How can tunneling be one-way? I was recently at a lecture given by Dr. Harry Gray, a biophysical chemist, where he talked about how proteins (specifically those involved in photosynthesis) are able to use various phenomena, like superexchange and tunneling, to move electrons through them, coupling redox reactions despite the extreme distances (~20 Å).  By using tunneling/superexchange, various redox proteins are able to move 300-2000 electrons/second over this distance, while the reverse rate is effectively zero.
I asked about why this barrier is one-way, but I don't have enough base knowledge to understand his reply; something about how the injected hole is chased through a series of several bonds (sigma and hydrogen)...  
In any case, I'm curious for a simpler case and explanation of this 'diode' effect about how electrons can be shuttled one way with ease but not the other
 A: The ratcheting in photosynthesis is due to energy loss at each step. If you go from a high energy to a lower energy state, where the difference exceeds kT by several multliples, you cannot go back, because the reverse process requires a thermodynamic conspiracy to take the energy out of many modes and transfer it back to one electron, and that's just not going to happen. The absorbed photon energy is about 20kT, so there is a lot of room for one-way processes.
The cases which are more mysterious are where you don't have an energetic photon supplying many kT's of energy--- this is the case during translation. If you preferentially bind a complementary anticodon to a codon during translation, you don't have an energy cost which is far enough above kT to guarantee fidelity of translation. So there are error correcting mechanisms, which cost energy, which repeat the binding several times, and use ATP at each step. This was predicted theoretically before it was observed by Hopfield, and is called Kinetic proofreading.
