The HORN fellowship facilitated the research exchange between Curtin University, Western Australia, and the Hyogo Prefectural University (Shosha Campus) in Himeji, Japan. The one-month tenure was focussed on development and purification of surfactants for use in solvent extraction, alongside lecturing at the university and attendance at an international conference.

Solvent extraction is a common industrial method used to separate metal cations in solution. A large number of surfactants are used industrially, depending on the metal being targeting and the composition of the ore. For example, phosphoric acids such as Cyanex 272 are commonly used for the separation of nickel and cobalt, and phenol oximes such as Acorga and LIX have strong affinities for copper. The main aim of the project was to synthesise 6-dodecyl-2-hydroxybenzaldehyde oxime (figure 1), which is a phenol oxime-type surfactant of the same tail length as the commercially available LIX 680 (5-dodecyl-2-hydroxybenzaldehyde oxime) but with the alkyl group shifted one position along the benzene ring, to investigate the possible effects of the position of the tail group on the effectiveness of the surfactant for metal extraction. Most commercial extractants take the para form, with the hydroxide and alkyl groups taking positions 2 and 5 respectively. For example, ACORGA 5640 (5-nonyl-2-hydroxybenzaldehyde oxime).

The synthesised surfactant, 6-dodecyl-2-hydroxybenzaldehyde oxime, displayed unusual behaviour when complexing with copper. Firstly, the extraction properties showed essentially no pH dependence through the range 0.98–5.26, with the maximum extraction being only 32% of the copper. In contrast, up to 50% of the copper could be loaded from the mixed Cu(II)/Fe(II) system, with the extraction greater at higher pH. Once extracted, however, stripping was unsuccessful at pH as low as 0.47. As the commercial surfactant LIX 860 is considered a particularly strong extractant and can be used below pH 1.0, the stripping difficulties are not altogether unexpected.
Interestingly, after complexing with copper, the colloid formed a solid precipitate in the organic layer. This has two implications. Firstly, it shows that the complex formation occurs either at the interface or in the organic layer itself, as the precipitate formed in the organic layer. Secondly, it shows a marked difference to the behaviour of the commercial extractant, where the alkyl tail is in a different position. The partially purified sample of ACORGA 5640 (containing 5-nonyl-2-hydroxybenzaldehyde oxime) remained in solution in the organic phase after complexing, allowing the copper to be stripped back into the aqueous phase by lowering the pH. This suggests that the position of the alkyl chain can significantly affect the arrangement of the complex.
Measurement of the diffusion coefficient in heavy chloroform confirmed that the larger 6-dodecyl-2-hydroxybenzaldehyde oxime is slower to move than the shorter 5-nonyl-2-hydroxybenzaldehyde oxime.

Five lectures were given during the tenure: two science lectures, for undergraduate and masters students, respectively, and three for undergraduate "English for science students". The lectures covered the differences between living in Australia and in Japan, with a particular focus on living as a fly-in-fly-out worker on a remote industrial site. From a technical perspective, the lectures also discussed the application of common laboratory techniques to an industrial setting. The more advanced lecture covered the specifics of my recent research and an in-depth discussion on industrial solvent extraction.