Think of two atoms colliding during a chemical reaction. Of the resultant molecules formed, how fast are these molecules: do they possess lots of translational energy much like cars bouncing off of each other during a crash; or rotate as if in a free wheel-spin? Are the molecules formed with their bonds vibrating like a car’s oscillating suspension?
Chemists studying the physical motions of molecules during a chemical reaction, a field known as reaction dynamics, try and answer these questions to elucidate how reactions really occur on an atomic scale, i.e. do molecules—with energy distributed into vibrations, rotations or translations—bounce off of each other unchanged after a collision, or is energy deposited into a molecular motion such that a chemical reaction can occur?
The reaction dynamics community are not so interested in what the outcome of a chemical reaction is, but how the products are formed, and why.
If one were to drive a quantum car, one might be perturbed by the fact that it could only go certain speeds: 50 or 70 km per hour—but no speeds in between. Our car would be said to possess discrete, or quantized, translational energy.
In quantum mechanics, discrete energy levels are imposed on molecules, which goes further to include quantization for vibrational and rotational motions too. Many molecules have distinct absorption spectra from one another, which underpins how reactions can be followed in real-time.
Absorption differences existing between chemical species—even molecules which exist for a very short time—means that marking the transformation from starting molecule to product can be brought about by probing changes in absorbed frequencies of initially identical molecules over time. Even unstable intermediates can often be observed provided they have a distinct spectral handle.
Aside from reactions of atmospheric importance, most chemical reactions performed in an academic or industrial setting are undertaken in solution. For those watching chemical reactions however, solution environments mean that little time is afforded a chemist in order to watch their molecules move.
The concentration of molecules in liquids is much larger than in gases, and the additional mobility afforded to the reactant molecules in liquids compared to solids increases the probability of reaction. But how does adding a solvent to a chemical impair scientists’ abilities at trying to understand what’s going on?
A complex feature of molecular absorption in solution is due to interactivity. How solutes interact with the solvent is subtly different for each molecule, even if these solute molecules are chemically identical; each solute molecule therefore absorbs light of a slightly different frequency to its chemically identical neighbour which—over the collective sample—gives rise to broad absorption peaks, an effect called inhomogeneous broadening.
Inhomogeneous broadening smears absorption features into each other, rendering elucidation of molecular motion extremely difficult. What would the absorption peaks look like without the blurring? Which extra information could be gained from the additional resolved peaks and fine detail?
Michael P. Grubb, current Assistant Professor at Fort Lewis College in Durango, Colorado, gained inspiration in how to lessen spectral blurring in solution from watching James Cameron’s 1989 sci-fi film, The Abyss.
“The film considers how to overcome breathing in deep-sea environments, not by using compressed air but by using perfluorocarbon liquids”. The idea of using perfluorocarbons for liquid breathing hinges on a 1966 Science paper by Leland C. Clarke and Frank Gollan, in which the loose, non-stick structures of perfluorocarbon liquids—which can be considered a type of liquid Teflon—meant that a large amount of gas such as dioxygen could be trapped in the open space between the solvent molecules for liquid breathing demonstrations with submerged cats.
Although the themes in The Abyss at first appear disparate from the field of reaction dynamics, in a series of experiments at the University of Bristol where Grubb was a postdoctoral researcher, Grubb posited “If a reaction dynamics experiment were to be carried out by trapping reactant gases in perfluorocarbon liquids, the reduced interactivity of the solvent and trapped molecules might mean that more spectral information could be gleaned directly following the reaction”.
In other words, the lack of solvent interactivity with trapped gases implies that inhomogeneous broadening will be lessened. Trapped molecules will exist in a similar solvent environment to their neighbours.
To initiate a chemical reaction, two lasers had to be used—one to initiate the chemical reaction by blasting a molecule apart, and one to probe the outcome. Absorption differences are then obtained, but to see this in real-time, both lasers were pulsed, and in this case the time duration between bursts of light was on the order of femtoseconds (million billionths of a second).
One can study chemical reactions with pulsed lasers of any time duration, or even with a continuous laser, however since individual chemical transformations occur on a femtosecond timescale then femtosecond pulses lasers must be used in order to watch the absorption changes and glean any idea of what is happening. This underpins the idea of temporal resolution.
To understand temporal resolution: take a car starting out at 70 km/hr in a straight line from a dealership. If we ignored the car for an hour and then observed it again, it could be 70 km away—but we would have no idea what it had actually had done in the meantime: whether it had crashed into another car after 50 km; stopped after half an hour, or had even decided to come back on itself.
The same is true of molecules. In order to follow a chemical reaction, a pulsed laser of suitable pulse division must be used in order to see what is happening: for our car, an observation every 10 minutes may suffice; for molecules, a laser flash every hundred femtoseconds.
In the chemistry femtolab at the University of Bristol, a pulsed UV laser was used to blast the BrCN molecule into bromide and cyanide radicals; a second laser was then used to follow the fate of the hot cyanide fragment absorption profile throughout the reaction. Observing relatively un-blurred spectral features, the group could observe the absorption peaks narrowing in time, which was attributed to translational cooling—the first time such an experimental verification of translational cooling has been made in this way.
Rotational dynamics were also directly observed. Rotational motions of trapped molecules have previously been observed in non-reactive liquid helium droplets (at a temperature of a few kelvin) although the applicability of molecular dynamics in cold helium to motion which occurs in room temperature liquids—a requisite temperature condition for many important reactions—was not clear until now.
“It has been known for some time that the very highly torqued products of the ICN and BrCN dissociation reaction, are metastable—they take an anomalously long time to cool”, says Professor Stephen Bradforth at University of Southern California. “This has been shown in high density gases and also in several room temperature liquids like water and alcohols in research going back several years at USC. Polarized femtosecond measurements from our lab followed the orientation of the CN fragment in real time, and showed that it took tens of free rotations of the fragment before water or alcohol as solvent finally developed the friction to relax the rotational motion of the diatomic.
The new work from Bristol beautifully shows the energy relaxation more directly—rather than observing the plane of rotation and how it tips over, the experiments in the perfluoroalkanes allow a direct read out of the quantum state distribution of the CN rotors and how that relaxes with time.”
“The next step”, Grubb mentions whilst aligning a laser, “is to study incrementally complex systems, perhaps ones in which the solvent fluorine atoms are replaced one at a time with more interactive groups. This would be done in order to see how more complicated and widely-used solvents behave”.
Ultimately, a desire for full comprehension of chemical reaction dynamics is sought because a complete knowledge of the comings and goings of a chemical system implies that scientists could design new reactions from scratch, a process whereby individual reactions could be tailored to give a wholly predictable outcome.