What Do Atoms Really "Look Like?"

Steven Dutch, Professor Emeritus, Natural and Applied Sciences, University of Wisconsin - Green Bay


First off, in order to have a "look," things have to reflect light. Since atoms are far tinier than light waves, we can't see them with light.

After physicists discovered that atoms were made of a positive nucleus surrounded by negative electrical charge, they began wondering why the charges remained apart. An obvious analogy was planets orbiting the Sun. One problem with this idea was that if you force electrons to travel curving paths, they emit radiation. Electrons should emit radiation, lose energy, and spiral into the nucleus. Physicists were forced to postulate that  for some reason, electrons around atoms simply didn't do this. Also, since the electrons had definite energy levels, they postulated that one of the rules of atoms was that electrons could only have specific energies, and nothing in between those energy levels.

Using these rules, the Danish physicist Niels Bohr came up with a model of atoms much like a solar system. Although textbooks usually show Bohr's atom as having circular orbits, physicists actually explained some of the subtleties of electrons in terms of elliptical orbits. Bohr was able to explain the energy levels of hydrogen atoms in great detail. Other atoms (like sodium) where there is a single outer electron well removed from the inner electrons, also behave much like hydrogen. However, nobody was ever able to describe atoms with multiple electrons in exact detail using a planetary interpretation.

By the 1920's, physicists had discovered that matter also has wave-like properties and that it just doesn't work at the atomic level to regard particles as tiny points with precise locations and energies. Matter is inherently "fuzzy." They gave up thinking of electrons as tiny planets altogether.

In a way, it's unfortunate that Bohr's hydrogen atom worked as well as it did. To this day, even advanced physics books use it as a way of introducing quantum mechanics. But it's just plain wrong. And the terminology that came with Bohr's atom, of "orbitals" and "spin," reinforces the image of electrons as tiny planets even though physicists have long since given up the literal imagery. At least the weird terminology that quark physicists use, of "color," "flavor," and "charm," carries no danger of being taken too literally. Nobody is likely to believe that quarks are really red, taste like chocolate, or flirt. So forget that you ever heard of the solar system model of atoms. Abandon all planetary concepts, ye who enter here.


Electrons occupy discretely different energy levels, and since these differ in size as well, there's no harm in using the traditional term "shell" to describe them. They used to be lettered K, L, M and so on with increasing distance from the nucleus, and x-rays emitted by electrons in different shells are still denoted by those terms. In fact, it was the discovery that atoms emitted x-rays in sharply defined energy levels that led to the discovery of electron shells in the first place. For most purposes chemists and physicists just number the shells 1,2,3.. out from the nucleus. 1 = K, 2 = L, and so on.

Within each shell, electrons occupy sub-levels, of which the most important are called orbitals. With increasing energy, they are called s, p, d and f orbitals (from sharp, principal, diffuse and fundamental, relating to the spectral lines produced by the alkali metals). These have size and shape, but attempting to portray them or describe them can lead to many misconceptions:

If you put a tank on a vibrating platform, you will see standing waves on the surface.

The wave is everywhere in the tank, all the time. The negative portions (troughs) are just as real as the positive portions. Picture an orchestra. The individual notes last long enough to fill the entire auditorium. Where are the flute notes? Everywhere in the auditorium. Where are the violin notes? Everywhere in the auditorium. Where are the oboe notes? Everywhere in the auditorium. All at the same time, and they are all distinct.

So let's say you just absolutely cannot get your mind around this notion. You've seen surfers in Hawaii, so you know waves have definite locations. So you decide to do an experiment to find out where the wave is. You mix dye with the water and you poke around with toothpicks to see where you pick up the dye. You will find that you get the largest number of hits at the crests of the waves, fewer on the flanks, none at all over the nodes. If you plot all the hits you get, you'll get dense rings of points separated by zones of no points. But you could still misinterpret your result as meaning the wave is traveling in discrete circles and the points are just where you happened to hit it.

Similarly, if you absolutely force the issue with electrons, by say firing gamma rays at atoms and seeing where you strike electrons, you'll get more hits in some places than others. The wave function of the electron describes the probability of that happening. But that doesn't mean the electron happened to be at that exact point when it was hit by a gamma ray. How can we picture what happens in detail? We can't.

The simplest orbitals are spherical and are called s orbitals. We schematically represent the "fuzzy" nature of the orbital with shading.

The first shell around an atom is just a simple sphere. The s orbitals for higher shells have a concentric internal structure (not shown) with layers of higher and lower probability of finding an electron.

Remember, the electron occupies the whole orbital at once. It does not travel a circular path.

Next most complex orbitals are p orbitals with a twin lobe structure. The innermost p orbitals are two simple lobes; p orbitals in outer shells have small additional lobes toward the nucleus (not shown).

Remember, the orbital does not travel two looping paths, or loop around one side, through the nucleus, then the other side. It occupies both lobes simultaneously.

Multiple p orbitals are oriented perpendicular to each other. Here we see two p orbitals.
Here we see three p orbitals, the maximum number allowed.
After p orbitals come d orbitals. There are five of them. Four have the four-lobed appearance shown and several different orientations, the fifth is two-lobed with a ring around the waist.

Higher-order orbitals have even more complex rings but the vast majority of the electron density is in the large lobes, so we are justified in considering only the large lobes.

Just as p orbitals occupy the same space as the s orbitals, d orbitals occupy the same space as both. Furthermore, d orbitals can't exist until there is a complete set of p orbitals. Trying to show the complete appearance of all this would be hopelessly complex.

The representation at left attempts to show the complete set of d orbitals. The colored dots represent the outer parts of the lobes. Large brightly colored dots are on the front side of the sphere, pale smaller dots on the rear side.

The lobes have the symmetry of an Archimedean solid called a rhombicuboctahedron (left). The coloring at left matches the diagram above.

Below is a three-dimensional sketch of the d-orbitals.

The remaining orbitals, the f-orbitals, don't come into play until we get to the rare earth elements, and they will be described later. The sketch below shows the arrangement of the p orbitals (yellow) and the d orbitals

What Do Atoms Really "Look Like?"
What Atoms of Hydrogen Through Xenon Really "Look Like"
What Atoms of the Heavy Elements Really "Look Like"
Scale Drawings of Atoms and Orbitals: Hydrogen Through Krypton
Scale Drawings of Atoms and Orbitals: Rubidium Through Xenon
Scale Drawings of Atoms and Orbitals: Cesium Through Radon
Scale Drawings of Atoms and Orbitals: Francium Through Lawrencium
What the Atomic Structures of Some Simple Materials Really "Look Like"

Return to Mineralogy-Petrology Index
Return to Thin-Section Index
Return to Crystals and Light Index
Return to Crystal Structures Index
Return to Mineral Identification Tables
Return to Professor Dutch's Home Page

Created 20 September 2005, Last Update

Not an official UW-Green Bay Site