Talk:Antoine equation

dimention-consistent form
The dimention-consistent form is desired for Antoine equation:
 * $$\log_{10} (p/p_0) = A-\frac{B}{C+T},$$

where it is common that $$p_0$$ is given by $$1\,\text{mmHg}$$ and $$T$$ is given in degree Celsius. Kkddkkdd (talk) 09:19, 15 September 2012 (UTC)
 * I suppose that you mean dimensionally consistent. The form you suggest is closer to being dimensionless, but still A, B and C have different units, so it is not much better than the original form.The form given is standard in chemical engineering, e.g. Poling et al.'s "The Properties of Liquids and Gases". AlanParkerFrance (talk) 12:08, 17 October 2012 (UTC)

Example Calculation - problems
As of Sept 9, 2013 the section Example Calculation claims: "This example shows a severe problem caused by using two different sets of coefficients. The described vapor pressure is not continuous—at the normal boiling point the two sets give different results. This causes severe problems for computational techniques which rely on a continuous vapor pressure curve." [ the two predicted VPs are 760.0 and 761.0 mmHg]

I have several problems with this: #1. (761-760)/760 = 1/760 = 0.13% ← this difference is not significant when we expect the prediction to only be accurate to within several percent, the difference does NOT show "a severe problem". #2. The example does NOT show that the described VP is discontinuous. (it does show it is not single valued (function), to show discontinuity requires a more sophisticated approach. I do not argue the two equations are continuous, just that the example does not "show" it #3. To repeat point #1, the "different" results are not significantly different - at least not without a stated accuracy/error measure and #4 any computational technique which relies on finite piece-wise continuous equations should easily be able to handle transitions between equations - especially if they overlap. (how this is handled depends on the heuristic the programmer chooses.) I do not disagree that with some programs, the discontinuity will cause problems, but so what? That is the fault of the program, not the underlying data.

It goes on to state: "Two solutions are possible:...." This is nonsense. Many solutions are possible, including just giving the user multiple answers, or giving the user a #N/A result. Interpolation or weighted interpolation between the two (or more) results is an obvious solution, yet is NOT one of the two "possible solutions". While the two solutions might be better (citations ?), the ease of using a weighted average technique is obvious. [ w_1+w_2 = 1, w_2= (T-T_lo)/(T_hi-T_lo), where T is the desired Temperature and T_lo & T_hi are the end-points of the range of overlap. P = w_1*P' + w_2*P" where P' & P" are the predictions from the two sets of Antoine parameters. This weighing runs from 1:0 to 0:1 continuously, and the first derivatives will approach the "nearest" curve's at the end-points. This can obviously be extended to non-overlapping equations by assuming a range of extrapolated overlap.] I am NOT creating/inventing new art; this is a COMMON technique, but I have no references for it, my bad. I suggest replacing "two solutions are possible..." with "Several solutions are possible..." and mentioning weighted interpolation as one of them.216.96.76.228 (talk) 15:03, 8 September 2013 (UTC)

Diagram
The diagram "Typical deviations of a parameter fit over the entire range (experimental data for Benzene)" seems to have some statistically fluctuating experimental data included. In order to understand the three prediction functions (Antoine, August, and ...), shouldn't we be comparing them with the best consensus values of the boiling point curve for benzene, rather than with uncured experimental data? 84.227.241.146 (talk) 10:53, 13 September 2014 (UTC)

Combination with Raoult law
Is this equation applicable only to pure compounds or it can be used for solutions?--5.2.200.163 (talk) 17:03, 24 March 2016 (UTC)

This source says that the equation can also be applied for solutions.--5.2.200.163 (talk) 17:09, 24 March 2016 (UTC)

derivation from Clausius–Clapeyron relation
The Clausius–Clapeyron relation is:
 * $$\frac{dP}{dT} = \frac{\Delta s}{\Delta v},$$

where $$\Delta s$$ is the specific entropy and $$\Delta v$$ the specific volume change of a particular phase transition.

In the article, it is stated that the Antoine equation is derived from the Clausius–Clapeyron relation. However to my knowledge this is not possible, as the Antoine equation is semi empirical. The best one can do is derive the August equation, which holds when the specific volume in a substances initial phase is very small compared to its gas phase.

If a derivation of the Antoine equation from the Clausius–Clapeyron relation exists, then it would be suitable to provide the derivation on this page. Should this be too long, a reference that provides it would be sufficient.

If it is not the case that one can be derived for the other, the claim that the Antoine equation can be obtained using the Clausius–Clapeyron relation should be removed. — Preceding unsigned comment added by Whole Oats (talk • contribs) 02:52, 2 January 2019 (UTC)