On Sequencing Transition Metal Chemistry

Transition metals complexes are a central topic in inorganic chemistry, but hard to sequence in the curriculum. On the one hand, they are so important that it is valuable to expose students to the chemistry of complexes early. On the other hand, they are best modelled using group theory to develop an appreciation of the splitting patterns characteristic of multicentre bonding. Group theory is normally taught later in the degree than introductory complex chemistry.

The traditional work-around to this has been to teach introductory transition metal chemistry using crystal field theory. This allows you to advance a model of splitting using a knowledge of the shapes of d-orbitals, rather than the irreps they span within the symmetry of a point group. This is normally accompanied with a faintly-embarrassed appeal to rote memorisation of the symmetry labels (“eg and t2g are just names - you’ll look at this more closely next year”).

The crystal field approach allows students to engage with the key ideas of field splitting, high/low-spin configurations, and the basic magnetic properties of metal complexes. At the same time, it is heavily limited by its basis in electrostatics, meaning that important rationalisations are impossible to access (perhaps most importantly for the spectrochemical series). It also wastes time in an already-pressured curriculum. Learning crystal field theory only to abandon it for ligand field theory is arguably a redundancy.

I think it would be more successful to sequence transition metal complexes after group theory, and want to use this blog to examine ways this might be accomplished.

Option 1: teach group theory in first year

I present this option so I can reject it. Group theory requires a solid grounding in maths and a fair appreciation of molecular geometry. Both of these things are reliable at the start of second year, and in my view this makes group theory a good fit at the start of second year.

I’ve seen people try to sequence point group assignments in first year (ammonia is C3v etc). I feel this is problematic: the point group assignment is not significant knowledge until you can link it to the behaviours of orbitals and SALCs (or stretching modes). “Why are we learning this?” is a blisteringly fair question.

Option 2: teach metal complexes in second year, and teach something else in first year

This is broadly my preferred option. There’s a substantial question about what you shuffle in to get the d-block out, though, and I want to explore a few options: defects, PXRD, and the lanthanides.

Crystal Defects

Simple crystal defects would link well to the ionic structures typically covered in first year, and provide a nice exploration of the key first-year themes of enthalpy and entropy. Some brief discussion of kinetic/thermodynamic approaches to solid state synthesis would maybe let you bulk this up to the size of a full lecture course, but its central purpose would be to review and consolidate the central thermodynamic considerations of ionic bonding.

PXRD

Indexing the PXRD patterns of (cubic) solids would let students do some good Bragg geometry, something which leads smoothly from derivations of Madelung constants for simple ionic solids. Simple isoelectronic ion pairs (CsI, MgO) and the mathematics of systematic absences (but n can’t be 7!) are really nice introductions to thinking critically about analytical work, something which would perhaps be lost when the spin-only magnetic moments of transition metal complexes were removed from the first year.

Lanthanides

The lanthanides are probably my favourite proposal.

The free ion model means that the topic can serve to review the central features of atomic energy levels and Hund’s Rules, as well as providing an interesting extension of the ionic bonding which sits so centrally in the first year (the contraction of the cation radii across the series, and its consequences on p-block elements such as thallium would be a sophisticated piece of first-year work).

The magnetic moments appropriate to systems with spin-orbit coupling arguably serve as a more general case than the spin-only analysis currently performed by first years on d-block complexes (and quenching could be introduced in second year as a corruption of the Lande formula rather than the normal state of affairs).

Conclusion

I have a lot of time for discussions of the spiral curriculum, but I also think we have failed to communicate the value of revisiting descriptive chemistry from different theoretical perspectives: when you ask a student what their experience of school chemistry was, for example, they often say things like “the teachers started each year by saying they lied to you last year and now we’re going to think about chemistry like this”. The spiral is described here - however light-heartedly - as deceptive, unproductive. And - broadly speaking - that’s because sometimes it is.

In the school curriculum, I see the value in teaching the shell model of electronic configurations before orbitals. The rationale here is the same as the rationale for scaffolding: you use a temporary structure to build something grander. Abandoning the scaffolding afterwards is the correct thing to do, just as constructing it in the first place was. Scaffolding is necessary.

But crystal field theory seems unnecessary: it’s just a way to tread water while we wait for symmetry. So why not just do metal chemistry cleanly once students have learned group theory? There’s lots of great, relevant material which could be put into first year instead.