Broadly speaking, this paper shows how the interaction between scientific practices may be subject to strategic constraints—that is, fruitful interaction between practices may require finding ways to adapt one practice to the strategies informing the other. One way, therefore, that the concepts within an interfield theory develop is by recrafting concepts originating in one practice in terms compatible with the strategies guiding the other. This sort of conceptual accommodation is often a prerequisite to fruitful interactions between practices initially directed toward different sorts of aims. I explore these general themes by considering both the strategic constraints operative in the interfield theory of quantum chemistry as well as the conceptual accommodation directed to meet them. To accomplish this, the paper supplies an account of the constraints put on the explanatory concepts of organic chemistry by the aims that have shaped the field. Because of the guiding interest in synthesis, I argue, the explanatory concepts of organic chemistry support a compositional strategy. This is how they manage to be outward-looking and thereby facilitate the reasoning by structural analogy crucial to crafting plausible syntheses of novel compounds. What makes a concept or classification “adequate or suitable” for use in organic chemistry is, therefore, its compatibility with the compositional strategy adopted by the discipline. Because quantum mechanics is responsive to different interests, quantum chemists have had to find ways to recreate this compositionality within their own framework in order to become broadly useful in support of organic chemistry. Orbital diagrams, for example, have been a useful tool for organic chemists at least in part because they are compatible with this compositional strategy. Several other ways of ensuring compatibility with compositionality that do not rely on diagrams or spatial representations have also been employed by quantum chemists. For instance, by introducing ‘localizing assumptions’ or using an ‘independent particle model’ quantum chemists have found ways to build their representation of a molecule out of pieces that can potentially recur in descriptions of distinct chemical structures. This allows insights generated in the account of one molecule to be potentially applicable to a class of other structures that contain some of the same structural components. In this way, quantum chemistry can potentially contribute to the project of understanding novel molecules in terms of structural analogies between their components and known compounds. I hope that by identifying compatibility with compositionality as the key feature of the explanatory concepts in organic chemistry, the strategic constraints operative within quantum chemistry can be brought to light. This in turn makes it possible to appreciate one significant form of conceptual accommodation that takes place in this interfield theory.

Does quantum mechanics accommodate chemical intuitions about bonds, or sweep them away in favour of something new? In this paper I look at foundational issues arising for two revisionary efforts to embed the chemical bond into quantum mechanics. Energy and structure: Since the 1930s, a standard way to explain the stability of a molecule within quantum chemistry has been to use correlation diagrams based on Molecular Orbital (MO) theory. Such diagrams allow estimates of the total electronic energy of a molecule and comparison with the separated atoms: H2 exists because its energy is lower than that of two isolated hydrogen atoms, while He2 does not exist because its energy would be higher than that of two isolated helium atoms. Hendry (2008) has distinguished the structural and the energetic conceptions of the chemical bond. The structural conception tries to identify the explanatory role of the covalent bond in chemistry, then work out what realises that role. The energetic view draws on physical rather than chemical explanation: how the stability of molecules is explained within quantum mechanics. Taking MO theory as a guide, this suggests that facts about bonds are determined by facts about bonding at the level of whole molecules. This would be a radical revision of a longstanding chemical concept, but perhaps quantum mechanics forces that on us: Weisberg (2008) has challenged the quantum-mechanical adequacy of the more retentionist structural view. The Quantum Theory of Atoms in Molecules (QTAIM): The late Richard Bader (1990) analysed the distribution of electron density within molecules, arguing that it is possible to construct well-motivated quantum-mechanical correlates of some classical chemical concepts, including bonds between atoms. However, Bader thought that QTAIM provides not bonds but ‘bond paths’ and that chemical intuitions which cannot be justified by quantum mechanics should be discarded. Esser (2019) has argued further that QTAIM provides a third ‘interactive’ conception of bonding. In this paper I will respond to Weisberg’s challenge to the structural conception, but also issue my own challenge to the energetic view: the identification of a bond with a change in energy breaks down for anything more complex than diatomic species. I also argue that Bader’s dismissal of classical ideas about bonds as mere ‘chemical intuition’ is too quick: this is a well-developed body of theory that has made important contributions to the development of chemistry. It should be given a robust defence, not a dismissal.

When philosophers investigate molecular concepts to determine whether particular accounts of molecules are satisfactory, they face a methodological challenge: they must make assumptions regarding the role(s) such concepts are intended to play. I suggest recognizing a distinction between explanatory and ontological concepts. While scientific theories intimately intertwine the explanatory and the ontological, they should not be equated. While all panelists in this symposium are discussing molecular concepts, I will argue we’re not all talking about the same thing. Recognizing this may dispel some apparent tensions that have seemed inherent in attempts to locate, or recreate, chemists’ molecules in quantum theory.

In quantum mechanics no specific molecular structure is assigned from first principles. Franklin and Seifert (2020) argue that this is due to the measurement problem. I explore the implications of this to the metaphysical understanding of structure. Specifically, I propose two metaphysical views: the dispositional and relational view. According to the first, isolated molecules maintain their structure only as dispositions. On the second, structure comes about only in relation to some environment. I evaluate how these two views match with the metaphysical implications of realist interpretations to quantum mechanics and conclude that both views radically revise our understanding of structure.