What exactly happens when molecules react? Given molecules’ flexibility and the constant fluctuation of atoms’ positions, perhaps it’s surprising that chemicals react with each other over and over in predictable ways.

Transition state theory, developed in the 1930s by the likes of Henry Eyring and Michael Polanyi, offers an explanation by linking together key atomic motions with the energy changes that drive chemical reactions. But the ultimate insight would surely come if researchers could directly observe the configurations of chemical species in the moments before, during and after a reaction.

That grand thought experiment chemists have used as a pedagogical tool for nearly 100 years, we can now do

Dwayne Miller, University of Toronto

That was the dream that inspired University of Toronto researcher John Polanyi – Michael’s son – and the California Institute of Technology’s Ahmed Zewail in their holy grail article. But how could scientists possibly track sub-angstrom atomic motions that take place in less than a millionth of a millionth of a second?

Full screen in popup

• Previous
• Next

This network graph shows how prominent researchers in this area are connected to each other – a link between two people means they have frequently published as co-authors.

For example, we can highlight Ahmed Zewail (the large dark blue dot) and see his connections. Darker coloured dots represent authors who published earlier. A larger dot means more publications.

We can go back to the period 1995-1999 and see that the whole network, and Zewail’s, was smaller.

As we go through the years in five-year periods, Zewail’s network grows. He won the Nobel prize in 1999.

The whole network, like Zewail’s, is much more developed by 2009. Most of the highly cited papers were published by around this point.

Zewail’s network does not change between 2014 and 2020 as he sadly died in 2016.

1/6

show caption

Setting the stage

In 1986, John Polanyi won a share of the chemistry Nobel prize for his work on reaction dynamics. He was recognised for work he began in the 1950s, using infrared chemiluminescence to measure vibrational energies during molecular reactions taking place in the gas phase.

Polanyi is still unravelling the secrets of the transition state today. He notes that a limitation of molecular beam techniques was that they fell short of targeting individual molecules. Recently, Polanyi’s group showed how crystalline surfaces could be used to precisely aim projectile molecules at molecular targets.

Using this technique, which Polanyi describes as ‘molecular archery’, researchers can now select whether molecules will collide head-on or slightly off-centre and analyse how this affects bond formation. ‘Defining the reactive collision – called “selecting the impact parameter” or equivalently the collision miss-distance – restricts the transition state geometry. That is the new development here,’ he adds.

Four years after the holy grail paper, Polanyi’s co-author Zewail would also win a Nobel prize for pioneering the field of femtochemistry. This involved the use of lasers to first trigger a chemical reaction and then probe the sample in the femtoseconds (10^-15 s) that followed. To illustrate the unimaginable timescales involved it’s worth noting there are more femtoseconds in a single second, than there are seconds in 31 million years. Zewail passed away four years ago, but his work on ultrafast laser techniques laid the foundations for following chemical reactions on timescales relevant to atomic motion. 

Lights, camera…

In terms of directly observing transition states, the recent investment in major XFEL facilities promises to light up biologically relevant reactions using ultrafast x-ray pulses. However, these projects are still in their infancy and face a number of technical challenges, in particular how to avoid the issue of multiphoton absorption complicating experiments.

Some of the most important advances have come in the field of ultrafast electron diffraction. Important early work was carried out in the 1980s by Anatoly Ischenko and his team, then at Moscow State University, when they used electron pulses to probe short-lived radical species formed during the photodissociation of gas-phase halomethane molecules. These first experiments were able to achieve time resolution down to a millionth of a second.

By the early 1990s, Zewail’s group was carrying out electron diffraction studies with time resolutions down to a few picoseconds (10^-12 s). These studies were able to capture transient structures during halomethane dissociations, but still weren’t on a short enough timescale to follow the key individual atomic motions.

A major stumbling block was generating a suitably bright electron pulse. University of Toronto researcher Dwayne Miller likens the process to shooting a movie in slow motion. Cameras need to detect enough light in each frame to give a high contrast image. But as the shutter speed gets faster and faster, fewer photons are captured and so bright flash bulbs or spotlights become essential.

In electron diffraction techniques operating at a scale of 100 femtoseconds, a powerful electron source is critical for achieving a high signal-to-noise ratio. But there was a fundamental problem: electrons are charged species and when thousands move together in a single pulse, they repel each other.

In 2002, Miller’s team adapted methods used by astrophysicists to simulate the movement of celestial bodies during galaxy formation, to precisely calculate the effect that repulsion would have on each electron in a 10,000 electron pulse. They showed how electrons at the front of the pulse would accelerate, while electrons at the back would lag behind, stretching the pulse into a ‘chirp’. With calculations in hand, the team could devise ways to re-bunch the electrons together.

‘If you can just use normal dispersion like prisms to make the faster electrons go further than the slow electrons, you can re-compress the sample. It’s called temporal compression,’ explains Miller. Another surprisingly simple technique is just to limit the distance between the electron source and the sample. ‘This is the number one trick for me: if you know electrons are going to repel each other, their pulse is going to broaden – just don’t let it propagate very far,’ says Miller.

In 2003, Miller’s team created the first ‘atomic movie’ by using a compact electron gun to image a solid aluminium sample melting as it was superheated with an ultrafast laser. Sub-picosecond electron diffraction was able to capture the moment that the solid lattice transformed into a disordered liquid phase, as the atoms literally shook themselves free.

In an interesting example of fundamental research leading to real-world innovation, insights on laser-induced phase transitions gleaned from this work have enabled Miller’s team to develop new tools for scar-free surgery. Using picosecond laser pulses tuned to the frequency of O–H bonds, water molecules in biological material can be driven into the gas phase faster than any other energy exchange that can damage the surrounding tissue. This technique allows excision with accuracy down to the single-cell level, and Miller believes it will eventually be adopted into most surgical procedures.

…Action

Solving the brightness issue has opened the door to studying more complex chemical systems with time-resolved electron diffraction. Sample preparation is now one of the biggest challenges for these experiments. The reactions studied need to be triggered by light to enable the precise timing of the electron beam to capture the process. And the crystalline samples of suitable substrates need to be prepared on a nanometre scale.

In 2013, for the first time, Miller’s team captured key atomic motions during a molecular reaction. They were able to identify four structural changes that are crucial for driving the photo-induced ring-closing reaction of a diarylethene derivative. In another example, they followed six essential movements during a metal-to-metal charge transfer in an organometallic molecular crystal, with spatial resolution down to a hundredth of an angstrom.

These movies have revealed how just a small number of atomic motions are key to driving reactivity. And Miller emphasises that these experiments are now providing insight into systems of real relevance to chemists.

‘We can now talk to our organic colleagues and say, “Look here are the key modes, push them around, you’ll find a critical point – that’s the halfway point defining the transition state. Now think about how you’d modify that”,’ says Miller. ‘So we now have a new intuitive basis that is a hybrid of structure and dynamics to think anew about how chemistry works. That’s where I think we’ve evolved from Michael Polanyi and Henry Eyring’s work in the 1930s ­– ­that grand thought experiment chemists have used as a pedagogical tool for nearly 100 years, we can now do.’

With the tools now available, it’s all about application. Miller explains that advances with nanofluidic cells should soon allow ultrafast electron diffraction to probe solution phase samples, opening up a way to examine the role of solvents during chemical processes. Solution phase studies also bring the tantalising prospect of watching biomolecules perform their functions in their native states.

Beyond this, the next challenge for the ultrafast imaging community will be to try and pull out even more detailed information from these experiments. With improved resolution and innovative approaches to diffraction analysis, Miller believes it may soon be possible to follow quantum interactions as reactions take place. ‘We know we are dealing with purely quantum objects, but we view them for the most part as balls and sticks – the atomic positions and bond lengths. It would be amazing to observe constructive interference along a reaction pathway leading to a product state,’ he says. ‘One can dream of pulling out the nuclear and associated electron probability distributions overlaid on top of the evolving structure to see these interference effects.’