Monday, October 24, 2011

Fuel cells, Thermoelectric Generators, Solar Photovoltaics, and the Origin of Life

In a previous post on fuel cells, I mentioned that you are a fuel cell.  Meaning that you are made up of billions of fuel cells, i.e. mitochondria. You are not a piston engine, nor are you a steam turbine. What's fascinating is that, of the multitude of different types devices that humans have designed over the years to generate electricity and transportation (such as turbines, rockets, reciprocating engines, MHD, solar cells, fuel cells, thermoelectrics, and thermionics), biological creatures have only come up with two type of generators of work: fuel cells and photosynthesis. To my knowledge, there has not been a single biological creature that uses mechanical generators, like piston engines or turbines, and it appears that there have not been any biological creatures that use thermoelectric generators to provide power from a temperature gradient. Why haven't non-human life forms developed mechanical combustion based systems to generate work and power? (Examples of work and power are: movement against friction, moving chemicals against a chemical gradient, and moving mass against the force of gravity.)
Instead, biological life forms use a combination of fuel cells in combination with photovoltaic cells to generate work.
Why did biological creatures only developed a narrow range of different generators of electrical or mechanical work until humans evolved means of generating work from the wide range of devices we use today? Is a high temperature combustion device out of the question for a biological creature?

This means that learning about fuel cells can help us understand more about ourselves. But while I think that fuel cells are fascinating devices and will find certain niche applications in the short term, I realize that what's worked in the past is not always what will work the best for the future.  In particular, fuel cells are unlikely to lower the cost of getting into outer space, and once we start sending self-replicating solar robots to the Moon, these robots probably won't be fuel cell based. They will likely be either solar photovoltaic or solar thermoelectric based self-replicating robots. Why?

The robots will have to self-replicate using the materials available to them. Biological fuel cells, such as bacteria, are unlikely to grow on the Moon, because of the lack of carbon and nitrogen. And human-designed fuel cells (such as PEMFC's or SOFC's) require platinum or other hard to mine metals, such as Yttrium, as well as a fuel source. Instead, it seems possible to make solar panels, thermoelectric generators, and batteries using only the most abundant chemicals on the Moon, such as silicon, iron, aluminum, titanium, and sodium. The silica on the moon can be used to make silicon, which is used in solar PV's, in thermoionics, and in batteries. The iron, aluminum and titanium would comes from iron oxides, alumina, titanium dioxide, and would form the casing and structure for the solar panels and batteries. The sodium would be used as the ion carrier in the batteries. Also, the iron would be part of the electrode of the battery. Phosphate might be one of the difficult materials to find, so there might need to be some change in the battery chemistry for these self-replicating solar robots, i.e. solar auxons.

While I suspect that it will be possible to build self-replicating robots on the Moon, I'm more optimistic about self-replicating robots on Mars. In fact, it might even be possible to terraform Mars so that we can grow fuel cell-based lifeforms similar to the type of lifeforms that survive on Earth. Only time will tell.

So now I'd note like to go back to an important point raised in my previous post on "You are a fuel cell." I'd like to emphasize the fact that what allows fuel cells at room temperature to generate work is the fact that a fuel cell is an electrochemical device. A purely chemical process at room temperature can not generate non-mechanical work. A purely chemical process at room temperature could generate mechanical work, such as moving a piston, but we've already seen that biological creatures don't have pistons or turbines built into their bodies.

In that post, I mentioned that a purely chemical reaction at constant volume and maintained at the temperature of the Earth's environment can't do work unless some of steps in the overall chemical reaction are electro-chemical. The reason why is that the exergy destruction of a purely chemical reaction would be exactly equal to the exergy in the original fuel. All of the exergy would be destroyed, and hence, there would be no exergy available to generate work.

Why can electro-chemical processes generate work when the system is at constant volume and maintained at the temperature of the Earth's environment? The reason why appears to be that, while the temperature of the bulk of the fuel cell is maintained at room temperature, the Fermi temperature of the metal electrodes is not room temperature, i.e. if you convert the Fermi energy of the metal into an equivalent temperature, this temperature is on the order of 10,000 degrees C. The electrons in the metal are constrained to be a high energy levels because of the Pauli exclusion principle, this means that adding or removing an electron to a metal requires a large amount of energy without the transfer of a significant amount of entropy (i.e. permutation symmetries.) If the electrons in the metal were not constrained to sit at high energy levels, then it would not be possible to generate work from such a 'fuel cell.' The generation of work in a 'fuel cell' is due entirely to the fact that electrons in metals can carry energy without carrying significant amounts of entropy. This is the main 'trick' of a fuel cell, and the other main 'trick' of a fuel cell is that the electrolyte can not be allowed to conduct electrons. (Transferring the chemical exergy of the fuel...methane, hydrogen, ethanol, sugars, etc... into electrical exergy of the electrons. The electrons can only generate work if they can carry energy without carrying significant amount of entropy. And the reason that they don't carry with them significant amounts of entropy...for a given amount of that an electron added or removed from a metal is constrained by the Pauli exclusion principle to be added or removed from the top of the stack.)

Biological creatures have figured out this trick and are doing a great job of generating work in a self-propagating manner via the use of fuel cells, and photosynthesis. Biological fuel cells, such as inside mitochondria, require electrodes that take advantage of the Pauli exclusion principle. While it's clear that mitochondria have proton or alkali conducting membranes, it's a little harder to understand what are the electrodes inside of mitochondria. It appears that the electrodes in biological cells are proteins that are carbon-nitrogen-iron or carbon-nitrogen-cobalt based. It appears that protein crystal can achieve a Fermi temperature as high as 3 eV (i.e. ~33,000 deg C), so it is likely that proteins attached to a proton-conducting membrane could have electrons sitting a high enough energy levels, such that the electrons can carry large amounts of energy with only small amounts of entropy.

Note: the chemical reactions in a piston-engine or a rotating turbine can produce work because the reactions are constrained so as to produce a force in one direction or torque in one direction. The motion of a piston against a gravitation field carries with it energy with zero entropy because gravitational energy can be described by one variable: the height of the center of mass. The logarithm of 1 is zero; hence, mechanical work that changes the gravitational energy of an object carries with it zero entropy.

One major question remains still in my mind: can bacteria or other life forms generate power by taking advantage of the exergy associated with a temperature gradient? My guess is: perhaps...and here's my thinking.

The two ways of generating work/power from a temperature difference are: 1) some mechanical method, such as a steam cycle, in which you constrain the motion of the turbine. or 2) a thermoelectric generator that takes advantage of the fact that the Fermi temperature of the electrons or electron-holes is much greater than the temperature of the phonons in the solid (Note: a large Fermi temperature is required in order to allow energy transfer with little entropy transfer, i.e. work.) If the temperature of the electrons and electron holes were the same as the temperature of phonons in the solid, then there could be no work/power generation.

Why?   Because there would be no energy carrier that could transfer energy without transferring a significant amount of entropy. If electrons and electron holes in the thermoelectric generator were the same as the temperature of phonons in the solid, then the exergy destruction of heat flowing across a temperature difference would be equal to the amount of exergy in the original temperature difference, and hence there would be no work generation.

So, it's unclear whether there are any life forms on Earth that have developed thermoelectric generators, but what is clear is that, if they have, the only way to generate power from a temperature gradient is to have an electrode like species at the anode and cathode with a very large Fermi temperature, and an "electrolyte" that does not conduct electrons, such as an electron-hole conductor.

Whether life on Earth has developed such a n-p-n type of structure is still debatable, but what would be interesting is whether or not humans can design self-replicating solar robots on the Moon that are the equivalent of thermoelectric generators. Such creatures would use solar energy to heat one side of a n-p-n surface (while cooling the backside of the device) to generate electricity. The electricity would be stored in a large scale battery, and the electricity would be used to build more solar robots from the materials on the Moon (such as silicon, iron, etc...). It's not clear to me right now whether thermoelectric based self-replicating solar robots could generate a larger rate of growth (i.e. rate of return on work invested) than photovoltaic based self-replicating solar robots. So, I'll post any results once I calculate the estimated growth rates of either case.


  1. Hey its a great post. It is as interesting as your previous post. I must say it seems you made a detailed study before writing this post. Appreciate you a lot.

  2. Thank you for sharing valuable information. Nice post. I enjoyed reading this post.