Also, is there really enough hydrogen ionizing at 400C?
The article says it is compressed in order to move it back to the high pressure side.
> Also, is there really enough hydrogen ionizing at 400C?
I don't believe (traditional) ionization is occurring. My guess is that it works at the molecular level like reverse osmosis works with salt water to form fresh water. http://en.wikipedia.org/wiki/Reverse_osmosis The dissociation of the hydrogen atoms into protons and electrons is through heat and pressure, facilitated by the semi-permeable membrane.
Note that the membrane is key and is very unique. Johnson [...] proved he could make a ceramic membrane capable of withstanding temperatures above 400 degrees Celsius. A semi-permeable plastic membrane is non-trivial. A semi-permeable ceramic membrane that passes protons, strips electrons, and blocks hydrogen atoms while running at 400^C has to be really, really tough to do.
The electrons and protons recombine on the cold side to reform hydrogen. That hydrogen has to be compressed sufficiently to move it back to the "hot" side of the system. This is probably a pretty substantial compression and will probably limit the efficiency of the system - it also raises the "perpetual motion machine" red flag in my mind, except the sun is adding lots of thermal energy to make it go for real.
I'm picturing the system working at a molecular level like reverse osmosis of seawater: http://en.wikipedia.org/wiki/Reverse_osmosis. Reverse osmosis requires a substantial amount of pressure (~1000 psi) to force the water through the semi-permeable membrane while the salt cannot pass through the membrane and thus stays on the salty side.
He isn't ionizing the gas, he is increasing the gas pressure by heating it. "Only instead of using those pressure gradients to move an axle or a wheel, he’s forcing ions through a membrane." See next for why that is important.
> Why don't the whole hydrogen atoms go through the membrane to the lower pressure area?
He uses a semi-permeable membrane http://en.wikipedia.org/wiki/Semipermeable_membrane. They also have use in chemical applications such as batteries and fuel cells.
The membrane blocks the atoms, but allows the ions (protons) through. It does require a substantial amount of pressure for it to work. That is where the heat -> pressure comes into effect, and why the reconstituted hydrogen on the "cold" size has to be compressed and pumped back into the "hot" side.
While this sounds a little like a perpetual motion machine, it isn't because the sun's heat is adding the energy that is being extracted via the electrons.
> How does he alternate which side is heated? How does he cool the other side?
He heats one side and cools one side, no alternating. The heating (and cooling) cause a pressure differential, which drives the reaction via the semi-permeable membrane.
---
I suspect there are some pretty substantial challenges remaining before this becomes a sizable / viable energy source.
It sounds like it is related to proton exchange fuel cells, except instead of using the catalyst + oxidant to drive the reaction, it is using pressure.
Refs:
http://en.wikipedia.org/wiki/Fuel_cell#Proton_exchange_fuel_...
http://auto.howstuffworks.com/fuel-efficiency/alternative-fu...