The results show that silk sericin adsorbed only gold over the entire concentration range of metal ions in the solution (i.e., 0.3–3 mM).
The adsorption capacity for gold from the binary solution is slightly lower, but LeVan and Vermeulen equation can still provide an accurate prediction of the metal adsorption
on silk sericin.
The LeVan and Vermeulen equation oversimplifies adsorbate–site interactions and assumes adsorbate–adsorbate interactions to be absent.
This could be the reason for observed underestimation of gold adsorption on silk sericin in Fig. 6a.
It is also possible that the lower adsorption capacity was a result of more complex interaction between copper ions and sericin resulting in chemical and structural changes of the protein. Fig. 7 displays the XPS data for C1s, N1s, Au4f and Cu2p3/2 of the fresh and spent adsorbents. It can be seen in Fig. 7a that sericin displays a broad C1s peak deconvolutes to carboxyl (287.8 eV), carbonyl (286.6 eV), alkyl (285 eV) and carbon (282.6 eV) that are typically
found in proteins.
The carboxyl and carbonyl peaks decreased after metal adsorptions suggesting possible interactions with these functional groups.
Metal adsorption affects not only the local chemistry but also the global hydrophilic and hydrophobic character of the protein.
This could impact protein folding and therefore the accessibility of the adsorption sites. The change in N1s (Fig. 7b) is more subtle with slight broadening of the peak after gold adsorption.
The Au4f5/2 and Au4f7/2 peaks shown in Fig. 7c belong to Au+ and Au3+ and are typical of gold salt, while the absence of the satellite peak in Cu2p3/2 (Fig. 7d) indicates that copper is adsorbed as Cu+ instead of Cu2+.