The pH of the internal aqueous solution containing a basic salt was theoretically calculated on the basis of the amount of penicillin G transported into the internal phase. The calculated results agreed with the experimental data well and were used to select a suitable type and concentration of a basic salt in the internal phase to give a pH within the range pH 5 and 8 where penicillin G was stable after the termination of extraction. Also, it was shown that $Na_2CO_3$ solution was the most suitable internal aqueous solution for the extraction of penicillin G by an ELM process.
ECA 4360J functioned as a carrier as well as surfactant for the extraction of penicillin G by the ELM process. Thus, the reaction equilibrium between penicillin G and ECA 4360J was investigated through two phase extraction experiments. The experimental results showed that the following reaction occurred at both the external and internal interfaces.
$B^E(org.) + 0.5H^+(aq.) + 0.5P^-(aq.) ≒ B^E(Hp)_{0.5}(org.)$
In the same way, the reaction between penicillin G and Amberlite LA-2 was obtained as follows.
$2B^L(org.) + 2H^+(aq.) + 2P^-(aq.) ≒ B^L_2(HP)_2(org.)$
A permeation model which was an extension of the model of amino acid extraction given by Teramoto et al. was presented in order to simulate the extraction of penicillin G by the ELM process using ECA 4360J. The experimental data were satisfactorily simulated by the proposed permeation model, in which the external mass transfer around the emulsion drop, the reaction in the external interface and the diffusion as well as the reaction equilibrium in the W/O emulsion drop were considered. Also, The ELM process using only ECA 4360J without another carrier could increase operation cost and bring about the losses of penicillin G in the internal phase because of slow extraction rate. Finally, another carrier, for example, Amberlite LA-2, must be used to facilitate the reaction at the external interface.
The extraction of penicillin G from si...