My freshman year was almost over. Final exams would take place in about two weeks and I was already beginning to prepare myself mentally. Almost as an afterthought, the chemistry professor assigned us one last chapter to read on solid-state chemistry. It wasn’t a real assignment – the professor never mentioned the chapter again and as I correctly surmised, it wasn’t going to be covered on the final exam. It should have been a meaningless blip in my academic career, quickly forgotten, but I still remember that chapter after all these years, or at least the section which covered the concept of non-stoichiometric materials.
I remember absolutely hating it.
We had just spent the entire year having the “Law of Multiple Proportions” hammered into our brains. “Atoms combine in ratios of natural numbers,” they would say. “If you can’t grasp this basic concept, you’ll never get a job as a chemist. You’ll just have to settle for being a doctor or lawyer or telemarketer.” Frightening words indeed! But now, just weeks before the final, I was discovering that this law was more of a suggestion.
I think I even remember one of the provided examples. It was NiO1.03. WTF? 1.03? What sort of sick joke was this? It looked like something a freshman engineer would write, one who hadn’t yet grasped the concept of rounding. This deviation from stoichiometry was within the experimental error associated with an elemental analysis. I cannot begin to imagine what my thesis advisor would have done had I submitted an article discussing the properties of the V10O27.976- ion. The beatings would have been severe. The whole idea seemed stupid to me.
Fast forward to the present and my mind is now quite a bit more receptive to this concept. The field of non-stoichiometric materials is huge, incredibly huge, due to their special properties (catalytic, electronic, and optical). As a transition metal chemist, I now understand that the many oxidation states available to most transition metals can lead to mixed oxides, many of which are non-stoichiometric. I’ve also come to the realization that over half of my projects over the years have involved non-stoichiometric oxides in some fashion. Examples would include ZrO2/CeO2 solid solutions, various doped metal oxide catalysts, zeolites, and, at the present time, fuel cell cathodes. (Strictly speaking, zeolites are not really considered non-stoichiometric materials since there are no mixed oxidation states available, but with formulae such as NaxAlxSiO(2+2x), where x can be < 0.01, I’m still counting them.)
The defect sites in these non-stoichiometric oxides make them wonderful catalysts, especially for redox reactions. The vacancies left by the loss of oxygen atoms in the crystal structure can create materials with the ability for ion conduction (usually at higher temperatures). This leads to their use in gas sensors, batteries, and fuel cells. (La1-xSrx)yMnO3-z is a typical oxide used in fuel cell cathodes. Non-stoichiometric oxides are here to stay.
And I'm loving it.
The moral of the story: Don’t dismiss new concepts in chemistry until you’ve had a chance to work with them first.