Lipase-catalyzed esterifications in biphasic media (heptane–water, 1:1) were conducted by using Thermomyces lanuginosus lipase (TLL) as biocatalyst. Different carboxylic acids (from acetic to lauric) were thus esterified with 1-butanol at different pH values (2–10). For all carboxylic acids tested, pH optima for the enzymatic esterifications were ca. 3–5, in clear agreement with previous literature, and quite different from optima pH of TLL in hydrolytic reactions (8–11). Interestingly, the interval of pH in which TLL was active in esterifications, varied markedly depending on the carboxylic acid. Thus, for long-chain acids (i.e. lauric), such pH interval was much wider than for short-chain carboxylic acids (i.e. acetic). To understand this effect, simple pKa values of carboxylic acids, retrieved from open literature, are not sufficient, since those values are measured in pure water. When a second phase is added, aspects related to partition coefficient of each carboxylic acid need to be considered as well, thus leading to the apparent pKa concept. Herein we performed theoretical calculations to obtain such app pKa of each carboxylic acid. When such calculations were compared with the pH interval, a clear correlation was observed. Overall, results confirm that lipases accept only the protonated form of the carboxylic acid (R–COOH) during esterifications in biphasic media.