My research is primarily concerned with the physical behavior of the condensed phase in general and liquid crystals in particular. In simpler times the condensed phase was understood as a dichotomy whereby there could be liquids which flow easily and crystals that do not. The liquid, with no regularity in any direction was the anti-thesis of the crystal which has regularity in all spatial directions. In 1888 Reinitzer found that a turbid liquid was formed when solid cholesteryl benzoate was melted and became a clear isotropic liquid upon further heating. When the turbid liquid was characterized by Lehmann it was found to be birefringent (angle dependent light refraction) and therefore anisotropic. These phases which are anisotropic and still exhibit some degree of fluidity are described as ‘liquid-crystalline’. So it seems as though the condensed phase was actually a trinity of isotropic liquids, anisotropic liquid crystals and solids but this, as it turns out, is not quite so. Specifically, certain organic materials do not show a single anisotropic phase between liquid and solid, but instead a cascade of transitions involving new phases. These phases are a rich source of fascinating physical and chemical behavior and play important roles in our lives since they form our cell membranes and the Liquid Crystal Display (LCD) screens of our smart phones, computers and televisions. As can be seen on the left side of figure 1, liquid crystals molecules can stack in layers where the director of each layer (the direction which molecules point toward on average) is shift from that of the layer above and below so as to trace out a helix. Since the components of light also travel as a helix, passage across the pixel will take place when the liquid crystal is in its helical superstructure making that part of the screen appear bright. However if this molecular architecture is destroyed with an electric field the light will be blocked and the pixel will appear dark as can be seen on the left side of figure 1.
The simplest liquid crystal phase is the nematic with one direction of orientational order which upon cooling can form layers or columns depending on the shape of the mesogen as can be seen in figure 2. The best way to study these systems, in my view, is by placing the various liquid crystals phases of interest into the very large magnetic fields of superconducting magnets (figure 3) and pulsing them with radio frequency light commensurate with the information desired. The information obtainable is highly accurate and vast in scope ranging from molecular structure, energies, orientational order, positional order, conformation populations and motions. Experiments of this type are known as Nuclear Magnetic Resonance (NMR) spectroscopy which describes the fact that atomic nuclei possess an intrinsic quantum mechanical property called spin angular momentum which resonates at a specific frequency according to the identity of the nuclei and the strength of the field.
Since proton NMR spectra of liquid crystals involve so many splittings of interacting nuclei and averaging over many molecular conformations much of my past work has been carried out using simpler well characterized solute molecules to probe the intermolecular liquid crystal environment [1,4,5,8,11,14*]. After asking what probe molecules can tell us about the bulk environment however one must ask the complementary question of how does the condensed phase effect individual molecular properties? To this end the effect of the condensed phase on molecular structure and internal rotational motion has been carried out after first computing the gas phase rotational potential with computational chemistry techniques that use parallel processing to approximate the Schrodinger equation. The result of these studies is the finding that the former is affected negligibly while the latter, perhaps surprisingly, is substantially affected [6,7,9,10,14*].
Another aspect of my research is the problem of pushing back the boundaries of the complexity of spectra that can be solved which as a consequence opens up the amount of information that can be obtained about the physical world. For simple molecules one can guess molecular parameters to calculate the matrix elements of the NMR Hamiltonian (the equation governing the behavior of nuclei in a magnetic field) followed by matrix diagonalization to calculate the nuclear spin energy levels and corresponding transition probabilities to generate a theoretical spectrum. The theoretical spectrum is then compared with the experimental and line assignments made followed by least squares variation of the spectral parameters. This last step is repeated until convergence is reached which will occur if the initial guess was close enough and if not a new guess must be attempted. This all works well enough (although very time consuming) for rigid molecules of higher symmetry and less than eight spins. But what if we have lower symmetry, internal degrees of freedom and more than eight spins? One way to proceed is with multiple quantum NMR or selective deuteration of the molecule of interest which can yield a better estimate of the single quantum spectrum for line assignment and least squares. But these techniques require much spectroscopy and/or synthesis experience and a lot of time and do not get around the problem of having to make line assignments. What if there are so many overlapping lines that they are no longer individually identifiable? This is the case for n-pentane dissolved in a liquid crystal whose NMR spectrum has about 20,000 transitions. To solve this spectrum it was necessary to use molecular dynamics simulations and genetic algorithms which use the principles of natural selection to find the solution at the global minimum of an error surface by searching a multi-dimensional spectral parameter space as can be seen in figure 4 [3,12,13].
While much of the above work is still ongoing, more recently and going forward I have begun studying more complex liquid crystal systems that could be used in LCDs, as light valves and optical modulators to name just a few applications. Currently under study via Deuterium solid state NMR is a discotic liquid crystal (figure 1) possessing unique charge transport (with the potential to be used as a nanometer scale wire in molecular scale computer processing and other electronics) and magnetic properties. The ordering of the molecules is to be obtained from quadrupolar spin echo and the various motions displayed will be deduced from observation of the spin-lattice relaxation times. Going forward I also plan to study how individual molecular motional degrees of freedom bestow unique properties onto the bulk chiral smectic C and biaxial nematic phases which also show promise in LCD technologies. Hopefully these few paragraphs have demonstrated the cornerstones of my research namely the posing of complementary questions in scientific inquiry and the interplay of theory and experiment.