And by round I mean the protons a neutrons arrange themselves in a generally spherical shape. At least thats how I have always seen them depicted. Why don't the protrons and neutrons arrange themselves in cubes or tetrahedrons or some other geometric shape; or some random shape for that matter? Do they arrange themselves in random shapes and we just depict them spherical?

Stu

Without knowing much about it I'd guess efficiency is the answer. Maybe along the same lines as why bubbles are spherical.

Spherical shapes minimize the potential energy of the inter-nucleon interactions. Its basically the same reason that planets and other bodies dominated by the gravitational force are more or less spherical.

Atoms don't really have "shapes" per se. They are "fuzzy" in the sense that they are best represented by a cloud of electron probability density. Since we can't know the precise location of electrons in the orbitals, we use this probability cloud model. Google some images of electron orbitals to get a better sense of this.

Do you mean atoms, or how quarks form to make protons and neutrons? If you are talking about atoms, they are not really round.

http://en.wikipedia.org/wiki/Atomic_orbital

Because of gravity?

That would be my guess.

Boundaries are just for our piece of mind; I don't believe they exist in reality.

Youre saying its a design feature?

He's more saying it's a result of entropy.

gogogo OP

The Bohr model FTW. Atoms are you know, like solar systems 'n stuff, more like flat but with fluffy dots in them -- totally circling and such.

At the scale of the atom, the force of gravity is extremely weak. I don't think we even need to consider it.

Do atoms have definite shapes? For instance, is one atom of uranium congruent to another atom of uranium(assume same isotope).

Stu

The answers so far have been that the shape of the outermost orbital of electronics is symmetric so the shape of the atom as a whole is usually a fuzzy sphere.

If you're asking specifically about the nucleus, the stable nuclei are approximately spherical most of the time (because of the strong nuclear force, not because of gravity.) The less stable nuclei bounce and seethe and spend a lot of their time somewhat distorted. The ones that bounce and seethe and distort so much that they have some probability of breaking apart, we call radioactive. The phrase "raindrop model" sticks in my head from a long-ago nuclear physics class, much the same way that "billiard ball model" sticks in my head from high school chemistry.

The nucleus is round for the reasons described above (i.e. for the same reason planets are round, substituting nuclear forces for gravity).

The shape of an atom could be considered to be defined by its outer electron orbitals. These aren't all round. Go here and read the section titled "The shape of orbitals".

Well, this is certainly true. All orbitals with angular moment > 0 (i.e. everything other than s orbitals) aren't spherically symmetric. Considering the next simplest type of orbital, p orbitals have a favored axis that they are stretched out along in sort of a figure eight looking deal. However, in the absence of anything to break the symmetry, there isn't anything to favor one axis over any of the others, and so if you were to average over a collection of atoms I'd think they'd still look pretty round.

Also - and I'm talking out my ass here a little bit - for the larger atoms where the weirdo orbitals start coming into play, I'd think the overall size and shape of the atom would be dominated by the lower filled n states. Once you fill all of the orbitals of a particular angular momentum I'm pretty sure you must recover spherical symmetry, and so if those are responsible for the bulk of the size, then whatever deviations from roundness you have from last orbital seems like they would probably be relatively minor.

No.

Atoms don't have definite shapes or sizes. Bonds in particular will warp the shapes and sizes of atoms a lot. A hydrogen atom in H₂O isn't the same shape or size as a hydrogen atom in H₂ or in NH₃.

Well, the "shape" of the overall atom would probably correspond to its largest (outermost) orbitals. And those would be for its highest n state. So this reasoning is probably not correct.

The "weirdo" orbitals are of much higher energy, so the highest filled shell never contains them (only s and p orbitals). Palladium may be an exception. At any rate, the electron distribution doesn't always resemble a sphere.

Also, orbitals don't really represents a "shape." There are no "boundaries" to the orbitals, and the boundaries typically drawn are based on an arbitrary percentage (usually 90%) that an electron will be found within the region. Electrons can move outside of the region, sometimes so far outside of the region that they leave the atom and become part of a different atom ("tunneling"). I'm not sure how quantum weirdness factors in, but I'd bet atoms don't really have "shapes" per se, and that if they did they would represent fluctuating "blobs" more than spheres.

Right, but come on, this represents a shape. To say that the shape represents a rigid object like a block or something is obviously wrong, but one can easily talk about the shape of the distribution, which is what is pretty clearly meant there.

Also,

you're right on the second one, but I feel like this is more of an argument for why you'd expect them to be roughly round (higher l states do tend to push out away from the center for centrifugal reasons, but I think higher n's push out too.) My point with the nesting and large n is that even if you did have some of the weirder orbital shapes, because there's a substantial spherically symmetric core, the proportional deviations from roundness are probably smaller than you'd expect looking at just the shape of the valence orbitals. EDIT: In other words, it would be the difference between a cigar and a ball with some bumps on it.

Agreed on all counts. I should have said they don't have a definite shape, not that they don't have a shape. "Ball with bumps" is how I'd visualize it. But in terms of the OP, I would consider that more of a "they arrange themselves in random shapes and we just depict them spherical." It's a bit semantic, but I think it's important to note that this is more complex than gravity (partly because tiny particles are weird, partly because there are repulsive forces involved, partly because electrons need to be in motion/can't be confined to a specific position).

Particles take the path of least resistance. A sphere is--for electrons at least--the shape in which they will usually be found when in their least energetic state. While I hold no advanced degree in chemistry, I have heard that there is evidence that suggests the presence orbitals like this within the nucleus as well.

But anyway, atoms usually have so many orbitals around them that they're never round. Hydrogen, yeah, but that can't hold any more electrons. In chemistry textbooks, you'll find that p and s orbitals for example are very different looking. My organic prof had beef with this, as one poster has already spoken about...they don't look very different IRL. I would imagine that an atom always strives to be as spherical as possible, given that a sphere is the easiest shape to be in.

It seems like over a time average, with a large number of electrons, the atom must be quite close to spherical, just to minimize potential energy, no? Doing anything different would seem inefficient, and I just don't know what forces would create such a lack of time-averaged spherical symmetry.

The problem is that electrons are attracted to the protons, but they are also repelled by one another. For a single electron, and even for a pair of them, this isn't all that complicated. But when you have more electrons involved, these attractions and repulsions start having patterns.

The patterns can be described using a nasty bit of math called a "wave function," and they can be experimentally observed (to a degree). After looking at atoms and crunching the numbers, we see that the electrons tend to exist in specific regions most of which are not spherical, because that minimizes the potential energy better than simple spherical orbits. The way it works is related to stuff like the wave nature of the electron, and it gets funky, but it all adds up to specific types of repeating orbitals.

The higher orbitals usually look like hourglass shapes.

Atoms are generally round in shape for practical purposes. The whole field of crystallography is based on this model and its predictions are very accurate (i.e., estimate the theoretical density of a material using the crystal structure and the atoms touching as hard spheres and it matches the actual density). Look at a plane of atoms using atomic force microscopy and you can "see" the atoms which resemble the hard sphere model.

The orbits of electrons must satisfy the Schrodenger equation. The solutions to the schrodenger equation with a central electric potential are products of angular distributions and radial distributions. The angular components form the shape in discussion. The solutions to the angular part of the three dimensional Schrodenger equation are called spherical harmonics and can take many shapes.

The potential that binds the electron to the nucleus is spherically symmetric. In other words, it pulls equally in all directions without any preferred direction. This is why spherical harmonics come into play.

Spherical harmonics aren't always spherical. They have many shapes each with different values of angular momentum. The "S" states are spherically symmetric. "P" states have total angular momentum 1 and are not spheres, rather they look like weird bulges coming out of the proton.

To agree with what has been said earlier, the electron orbitals are wavefunctions. They don't describe the shape of the electron's orbit, but rather they describe the probability distribution of the electron. They are defined over all space, but have maxima at certain radaii and angular distributions. It is the contours of equal probability which we describe as having a particular shape.