Understanding colloidal instability in white wine model solutions: A study focused on the effect of polysaccharides and salts onto bentonite efficiency

Abstract

A white wine model solution (12% v/v ethanol, 4 g/L tartaric acid, pH 3.2) was used to assess wine colloidal instability as well as the influence of several wine components on bentonite performance in protein removal. BSA was used at a maximum concentration of 500 mg/L [1] as a protein standard, and sodium-activated bentonite was chosen due to its high efficiency demonstrated in previous studies [2]. For all tests, samples were let to react for 24 hours, protein content was measured using the Bradford assay [2], while polysaccharide content was measured using the phenol-sulfuric method [3]. Adsorption curves with increasing levels of bentonite showed that 100 g/hL of bentonite was required for the total removal of protein. The influence of polysaccharides was also tested. To achieve this, naturally occurring polysaccharides were added to the model solution at concentrations equivalent to the sum of their monosaccharides [2]: arabinogalactans (gum arabic), mannoproteins, and pectin. With the addition of polysaccharides, bentonite efficiency decreased at lower concentrations (20, 40, and 60 g/hL) due to polysaccharide/bentonite interactions that occupied bentonite adsorption sites. Additionally, arabinogalactans and pectin showed protein interactions by themselves, as evidenced by the reduction in protein content to 350 and 287 mg/L, respectively, compared to 436 mg/L in the control. This, combing with the fact that polysaccharide content dropped near 0, showed that this interaction was likely due to surface charge interactions. To further characterize the adsorption process, adsorption isotherm models were performed by maintaining a fixed bentonite concentration (20 g/hL) while varying the concentration of each polysaccharide and protein from 500 to 1 mg/L. Except for pectin, Freundlich isotherm models were applicable in all cases, indicating that adsorption occurred on a heterogeneous surface with interactions between adsorbed molecules [4]. Furthermore, arabinogalactan, mannoprotein, and BSA had 1/n values of 0.9460, 1.0997, and 0.7943, respectively, suggesting that BSA had the most favourable adsorption (1/n < 1) onto bentonite, whereas both polysaccharides exhibited (near) irreversible adsorption (n ≈ 1). Pectin showed no adsorption capacity onto bentonite, which was expected due to its neutral nature. On the other hand, the addition of both KCl and K₂SO₄ (at the maximum authorized concentrations by IVV of Cl^– and SO₄^2–, respectively) reduced the efficiency of bentonite. Based on particle size measurements obtained via DLS [5], the addition of KCl and, at higher levels, K₂SO₄ induces protein aggregation, which makes protein/bentonite absorption more difficult. These results reveal a complex colloidal system, shedding light on the interactions among major wine components and explaining the variability in bentonite efficiency observed in the industry.

Acknowledgement

The authors gratefully acknowledge the financial support from FCT—Portugal related to the PhD grant with reference 2023.02067.BD. and to CQ-VR (CQVR-UIDB/00616/2025)

References

[1] Cosme, F.; Fernandes, C.; Ribeiro, T.; Filipe-Ribeiro, L.; Nunes, F.M. (2020). Beverages, 6, 19.

[2] Arenas, I., Ribeiro, M., Filipe-Ribeiro, L., Vilamarim, R., Costa, E., Siopa, J., Cosme, F., Nunes, F.M. (2021). Foods, 10, 608.

[3] Nielsen, S. S. (2010). Food analysis laboratory manual, 47-53.

[4] Filipe-Ribeiro, L., Cosme, F., Nunes, F.M. (2020) Food Research International, 129, 108855.

[5] Burken, O.; Sommer, S. Journal of Food Science, 89, 6413–6424.