Do molecules really have shapes?

I don’t think I need to explain to chemists the importance of molecular structure. It is one of the most central properties invoked to understand natural phenomena, and one that has been pictorially represented like no other, in both research and educational settings.

As such, no one really questions that molecules have structure. But maybe they should?

I admit, I am intentionally provocative. No one, myself included, really doubts that molecules have a shape. Its representation on paper may be idealised in that it corresponds to a still picture of particles in space, whose movements concentrate in specific regions in such a way so as to give an abstract image of a shape. This, even if we don’t always point it out, is uncontroversial.¹ However, it does not undermine the reality of structure. In sufficiently large time and length scales, and with the appropriate technology, it is visible. Not only that; structure acts as the explanation of many important phenomena. It figures in counterfactual statements and helps predict how materials behave and transform.

The ground state of isomers does not correspond to a specific observed isomeric structure

In philosophy, when entities or properties have such explanatory and predictive capacities, we say that this implies them having causal powers. Following Alexander’s dictum that ‘to be is to cause’, this suffices (at least according to some) for them to exist.

Nonetheless, there is a context in which the existence of structure can be meaningfully questioned. This context concerns isolated molecules. In 1927, Friedrich Hund was among the first to point out a certain paradox with isolated molecules, and specifically with enantiomers.² He noticed that the ground state of such molecules corresponds to a superposition of its possible structures. This is a paradox because in chemistry we observe either one of the two possible enantiomeric structures. Later on, chemists as well as philosophers, noticed that a similar problem arises when one describes quantum mechanically other kinds of isomers too.³ Specifically, it was pointed out that the ground state of isomers does not correspond to a specific observed isomeric structure. Instead, one has to apply the Born Oppenheimer approximation for the quantum mechanical description of such systems to correctly identify the observed structure.⁴

For a long time, this situation was invoked to show that there is a certain rift between chemistry and quantum mechanics.⁵ It was believed that this is evidence of the non-reducibility of chemistry to physics. However, this was recently contested, by none other than yours truly!

A structural paradox

In a paper I co-wrote with King’s College London’s Alexander Franklin, we argue that this situation is just another instance of a more general problem in quantum mechanics: the measurement problem.⁶ Assuming that quantum mechanics offers a complete description of systems and that superposition states are also candidate physical states, then why do the properties we measure always have determinate outcomes?⁷ This conundrum, we argue, seems to arise when describing isolated molecules like isomers, and explains the paradox pointed out by Hund.

If the ground state corresponds to a superposition, one could claim the system has no structure whatsoever

Now, how does this undermine the existence of structure? Well, there are certain interpretations to quantum mechanics that take superpositions of states as meaningfully existing in some way. But if the ground state of a system corresponds to a superposition of structures, then one could claim that the relevant system has no structure whatsoever.

The possible implications are very exciting from a philosophical perspective. This is because shape is considered one of the most uncontroversial properties of objects. In fact, Locke considered it an intrinsic property of objects.⁸

But maybe this is incorrect? If molecules do not exhibit structure in isolation then we should cease thinking about it as an intrinsic property. Instead, structure is something that objects actualise in relation to something else. Put in philosophical terms, it is a relational property: a property that comes about when interacting with something. This something could be anything: an entity, an energetic exchange or whatever else. If you think about it, from a chemical perspective, this is not particularly surprising. After all, it is well known that the environment in which a molecule is found not only influences but partially determines how that molecule is structured.⁹

This is a preliminary idea, but it is an exciting one to explore! After all, it seems so natural to think that things have shape regardless of there being anything around them. This may no longer be the correct way of understanding shape. Of course, this all depends on how we understand quantum mechanics (will we ever?). But it also hinges on philosophical assumptions that we all make, even if implicitly, about the nature of properties and of shape in particular.