Assume a cubic diamond crystal with hydrogen-terminated surface bonds. The terminating hydrogen atoms will form pairs that are geometrically close together in alternating diagonal pairs (think of the rectangles on corrugated walking steel).

Next, picture a formaldehyde molecule being pushed oxygen-first into one of these terminating hydrogen pairs. If the oxygen atom can be made to react with the hydrogen pair, a water molecule would be expelled and the remaining $\ce{CH2}$ group would presumably bond to the two bonds of the diamond lattice.

If you repeat this process over the entire cubic face, the result will be a new layer of diamond that is terminated with the same paired hydrogen atoms as the original layer. The entire layer-growth process thus in principle could be repeated indefinitely, resulting in a diamond grown at or near room temperature from an aqueous or other low-temperature solution.

Question: Can anyone think of a fundamental reason why the above process, or some variant of it, could not be used to grow diamond crystals at room temperature? (A "variant" might start with carbohydrate units larger than $\ce{OCH2}$, for example.)

One or more geometrically sophisticated catalyst molecules would be required. Each catalyst molecule would snag, carry, and orient a formaldehyde molecules onto some part of the textured surfaces provided by the cubic diamond face, such as the face itself or along a growth edge.

I did this a long time ago in high school, but my recollection is that the reaction should be mildly exothermic. But even if it is not, energy-source molecules (think of the role of ATP in living systems) could provide energy to enable an endothermic reaction.

So... thoughts, anyone? Is there any deep reason why this approach to room-temperature or near-room-temperature diamond synthesis is clearly not possible?

  • 2
    $\begingroup$ I'm not saying this is impossible, but this requires to rehybridise the formaldehyde carbon from an sp2 to an sp3, and there's no way I can imagine this taking place in a geometrically sensible way in aqueous media under STP. The formaldehyde carbon does have a free p orbital, but still, the intermediate would have to be twisted, and sufficiently stabilized by water molecules to live long enough for the reaction to complete. I also predict an entropic handicap to this reaction. The only method I know of that can produce diamond at (lower than) 101,3kPa is CVD. Interesting question. $\endgroup$
    – CHM
    May 23 '12 at 3:55
  • $\begingroup$ CHM, thanks on both comments. Terminating H atoms would be by construction, e.g. via a thin atomic hydrogen plasma against a cubic face might do it. Also, considerable generalization both of the liquid medium and of STP are permissible, since the underlying constraint is allotrope selection via "conveyance" of the allotrope-distinguishing surface geometry presented by the cubic face (only). So, if a catalyst can both (a) survive and (b) orient the surface reaction with a sufficiently high probability, permissible temperatures could go up substantially. STP as a start supports high orientation. $\endgroup$ May 24 '12 at 2:31
  • $\begingroup$ I would like to say that this is an interesting idea. I don't really think we can say it is impossible, but I don't think anyone right now can rationally design a system to do this. It would be found by dumb luck. Regarding the C-H terminated surface. These surfaces can certainly have dangling bonds, especially when we're talking about <111>, because something is necessary to stabilize the structure. $\endgroup$
    – Chris
    May 24 '12 at 17:35
  • $\begingroup$ @CHM, to be precise it would be a very large hydrocarbon molecule that's a more than a tad heavy on the carbon... :) The current plasma based synthesis methods terminate the surface bonds with something -- I've always assumed hydrogen -- so that part should be fairly easy. I think the harder part is the other issue you mentioned, which is whether simply aligning molecules would really force reactions. I'm going by the very rough (and old) biological enzyme rule of thumb that if you align molecules into roughly the shape of whatever intermediate form is needed, it tends to make them react. $\endgroup$ May 25 '12 at 0:29
  • 1
    $\begingroup$ I see no reason that a properly designed protein could not catalyze such a reaction. The problem is designing such a protein. $\endgroup$
    – Dale
    Sep 13 '12 at 13:26

The reaction you propose would be formally equivalent to the protection of aldehydes with 1,3-propanedithiol to form the correspondent dithiane. There are many organic reactions in which aldehydes participate this way, but all of them have something in common: the other species(lets call it donor) must have a free pair of electrons to attack the carbon atom in the aldehyde. In case this pair is not present, but the donor has an acidic hydrogen, we could use a strong base to remove it and form in situ the pair of electrons. (See aldol reactions)

Here's the problem: diamonds have no free pairs of electrons. And terminal hydrogens in a diamond lattice are by no means acidic, hence can't be removed by any base to the best of my knowledge. So there's no way a diamond will react with aldehydes.

  • $\begingroup$ Thanks! And yes, that oxygen isn't going to react easily. So, I think my question amounts to whether close, careful alignment of the oxygen near the pair might make possible an otherwise vanishingly unlikely reaction mode. To be frank I doubt it too, but I don't know if modern quantum models of that sort of thing have advanced to the point where they can gives more definitive answers to such oddball questions. (More thoughts welcome, but I'll likely give you the answer nod in a couple of days either way.) $\endgroup$ Sep 14 '12 at 3:05
  • $\begingroup$ You also have the problem that diamond is not the stable form of carbon at normal temperature and pressure. So it is very likely that energetics are not favorable for the formation of diamond this way. $\endgroup$ Jan 17 '13 at 1:18
  • 1
    $\begingroup$ That point I'm pretty sure is not a problem, since the energy difference between diamond and graphite is very small, quite trivial in comparison to some of the energetically non-optimal configurations that enzymes produce all the time. $\endgroup$ Jan 18 '13 at 2:08

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.