Click here to edit subtitle

What Are Molecular Gels

An organogel is a gel with the liquid phase being oil as opposed to a hydrogel, which is a gel that has a liquid phase of water.  Low molecular weight organogelators (LMOG) form gels which are small molecules that assemble into fibers, rods, liquid crystals, micelles or ribbons to immobilize oil.  Finally, a self-assembled fibrillar network (SAFiN) is a specific type of LMOG which forms crystalline fibers that entrain liquid oil.  Before we proceed to discuss SAFiNs and molecular gels, we need first to define them. More broadly, the definition of a “gel” has been evolving for more than 100 years since Thomas Graham in 1861 attempted a loose definition.  65 years later, Dr. Dorothy Jordan Lloyd, stated, that “the colloid condition, the “gel”, is one which is easier to recognize than to define” (Jordan Lloyd, 1926).  Her major contribution was in recognizing that all gels must be comprised of at least two components- a liquid and a gelling substance (i.e. a solid) and that the entire system must have the mechanical properties of a solid (Jordan Lloyd, 1926).  The major downfall to this definition is that not all colloids are gels and not all gels are colloids.  Over the next twenty years the definition progressed to the point where Hermans proposed that gels are: coherent colloid dispersed systems of at least two components; they exhibit mechanical properties that are consistent of the solid state; and both the dispersed (gelator) phase and dispersion medium must extend continuously throughout the whole system (Hermans, 1949).  Because of the exclusivity of this definition, Ferry offered a more descriptive definition of a gel; “a gel is a substantially diluted system which exhibits no steady state flow” (Ferry, 1961).  Another common definition of a gel is “if it looks like Jello, it’s a gel” (Jordan Lloyd, 1926). From this point, a gel must contain two features: 1) has a continuous microscopic structure with macroscopic dimensions that is permanent on the time scale of an analytical experiment and 2) is solid-like in its rheological behaviour, despite being comprised mostly of water (Weiss and Terech, 2006).

Molecular gels occur when the low  molecular weight organogelator (LMOG) undergo supramolecular aggregation and the corresponding SAFiN forms in a solution (sol) at low concentrations of gelator molecules (£2wt%) (Weiss and Terech, 2006).  In brief, as the sol is cooled, the solution becomes super-saturated causing a chemical potential, which is the driving force for phase separation and nucleation.  Gelator molecules begin to self-assemble in stochastic nucleation events which have highly specific interactions, promoting one-dimensional growth.  Crystallographic mismatches vary the degree of branching in the network structure.  Fibers form due to the one-dimensional growth and they are able to interact with each other forming a three-dimensional network (Figure 2).  Figure 2 shows the resulting micrographs from a system which is has a high degree of crystallographic mismatches (left) and a low degree of mismatches (right).

SAFiN Gelation

The driving forces for surfactant aggregation in organic solvents are different than the “hydrophobic effects” that govern aggregation in aqueous systems (Terech, Rodriguez, Barnes and McKenna, 1994).  As an organogelator is cooled in solution, a super-saturated solution forms eventually causing the gelator molecules to microscopically phase separate and to self-assemble via stochastic nucleation events. This is different than conventional crystallization, which involves macroscopic phase separation (Terech and Weiss, 1997).  Organogelator molecules self-assemble in stochastic nucleation events with highly specific interactions promoting one-dimensional growth (Weiss and Terech, 1997).

In terms of a molecular gel, the sol is defined as a dispersion of solid particles in a liquid colloidal solution.  The nucleation process requires extremely specific interactions to induce preferential one-dimensional growth.  This one-dimensional expansion results in fiber formation which has been described as ‘crystal-like’ (Terech, Furman and Weiss, 1995).  SAFINs serve the same function as (polymer) chains in a polymer gel (Terech, Furman and Weiss, 1995).  The junction zones and branching between these polymer-like SAFiN strands are responsible for the rigidity of the networks (Terech, Furman and Weiss, 1995).

There are three levels of structure described for SAFiNs, ranging from the microscopic scale to the macroscopic scale (Wang, Lui, Narayanan, Xiong and Li, 2006).  The aggregation of the gelator molecules builds the primary structures via non-covalent bonds. The ability for these molecules to self-assemble into rod-like structures is not well understood.  It requires a balance among opposing parameters such as solubility and those parameters that control epitaxial growth into axially symmetric elongated aggregates (Weiss and Terech, 2006; Suzuki, Nakajima, Yumoto, Kimura, Shirai and Hanabusa, 20036).  For instance, in sodium stearate gels, decreasing solvent polarity has been shown to induce morphological transformations from fibers to plates (Liang, Ma, Zheng, Davis, Chang, Binder, Abbas and Hsu, 2001).

SAFiNs create a three-dimensional network structure (secondary structure) by self-organizing  the rods, tubes or sheets into three dimensional networks through non-covalent interactions, including hydrogen bonding, van der Waals interactions, p-p stacking, and metal coordination bonds (Suzuki, Nakajima, Yumoto, Kimura, Shirai and Hanabusa, 2003).  The tertiary network structure relates to how the strands then interact to form the supramolecular structure. This is reportedly the most complicated level of structure to modify, but is the structure which affects the macroscopic properties of the material (Wang, Lui, Narayanan, Xiong and Li, 2006) such as hardness, stability and oil mobility.

There are two types of junction zones identified in SAFINs, i.e., transient and permanent (Wang, Lui, Narayanan, Xiong and Li, 2006).  Transient bonds are able break and reform and occur in 12HSA/canola oil gels and are based on two predominant forces, including hydrogen bonding and van der Waals forces.  These interactions are strongly influenced by the nature of the liquid portion of the gel, which can either promote or interfere with interactions between SAFiNs.  The permanent junction zones arise due to crystallographic mismatches at the interface of growing fibers, which result in a branched fiber (Wang, Lui, Narayanan, Xiong and Li, 2006). The supersaturation-driven crystallographic mismatch branching (also called non-crystallographic branching) is governed by the nucleation and growth of a gel network (Wang, Lui, Narayanan, Xiong and Li, 2006).


Our latest work, published in the Journal of Materials Chemistry focuses on applying solubility parameters to molecular gels. Hildebrand solubility parameters are governed by the free energy of mixing of the gelator and the solvent.   It is assumed, in polymer physics, that the dissolution of the polymer is accompanied by a minor increase in the entropy, and the enthalpy is the deciding factor in the Gibbs free energy change. Therefore, the Hildebrand solubility parameter proposed in two seminal papers by Hildebrand 1959 and Scott and Scatchard, 1949, relies solely on the enthalpy.   The HSP decomposes the cohesive energy density according to the dispersive interactions (dd), polar interactions (dp), and hydrogen-bonding interactions (dh).

The objective of this research is to scrutinize these parameters for a much wider class of solvents and to observe if significant trends exist that may correlate the individual HSPs to the critical gelator concentration. The ability to predict gelation behavior of organogels has been elusive due to the meticulous balance of contrasting parameters including solubility and the intermolecular forces controlling epitaxial growth.  The solubility parameters, of the selected organic solvents, cover a vast breadth of static relative permittivities (1.82 to 37.5) as well as a large portion of Hansen space (Figure 1).  Due to the apolar nature of the solvents, the dispersive component of Hansen space is restricted, even though almost all of solvents fall within a narrow range of dispersive components (14 < dd < 20).  Typically, if the solvent has a dispersive component below 14 MPa1/2 it is in the gaseous state at atmospheric pressure and above 20 MPa1/2 the solvent is a solid at 20 to 30 oC. The dispersive component of HSP is linearly correlated to the carbon length, while the polar and hydrogen-bonding HSP are inversely correlated. 

For the 56 tested solvents, 32 are capable of forming organogels at concentrations below 3 wt%. An increase in the solvent aliphatic chain length increases the likelihood that a molecular gel will develop. However, beyond this there is no clear correlation between solvent type and the ability to gel, with the exception that neither alcohols nor carboxylic acids form molecular gels. Although the predictive nature of the HSP is much stronger than originally anticipated for the wide range of solvents. The parameter becomes even more powerful when observing the CGC as a function of the individual HSP within the individual classes of solvents.  As the carbon length of the solvent increases, the static relative permittivity, the HSP and the dispersive component of the HSP increase while the distance in Hansen space decreases.  On the other hand, for polar and hydrogen-bonding solvents, as the chain length increases the static relativity permittivity, the hydrogen-bonding HSP and polar HSP decrease while the dispersive HSP increases. As the chain length of the aliphatic solvents increase (i.e., increased static relative permittivity) the CGC decreases.


Hydroxy fatty acids were synthesised in collaboration with Dr. Richard Weiss, Georgetown University where the hydroxy position on octadecanoic acid was varied between the 2, 3, 6, 8, 10, 12 and 14 positions. Early work on 12HSA organogels indicated the presence of a SAFiN whose nature is dependent on the ability of the carboxylic acid head groups to dimerize and the secondary hydroxyl groups on the fatty acid backbone to form hydrogen-bonding arrays. All of the HSA/mineral oil dispersions, whether viscous liquids or gels, were opaque. The opacity or transparency of an organogel is directly related to the cross-sectional thickness of the crystalline aggregates, the number of junction zones capable of diffracting light, and the number of crystalline aggregates within the self-assembled network. The smaller the constituents of a SAFiN network, the more transparent the gel.

Inter-/intra-molecular hydrogen bonding of a series of hydroxystearic acids (HSAs) are investigated. Self-assembly of molecular gels obtained from these fatty acids with isomeric hydroxyl groups is influenced by the position of the secondary hydroxyl group. 2-Hydroxystearic acid (2HSA) does not form a molecular dimer, as indicated by FT-IR, and growth along the secondary axis is inhibited because the secondary hydroxyl group is unable to form intermolecular H-bonds.  As well, the XRD long spacing is shorter than the dimer length of hydroxystearic acid.  3-Hydroxystearic acid (3HSA) forms an acyclic dimer and the hydroxyl groups are unable to hydrogen bond, preventing the crystal structure from growing along the secondary axis.  Finally, isomers 6HSA, 8HSA, 10HSA, 12HSA and 14HSA have similar XRD and FT-IR patterns, suggesting that these molecules all self-assemble in a similar fashion.  The monomers form a carboxylic cyclic dimer and the secondary hydroxyl group promotes growth along the secondary axis.