The Primordial Haze?

by Nathaniel Virgo

Here’s an interesting fact: apparently, chemical self-reproduction is easier to achieve in gases than in liquids.  This leads me to an interesting idea: maybe the very first steps in the origins of life took place not in the oceans  but in the atmosphere.  The mechanisms by which molecules can produce more of themselves are interesting, and in this post I’ll explore a bit about how such molecular reproduction (or, to use the technical term, autocatalysis) works.

I’ve been reading this classic paper by G.A.M. King called “Autocatalysis” (Chemical Society Reviews, 1978, unfortunately not freely available). He makes the interesting point that for chemical reactions happening among gas particles it’s very difficult to have a reaction where two molecules collide and stick together, producing a single molecule of product.  This is because in order for the two molecules to stick together, some kinetic energy has to be lost, and unless the molecules are quite big and floppy (i.e. they have many internal degrees of freedom), there’s nowhere for that energy to go.  Consequently, he says, most gas-phase reactions consist of two particles colliding and exchanging something, so that the product consists of two different molecules, each moving away from the other in such a way that energy is conserved.  (For reactions happening in solution, the energy can go into the surrounding molecules of solvent, which is why this is a particular feature of gas chemistry.)

This has an interesting consequence. It means that in gas chemistry it’s quite easy to get a self-reproducing set of chemical compounds.  These self-reproducing sets of molecules are called “autocatalytic cycles”.  I’ll explain shortly how they work.

I’m going to break with tradition and represent chemical reactions like this:

The coloured circles represent chemical species (i.e. types of molecule). They’re not boron (B) and carbon (C) and so on, they’re just arbitrary letters filling in for chemical formulae like CH4 or H2O.  The grey box represents a chemical reaction, which in this case combines a molecule of A with a molecule of B to produce a molecule of C, so it’s an example of the type of reaction that King says you don’t really get in gases.

This type of reaction is OK, though, because it combines two molecules to produce two different molecules:

This type of thing is also OK:

Where the double arrow means that two molecules of C are produced rather than one.

The top reaction above can be written A + B → C, the second A + B → C + D, and the third A + B → 2C.  (By the way, you couldn’t have these reactions happening in the same system as each other, as this would violate the conservation of mass. The “A”, “B” and “C” stand for different molecules in the three reactions above.)

Now that we have a neat way to draw chemical reactions, let’s look at how catalysis works.  The most basic kind of catalytic reaction looks something like this:

In symbols this is written A + B → A + C.  The “A” species is required in order for the reaction to take place, but it’s also a product of the reaction, so it doesn’t get used up.  The idea behind catalysis is that without this reaction, B might turn into C very slowly, or not at all, so if you want to speed the process up you can add some A to the system. A good industrial catalyst also has the property that it can be filtered out again afterwards, so you can use it for the next batch. (But we don’t have to worry about that here.)

The above diagram represents a reaction where a molecule of A collides with a molecule of B and then transforms the B molecule into a C molecule, without the C molecule being affected. (Or maybe the A molecule got transformed into a C, but the B got transformed into another A. That can probably happen.)

However, most catalysis reactions don’t actually happen in a single step like this.  Here is a slightly more realistic example:

This diagram now represents two reactions, B + C → D + E and A + D → C, joined together into what I like to call a “chemical reaction network.”  This particular network contains a cycle, which turns out to be a defining feature of catalysis.

Let’s trace our way around the cycle, starting with the bottom reaction.  First, a molecule of A collides with a D molecule, and they stick together to make a molecule of C.  This C molecule then collides with a B molecule. When this happens it turns back into a D molecule, but it also releases a molecule of E. So after all this we’ve used up an A and a B molecule, and created one of E.  The D got turned into a C, but then it got turned back again, so overall it doesn’t get created or used up.  So we can think of this network as representing an overall reaction A + B → E, catalysed by D, i.e. A + B + D → E + D.  If you want you can think of the “C” molecule as just being a D molecule with an A molecule weakly stuck onto it, so that when it collides with a B, the B reacts with the A to form an E molecule, which then separates from the D molecule, returning it to its original state.

This sort of thing probably happens all the time in wet chemistry.  But this network contains a reaction of the type King says doesn’t happen much in gases, where two molecules stick together to make one molecule.  What happens if we try to solve this by changing the reaction A + D → C into A + D → 2C, so that two products are created instead of one?  Then we get this:

Let’s work our way around the cycle again.  First, A collides with D to make C as before, but this time two molecules of C are created rather than one.  Now each of these C molecules collides with a B molecule to create an E molecule and a D molecule.  So we’ve used up one A, two B’s and a D to create two E’s and two D’s. We’ve ended up with one more D than we started with.

In this system, D and C still act as a catalyst – if neither is initially present in the system then none of these reactions will happen.  But this catalyst catalyses not only the creation of E out of A and B, but also the creation of more of itself.  Like a living organism, it uses up “food” (A and B) to produce more of itself, along with a “waste product” (E).  This self-reproduction can take place as long as there’s an excess of food over waste (or more technically, if the chemical free energy of the food is higher than that of the waste).  The analogy to life is not a coincidence, because an organism’s metabolism is essentially a very large and complex chemical reaction network that has this property of self-reproduction or autocatalysis.

Now, finally, I can get to the point I wanted to make.  King says that, because of the non-sticking-together property of gas reactions, it’s much easier to get these kinds of autocatalytic loops in gas chemistry than in solution.  The example above should give you an intuitive idea of why.  This property of gas chemistry also makes it harder for other “parasitic” reactions to occur that destroy the autocatalyst before it has a chance to take its effect.

Now, in order for life to get going, you need some kind of autocatalytic network: life (or proto-life) has to be able to make more of itself before it can get around to doing anything else, like evolving.  Some people think this must have been a very simple network involving a comparatively complex type of molecule, like this single-step autocatalytic reaction:

This is known as the “information first” theory of the origin of life, in opposition to the “metabolism first” one.  (My diagram is probably an unfair caricature, but it gets across the point that even in the most extreme version of the information-first theory, we’re still talking about autocatalysis, even if it isn’t called “metabolism”.)  But to me this seems kind of naive.  It seems to me that if you have a big enough network of reactions, you’re bound to get some autocatalytic cycles, but they won’t necessarily be small or simple.  A complex autocatalytic system built up out of simple molecules seems much easier to create than a simple autocatalytic network made up out of complex molecules.  Life, of course, is a complex network made out of complex molecules, but I think the complexity of the network came before the complexity of the molecules.

In order for these autocatalytic reactions to go on to develop into life they have to survive being used up by side reactions.  King’s claim is that this can happen much more easily in the gas phase.  In the liquid phase, catalysts are often fairly complex (e.g. enzymes), but in the gas phase they can just be a collection of simple molecules.  So my idea is just that the “original” autocatalytic network, the one that produced the first complex organic molecules that eventually went on to form into living cells, might have been not in the open oceans, nor in hydrothermal vents, nor in shallow tidal pools, but in the early Earth’s atmosphere.  Such a network would require a source of power – this could have been sunlight (via photochemical reactions – without the ozone layer there would have been plenty of UV light available), or outgassing from tectonic activity, or both.  Lightning and meteorite impacts have been suggested as well, but I reckon those are pretty tiny power sources compared to the other two.

This isn’t such a crazy idea.  The Miller-Urey experiment showed that complex organic molecules can be produced from gas chemistry, and we know that the orange atmospheric haze on Saturn’s moon Titan is caused by photochemistry producing complex molecules.  I don’t know for sure but I suspect that many of these complex molecules are produced autocatalytically in each of these cases.  (I often wonder whether, on Titan, those complex atmospheric haze molecules precipitate onto the surface, wash into Titan’s methane lakes, undergo reactions where they split back up into smaller molecules, and then end up back in the atmosphere, closing the cycle. Such a process would take place very slowly due to the low temperature, but could nevertheless involve very interesting and complex chemistry – perhaps even complex enough to be called life, Jim, but not as we know it.)

I want to finish off by saying that, currently, I don’t know of a precise formal definition of autocatalysis.  It’s not just a case of tracing around a single cycle as above, because autocatalysis can also happen in a case like this

where you need to follow several cycles in order to see that the catalyst (A and C) produces more of itself.  I don’t know of a definition that can capture both this and the example above, but I’m working on developing one, with the intention of applying it to real chemical reaction networks to look for the autocatalytic sets. That data should be interesting.

Edit: It turns out that the phrase “primordial haze” has been used before to refer to this type of idea (though without the emphasis on autocatalysis) – see here and here for some interesting reading on the subject.

About these ads

7 Responses to “The Primordial Haze?”

  1. I think that the problem of defining autocatalysis lies, in part, in defining a molecular species – it’s not unlike the defining life problem.

    Often, a species is taken to be a minimum in energy, (maximum in the case of a transition state), but this of course, it is only a minimum or maximum when constraints are added i.e. when we only look at a particular projection of a larger state space. We do this because the state tends towards a singular determined point, and we wish for a richer description – we gain this from ignoring some aspect of the system. As we add more and more degrees of freedom to a model of reaction kinetics, more and more local extrema appear in some projection, meaning that the idea of stability, and of a molecular species becomes more and more problematic. It think that that we would like to exclude a lot of cases that might fit this description.

    So, my only point is that

    • Hmm, I hadn’t thought too much about that, but it’s undoubtedly a problem. My approach, at least so far, is to leave such judgements up to the chemists (not being one myself). I think it should be possible to formally define autocatalysis, but only given a set of species and reactions, though even that part is surprisingly tricky.

      (I think transition states are more like saddle points than maxima. Kind of like a mountain pass – you don’t go straight over the mountains, but instead take the route that minimises the height you have to climb. Except it’s all stochastic and they probably go higher than they have to a lot of the time.)

  2. Urg, I thought I wasn’t going to post that, but it happened anyway


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 483 other followers

%d bloggers like this: