We were debating whether a lab experiment can demonstrate abiogenesis, even in principle. In my opinion, no experiment would suffice, because such a demonstration cannot establish naturalness of a process. Without the latter, the experiment is pointless. This brought to my memory another topic; it is related through similar difficulties.
We are made of organic molecules. What is the most abundant organic molecule in the Universe? If you do not know the answer, do not try to guess. Asking around would not help you either. But I know someone who knows the answer.
The answer is unknown. As far as we can tell this molecule (or, more correctly, a group of closely related molecules) is everywhere across the Galaxy, and it captures 10% of all cosmic carbon (organic and inorganic). What is it, no one has a clue, 87 years after the first observation. There are more than 100 theories trying to explain the nature and the origin of the compound. None of these theories appear to work. It is the greatest open problem in astrophysical spectroscopy and an embarrassment. This stuff is not exotic, it is exceedingly common. We understand marginalia better than the garden variety. Many people have heard about molecular clouds that include numerous organic molecules, from ethanol to amino acids. That's chicken feed in comparison with this chemical. These molecular clouds are dense and, therefore, relatively rare and small, accounting for a tiny fraction of the cosmic carbon. This diffuse chemical is thinly but widely spread, and it wins big over the rest. Intriguingly, it is almost never seen near these dense molecular clouds; there is no evidence that it is produced where the complex organic molecules are produced in space. It is in a class of its own.
This mysterious organic molecule is called a DIB carrier. DIB stands for diffuse interstellar bands that are seen against young O and B reddened stars in any direction in our Galaxy, and in close by galaxies, like M31. There are about 200-300 of these bands. Not a single one of these has been identified. It is known that these bands come from isolated molecules rather than dust or grains and that these molecules contain C and H and occur in hydrogen clouds. They tend to be linked to C-producing stars spewing soot. That is about all that is known with certainty. About 50,000 organic compounds have been tested as a candidate species. None has matched these bands. People are at the end of their wits guessing what this molecule might be, going through all kinds of exotic options. Kroto, the co-discoverer of C60 fullerene, has discovered it in an effort to identify the DIB progenitor – he spent many years on this problem. The C60 did not match -- and so did every other fullerene, carbon nanotube and countless other clever suggestions. The only molecules that come any close are not, actually, molecules but positive ions of polycyclic aromatic hydrocarbons. There were popular as a candidate for about a decade and then began to fade out of fashion, as it appears that small ones do not quite match and the big ones should be inconsistently protonated, yielding too numerous bands. The molecules tend to ionize because these are bombarded by UV photons from the young stars, day in and day out, for thousands of years; some of these photons are sufficiently energetic to ionize anything.
The clouds are very diffuse, so the excess energy gained by the absorption of a photon cannot be dissipated by collisions. If the energy is not very rapidly converted to heat, the molecule breaks apart after so many excitation events. There is Darwinian selection occurring in interstellar space bathed by radiation from the young stars: the molecules are continuously broken apart and the fragments cross recombine until a molecule is formed that converts the UV photoexcitation to heat so rapidly that it does not fragment. It survives. You can blast this survivor for a million years, and it is as good as new. These molecules/ions gradually accumulate while everything else fragments more and more, and over the long stretches of times the survivor molecule becomes predominant. Or so says the theory that I like best (there are many more). The reason I like this one is that I provided the first experimental proof that polycyclic aromatic hydrocarbons do have positive ions that exhibit extremely rapid nonradiative relaxation when photoexcited. Furthermore, these lifetimes corresponded well with the observed band line widths in the DIBs. Alas, suggestive as it may be, the positions of these bands did not quite match. Yet I think the main idea is correct: it is the survival of the fittest large organic molecule in the interstellar space.
Naturally, reproducing this process in a lab is impossible, as we do not know what this selection process begins with and how the selection occurs over the millions of years. Guessing this molecule proved counterproductive. Even if tomorrow someone comes with a mixture of chemicals that matches some of the bands (that happened repeatedly), one still has to prove that these chemicals occur in the clouds and are the most common organic compound. But there is no way of proving that. I remember, there was a nutty guy who claimed that some spectral lines matched those of specially prepared bacterial spores. Suppose he is right (he was not). Now you have to explain just how is it possible that 10% of the cosmic carbon is in the form of bacterial spores floating in interstellar space near young stars.
I remembered the DIB progenitor problem for its uncanny resemblance to the problem of the origin of life. Both of these deal with the same issue: inaccessibility of the object of one’s study (in one case, in space, in another, in time) and demonstrating naturalness. Even if tomorrow I’d show that I can make life in a test tube that does not prove that this can happen in nature -- unless you demonstrate that the test tube conditions were probable. If tomorrow I find a chemical matching the bands, this does not prove that the DIB carrier is this chemical unless its generation in the interstellar space is probable. But how to prove that it is probable? In one case, you need a large number of young planets looking like the early Earth and survey them for the occurrence of such conditions over the millions of years. In another case, you need to observe the fate of a huge diffuse cloud as it evolves driven by radiation and arriving at the postulated survivor molecule. Neither approach is practicable. A lab experiment can provide something that looks like it might be an answer. But it cannot prove that it is the answer. For that you need field tests, but those are impossible to do.
At the peak of my DIB infatuation I had a dream: the devil himself appeared before me in all his devilish glory and offered to tell the identity of this molecule if I cut my right hand. Without missing a batt I darted to the kitchen, got a carving knife and started sawing it off, as he opened his mouth to tell me the Great Satanic Secret: "The most abundant organic molecule is..." --- but at that moment I woke up. I was THAT close to knowing it... I've never regained my resolve and stopped working on the problem after a year. I understood that no matter what I'll do in the lab, I would not be able to solve this problem. To the existing 5000 experimental and 500 theoretical papers I will add a few more, and that would be it. I did not want to write more papers; I wanted to know the answer.
I still do. What is the most common organic molecule in the Universe?
Further reading: Ted Snow's The DIBs