Magic molecules

Ever since science first stumbled across them, molecules have mainly been used to produce things, from materials to drugs to pesticides. Increasingly, they are now being called on to perform other tasks, from operating entirely new forms of computing to probing the secrets of the universe, as Jon Evans reports

Image caption: The cryogenic storage ring at the Max Planck Institute for Nuclear Physics, with some of the walls open to reveal the inner ion beamline 

 ‘Molecules are the best,’ asserts Danna Freedman, chemistry professor at Northwestern University, US, speaking at the virtual AAAS annual meeting in February 2021. ‘Molecules are amazing, because you can come up with a question and put every atom exactly where you want it to answer that question.’

Freedman and her group have been developing molecules for use as qubits in quantum computing. Qubits are the quantum version of the bits in conventional computing. Whereas a single bit can adopt values of either 1 or 0, a single qubit can take advantage of the indeterminacy of the quantum world to adopt values of both 1 and 0 at the same time.

This should allow quantum computers with arrays of qubits to perform multiple calculations simultaneously, rather than consecutively, as happens with conventional computers. Quantum computers will thus be able to answer questions that are difficult or even impossible for current computers to answer, such as questions that require considering countless different possibilities.

Most of the qubits developed so far have been ‘spin qubits’, meaning the values of 1 and 0 are represented by a quantum property known as spin, possessed by subatomic particles such as protons and electrons. Spin is a perfect property for qubits as it can exist in just two states: up and down.

Scientists have adopted various approaches to producing spin qubits, utilising different particles in different materials, but they all tend to struggle with producing qubits that are the same. For example, one approach is based on inserting nitrogen atoms into diamond to produce what are known as nitrogen-vacancy (NV) centres. These NV centres can then act as a spin qubit, based on the spin of the nucleus of the nitrogen atom, which can be manipulated and read out with laser light. But because the process of inserting nitrogen atoms into diamond is inherently variable, it is difficult to ensure all the qubits are identical and behave in the same way.

Working with colleagues at the University of Chicago, Freedman and her group decided to try replicating this approach, but with a designer molecule. Rather than insert a nitrogen atom into a diamond crystal, they synthesised a crystal containing a central metal atom. In both cases, however, an atom containing a subatomic particle acting as a qubit is surrounded by a crystalline material.

‘If we could take a molecule and imbue it with the same properties, we could immediately take that molecule and integrate it with the spectroscopic interface that’s been built up to address NV centres in diamond,’ said Freedman. To do this, Freedman and her colleagues produced crystals with chromium as the central metal ion surrounded by aryl ligands, which had to be arranged in just the right way to produce the electronic structure required for a spin qubit (Science, 2020, 370, 1309).

‘We can take a metal and surround it with organic species,’ she explained. ‘We can then modulate the chemistry of the organic species and use them to modulate the electronic structure of [the metal atom]. Being able to design the electronic structure enables us to control where the spins are in the system.’

Like NV centres, the spin state of the metal atom in this crystal could be read out using a laser, because the polarisation of the light it emits in response to excitation by laser light depends on its spin state, but that spin state could also be manipulated using microwaves. This means the molecule can function as a controllable, readable spin qubit, just like NV centres.

But unlike NV centres, multiple identical versions of this molecule can be produced, all with the same properties. This should make it easier to link them together into an array of qubits for proper quantum computing, which is what Freedland and her colleagues are now working on.

As well as quantum computing, they are also investigating using these molecules for quantum sensing, because, by their very natures, quantum entities are highly sensitive to their external environment.

Their crystal molecules could be particularly appropriate for this work because they are soluble in water, potentially allowing them to be used for applications such as imaging proteins inside the body.

Universal matters

Other researchers are using the quantum properties of molecules to probe the secrets of the universe. One such secret is why the universe exists at all, at least the universe of matter that we know best. For our physical laws suggest that matter and antimatter should have been created in equal amounts by the Big Bang, in which case they would have quickly annihilated each other. But the fact that we live in a universe full of planets and stars shows that this did not happen, as does the fact that antimatter is very rare in the universe, accounting for about one part per billion.

So what explains this lack of antimatter? For answers to the asymmetry between matter and antimatter, scientists are looking for other asymmetries in fundamental forces and particles. But the asymmetries they have discovered so far, such as in some forms of nuclear decay, are too small to account for the huge imbalance between matter and antimatter.

Until recently, the search for these asymmetries had mainly been conducted in particle accelerators such as the Large Hadron Collider (LHC) – huge, highly expensive instruments for smashing atoms and subatomic particles together at close to the speed of light. But thanks to advances in the quantum control of molecules over the past decade, scientists such as Nick Hutzler, assistant professor of physics at the California Institute of Technology (Caltech), US, who also spoke at the AAAS meeting, are searching for these asymmetries at much smaller scales.

Image caption: Bright green lasers used to create molecules at Caltech

‘The LHC is very exciting because it has very general-purpose particle detectors, it can discover an extremely wide range of new particles and forces,’ said Hutzler. ‘For our searches, we’re narrowly focused on new particles and forces that violate the fundamental symmetries needed to explain a matter-dominant universe. Because we can take advantage of the quantum-coherent properties of molecules, we’re more sensitive than the LHC for these particular particles and interactions.’

The basic idea behind their approach is that certain asymmetries should have detectable effects on electromagnetic fields, and it turns out that the most intense electromagnetic fields occur inside molecules, especially polar molecules. Hutzler and his team are specifically looking for evidence of violation of the charge conjugation-parity (CP) symmetry, which holds that physical laws should act the same on particles and antiparticles that differ only in having opposite charges and being mirror images of each other. Any violations of CP symmetry should alter the electromagnetic fields of molecules in predictable ways.

Ytterbium is very heavy; it has 70 protons in it. That means the electric field is very large, especially near the nucleus. YbOH also has a very simple electronic structure, with just one valence electron, which makes it easy to optically control using lasers.
Nick Hutzler assistant professor of physics, Caltech

Again, however, this only works with very specific kinds of molecules. ‘It turns out that not all molecules are sensitive to these effects,’ says Hutzler. ‘There are particular structural features that you want to engineer in your molecule to make them sensitive to the effects. We want the molecule to have a very large internal electromagnetic field and to possess single, unpaired electrons that you can very easily control using a laser or a magnetic field.’

The larger the molecule, the easier it is to give it all the desired properties. But larger molecules are more difficult to control and inherently more variable, making it harder to distinguish effects caused by CP symmetry violation from effects caused by more mundane factors. Until recently, scientists were only able to work with the simplest molecules, made up of just two atoms, but such molecules are too simple to possess the right mix of properties. So now scientists such as Hutzler are trying to work with molecules made up of more than two atoms. Specifically, they have been working with molecules containing heavy species like radium and ytterbium, such as ytterbium monohydroxide (YbOH).

‘Ytterbium is very heavy, it has 70 protons in it,’ said Hutzler. ‘That means the electric field is very large, especially near the nucleus. YbOH also has a very simple electronic structure, with just one valence electron, which makes it easy to optically control using lasers.’

The first molecule

Within the constraints of current capabilities, both Freedman and Hutzler are free to design the most effective molecules for their purposes, with the optimum mix of properties. But Oldrich Novotny at the Max Planck Institute for Nuclear Physics (MPIK) in Heidelberg, Germany, another speaker at the meeting, is more constrained. Like Hutzler, he is using molecules to explore the very beginning of the universe, but he is specifically interested in the very first molecule to appear.

The Big Bang produced just four elements: lots of hydrogen and helium, and tiny amounts of lithium and deuterium. The remaining elements would have to wait for the formation of stars before they could appear. As such, the first molecule to form was most likely the helium hydride ion (HeH+), produced by helium atoms binding with hydrogen nuclei.

As the first ever molecule, scientists think HeH+ played a central role in the formation of the first stars and thus they would like to know how much HeH+ existed in the early universe. This depends on the difference in the rates at which HeH+ formed and disassociated, by binding to a free electron to form neutral atoms of hydrogen and helium, in the early universe. Until recently, scientists had to estimate these rates, but now Novotny and his colleagues have managed to determine the disassociation rate experimentally.

This required them to replicate the conditions that existed in the early universe, before the formation of any stars, which are basically the same extremely cold, vacuum conditions that exist in the depths of space today. But replicating these conditions was exactly why the MPIK constructed the cryogenic storage ring (CSR), which began operating in 2015. This is a 35m-long vacuum tube, forming a large square, which can be cooled with liquid helium to temperatures of below 6K.

Novotny and his colleagues fired a beam of hot HeH+ into this tube. The ions then circulated around the CSR for hundreds of seconds, guided by electrostatic deflectors, during which time they gradually cooled. At this point, the scientists introduced electrons into the CSR travelling at similar speeds in the same direction, causing them to collide gently with the HeH+ ions and induce their disassociation. By detecting the resulting hydrogen and helium atoms, Novotny and his colleagues were able to determine the disassociation rate under these conditions. They found that this rate was 20 times slower than previous estimates, meaning the early universe contained 20 times more HeH+ than had been thought (Science, 2019, 365, 676).

Although not needing to replicate conditions in the early universe, Freedman and Hutzler also need very low temperatures, close to absolute zero, and vacuum conditions to perform their experiments. The low temperatures are needed to make their molecules as inactive as possible, so the tiny effects they are looking for are not swamped by the molecular vibrations and random thermal motions that occur at higher temperatures. The vacuum conditions are required to ensure their molecules do not interact with any other molecules.

Freedman used a smaller version of the CSR to conduct her experiments at 5K, while Hutzler used cryogenic cooling to reduce the temperature of the YbOH molecules to around 4K. Hutzler’s collaborators in John Doyle’s group at Harvard University have gone even further, by reducing the temperature of YbOH molecules to less than 1K. They did this with laser cooling, which works by firing a laser at molecules to slow their movement, as photons possess momentum and will thus slow down molecules moving in the opposite direction. Because temperature is just movement, this slowing reduces the temperature of the molecule.

With these kinds of advanced technologies, Freedman, Hutzler and Novotny are managing to get molecules to perform some unprecedented tasks, but they would all like to take it further. Freedman, for example, would like to work with single molecules, rather than the ensembles she uses now. That’s the next task.