[Nimmie (OP)]

Why is the atomic structure always described as spherical or using a sphere analogy?From Rutherford's model to Bohr's model, both described an atom using a sphere, why? could it not have been triangular, square or any other shape?

[Kryptid]

It isn't always a sphere. The shape of an atomic orbital is determined by quantum physics and changes depending upon its energy level. The s orbitals are spherical, but p orbitals, d orbitals, f orbitals and so on are more complex than that: https://en.wikipedia.org/wiki/Atomic_orbital

It is also important to note that atoms are not to be mistaken for "miniature Solar Systems" where the electrons go whizzing around the nucleus like a tiny ball in a planar orbit. Electrons have wave-like properties and that contributes significantly to the shape of atomic orbitals.

[alancalverd]

For some purposes tt's a reasonable model of an isolated atom but as Kryptid says, it hasn't used to describe atoms in molecules for the last 100 years or so because it is manifestly wrong.

[evan_au]

Why is the atomic structure always described as spherical or using a sphere analogy?

I purchased a chemistry model kit, with balls of various sizes & colors, and sticks of various lengths.
You can assemble models of molecules by joining the balls with the sticks (there are little holes in the balls).

Although these model atoms are spheres, if you look closely, you will see that:
- Different atoms have a different number of holes, eg Hydrogen has 1 hole and Oxygen has 2 holes. This reflects the different number of covalent bonds that this atom can form, in an attempt to fill its outer electron shell.
- Different atoms have the holes at different angles. This reflects the fact that the orbitals are positioned at specific angles. So water has a boomerang shape. Nitrogen and Carbon can both form little pyramids, but the bonding angles are different.
- Some atoms have multiple versions. For example, some Carbon balls have 3 holes and some have 4, reflecting the different ways that carbon can form atomic bonds.

So atoms are modeled as little spheres (it helps them pop out of the moulding machine more easily), but to make more realistic models of molecules, the holes are oriented to reflect the more complex underlying structure.

[chiralSPO]

A single atom is best considered as spherical. The positively charged nucleus is at the very center, and the negatively charged electrons are distributed around it. The electrons are attracted to the nucleus and repel each other. Because the nucleus is so incomprehensibly small, even compared to the size of an atom, it can be thought of as a single point (from the perspective of an electron). In the absence of any external influence (like an electric or magnetic field, or other nearby atoms) a sphere is the only reasonable shape for such a thing to have, because the only interactions are within the atom (between the point-like nucleus and the electrons), none of which have any way to define a special direction (atoms are isotropic, and spheres are isotropic).

Even orbitals, as mentioned above by Kryptid, are actually spherical in an isolated atom. For the purposes of intuition and making the math easier, we can break the orbitals up into linear combinations of sub-orbitals. Depending on which linear combinations are used, each suborbital can have a shape that is different from spherical, but sum of all of the sub-orbitals will form a spherically symmetric orbital in a lone atom.

Once we go to an atom that is interacting with anything else, the assumptions of sphericality all go out the window. For instance, if we place an atom in a strong electric field, the atom will distort into an ellipsoid (elongating along the axis parallel to the electric field lines).

Then, once we consider molecules, which have many nuclei (at least two), molecular orbitals become the important construct (sometimes one can think of this in terms of hybrid atomic orbitals), and these come in all types of shapes (none of which are spherical).

[chris]

to make more realistic models of molecules, the holes are oriented to reflect the more complex underlying structure

So you couldn't use the same model oxygen atom in a water molecule and a carbon dioxide molecule then?

[yor_on]

Spheres are interesting. For example, is a 'point' spherical? And it is a very special figure of geometry in that you can't define a 'start', nor an 'end' to it. At a quantum level they lose their definitions though.
=

Actually it might have to do with classical thoughts of existence and geometry. We always stand on others shoulders

[alancalverd]

So you couldn't use the same model oxygen atom in a water molecule and a carbon dioxide molecule then?

No - and it poses quite a problem!

The H/O\H molecule is planar, with a bond angle of about 104.5°, so easy to model with a "water oxygen" atom singly bonded to two hydrogens. This is a fairly good model as it demonstrates the possibility of multiple modes of vibration and stretching, and an electrostatic asymmetry that allows temporary polymers and mutiple crystal structures to form, hence the anomalous properties of water and its exceptionally broad infrared absorption spectrum. 

Where carbon is singly-bonded, as in aliphatic compounds, the bonds lie towards the apices of a tetrahedron, so your carbon atom model usually has four holes at 109° in 3 dimensions. You can't use rigid rods to get this to double-bond with two oxygen atom models, whether they are "water", "linear", or "parallel" models, so the kit usually contains springs. Thus when you assemble the O=C=O model you get  something flexible, with multiple bending and stretching modes. This suggests that CO₂  would have a broad infrared absorption spectrum, but the fact is that the real molecule is very rigid, with the valence electrons forming a linear cylinder, and has a very limited IR spectrum.

The result is much confusion in classrooms where anthopogenic climate change is preached, and a whole generation grows up believing a profound untruth.   

In the words of Stafford Beer "a dead mouse is a very good model of a live mouse, but only for a very short time". Beware of models!

[chiralSPO]

Where carbon is singly-bonded, as in aliphatic compounds, the bonds lie towards the apices of a tetrahedron, so your carbon atom model usually has four holes at 109° in 3 dimensions. You can't use rigid rods to get this to double-bond with two oxygen atom models, whether they are "water", "linear", or "parallel" models, so the kit usually contains springs. Thus when you assemble the O=C=O model you get  something flexible, with multiple bending and stretching modes. This suggests that CO₂  would have a broad infrared absorption spectrum, but the fact is that the real molecule is very rigid, with the valence electrons forming a linear cylinder, and has a very limited IR spectrum.

The result is much confusion in classrooms where anthopogenic climate change is preached, and a whole generation grows up believing a profound untruth.   

I doubt very much that the global perception of CO₂ as a greenhouse gas has been unduly swayed by the use of model kits which employ springs for double bonds and sticks for single bonds.

[Bored chemist]

Thus when you assemble the O=C=O model you get  something flexible, with multiple bending and stretching modes. This suggests that CO2  would have a broad infrared absorption spectrum, but the fact is that the real molecule is very rigid, with the valence electrons forming a linear cylinder, and has a very limited IR spectrum.

The result is much confusion in classrooms where anthopogenic climate change is preached, and a whole generation grows up believing a profound untruth.   

Not really.
The stiffness of the springs affects the frequencies of IR absorbed, but not the strength of the absorption.
The fact that it's linear does,  make a difference to the number of absorption bands. But it's an effect of symmetry rather than stiffness.


The breadth of the peaks is due to rotational quantum levels and those are affected by the symmetry the effect's not overwhelming.
In any event, CO2 absorbs IR.
Some of those absorption bands aren't quite saturated so a change in CO2 concentration gives rise to a change in the amount of energy transferred, and where that transfer happens.



You seem to be right about a poor understanding giving rise to mistaken beliefs about climate change.

[Bill S]

I suspect that when it was first discovered that the atom was not a fundamental particle, there was no known reason to think that the spherical model (which would have been the simplest) would not provide an adequate image.  It serves so well for so many teaching purposes (I first met it in O-level geology, at which level it still provides a workable aid to understanding minerals) that I think it will be around for some time to come. So it is probably not a "dead mouse".

If your ladder reaches the roof, why insist on having one ten feet longer? 

[evan_au]

So you couldn't use the same model oxygen atom in a water molecule and a carbon dioxide molecule then?

My model doesn't have springs, but it comes with somewhat flexible plastic rods. So the same red plastic ball representing Oxygen can be connected to:
-  two Hydrogen atoms to form a boomerang-shaped Water molecule
- or two of them to a Carbon atom to form a rigid, linear Carbon Dioxide molecule. The flexible bonds bulge out a bit, but the shape of the molecule is representative
- Similarly, you can join two carbon atoms together (with some hydrogen atoms) to form ethane (single bond), ethene/ethylene (double bond) and ethyne/acetylene (triple bond - this molecule is linear). This uses the 4-hole version of carbon.

Water_CO2_acetylene_small.jpg (113.11 kB . 640x853 - viewed 17141 times)
But when it comes to forming the 6-carbon benzene ring, you need to use the 3-hole version of carbon to get the observed planar molecular structure with conjugated bonds.

[Bored chemist]

It's kind of tricky to talk of the "shape" of something when it doesn't have a well defined surface.

As ChiralSPO pointed out, the atom is held together by the +ve charge on the nucleus and the field round that nucleus is spherical.

[chris]

So you couldn't use the same model oxygen atom in a water molecule and a carbon dioxide molecule then?

My model doesn't have springs, but it comes with somewhat flexible plastic rods. So the same red plastic ball representing Oxygen can be connected to:
-  two Hydrogen atoms to form a boomerang-shaped Water molecule
- or two of them to a Carbon atom to form a rigid, linear Carbon Dioxide molecule. The flexible bonds bulge out a bit, but the shape of the molecule is representative
- Similarly, you can join two carbon atoms together (with some hydrogen atoms) to form ethane (single bond), ethene/ethylene (double bond) and ethyne/acetylene (triple bond - this molecule is linear). This uses the 4-hole version of carbon.

Water_CO2_acetylene_small.jpg (113.11 kB . 640x853 - viewed 17141 times)
But when it comes to forming the 6-carbon benzene ring, you need to use the 3-hole version of carbon to get the observed planar molecular structure with conjugated bonds.

Those look like good modelling atoms - what's the brand please?

[evan_au]

I just found it on eBay - this one was called "267pcs Molecular Model Set Links Kit - General And Organic Chemistry Science ".

There is a single sheet of instructions, but apart from the chemical names "H", "C", etc, it is all in Chinese, plus a few drawings.
- I managed to puzzle it out with the help of a colleague who spoke Chinese, but who didn't really know any chemistry.
- The key is: Hydrogen is white, Carbon is black, Oxygen is red, but for the rest, I still have to look at the translated sheet!