Escapement mechanisms and the conversion of disequilibria; the engines of creation: "contrary to general belief, it is never the ‘‘use’’ or ‘‘consumption’’ of energy that powers life, nor indeed any other powered activity in nature"

Amazing paper. Abstract:

Virtually every interesting natural phenomenon, not least life itself, entails physical systems being forced to flow thermodynamically up-hill, away from equilibrium rather than towards it. This requires the action of a mechanism, acting as an “engine”, which lashes the up-hill process to a more powerful one proceeding in its spontaneous, down-hill direction; in this way converting one disequilibrium into another. All organized and dynamic elements of creation, from the galactic to the atomic, can be viewed as powered by, or being the result of, engines of disequilibria conversion; each a link in a great hierarchical cascade of conversions. There is, however, widespread misunderstanding about how disequilibria conversions happen–and indeed about what physically causes them to happen–especially regarding the role of energy and of the physical meaning of free energy. We attempt here to describe and justify what we assert is the correct alternative view of how phenomena are powered in nature, focusing especially on the molecular-level conversion processes (often called “energy conserving”) that power life and that must, then acting in an entirely abiotic context, have driven it first into being.


The essential point is that ATP is not simply synthesized (e.g. by ATPsynthase) and then consumed as a chemical reagent. It is instead, and as a categorical necessity must be, driven out of equilibrium with respect to its hydrolysis products ADP and Pi (orthophosphate); were that not so it would be useless to the cell as a carrier of free energy no matter what its concentration, since it is the disequilibrium itself, not the molecule ATP, that ‘‘carries’’ the free energy. In actual fact it is driven to a quite astronomical disequilibrium. The Gibb’s free energy3 of the ATP disequilibrium maintained in cells typically stands at between 20 and 24 fold above kBT [32] meaning simply that the ratio of the concentration of ATP to that of its hydrolysis products: [ATP]/[ADP][Pi], is of the order of 10^9–10^10 higher than it would be at equilibrium.

These magnitudes and their derivation suggest several general conclusions, all of which we later make some attempt to justify:

  1. In its physical content ‘‘free energy’’ is a somewhat misleadingly named measure of the extent of a disequilibrium (equivalently, and arguably more revealingly, a (log) measure of the ratio of a reaction’s forward to backward rates (in Section 12.2 we discuss this point, and seeming exceptions to it).4
  2. It is not the bond energy of the terminal phosphoanhydride bond of ATP whose liberation upon hydrolysis is what powers (by supplying energy to it, it is commonly, and incorrectly, supposed) the thermodynamically ‘‘up-hill’’ (endergonic) reactions to which the hydrolysis is coupled (much more on this point later); that bond energy is of course liberated whether the hydrolysis occurs at equilibrium or not. But more importantly, as we will see, the liberated bond energy is itself in no sense transferred to the driven reaction. Further, in all adequately explicated cases, (1) the free energy available in the hydrolysis of ATP is realized only after, and is contingent upon, the completion of the driven reaction, and (2) is necessarily dissipated, not ‘conserved’, since it is the dissipation itself that ‘drives’ the conversion.
  3. The magnitude of the ATP disequilibrium, and the incredible biochemical machinery and metabolic effort involved in its maintenance, is compelling testimony that life hangs by the thread of, and is in general ‘‘rate limited’’ by, an extreme state of dynamic disequilibrium – one which it perforce must strive quite ‘desperately’ to sustain; it is the disequilibrium itself, particularly its magnitude, that is essential – and that powers life.
  4. Specificity is critical. Chemically non-specific states of disequilibria are not simply useless to life, they are implacably incompatible with it. Living systems interconvert extremely specific pairs of disequilibria, and are at immense pains to achieve that specificity. Life never employs, nor could it, non-specific means (e.g. temperature or hydration excursions) of generating the disequilibria it requires.
  5. The disequilibria economies of life are inherently dynamic. That we turn over our ATP inventory in the order of a minute, and consume nearly our body’s weight of ATP per day, speaks forcefully to the point that it is not the static state of disequilibria that is the essential point, but the dynamics of its generation and dissipation—and that the race is to the swiftest.

However, implicit in life’s obligate deployment of disequilibria to power its activities lies a linked pair of deeper truths that are the main point of this piece. The first is that, contrary to general belief, it is never the ‘‘use’’ or ‘‘consumption’’ of energy that powers life, nor indeed any other powered activity in nature; it is, and in principle can only be, the dissipation of a disequilibrium. The second reflects the almost paradoxically opposed fact that virtually every relevant instance of something happening in nature, let alone in life, involves the creation of a disequilibrium, not just the dissipation of one; a feat, which if it happened by itself, would violate the inviolable 2nd law. The upshot is that the entire drama of the dynamic universe, life quintessentially, lies in activities that couple thermodynamically opposed processes in order to create one disequilibrium at the expense of dissipating another. How does this come about?


Also, hmmmmmm:

In all cases of which we are aware wherein the inference can be drawn, this requirement is met in the following somewhat counterintuitive manner: (1) the driven process fully completes before the driving one is permitted to do so, (2) the state in which the driven process is fully complete is what triggers, or ‘‘gates’’, the completion of the driving process, and (3) the state in which both processes are fully complete functions as the necessary pre-condition ‘trigger’ for the engine to return to its starting configuration. In such engine designs, the rule is, therefore, ‘pay upon completion’, whereby a cycle’s worth of disequilibrium creation must be completed before a cycle’s worth of the driving disequilibrium is allowed to be dissipated.

We pause here to note by way of illustration a particularly compelling example, that of the nitrogenase enzyme. In each of its conversion cycles this engine reduces one molecule of molecular nitrogen N2 to two molecules of ammonia NH3; an eight-electron reduction. It performs this by sequentially executing eight subcycles each one of which performs a one electron reduction. Impressively, each of these single-electron subcycles is driven by the quasi-coincident hydrolysis of 20). two molecules of ATP (providing an ‘irreversibility bias’ of ~10^-20. And, tellingly for our current concerns, these paired hydrolyzes occur after the individual electron reductions and are triggered by them, a point only recently established. The relevant paper, by Duval et al. [43], has the title ‘‘Electron transfer precedes the ATP hydrolysis during nitrogenase catalysis’’ where the term ‘‘precedes’’, carries the paper’s primary, highly counter-intuitive, and highly contentious, burden. Tellingly also, the finding leaves the authors compelled to wonder where the ‘‘energy’’, which, they believe, must be needed to force the electrons on to the nitrogens, could be coming from. A quest which, the present paper is largely devoted to establishing, is inherently misdirected.


First because the engines themselves are individual macromolecular complexes (protein or protein–RNA complexes) operating in an aqueous environment, viscous forces (kinetic ‘‘damping’’) completely overwhelm inertial ones (the extreme low Reynolds number limit); to the extent that there are no significant inertial effects and molecules never ‘coast‘. In consequence they are everywhere in mechanical equilibrium [24,44];

(I’m going to end highlights here bc I’m about to copy-paste the entire paper into this thread)

It’d be interesting to formalize the notion of an escapement in the general case such that one could detect them automatically given the dynamics of the system (e.g. MD simulation => list of escapement mechanisms).

Can you expect on this? I don’t think I understand what you’re trying to say.