Back to blog index





The chemical structure models that we build and what they are for


We specialise in ready-made ball and stick models, which represent just one possible type of model for representing molecules and crystal structures. Models are invaluable in allowing researchers to easily illustrate how and why their structure functions in the way that it does, and to better explain their research to visitors. It seems like a ridiculous question if you are used to using molecular models, but have you really thought about what molecular models are?


When I need to explain something, I find it helps to re-examine its dictionary definition.  In our case, the Oxford English Dictionary defines a model as:

“a representation in three dimensions of some projected or existing structure, or of some material object, artificial or natural, showing the proportions and arrangement of its component parts”.

Models are common in our everyday lives. Anyone who enters a museum is likely to be surrounded by dozens of them, new public buildings often have an architectural model in their foyer, and many people still build the "Airfix" or "Revell" style of transport model. These models are almost always scaled-down versions of larger structures, and are structurally and aesthetically very similar to the objects that they represent – they're just smaller, having been scaled down by factors of between 10 and 1000.
Models of chemical entities are not like these, though, and it is all too easy to think that they are.  The first difference is only obvious when we think about it - we are not scaling down, we are scaling up - and by huge factors: typically by the order of 1010.

The direction and magnitude of the scaling, while curious, isn't a significant difference. The important distinction is that we're not creating enlarged facsimiles of the subject matter, because what we are modelling doesn't exist in the same form as the macroscopic world.  We are attempting to produce models of concepts that are the result of mathematical descriptions (take a moment to get your head around that).  Those mathematical models have assumptions and approximations imposed on them, such as larger atoms having hydrogen-like orbitals.  We are creating physical models of a scientific model of a mathematical approximation.  These aren’t models in the engineering sense; these are now illustrations of descriptions of reality.


Models of chemical structures, then, can never be perfect replicas of the microscopic structures that they represent.  Unlike engineering models, the qualitative difference between the quantum world and the macroscopic world is an unbridgeable gap because we cannot create a scaled replica of that particular reality - and if we could, it wouldn't really help much anyway.


What do we want from a molecular model?

When we construct a model of a crystal structure or a molecule, therefore, we must first decide on the purpose of the model.   Any worthwhile model has a purpose - and if the purpose of a molecular model cannot be that of a facsimile of the real thing, then it must lie in the representation of specific aspects of the molecule.


There are a number of different types of molecular model, and each can be used to illustrate different aspects of the objects that they represent, but there are two fundamental things that they show: either the position of the nuclei and the connections between them, or the volume occupied by the electrons. The former are represented by ball and rod structures or framework models, and allow visualisation of the atom positions in relation to one another and whether or not there are bonds between them. Space-filling models, on the other hand, allow visualisation of the volume occupied by the molecules (but bear in mind that molecules comprising diffuse electron clouds don’t have hard boundaries). There are other specialist forms, of course, such as polyhedral structure models, that are used to illustrate the coordination polyhedra around cations.

School is where most of us start learning about molecular structures. We are taught that we can draw structures comprising element symbols that we join together with lines and, later, that we can combine some groups in units, such as -CH3. At this stage of a science student’s education, space-filling models usefully show how the simple 'structures’ written on paper representation three-dimensional electron clouds that are held in check by quantum properties (although we just tell school students that they have these shapes and they should accept that for now…).  As the nature of electron orbitals and chemical bonds become more familiar to us, the size and shape of the electron clouds in molecules (and hence the shapes of the molecules themselves) become implicit in our minds as we scribble down a chemical structure. It's unlikely that many of us give a second thought to this process, but just think for a minute - when you draw a trivalent nitrogen structure, don't you just know there's a non-bonding electron pair in there, too? Most of the time, then, it's enough to represent the relative positions of the nuclei – just as a paleozoologist can look at a skeleton of an extinct animal and visualise the muscles of the living creature, how it held itself and how it moved, chemists automatically ‘fill in’ the electron clouds and corresponding bonds for themselves.  This is incredibly useful – that we can, to a significant extent (and without having to spend much conscious thought), extrapolate the three-dimensional electron clouds in a molecule from the positions of the nuclei, enable us to learn more from these ‘skeletal’ structure models than might appear to be the case.  

This explains why ‘ball and stick’ (or ‘ball and spoke’) models can be so useful in our teaching, and why they remain the most common form of molecular model. Chemical function is affected by shape and form just as much as it is by aspects such as charge separation - and because you can see right through a ball and rod structure, and you can see 'atoms' on the far side of the model means that the model’s teaching potential is enhanced. The user is able to clearly see the nature and type of atoms that make up the structure through the whole structure, along with the the manner in which the atoms are joined together to form the shapes that those linkages produce. If we represent the structure with space-filling models, we would normally be unable to see beyond the outermost layer of atoms. Being able to look through the structure and into its interior allows huge amounts of structural information to be conveyed in a simple, elegant and efficient form.

Our models

This is where companies like ours come in. In previous times, many chemists made their own models, but that was in an age when research worked more slowly, academic pressures were lower and people had more time to spend making the models. Our role is to make those models for businesses, museums, academics, individuals, lawyers, artists - anyone who wants a fine-looking model.

Ours is a business that was started by Arnold Beevers, an early crystallographer, who started the business as a unit around forty years ago within the University of Edinburgh. After Arnold's death and some uncertainty about its future, I had the opportunity to take over 'Beevers Models’ - the alternative was that it would have simply ceased to exist. That would have been a tragedy - Beevers models are found in museums, universities and businesses around the world. As the late Victor Kiam, former owner of Remington put it; "I liked [the product] so much, I bought the company". We employed new staff, and continued for while in the University of Edinburgh’s School of Chemistry, where I lectured for several years. A few years ago, we moved out and are now based in the Scottish Borders, completely separate from the university.

Our business focuses on ball and spoke models of inorganic structures and small organic molecules. We now build giant models, metal models, Perspex (Lucite) models, 3d printed models, medium scale model kits, and any new challenges that customers like to throw at us.


The end of molecular models?

It's a common question - how long do we think that model-making will last in a world of increasingly realistic computer graphics? There is an issue with computer graphics, though, because they don't perfectly represent 3d objects to our eyes - our vision results from a far richer experience than two slightly dissimilar images viewed in each eye. Our perceived depth of field and the need to be able to instantly refocus enhance our experience. The three dimensional nature of physical models is more easily viewed than even the best computer graphics.


It is a wonderful irony that we sell a good number of models to theoreticians who can’t easily visualise the interactions in, or geometry of, the molecules that they are studying, or can't easily show them to visitors.  We get many, many, emails from customers thanking us for their models - most often for their beauty but often for new insights that they have given them.  


Ultimately, many people simply take pleasure from owning tangible and elegant structures that illustrate their research interests.  It’s part of being human - people have made and valued decorative objects for tens of thousands of years and that's unlikely to change soon.   For myself, I've enjoyed and appreciated molecular models for as long as I can remember - it’s why I build them, it's why I run this company, and it's why we do our best to produce the best possible models for others to enjoy.   As long as people want to have their crystal structures, and molecules in the form of sculptures that they can display, we’ll be here to make those models for them.



Molecules with granite base

Face Centred Cubic structure model of Copper



Next posting


"Molecular model colours"