well.
technically the correct way of saying that is "the vapor pressure of ethanol at the distillation temperature is higher than the vapor pressure of water".
vapor pressure being the partial pressure of the vapor in the atmosphere above the liquid.
the vapor pressure of one component of a mixture of liquids b and c is given by Raoult's law:
P_b = (x_b,l)(P*_b)
where P_b is the vapor pressure of b, x_b,l is the molar fraction (= n_b,l/n_b,l+n_c,l) of liquid b in the mixture, and P*_b is the vapor pressure of pure b at the temperature T of the system.
Raoult's law can also be used to find the composition of a mixture of vapors above a liquid:
x_b,v = (x_b,l)(P*_b) / P, and
x_c,v = (x_c,l)(P*_c) / P
where P is the sum of the partial vapor
pressure of the system:
P = P*_bP*_c / x_b,v(P*_c - P*_b) + P*b
exciting innit?
well, you can then use this to plot an incredibly, almost unbelievably exciting curve--actually, an area bounded by two curves, which you can use to easily predict the composition of the vapor!
but only at constant temperature, yo.
so you have to adapt this and get the constant-pressure version:
x_b,l = P* - P*_c(T) / P*_b(T) - P*_c(T)
and
x_b,v = P*_b (T) P* - P*_c(T) / P* P*_b(T) - P*_c(T)
which you can then use to predict the dependence of the composition of the vapor upon the boiling point of the mixture!
yup, you can actually trace lines up from x°_b to the lower curve of the resultant graph, then over to where it intersects the upper curve, then back down, and that's your new x_b,l2.
this is when things start to suck.
because a simple distillation will result in a mixture having molar fraction of b = x_b,l2 < 1. less than 1? right. this is a problem for those of you who want a) 95% ethanol from rotting grain-ethanol-water b) absolute ethanol from ethanol-methanol.
so you have to do a fractional distillation. you don't actually *use* plates like an oil refinery, you use a column packed with glass rings or something, and allow the vapor to rise slowly, creating a heat gradient wherein the vapor refluxes slowly as it climbs upward, resulting essentially in a large number of distillations equivalent to a certain number of "theoretical plates".
until you get to 95% ethanol, and then intermolecular attraction causes ethanol and water to have the same boiling point. look, it's an azeotrope!
now, since you're probably going to dilute this, it doesn't really matter, but if there's any methanol in there you're going to have a problem.
i don't know the vapor pressure of methanol, and i wouldn't do the calculations anyway.
the basic idea is that roches himself has been called on the following:
a) saying that the boiling point of a substance is the point at which the vapor pressure of the gas phase above the mixture is equal to the atmospheric pressure, when it is in fact the TEMPERATURE at which the vapor pressure of the gas above the mixture is equal to the atmospheric pressure, and
b) pronouncing "azeotrope" incorrectly.
this is especially interesting in light of the fact that no one else in the class knew either one.
and the end result of the abovementioned facts is that you, even if you were me and had a fractional distilling apparatus, will inevitably arrive at one of the following:
a) you drink methanol, go blind, and wish you had spent the $4.25 on a 40 of OE 800,
b) you do the extraction correctly, still end up with some methanol cause you didn't do the fermentation correctly, but it's not enough to hurt you, or
c) you wonder exactly why you're distilling ethanol from water when you have a fractional distillation column, because you're in the same room as a 500 gram jar of o-chlorobenzonitrile and no one's looking.