Antoine's equation is an approximation, and in the case of ethanol and water, not the greatest, at least not with a single set of coefficients. There are higher order extensions to Antoines' equations that will give more accurate values if you really wanted to do it this way. Clearly the approximation used by Nixon is in error since there is no longer an azeotrope of ethanol and water with this approximation.

Also, Mike was a little careless with his use of both mole fractions and ABV, a conversion which is not straight forward and requires an additional calculation. For example a 49.6% ABV solution has 391.1 g/L of ethanol in it. This is 42% alcohol by weight and has a mole fraction of 20.1%. (since volumes are not additive it make the conversions tricky) (From CRC) I think that this is where the largest deviations come from when comparing Nixon's Antoine equation numbers with other calculators out there.

-Loneswinger

Edit: I checked the tabulated data on

http://homedistiller.org/calc.htm and it's consistent with the measurements in Perry. The calculator on that page is based on a polynomial fit to the VLE data. However, I got slightly different conversions than what is on there going from %mass to %volume. The %molar to %mass conversion appears to be correct though (the Perry data is listed as mole fraction). The differences between mine and what is on the site are less than 2% so I am not sure where the difference is coming from. The calculator on this page

http://homedistiller.org/ under the theory tab, gives slightly different values from the other also, so there are some inconsistencies but all less than ~ 2%.

There are probably updated VLE data that should be used for more up to date calculators. But, I think the ones on the HD site are sufficient for our purposes.