123

There are plenty of examples of people doubting ([1], [2], [3], [4], [5]) or denying ([6], [7], [8]) the existence of atoms.

Is there any compelling evidence for the existence of atoms? Has anyone ever actually seen one? Are there any pictures of atoms?

I am skeptical that atoms are anything more than a useful mathematical model.

Brythan
  • 10,162
  • 5
  • 46
  • 53
nima
  • 995
  • 2
  • 7
  • 9
  • 42
    The atomic theory of matter is one of the best validated ideas in modern science but the ability to "see" atoms is not what convinced people it was a good hypothesis (and seeing atoms is very recent technology). That worked was the way the atomic theory explained things such as many observations in chemistry such as the ideal gas law and the nature of chemical reactions. – matt_black Feb 02 '12 at 22:21
  • 9
    I think the "existence" in case of physical (and other) non obvious phenomenons is not a well defined term and the question is not answerable. Does thought exist? Does light exist? Do gamma rays exist? Do muons exist? Does anti-matter exist? There is plenty of concepts which are useful and most people would agree they "exist", but their existance is not something one could "see". – Suma Feb 03 '12 at 12:27
  • 5
    We have addressed questions skeptical of many other obvious matters. The fact that it's basic physics is not a good reason for closure. See also questions on memory of water, age of the universe and moon trips. – Sklivvz Feb 04 '12 at 00:36
  • 5
    This question has been through significant revision, and has been gathering close votes along the way, even though I personally think it has become a good question now. I assume the reason is people don't think anyone disbelieves in atoms. I have added some links (credit to Sklivvz for finding them) to overcome this. If there is another reason for the close votes, please comment. – Oddthinking Feb 04 '12 at 01:43
  • 3
    The links do not show real doubt about the atomic theory, they are concerned with the "real, or a model" question. More [in the meta](http://meta.skeptics.stackexchange.com/a/1394/79). – Suma Feb 04 '12 at 04:46
  • @Suma that is exactly what the OP is skeptical about (look at previous revisions - and read between the lines) – Sklivvz Feb 04 '12 at 10:55
  • I don't think questions being skeptical of well established facts are on topic. –  Feb 04 '12 at 15:12
  • 3
    @JohnSteeley *basically all our questions* are skeptical of well established facts. In fact, we answer them with references to articles published in well known scientific magazines... :-) – Sklivvz Feb 04 '12 at 23:34
  • 5k views in 2 days? – TheTechGuy Feb 05 '12 at 13:22
  • 1
    @Believer: thanks to revisions by Sklivvz and others, in fact my original question and title deserve negative feedback since my grammar is so bad and question is not clear, but it gets better and closer to my point by each revision. 8) – nima Feb 05 '12 at 21:10
  • Was it asked three days ago as it says or it is actually an older question? – TheTechGuy Feb 06 '12 at 00:15
  • 1
    @Believer it was asked 3 days ago, went on top of the supercollider and got tweeted by Jeff Atwood. – Sklivvz Feb 06 '12 at 07:53
  • 14
    I'm surprised that some people really doubt the existence of atoms. Reminds me on the one person which told me that it can't be true that all things are made from atoms, because you can make bombs out of them and therefore everything would then be explosive ... – Martin Scharrer Feb 23 '12 at 15:39
  • If wavelength is an issue why not use shorter wavelengths? –  Jun 10 '12 at 10:46
  • @JimThio: Because you can't lens shorter wavelengths, you can only diffract them. – Ron Maimon Oct 22 '12 at 18:59
  • Is it possible to add the direct visibility of alpha-particles hitting a scintillation screen? This is one of the earliest cases where a person can see individual atoms directly. – Ron Maimon Oct 22 '12 at 19:00
  • The answer depends on the framework you are working in (physics vs quotidian vs linguistics vs philosophy). –  Oct 24 '17 at 20:41

5 Answers5

231

Atoms are too small to see with microscopes that rely on light (also due to problems such as the Abbe diffraction limit). Essentially, with a conventional microscope you cannot distinguish points that are too close.

Even the recently introduced superresolution microscopes, which can surpass this limit, are still not able to image single atoms (they can, however distinguish single molecules, with a distance of <20nm).

This is mostly due to the fact that these microscopes rely on light, which has a wavelength that is too long to "catch atoms" (pardon the non-scientific wording of this sentence).

However, several other techniques exist that allow us to "see" atoms. I would like to stress that the very nature of atoms prevents us to actually have a photo of them like we take a photo of a macroscopic object.

Various techniques exist:

High resolution electron microscopy (HRTEM), which uses the phase-shift of a wave of electrons thrown at the sample.

This is an example of an indium nanoparticle imaged with HRTEM:

Indium nanoparticle imaged with HRTEM (From: "Current Research", Prof. Dr. Nicola Pinna )

These are HRTEM images of a superconductor:

superconductor HRTEM (From: "Structural order and disorder in Co-based layered cuprates CoSr2(Y,Ce)sCu2O5+2s (s = 1-3)", National Institute of Material Science )

Another technique is X-ray crystallography. I am not going to explain the mathematics and physics beyond it, as they are quite complex but essentially you take a crystal of the molecule you like, shine an X-ray beam on it and look at how the X-rays are scattered by the atoms in the crystal.

The diffraction pattern does not per se show the single atoms, but mathematical analysis of the diffraction spots allows for reconstruction of an electron density map and subsequently of the structure of the molecule.

Probably one of the most famous X-ray diffraction patterns is Photo 51 made by Rosalind Franklin of DNA, which was then used by the Nobel prize winners James D. Watson and Francis Crick.

Photo 51 - X-ray diffraction of DNA

See also some published papers at this regard.

Finally another technique is the Atomic force microscope. The principle is to have a very small cantilever that can pass over the sample and gets pushed up and down due to interaction with the electron fields of atoms. An enhanced version of AFM was developed in 2009, which allows to see atoms and bonds in a molecule.

The Chemical Structure of a Molecule Resolved by Atomic Force Microscopy - Gross et al. - Science 2009

SFM of a pentacene molecule

Pentacene molecule. Top: ball and stick model, middle: classical SFM, bottom: enhanced SFM. Scale bar 5 angstrom.

Update: Scientests have now taken a photograph of a single ytterbium atom, showing the shadow created by a laser hitting it.

Ytterbium Atom

Glorfindel
  • 1,452
  • 1
  • 17
  • 28
nico
  • 6,300
  • 1
  • 40
  • 34
  • 3
    As a humble offering in addition to this fantastic answer, may I mention the [Geiger-Marsden experiment](http://en.wikipedia.org/wiki/Geiger%E2%80%93Marsden_experiment). – DJClayworth Feb 02 '12 at 17:01
  • @DJClayworth: good addition. There are actually plenty of other methods that indirectly give us "photos" of atoms. For instance [NMR](http://en.wikipedia.org/wiki/Nuclear_magnetic_resonance) or [RAMAN spectroscopy](http://en.wikipedia.org/wiki/Raman_spectroscopy), although probably SFM is the most "convincing" one from a visual and conceptual point of view. – nico Feb 02 '12 at 17:15
  • Beautiful answer! – Rotemmiz Feb 02 '12 at 20:56
  • 9
    I created an account just to upvote this post :P – wim Feb 03 '12 at 00:50
  • @wim: wow, I did not expect such a big return :) – nico Feb 03 '12 at 07:29
  • 9
    Young people have it easy these days. All this fancy modern technology that gets pretty close to *visualising* atoms and molecules makes things far too simple. Think what it was like for the people who derived the original atomic theory who managed to get a powerful theory without any of this. Or, more recently, Mendeleev who created an accurate periodic classification of the elements before modern science had worked out the structure of the atom. The last major component of the atom, the neutron, wasn't discovered until the 1930s. Yet chemistry books still had atoms! – matt_black Feb 05 '12 at 22:18
  • @matt_black: I don't think I will ever put X-ray cristallography in my list of "easy things to do". Definitely not. :) Plus neither TEM nor cristallography are really "modern technology", they were both invented at the beginning of last century! – nico Feb 05 '12 at 23:01
  • 8
    Of course, none of these things 'prove' that atoms are anything other than a 'useful mathematical model'. But then you have to start delving deep into epistemology and the nature of the relationship between our knowledge of the world and the reality of it. For all practical purposes, our mathematical models in which atoms play an important role have such a powerful predictive value that it's kind of ridiculous philosophical hair-splitting to debate whether or not they exist. – Omnifarious Feb 08 '12 at 17:33
  • @Omnifarious: I think these experiments prove that matter is made up of atoms. What they do not prove (because that is probably not true), is that our current model of atom is *correct*. Our model of atom is a model, hence it is a simplified representation of reality. With this in mind we can think of it as a useful mathematical tool that correctly predict what happens in real life the great majority of the times. :) – nico Feb 08 '12 at 18:34
  • 1
    @nico - No, it proves that if you interact with matter in certain ways it behaves as if it's made up of atoms. I could, for example, devise a series of experiments 'proving' that light is made out of particles. I suppose you could say that I have incomplete model of light then, but that I've still 'proven' that light is made of particles. – Omnifarious Feb 08 '12 at 19:32
  • @Omnifarious: but then you would be (partially) right. Light is made of particles. And it is a wave at the same time. So your experiment (which has been made a century ago) does prove indeed that light is made of particles. That does not exclude that its nature is more complex, don't you think? But that's probably more of a question for Philosophy SE :D – nico Feb 08 '12 at 20:00
  • 1
    @nico - Yes. As I said, hair-splitting. Also, how do we know that atoms aren't just features of a simulation. And if so, do they still 'exist'? Again, philosophy question. It's just that I think a lot of people have a hard time making the leap from 'this mathematical model says these things exist' to 'these exist'. Heck, the neutrino and the quark are modern day examples of actual working physicists having this issue. Of course, as the model required to describe them became more and more integral physicists decided they 'existed'. – Omnifarious Feb 08 '12 at 23:54
  • @nico: Raman spectroscopy senses *molecular* vibrations (or phonons in a crystal - as a chemist, both are fundamentally different from "an atom" to me as they involve multiple atoms). You won't get a Raman scattered signal from a single atom. But talking inelastic scattering, of course you may get Compton scattering from a single atom (or rather, one of its electrons). – cbeleites unhappy with SX Oct 25 '17 at 15:03
  • 1
    Might want to update the answer with a "cleared" photography: https://petapixel.com/2018/02/12/picture-single-atom-wins-science-photo-contest/ – LordHieros Mar 21 '18 at 09:37
104

No discussion of atoms would be complete without mentioning Brownian motion.

Brownian motion is the strange, seemingly random movement of small particles (like dust) when suspended in a fluid. This phenomenon went unexplained for almost 80 years, until Einstein published a paper in 1905 (the same year he published his paper on the special theory of relativity) showing that, if we assume atoms exist and that Brownian motion is due to atoms bombarding the particle, the movement will behave in a very particular way. He even gave equations for estimating the size of the atoms!

At first the paper was ignored, as it didn't agree with the observed particle movement; however, it turns out earlier experiments were flawed, and through a series of very careful experiments in 1908-09, Jean Perrin verified that the particles did indeed move in exactly the way Einstein predicted. For this work, Jean Perrin won the Nobel Prize.

This is usually considered the turning point of atomic theory - before this work, the scientific world was divided about the existence of atoms, but after this their existence became very hard to deny.

BlueRaja - Danny Pflughoeft
  • 8,567
  • 3
  • 29
  • 44
  • 4
    A very interesting blog post about Perrin points out that he demonstrated the behavior of soap films only makes sense if molecules exist. Key phrase "Similarly, Perrin demonstrated that the difference in soap film thicknesses was always an *integer* multiple of the size of the soap molecule. " http://skullsinthestars.com/2012/05/16/the-secret-molecular-life-of-soap-bubbles-1913/ (emphasis my own) – JayC May 23 '12 at 23:28
61

Yes atoms undoubtedly exist.

I also had the doubt a few years back - are they real-real or just abstractions? - until the following, interesting and very convincing peer-reviewed literature on the topic came out!

D.M. Eigler, E.K. Schweizer. Positioning single atoms with a scanning tunneling microscope. Nature 344, 524-526 (1990).

Here we report the use of the STM at low temperatures (4 K) to position individual xenon atoms on a single-crystal nickel surface with atomic precision. This capacity has allowed us to fabricate rudimentary structures of our own design, atom by atom. The processes we describe are in principle applicable to molecules also. In view of the device-like characteristics reported for single atoms on surfaces, the possibilities for perhaps the ultimate in device miniaturization are evident.

The following image is given. Yes, it's a digital image because that's how scanning-tunnelling microscopes work. All the data represented is real (just like a digital picture).

Atoms writing IBM

After the landmark IBM paper of 1990 many other groups have reproduced the results. IBM Research maintains a whole site dedicated to STM pictures here:

http://www.almaden.ibm.com/vis/stm/gallery.html

http://www.almaden.ibm.com/vis/stm/gallery.html (NOTE: wayback machine)

Furthermore, a single strontium atom has been photographed using an ordinary camera by a group at Oxford university, here's the picture -- click to zoom.

Photo credit - David Nadlinger

Photo credit - David Nadlinger

Sklivvz
  • 78,578
  • 29
  • 321
  • 428
15

Would an actual picture of a single captured atom convince you?

enter image description here

Sklivvz
  • 78,578
  • 29
  • 321
  • 428
Snowhare
  • 259
  • 1
  • 3
13

IBM had a great project that provided pictures and even a video of atoms. The video is called "A Boy and His Atom." It holds the Guinness World Records record for the World's Smallest Stop-Motion Film, which isn't the most prestigious and convincing award, but it's something. You can see how it is made in the video, "Moving Atoms: Making The World's Smallest Movie."

The movie and images were made with a Scanning tunneling microscope (STM), which earned it's inventors, Gerd Binning and Heinrich Rohrer, the Nobel Prize in Physics in 1986.

Here's a quote explaining the technical abilities of the STM:

C. Bai (2000). Scanning Tunneling Microscopy and Its Application. New York: Springer Verlag. pp. 2-3.

STM can achieve atomic-level resolution. The lateral and vertical resolutions can reach 0.1 nm and 0.01 nm, respectively, i.e., individual atoms and molecules can be resolved.

STM can be employed for the modification of a surface and for the manipulation of atoms and molecules through tip-sample interactions, opening up the prospects of constructing atomic- or melecular-scale devices.

The project's homepage has much more information about the project, including some Star Trek atom art:

Star Trek logo http://www.research.ibm.com/software/IBMResearch/images/atomicmemory/star-trek-logo_image.jpg

live long

USS Enterprise

Glorfindel
  • 1,452
  • 1
  • 17
  • 28
Austin Moore
  • 231
  • 2
  • 2