Inorganic salts in their dilute aqueous solutions usually demonstrate simple phase transitions – between a homogeneous solution and a macrophase separation (aka. soluble and insoluble). When dealing with soluble ions with larger sizes, e.g., on nanometer scale (macroions), due to their large sizes and multiple charges, macroions are known to moderately attract counterions around the macroions (counterion association) at moderate level. This delicate electrostatic interaction leads to many more fascinating solution behaviors not seen in simple ionic solutions or colloidal suspensions.
Polyoxometalates (POMs) are a type of metal-oxo clusters with inherent charges and large surfaces; many of them stay as macroions with fully hydrophilic surfaces, uniform shapes, and intact molecular structures in dilute solution, which make them ideal models for exploring of the solution properties of macroions. The POM clusters, like many macroions, can loosely attract their counterions, selectively favoring those with higher valance or those monovalent ions with smaller hydrated sizes accurately. This counterion association is known to generate counterion-mediated attraction among the POMs and often form 2D nanolayers. However, this general trend would not be applicable with poorly charged clusters. With carrying only a few charges, they cannot form 2-D sheets via counterion-mediated attraction, especially with multivalent countercations. An important question is whether and how such clusters attract with each other, and if yes, what are the consequent self-assembly and phase behaviors in their aqueous solutions?
Herein, we report the discovery of pure inorganic polyoxometalates hydrogels without carbon element. More importantly, a complex phase transition behavior has been observed and analysed: clear solution – macrophase separation – gelation – solution – macrophase, in an aqueous solution containing polyanionic clusters [Mo7O24]6− (Mo7) with the addition of simple electrolytes (Fe(NO3)3). The distinct phenomena can be attributed to the continuous addition of Fe3+ causing charge-inversion of Mo7 macroions, and then followed by a 1-D fiber-like supramolecular structure formation, as confirmed by various techniques and molecular dynamics simulations.
This upper solution phase covers a broad range of Fe3+ concentration in the phase diagram (Fig. 1) until after ~3.5 mol/L Fe3+ ions were added, phase separation was observed again. Two issues need much attention in the phase diagram – the gelation and the two separated solution phases at different Fe3+ concentrations. The question is what driving forces are responsible for triggering the self-assembly and consequently the hydrogel formation.
Fig. 1 (a) A photograph of Mo7-Fe systems at different molar ratios between Mo7 and Fe3+. (b) Phase diagram of Mo7-Fe system in aqueous solution.
To address this, the electronic and coordination structures of the metal components in Mo7-Fe complexes were explored by using X-ray absorption spectroscopy (XAS). The Mo-O-Fe bond was not found in the Mo7-Fe hydrogel, illustrating that the hydrogel formation isn’t based on covalent bonding between Mo7 and Fe3+ (Fig. 2b and 2d). Meanwhile, the hydrogel is also observed when titrating Y(NO3)3 into Mo7 aqueous solution, while Co2+ and Ni2+ cannot trigger gelation and the aqueous solution remains in its liquor state. As indicated by Coulomb's law, the electrostatic force is proportional to the charge of two ions. We assumed that due to the less charge (compared to Fe3+), the divalent Co2+ and Ni2+ ions cannot form gel with Mo7 in an aqueous solution. Static light scattering (SLS) and dynamic light scattering (DLS) measurements indicated the formation of larger supramolecular structures in Mo7 solution with the presence of Fe3+. Measurement of zeta potential of the Mo7-Fe3+ aggregates indicated the gelation region occurs around the region of neutral charge, suggesting the critical role of electrostatic interaction on the phase transitions. Meanwhile, in the upper solution region, the Mo7 clusters are clearly positively charged in nature; indicating that the enhanced solubility of Mo7 due to abundant positive charges is responsible for the upper solution phase.
Fig. 2 Electrostatic interaction induces Mo7-Fe assemblies. (a) Mo K-edge XANES spectra and (b) Fourier transforms of the Mo K-edge EXAFS spectra of Mo7-Fe hydrogel with the references. (c) Fe K-edge XANES spectra and (d) Fourier transforms of the Fe K-edge EXAFS spectra of Mo7-Fe hydrogel with the references. These experimental XAS spectra are shown to prove the electrostatic interaction between Mo7 clusters and Fe3+ ions. (e) Time-resolved scattering intensity plots, representing the increment of the scattered intensity throughout the self-assembly process of Mo7-Fe hydrogel. (f) Time-dependent size distribution by CONTIN analysis of the DLS data. (g) Zeta potentials upon reaction of 6.07×10-5 mol/L Mo7 clusters by the adding of solution, demonstrating the charge inversion process of Mo7 anion clusters.
The cryogenic scanning electron microscopy (cryo-SEM) (Fig. 3a) and cryo-TEM images (Fig. 3b) revealed extensive entanglements of uniform long-chain structures of Mo7-Fe aggregates in the hydrogels, which should be responsible for the gelation. The branched structures could be conserved even in the dried Mo7-Fe gel shown by the SEM. In small-angle X-ray scattering (SAXS) studies, the form factor of Mo7 cluster remains intact in the data of Mo7-Fe hydrogel without crystalline diffraction peaks, implying that Mo7 clusters are uniformly dispersed in the chain structures. Meanwhile, in situ optical microscopy studies also conformed these results (Fig. 3d).
Fig. 3 The morphology change of assembles in Mo7-Fe system. (a) Cryo-SEM images and (b) cryo-TEM images of Mo7-Fe hydrogel with a Mo7/Fe3+ molar ratio of 1:10, indicating the formation of branched structures in the Mo7-Fe gel networks. (c) SAXS spectra monitoring the gelation process with different molar ratios of Mo7/Fe3+. (d) In situ optical microscopy of the aggregates formed by Mo7 and Fe3+ in aqueous solution.
Overall, the mechanism of gelation of Mo7-Fe3+ in aqueous solution (Fig. 4) can be proposed as follows: (i) the anionic Mo7 clusters could instantly interact with Fe3+ cations to form Mo7-Fe aggregates due to the strong electrostatic attraction; (ii) with the continuous addition of Fe3+ cations and NO− 3 anions, the branched Mo7-Fe aggregates overcharged, turned to positively charged, and then rearranged into 1-D fiber-like supramolecular structure, and then formed homogenous Mo7-Fe hydrogels; and (iii) when adding excessive Fe3+ and NO− 3, the Mo7-Fe cationic complexes becomes more soluble, making the Mo7-Fe hydrogels redissolved and be back to solution phase.
This discovery advanced new methodology of preparation of inorganic powders to continuous bulk materials so as to solve the long-lasting problem of inorganic materials being difficult to process and to be patterned. The distinct phenomena, successive unusual transitions from macrophase separation to homogenous hydrogel and then the solution phase again in the Mo7-Fe system, was investigated by various techniques (e.g., XAS, DLS, Zeta potentials, ITC, SAXS, cryo-EM, DWS) and molecular dynamics simulations. Our results not only emphasize the importance of phase engineering in materials science, but also afford a simple and general approach to prepare inorganic bulk materials.