How many different organic structures (from the pure theoretical viewpoint) can be drawed with only 4 (exact) carbon atoms and with/without hydrogen? Polycyclic compounds and bridged compounds are also allowed. Please, name them all as well! :)

Remark: I have drawed the chemical graphs in a piece of paper (the skeletons) and I get 37 graphs (including linear with simple, double and triple bonds, and planar and nonplanar compounds), without geometric isomerism (cis/trans). Is that OK?

  • 1
    $\begingroup$ I top out at 23. I'd be curious to see your list. $\endgroup$ – jerepierre Nov 13 '14 at 18:25
  • 1
    $\begingroup$ There is also cis/trans 2-butene in real world. In theoretical world you could draw cis/trans isomers for some cyclobutenes too $\endgroup$ – K_P Nov 13 '14 at 21:10
  • 1
    $\begingroup$ @Ron Because carbon can bond hydrogens, not just other carbons. $\endgroup$ – DavePhD Nov 14 '14 at 12:14
  • 1
    $\begingroup$ @riemannium I moved your answer to a comment up here. If it was directed at anyone in particular, you can copy and paste it as a comment to the answer that you were replying to. $\endgroup$ – jonsca Nov 14 '14 at 20:22
  • 1
    $\begingroup$ In general, this kind of question is a bit broad, but it's gotten some good answers, so I will leave it be. $\endgroup$ – jonsca Nov 14 '14 at 20:23

Final update, all earlier edits incorporated.

Groundrules: Considering compounds with:

  • only carbon and hydrogen
  • only 4 bonds to carbon

There are 37 isomers without considering trans isomers; 49 when trans isomers are included. Also, many of these compounds seem extremely unstable and therefore unlikely to exist.

Note to self: check back in 20 years and see how many of the unlikely ones have been detected.

enter image description here

Names by row:

  • butane, isobutane
  • but-1-ene, but-2-ene, 2-methylpropene
  • buta-1,3-diene, buta-1,2-diene, buta-1,2,3-triene
  • but-1-yne, but-2-yne, but-1-ene-3-yne, buta-1,3-diyne
  • methylcyclopropane, 2-methylcyclopropene, 1-methylcyclopropene, methylenecyclopropane, methylenecyclopropene, methlycyclopropadiene
  • methylcyclopropyne, methlenecyclopropyne
  • cyclobutane, cyclobutene, cyclobuta-1,2-diene, cyclobuta-1,3-diene, cyclobutatriene, cyclobutatetraene, cyclobutyne, cyclobutenyne, cyclobutadiyne
  • bicyclobutane, bicyclobut-1(3)-ene, bicyclobut-1(2)-ene, bicyclobuta-1,3-diene, bicyclobuta-1,2-diene
  • tetrahedrane, tetrahedrene, tetrahedradiene
  • $\begingroup$ Comparing with my skeletons, you lack a 1methyl-cyclopropene, and the cyclobutatriene with 3 double bonds (irrespectively it existed or not, you have not drawed it). Also, you have not drawed the C4 (the square with 4 double bonds). Thus, I get 36 (or 37, if I include the cis/trans versions of 2-butene, that I had not considered diferent so far). $\endgroup$ – riemannium Nov 14 '14 at 0:24
  • 1
    $\begingroup$ @ron You might want to include a note for posterity that some of these "compounds" are incredibly strained and likely unstable. (This was why I was including ?? marks in my answer.) $\endgroup$ – Geoff Hutchison Nov 14 '14 at 0:45
  • 1
    $\begingroup$ @GeoffHutchison There is a higher than normal barrier to rotation about the 2,3-bond in 1,3-butadiene due to a contribution from the resonance structure with a 2-3 double bond. However, the barrier is still low enough that the s-cis and s-trans forms are generally considered rotamers. $\endgroup$ – ron Nov 14 '14 at 0:58
  • 1
    $\begingroup$ Is it legitimate to refer to cyclobutatetraene and cyclobutadiyne as different compounds, or are they just different resonance structures? Both are just a ring of 4 carbon atoms and no hydrogen. $\endgroup$ – DavePhD Nov 14 '14 at 15:31
  • 1
    $\begingroup$ @PrittBalagopal It fits within the stated "ground rules". $\endgroup$ – ron Jun 3 '17 at 19:59

How many different organic structures (from the pure theoretical viewpoint) can be drawed with only 4 (exact) carbon atoms and with/without hydrogen?

We could make strict rules like each carbon has exactly 4 bonds and get a specific answer, but this is not reality. There can be lone pair electrons and unpaired electrons. The octet rule is not stictly followed.

$\ce {C_4}$ actually has been observed and is linear.


$\ce {C_4}$ has been the subject of numerous theoretical and experimetal papers because of its possible occurrence in nebulae. It has been debated whether a singlet linear, triplet linear or rhombic (kite shape) state is the lowest energy state. http://www.sciencedirect.com/science/article/pii/S0009261400005765

Neither the linear nor rhombic states follow the naive rules.

Linear $\ce{C_4H}$ has been observed both in the lab and outer space.

In fact according to the University of Kohn lists Molecules in Space, linear $\ce{C_4H}$ is one of only 61 molecules and molecular ions found in extragalactic space and of only 190 found in the interstellar medium or circumstellar shells as of 2016.

For $\ce{C_4H_2}$ linear butadiyne in known. Cyclobutatriene, cyclobutenyne and tetrahedrene have been ruled out theoretically as not represtenting any actual potential energy local minimum, while similar to $\ce {C_4}$, structures having lone pair or unpaired electrons and not following the octet rule (such as carbenes) have been calculated to represent actual minima. See the following references for theoretical cyclic $\ce{C_4H_2}$ structures:




In outer space, not only has the usual HCCCCH isomer been found but also $\ce{H2CCCC}$

See Observations of cumulene carbenes, H2CCCC and H2CCC, in TMC-1

  • $\begingroup$ Yes, I was wondering as well how many of these structures are in fact known from the experimental side... $\endgroup$ – riemannium Nov 15 '14 at 19:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.