Poly-liquid Behaviors of Self-associating Liquids and the Anomalous Mesoscopic Aggregation in Liquid Solutions

According to standard notions of phase equilibrium, distinct phases that can spatially coexist for extended times should ordinarily occupy macroscopic regions, consistent with everyday experience. Contradicting this expectation, submicron-sized inclusions of a solute-rich liquid are sometimes found in equilibrated solutions of proteins and other molecules well outside the region of macroscopic liquid-liquid separation commonly observed when crystallization is kinetically inhibited. According to the classical density-functional study of Chan and Lubchenko, a steady-state ensemble of finite-sized droplets of a metastable solute-rich can emerge in a solution if the solute molecules can form long-lived complexes. Motivated by these notions, we solve for the thermodynamics of an explicit liquid model in which particles can also form transient dimers. This model exhibits a rich phase behavior stemming from the interplay of binding and condensation. In particular, we determine ranges of parameters where two distinct dense liquid phases, one dimer-rich and monomer-rich respectively, can exist. These two dense liquid phases can also co-exist, both with each other and, individually, with the solute-poor phase. We find that within a carefully chosen range of the dimer’s binding strength, the two dense phases become well separated in terms of stability. At the same time, the conditions for the existence of the clusters, as postulated by Chan and Lubchenko, become satisfied within a broad range of the solution’s density. The width and location of this range on the phase diagram are consistent with observation. We further predict that the clusters are comprised of a metastable monomer-rich dense liquid, while the bulk solution itself can be thought of as hosting pre-critical density fluctuations and is surprisingly dimer-rich. In addition, we predict that, surprisingly, the dense phase commonly observed during the macroscopic liquid-liquid separation is actually complex-rich. These findings provide strong evidence for the complexation scenario and suggest new experimental ways to test it. We have begun testing these predictions by performing direct molecular modeling of phase equilibrium of anisotropically interacting particles that can form transient complexes.Chemistry, Department o