hydrometer measurements of wine with ethanol. I think that might work. You could check it against the unadulterated wine to see if it really works out to 15%. I could just be dreaming. I'm not a scientist, I just play one in the kitchen sometimes.
I went to the sources listed by participants in this thread, i.e., CRC etc. and was able to find density and specific heat for ethanol-water mixtures as a function of temperature but no RI. And, as you've pointed out several times, wine is not a simple ethanol-water mixture. There are some fancy hydrometers requiring only a 100uL sample volume- cost \$3500.

Call me an optimist, but if the new refractometer isn't DOA, I'll try my plan to see what the PA of the undiluted wine measures in at, then dilute a sample with water multiple times and see what the meter returns for PA. In the limit, as the dilution approaches all water, the PA should approach zero. If I plot the measured PA as a function of ABV in the diluted samples, maybe we can learn something. I doubt the PA vs. ABV function will be linear, but if PA is tracking ABV closely this should be a 1-1 relation. Keep dreaming, right. Even if it's non-linear, we may be able use the non-linearity as a correction factor.

JT Smith
Call me an optimist, but if the new refractometer isn't DOA, I'll try my plan to see what the PA of the undiluted wine measures in at, then dilute a sample with water multiple times and see what the meter returns for PA. In the limit, as the dilution approaches all water, the PA should approach zero. If I plot the measured PA as a function of ABV in the diluted samples, maybe we can learn something.

PA is just another unit for sugar content. It really only makes sense in the context of unfermented juice where sugar is the main ingredient. In your wine the index of refraction differs from that of pure water not because it's a dilute syrup but because of ethanol and other components. Your refractometer is not designed for what you want it to do.

If you dilute the wine you will reduce the concentration of both the ethanol and the other dissolved solids by the same proportions. So you'll end up with a plot that describes the apparent Brix (or PA if you prefer) versus dilution level of your wine. Armed with that plot you could determine the dilution level by measuring a diluted wine. But it won't tell you that the wine started out at 15%. And if you boil off some of the ethanol it won't be able to tell you how much is left.

Fra
The alcohol-reduced wine will be used for cooking.
This is probably missing youre point, but as there may be several reasons of flavouring for using wine in cooking.

Isn't there some sort of rice vineager or balsamico hybrid, that you can use off the shelf to get a similar effect?

I have no flavour experience with chinese rice wine, but I often use balsamico vinegar or balsamico reduction in cooking. Not quite the same as wine of course, but does add some deep grape flavour and some balanced acidity in a simpler way.

/Fredrik

JT Smith
Yeah, it's not the same thing as wine...

JT Smith
I think what you really want is a plot of alcohol % of your wine versus refractometer output. I have an idea that might work to produce that, probably it won't. Maybe I will try it tomorrow. It's fun to be a little kid with a chemistry set.

ArtZ
This is probably missing youre point, but as there may be several reasons of flavouring for using wine in cooking.

Isn't there some sort of rice vineager or balsamico hybrid, that you can use off the shelf to get a similar effect?

I have no flavour experience with chinese rice wine, but I often use balsamico vinegar or balsamico reduction in cooking. Not quite the same as wine of course, but does add some deep grape flavour and some balanced acidity in a simpler way.

/Fredrik
Thanks but no. Shaoxing has a very unique flavor profile.

ArtZ
I looked at the numbers I came up with yesterday and realized I had been reading the wrong column in a table. When I did it again I got this:

8.7% Brix - original wine measurement
5.2% Brix - what a 14.2% w/w ethanol-water solution would measure

Subtract those two and you get 3.5% Brix, which is what I measured in my reduced/reconstituted wine. A naive assessment of that would be that 3.5% Brix is the residual "sugar" of the wine and all of the ethanol was gone. I'm pretty sure that isn't true so either I botched the experiment somehow or the hopeful assumption that a dealcoholized wine would refract light as if it were a sucrose solution is pretty far off.

Maybe a hydrometer would work better. But I still think using about 1/3 as much as wine in your food is the best solution.
Using about 1/3 as much wine in the recipe may be a good solution. For different recipes, the wine component can be as small as 1 tsp or as much as 2 tbs. Splitting these measures seems tedious.

ArtZ
Someone mentioned earlier of just simply reducing the ABV by adding water to a given volume of wine. With this in mind, using the simple relation C1*V1 = C2*V2 and solving for V2: V2= (C1/C2)*V1. V1 will be the starting volume. V2 will be the increased volume accounting for the added water. So for reducing from 15% ABV to 10%, with a starting volume of 250mL we get V2=(.15/.10) *.250= .375 so we'll be adding 125mL of water. This may not be radical.

If we do this again, using this time the same 250mL starting volume but now reduce the 15% ABV to 5%, we will be adding quite a bit of water. V2=(.15/.05) *.250= .750. We would need to add 500 mL to the 250 mL sample to achieve the ABV reduction. If then using this 1:1 in a recipe, seems that the flavor would be lost.

JT Smith
Someone mentioned earlier of just simply reducing the ABV by adding water to a given volume of wine....

...If we do this again, using this time the same 250mL starting volume but now reduce the 15% ABV to 5%, we will be adding quite a bit of water. V2=(.15/.05) *.250= .750. We would need to add 500 mL to the 250 mL sample to achieve the ABV reduction. If then using this 1:1 in a recipe, seems that the flavor would be lost.

Yes, it will have less flavor if you use less. You'll have to figure out the tradeoff between less unadulterated wine and more cooked down wine. Try cooking some wine and taste it.

Another thing: ethanol is a flavor component. Remove the ethanol from wine or beer and it doesn't taste the same. Ethanol is sweet. It's probably why my reduced wine tasted so sour. It was out of balance without the ethanol.

JT Smith
I did another experiment this morning. I did a few things actually.

First I measured vodka at a few different dilutions with my refractometer, just as a consistency check. The data lined up with the numbers I found on the internet.

Then I opened an old 750ml bottle of red wine (14.4%). I measured it with my refractometer: 8.4% Brix. Then I sealed some in a small jar and put it in boiling water for an hour. I wanted to see if simply heating the wine would change the refractive properties. It didn't.

At the same time I took the remaining 650ml and reduced it in a double boiler for a little over an hour, reducing it to 300ml. After cooling it I split it up into five portions, reconstituting each to the correct volume. But I didn't just use water; I also included varying measured amounts of 40% vodka. That gave me a series of samples at varying ABV. I didn't know what the values were but I knew by how much they differed. When plotted they displayed essentially the same slope as the ethanol-water (or diluted vodka) samples. And that slope is roughly 2.4% ABV/%Brix.

So I could figure out what how much my wine was reduced to. Reconstituted it was 2.3% ABV. The original reduced wine, before I added water, would have been 5.0% ABV.

Since the slope roughly matched pure (or nearly pure) ethanol-water I suspect that this is not something that will only work with the wine I chose, or even red wines in general. Probably it would work for yours as well. But I don't know that. You could always duplicate what I did this morning. Or just figure it's close enough.

The cooked wines sure didn't taste that nice though. The best of the bunch was the one I added the most vodka to. That one, which was just barely shy of the original strength of the 14.4% wine, had a sweetness the others lacked.

ArtZ
I did another experiment this morning. I did a few things actually.

First I measured vodka at a few different dilutions with my refractometer, just as a consistency check. The data lined up with the numbers I found on the internet.

Then I opened an old 750ml bottle of red wine (14.4%). I measured it with my refractometer: 8.4% Brix. Then I sealed some in a small jar and put it in boiling water for an hour. I wanted to see if simply heating the wine would change the refractive properties. It didn't.

At the same time I took the remaining 650ml and reduced it in a double boiler for a little over an hour, reducing it to 300ml. After cooling it I split it up into five portions, reconstituting each to the correct volume. But I didn't just use water; I also included varying measured amounts of 40% vodka. That gave me a series of samples at varying ABV. I didn't know what the values were but I knew by how much they differed. When plotted they displayed essentially the same slope as the ethanol-water (or diluted vodka) samples. And that slope is roughly 2.4% ABV/%Brix.

So I could figure out what how much my wine was reduced to. Reconstituted it was 2.3% ABV. The original reduced wine, before I added water, would have been 5.0% ABV.

Since the slope roughly matched pure (or nearly pure) ethanol-water I suspect that this is not something that will only work with the wine I chose, or even red wines in general. Probably it would work for yours as well. But I don't know that. You could always duplicate what I did this morning. Or just figure it's close enough.

View attachment 323886

The cooked wines sure didn't taste that nice though. The best of the bunch was the one I added the most vodka to. That one, which was just barely shy of the original strength of the 14.4% wine, had a sweetness the others lacked.
If you sealed some of the wine in a small jar and put it in boiling water for an hour, I wouldn't expect that it would change the refractive properties. After all, it was a closed system. All evaporated components would be condensed and returned to the wine, right? And at boiling temperature, I wouldn't expect anything flavorful.

Yah, I want to try out what you did- really awesome, thanks!

Ebay says I'll get my refractometer on Weds, BTW, Amazon credited my account and did not request a return of the defective refractometer which is surprising considering how much it cost. I have been in touch with the manufacturer- they requested the SN of the defective unit, which I provided. No word back from them yet.

JT Smith
If you sealed some of the wine in a small jar and put it in boiling water for an hour, I wouldn't expect that it would change the refractive properties. After all, it was a closed system. All evaporated components would be condensed and returned to the wine, right? And at boiling temperature, I wouldn't expect anything flavorful.

I wasn't sure. I thought that it was possible that there were chemical changes in cooked wine that might matter. I figured I had to eliminate that variable.

As for the taste, what's the difference between 80°C and 100°C? Both of those are hot enough to drive off aromatics and probably induce other changes in the wine. The wine I reduced in a double boiler stayed within the range 65-85°C. It didn't taste very good either. Flat, dull, and sour. Undrinkable as a beverage.

Maybe in the context of cooking it doesn't matter. I don't know. I was taught to never cook with wine I wouldn't be happy to drink. But it probably depends. Wine added to a dish that simmers for an hour -- what's the difference? But if you're adding a very aromatic wine in the last seconds of a stir-fry then heating it for tens of minutes prior might ruin it.

ArtZ
I wasn't sure. I thought that it was possible that there were chemical changes in cooked wine that might matter. I figured I had to eliminate that variable.

As for the taste, what's the difference between 80°C and 100°C? Both of those are hot enough to drive off aromatics and probably induce other changes in the wine. The wine I reduced in a double boiler stayed within the range 65-85°C. It didn't taste very good either. Flat, dull, and sour. Undrinkable as a beverage.

Maybe in the context of cooking it doesn't matter. I don't know. I was taught to never cook with wine I wouldn't be happy to drink. But it probably depends. Wine added to a dish that simmers for an hour -- what's the difference? But if you're adding a very aromatic wine in the last seconds of a stir-fry then heating it for tens of minutes prior might ruin it.
I think that there's a lot to be said for the loss of volatile aromatics whether heat induced or just from evaporation. I know that leaving an unfinished glass of wine out overnight yields a sour, flat undesirable drink. And then, when constituent compounds are denatured with heat, it can't be good. Different recipes call for the addition of wine at different times. Several recipes I make call for the addition of the wine to the meat marinade which typically consists of soy sauce, wine, the meat, aromatics like ginger and garlic, cornstarch, and vegetable oil. In this scenario, the high ABV wine is concentrated in a small volume. Marinating times of up to an hour allow the alcohol 'flavor' to permeate the meat. These types of recipes are where the alcohol 'flavor' becomes obvious.

JT Smith
I think wine changes primarily due to oxidation not just loss of aromatics. But for sure both aromatic compounds and ethanol will evaporate preferentially even at room temperature. You could remove ethanol with a low temperature distillation, aided by a vacuum pump. But you'd still lose the aromatics. One strategy for dealing with that is to condense and collect the initial distillate. It will be enriched in alcohol and also the aromatic compounds. You could keep that separate and then add it back to the mixture after removing most of the alcohol. I know this approach is used for making freeze-dried coffee, at least the better versions of it. It's probably tricky figuring out the right fraction to keep.

All the different uses for wine in food make it more complicated both in how well it holds up as well as how it is perceived.

I wonder: How did you come up with the number 4% ABV for wine that would be okay for your wife? Isn't the total amount of alcohol consumed the issue? It seems like that would vary a lot depending on the recipe and serving size.

ArtZ
I think wine changes primarily due to oxidation not just loss of aromatics. But for sure both aromatic compounds and ethanol will evaporate preferentially even at room temperature. You could remove ethanol with a low temperature distillation, aided by a vacuum pump. But you'd still lose the aromatics. One strategy for dealing with that is to condense and collect the initial distillate. It will be enriched in alcohol and also the aromatic compounds. You could keep that separate and then add it back to the mixture after removing most of the alcohol. I know this approach is used for making freeze-dried coffee, at least the better versions of it. It's probably tricky figuring out the right fraction to keep.

All the different uses for wine in food make it more complicated both in how well it holds up as well as how it is perceived.

I wonder: How did you come up with the number 4% ABV for wine that would be okay for your wife? Isn't the total amount of alcohol consumed the issue? It seems like that would vary a lot depending on the recipe and serving size.
This alcohol 'witch hunt' is somewhat convoluted and perverted. I started cooking when I was 7 years old. My heroes then were Julia Child and Joyce Chen. Both chefs used wines in their cooking; quite generously sometimes. As I have been cooking continuously over all these years, I never thought twice about alcohol (wine or spirits) to a cooked dish.

My second wife was Japanese and taught me a lot about Japanese cooking which invariably contains alcohol in some form such as Sake or Mirin, etc.

So, fast forward 10 years and a new wife. Though an American born Chinese lady, whose family was running a Chinese restaurant, (22 years) she is a total non-drinker - averse to alcohol in any form. When it became clear that I was the chef at home, I unrolled my repertoire of world-wide of food creations.

Needless to say, she was able to detect almost any residual alcohol in my dishes. The bottom line was that she would not eat the food.

Aah, so where did the 4% ABV come from? Part SWAG and some experience with some low ABV wines, i.e., Sake and Sherry. With those low ABV wines, she did not complain of the 'alcohol taste' as she calls it.

It's interesting though that she never complains of the 'alcohol taste' when we get Chinese restaurant food. Go figure. :-)

JT Smith
Thanks for the story. After living on Kraft Mac & Cheese, hot dogs, and canned baked beans through college I developed an interest in cooking and nutrition. I had not heard of Joyce Chen before but I used to watch Julia, Jeff Smith, Martin Yan, Jacques Pépin, Justin Williams, and a couple of others. There was a stream of cooking shows on PBS on Saturday mornings and I used to check out cooking books from the library to study. I'm not a good cook by nature but at least I know something about it now. My wife happily leaves the job to me, pretending that she is incompetent.

But one thing you wrote doesn't make sense to me. Both Sake and Sherry are normally stronger than 15% alcohol. I've never come across low alcohol versions of them. That's not to say they don't exist.

ArtZ
Thanks for the story. After living on Kraft Mac & Cheese, hot dogs, and canned baked beans through college I developed an interest in cooking and nutrition. I had not heard of Joyce Chen before but I used to watch Julia, Jeff Smith, Martin Yan, Jacques Pépin, Justin Williams, and a couple of others. There was a stream of cooking shows on PBS on Saturday mornings and I used to check out cooking books from the library to study. I'm not a good cook by nature but at least I know something about it now. My wife happily leaves the job to me, pretending that she is incompetent.

But one thing you wrote doesn't make sense to me. Both Sake and Sherry are normally stronger than 15% alcohol. I've never come across low alcohol versions of them. That's not to say they don't exist.
Glad you enjoyed the story. :-) When I tell people that I have over 200 cookbooks in my library, yes, friends are astounded. Bargain table books was where many came from. Most of these are stuffed with mini post-it notes that denote recipes I plan to make, someday. :-) I've got to say that I don't recall in what era of my cooking tenure that low alcohol wine became important. Low alcohol wine products availabilty wax and wane depending on depending on demand. Right now, low alcohol wine and beer products are becoming more prevalent. You are right about the Sake and Sherry- don't know what era the low-alcohol versions were available commonly.

Serendipitously, I talked to a long time friend who is an avid camper (former Eagle Scout) and outdoorsman. Somehow we latched onto a conversation about water purification while camping. He said that his first choice is always to boil unknown water if possible. His next choice are portable RO systems using a .01 micron filter. I told him what I was trying to do with the cooking wine and he explained:

In the reverse osmosis process, the permeate is the part that passes through (permeates) the filter, and the retentate is the part retained by the filter. For purifying camping water, the permeate is what you drink, discarding the retenate which has the bad cooties.

In your case, if you want to retain the wine flavor and reduce the alcohol, do the following: put your wine through the RO filter, retaining the retentate that did not pass through the filter. If you want alcohol free wine, simply re-hydrate the retentate and you'll have your alcohol-free wine. If you want some alcohol, you can distill the permeate to the alcohol level you desire and add it back to the retentate obtained in the RO filtration.

RO systems are avalable for camping- lots of them. I just don't know how they handle the retentate.

Tom.G
JT Smith
Yes, I am aware of the use of RO for dealcoholization. I didn't mention it because I thought it even less likely a DIY project than vacuum distillation, which is also probably not going to happen in your kitchen. Maybe I'm wrong but I got the impression from what little I read that it requires very high pressure to achieve adequate separation. I'm not sure how you do that with a gravity or hand pump powered camping water filter.

Here's a paper that discusses the many different ways that low alcohol wine is produced:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8535880/pdf/foods-10-02498.pdf

ArtZ
Thanks for the paper; it was very informative! Many ways to skin the proverbial cat. What was conspicuously absent though was a heat only methodology for alcohol reduction. The closest, which you have mentioned in prior posts, is vacuum distillation. The looming question for me is what to move forward with? Tomorrow, I'll receive my replacement refractometer. Should I continue with the experiments I planned? I don't know.

As a career-long researcher, I have known when to call it a day when contradictory and corroborated evidence derails the current direction of pursuit. I guess I would ask now: what is the simplest way to get to my endpoint?

ArtZ
Glad you enjoyed the story. :-) When I tell people that I have over 200 cookbooks in my library, yes, friends are astounded. Bargain table books was where many came from. Most of these are stuffed with mini post-it notes that denote recipes I plan to make, someday. :-) I've got to say that I don't recall in what era of my cooking tenure that low alcohol wine became important. Low alcohol wine products availabilty wax and wane depending on depending on demand. Right now, low alcohol wine and beer products are becoming more prevalent. You are right about the Sake and Sherry- don't know what era the low-alcohol versions were available commonly.

Serendipitously, I talked to a long time friend who is an avid camper (former Eagle Scout) and outdoorsman. Somehow we latched onto a conversation about water purification while camping. He said that his first choice is always to boil unknown water if possible. His next choice are portable RO systems using a .01 micron filter. I told him what I was trying to do with the cooking wine and he explained:

In the reverse osmosis process, the permeate is the part that passes through (permeates) the filter, and the retentate is the part retained by the filter. For purifying camping water, the permeate is what you drink, discarding the retenate which has the bad cooties.

In your case, if you want to retain the wine flavor and reduce the alcohol, do the following: put your wine through the RO filter, retaining the retentate that did not pass through the filter. If you want alcohol free wine, simply re-hydrate the retentate and you'll have your alcohol-free wine. If you want some alcohol, you can distill the permeate to the alcohol level you desire and add it back to the retentate obtained in the RO filtration.

RO systems are avalable for camping- lots of them. I just don't know how they handle the retentate.
Yeah, that's my question too. what do these systems do with the retentate. You are right, it's the rentent that I want. But, making this more complicated, there may be volatile aromatics in the permeate. that should be fractionally distilled off. How to determine this?

JT Smith
I guess I would ask now: what is the simplest way to get to my endpoint?

Move on to wife #4?

Tom.G
Why don't you just buy some non-alcoholic wine? It's readily available, and done by people with better equipment than you are likely to have.
Shaoxzing wine has a very distinctive flavor profile, so you would not want to substitute some generic non-alcoholic wine.

Aah, so where did the 4% ABV come from? Part SWAG and some experience with some low ABV wines, i.e., Sake and Sherry. With those low ABV wines, she did not complain of the 'alcohol taste' as she calls it.
How are Sake and Sherry low ABV?? The ones I know are in the 15-20% ABV range.

ArtZ
Move on to wife #4?
You know, I just cracked up when I saw your last response! LOL! No, she's a keeper. I am trying to address a larger audience. Maybe I don't need to, I don't know. The separating of food for consumption has grown to be, in my opinion, ridiculous. At the research lab I worked at before I retired, there were separate accommodations made for vegetarian, non- vegetarian, vegan, halal, kosher, food cooked with alcohol, and many more! This included separate refrigerators, microwaves, storage cabinets, etc. Also, washable utensils (dishwasher washed) were required to be separated by use category before going into the dishwasher. Only like with like. (vegetarian, vegan, etc.) We ran dishwasher loads with 1 fork, a plate, and 5 spoons to accommodate this. Any alcohol touched glassware or utensils were washed separately in the dishwasher, alone.
So, I don't think I'm alone with pointing out these outrageous food fetishes. These were all PhDs! (from around the world) I whizzed past all this by eating all my lunches out!
How are Sake and Sherry low ABV?? The ones I know are in the 15-20% ABV range.
Oh, you are totally right these are in the 15-20% ABV range. Over the years, I was able to obtain low ABV versions of these wines in the 3.5 -5 % range. Cooking with these at the time, no one reported the 'alcohol taste.'

JT Smith
I think it's a tough problem you're considering.

Last year I decided I'd try some dealcoholized beers. The number of them has exploded and I was able to buy 18 different types without much trouble. Many of them were horrible and went down the sink quickly. Some were okay but odd in character. And a few were actually pretty good. I liked them as refreshing beverages but there was no mistaking them with actual beer. No way!

There are different ways these beers are made. I think many have the alcohol removed via RO. One brewer claims theirs are not dealcoholized. Instead the beer ferments without producing very much ethanol. It is a proprietary process but I imagine it's a combination of mash profile and yeast selection.

You could brew your own Shaoxing-style wine. From what I just read it sounds very similar to sake production and doesn't look hard to do. It would take time and it's probably not trivial to do well. But it would open the door for experimenting with different yeast strains. Another trick is to arrest the fermentation to limit the alcohol. You end up with a sweeter wine that way. But in a cooked dish a little extra sweetener might be acceptable.

I'm going to have to go and buy some Shaoxing wine. I wonder if it's hard to find? I went looking for brewed Mirin (yet another rice wine) last year and discovered that it is either super hard to find or simply not available in the San Francisco Bay Area. Lots of inexpensive fake Mirin though.