Controlling molecular rotations in a gas-surface collision

Occasionally a small change can make a big difference. The small change we made was to alter the rotational projection quantum state of a deuterium molecule that hit a Cu(111) surface. And the big difference? To change the probability the molecule stopped rotating as a result of the collision.
Published in Physics
Controlling molecular rotations in a gas-surface collision
Like

Sometimes experiments don’t work, but if at first you don’t succeed, then try, try and try again. The breakthrough in our study came after about a month of trying when we got an unexpected result. Our experimental approach is based on using inhomogeneous and homogeneous magnetic fields to control and manipulate the rotational orientation projection (mJ) states of molecules, which can be considered (classically) to correspond to whether the molecule is rotating like a helicopter or a cartwheel, as shown in Figure 1.  The idea of the experiment that we ran was to determine whether we had this control over the mJ states of a D2 molecule, which we were pretty sure we wouldn’t under the conditions that we were using to perform the experiment. To be absolutely sure, we left the measurement running overnight, fully expecting the result to be a flat line (at least within errors). We came in the next morning to discover that the signal did in fact oscillate. Not only did this reward our patience but also showed us that we could control the rotational projection states of D2 and therefore could perform the measurement we wanted to do.

Figure 1. The rotational orientation projection (mJ) states describe the projection of the rotational angular momentum (J) of the D2 molecule onto the surface normal (grey line). Classically, when mJ = J this corresponds to a molecule rotating like a helicopter (left), and when mJ = 0 to a molecule rotating like a cartwheel (right).

This was the start of over a week of having the experiment running 24 hours a day collecting the data which produced the oscillation curve (as it is called in a paper), or wiggle curve (as it is known in our lab) presented in the publication and shown in figure 2 below. And what this wiggle curve shows is really quite cool (although I might be biased!). As we scan the magnetic field in the experiment, we are manipulating the mJ states of the D2 molecules that hit the Cu(111) surface. The fact that the signal oscillates demonstrates that this matters for the particular scattering channel we are measuring, which in this case corresponds to stopping an initially rotating D2 molecule. Very simplistically, this oscillation could be interpreted as a molecule that is rotating like a helicopter when it collides with the surface has a different probability of being stopped from rotating by the collision than when it is rotating like a cartwheel.

Figure 2. The oscillation (or wiggle) curve measured for D2 molecules that are initially rotating but stopped rotating when they collide with a Cu(111) surface.

Unfortunately, this intuitive picture does not give a complete description of what is going on, with it not being possible to reproduce the oscillation pattern we measured just by considering the population of the different mJ states. Instead, our collaborator at Leiden University in the Netherlands ran scattering calculations which could be used to predict what the experimental signal should be. The results of these calculations suggested that only molecules that are initially rotating like a cartwheel would undergo rotational de-excitation, with helicopter states being far less likely to lose their rotational energy. Whilst the calculations correctly predicted the general phenomenon that we were measuring, that manipulating the rotational orientation states would lead to oscillations in the measurements we performed, the signal that was calculated using this prediction failed to reproduce the experimental data. The Swansea University team also performed data analysis and ran fits of the experimental data. These fits reproduced the wiggle curves far better than either the simplistic model that only considers mJ state populations or the results from scattering calculations. Most significantly, they showed that both D2 molecules that are rotating like a helicopter and a cartwheel must be able to undergo rotational de-excitation to be able to properly describe the experimental data.

The phenomenon that we are modulating, stopping D2 molecules from rotating, requires the molecules to lose their rotational energy. In this case, this energy is transferred to translational energy, which simply means the molecules are travelling faster after the collision than before it. We are controlling this energy transfer by scanning the magnetic field which manipulates the mJ states of the D2 molecules, which changes their energy by about a billion (or a 1000 million) times less than the rotational energy that is transferred as a result of the collision. The fact that a manipulation which uses such a tiny energy difference to change the probability that a significantly larger energy transfer happens is counter-intuitive, as we are used to control schemes which require an energy which is equal to, or greater than, that being transferred.

Figure 3. The machine used to perform the experiment, with Yosi Alkoby at the controls.

The machine that allowed us to do these measurements is shown in the photo in figure 3. We only used the ‘first arm’ of the machine to control the D2 molecules in this study, which corresponds to the metal tube on the right-hand side of the photo, the second part of which contains the tuneable magnetic field that we scan to perform the measurements. At the end is the ultra-high vacuum (UHV) chamber, which houses the surface. The tube which runs from this to the left of the photo is the ‘second arm’ of the machine, at the end of which is the detector (which is not a giant roll of tin foil despite appearances!). Whilst this can be the most infuriating and attention-seeking machine, always finding a new and novel way to break (usually in a way you couldn’t possibly dream of), sometimes everything comes together and it just works. It’s those occasions that make the battle worthwhile, and it is one of those occasions that led to the publication presented in summary here, which can be found by following the link below.

Stopping molecular rotation using coherent ultra-low-energy magnetic manipulations.

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Subscribe to the Topic

Physics and Astronomy
Physical Sciences > Physics and Astronomy

Related Collections

With collections, you can get published faster and increase your visibility.

Applied Sciences

This collection highlights research and commentary in applied science. The range of topics is large, spanning all scientific disciplines, with the unifying factor being the goal to turn scientific knowledge into positive benefits for society.

Publishing Model: Open Access

Deadline: Ongoing