I'm interested in the denaturation of proteins by alcohols, specifically by ethanol.

I have devised a simple experiment in which I inject solutions of ethanol (of variable concentrations) into solutions of egg albumen (containing proteins such as conalbumin). I have observed a marked pH change in solution after injection of ethanol. The rate of change of pH is in fact proportional to the concentration of ethanol injected (as expected).

However, I am having trouble explaining the theory behind why ethanol's disruption of tertiary protein structure increases pH. As I understand, ethanol disrupts proteins such as conalbumin by disrupting hydrogen bonds in the tertiary structure of the molecule.

But why does this cause the pH to increase (in effect, the concentration of H+ ions to decrease)?

  • 1
    $\begingroup$ Just to be sure, is the pH change compared to pH change of a blank sample without albumine ? Is pH change caused by ethanol addition eliminated ? $\endgroup$
    – Poutnik
    Commented Mar 3, 2021 at 16:09
  • $\begingroup$ @Poutnik I’m a chemistry student, so forgive me if I make mistakes with nomenclature and such… but I basically have a pretty-made solution of albumen protein into which I inject ethanol. pH is measured first of the albumen without ethanol and then the change is measured when ethanol is injected. $\endgroup$
    – Joseph
    Commented Mar 3, 2021 at 16:26
  • 2
    $\begingroup$ Be aware the ethanol presence itself affects the measured pH value even without albumine. This effect has to be eliminated. I would try simulation with some aminoacid with comparable mass concentration as albumine, with the same initial pH. $\endgroup$
    – Poutnik
    Commented Mar 3, 2021 at 16:41
  • 1
    $\begingroup$ Yes, you misunderstood it - change in pH is a result of that you have a different solvent - any pH in water-ethanol mixture is different then in pure water. I'm also disappointed that answerers didn't consider it. $\endgroup$
    – Mithoron
    Commented Mar 3, 2021 at 17:37
  • $\begingroup$ @Mithoron Good point:pubs.acs.org/doi/10.1021/j100803a022 Well, that's why the site exists I suppose, to aggregate and pass judgement over answers :). $\endgroup$
    – Buck Thorn
    Commented Mar 3, 2021 at 20:12

2 Answers 2


The main component of egg white, ovalbumin, contains 93 negatively charged residues (Asp + Glu) and 68 positively charged residues (Arg + Lys), as well as 14 His. If you make the solvent less polar by adding ethanol to water, they will tend to form the neutral species (Arg+ and Lys+ will lose a proton, and Glu- and Asp- will pick up a proton).

If you had the protein unfolded already, you would expect that the solution gets more basic if the effect on all charged amino acids is the same because there are more negatively charged ones.

To test whether the observed effect has to do with denaturation of the proteins or with changes in protonation state of side chains already exposed to water, you could measure the pH of amino acids (preferably with a modification of the main chain carboxylic acid and amino functions so only the side chains show acid/base reactions) at different ethanol concentrations.


Most proteins are polyelectrolytes because the constituent aminoacids may have ionizable groups, typically carboxylic and basic (imino, guanidino) groups on the sidechains, in addition to the terminating amino and carboxylic groups (when not modified). The degree of ionization depends on the fold of the protein and the pH of the solution. Denaturation of a protein can expose to the solvent buried sidechains with ionizable groups, resulting in acid/base reactions with water.

If the solution is buffered and the protein is not present at high concentrations, then the degree of ionization of the protein will depend on the solution pH, but the solution pH will not be particularly sensitive to the presence of the protein. On the other hand, if there is a lot of protein in the solution and/or the solution is not buffered so as to resist changes in the degree of ionization of the protein, then such changes will translate in changes in pH. Concretely, changes in secondary or higher structure of the protein will translate into changes in pH, as it appears you observe.

Unfortunately (as noted in the comments), the measured pH of the sample may vary strongly with the concentration of ethanol even in the absence of protein and in the presence of buffer. This is in addition to the effect which the organic solvent will have on the dissociation constant of ionizable groups on the protein noted by Karsten Theis. It is quite likely that those effects will exceed in importance that of denaturation, since charged groups are most often exposed on the protein surface. The extent of exposure and the dissociation constant may increase with denaturation, of course, and disentangling the different effects is not simple, requiring accurate measurements.

To start with you might check the extent to which the addition of alcohol may be responsible for the increase in pH. A lot of work has already been done in measuring such effects, and the following table compiles some results reported in the IUPAC Compendium on Analytical Chemistry (Ref. 1) for the 0.05 m potassium hydrogen phthalate (KHPh) buffer.

Other posts on this site address measurement of pH in organic media.

$\begin{array}{c|c} \hline \text{mole fraction ethanol} \\ \hline \begin{array}{c|c|c|c|c} T (°C) &0.0416 & 0.0891 & 0.2068 & 0.4771 \\ \hline -5.0000 & 4.2660 & 4.5700 & 5.1120 & 5.5270 \\ 0.0000 & 4.2490 & 4.5440 & 5.0760 & 5.5000\\ 10.0000 & 4.2350 & 4.5130 & 5.0260 & 5.4690\\ 25.0000 & 4.2360 & 4.5080 & 4.9760 & 5.4720\\ 40.0000 & 4.2600 & 4.5340 & 4.9780 & 5.4930 \\\hline \end{array}\end{array}$


  1. Web edition of the Compendium on Analytical Nomenclature (The Orange Book) (2002). https://media.iupac.org/publications/analytical_compendium/
  • 1
    $\begingroup$ Thank you so much for your input. I hugely appreciate the recent edit, in which you added methodology for mitigating ethanol's own effect. There are little (if any?) straightforward resources on the net that answer this question, so this page will surely come in handy for many people. I am always surprised at the sheer quality of answers on the Stack Exchange network :) $\endgroup$
    – Joseph
    Commented Mar 5, 2021 at 9:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.