Last time, I closed with a mention of thorium reactors for the necessary 24/7 "storage" baseload, in a final transition to a greener, PV-dominated energy supply. (The utility baseload is the constant power that runs 24/7, at the lowest point of power use, usually in the wee hours of the night). Some folks may find it strange that a solar roller would be considering nuclear power as the baseload for a PV dominated energy mix. So I thought I'd discuss the storage issue a bit.
Storage is the old bette noir of renewable energies, direct solar in particular. Wind, tidal, etc. have their intermittency issues also, but none with the regularity of solar (the wind and tides can still move in the darkness--photons are always bedded down for the night). The solar storage issue affects both the thermal and PV fields. Here I'll focus on PV.
A utility PV energy contribution to a utility power mix does not need storage until its Wp (Watts peak) grows to about 10% of the generating capacity of the utility. This is because the peak electrical power use in an industrialized society does not occur until around 2 PM standard time—that is, about 2 hours after solar noon. Hence the solar input curve vs. time is a reasonably close fit to the peak utility power load curve. But it is not exact, and the solar input does not help the utility baseload at night. Hence the 10% limit of the non-storage PV Wp to the utility mix.
Years ago I wrote several papers on "solar breeders" based on net energy analysis. I was interested in whether solar energy systems could "breed" themselves, by producing energy of sufficient quality and quantity to provide for their manufacture. A study of the energy spectrum of manufacture--the amount of energy at given temperature—revealed that a majority of industrial energy is below 250°C, basically process steam. However, only the electricity-producing solar energy systems produce the high quality energy that produce all the energy needed at very high temperatures that is embodied in their manufacture. Many solar thermal technologies--such as passive homes and flat plate collectors for heating cannot reproduce the energy spectrum needed in their manufacture--the glass and metal. These lower-temperature solar technologies are of course very efficient energy extenders, acting as conservation technologies.
I focused my papers on the PV solar area. The long report, where energy storage was discussed, was archived at the Iowa State University library:
D.P. Grimmer, "A Comparison of Solar Photovoltaic and Nuclear Fission Breeders as Self-Replicating Power-Generation Systems", IPRT Technical White Paper 89-2, Iowa State University Library, Dept. of Special Collections, University Archives Record Series 18/1/0/4.
I found that, at least for conventional uranium/plutonium fission breeder reactors, the long lead times necessary for the serial steps -- needed for safety-- would not allow them to be bred at as fast a rate as PV breeder systems, where construction could be done in a parallel unit process. As we have seen at Chernobyl and now Fukashima, an inherently unstable technology is err to all sorts of safety problems, both man-made and naturally occurring. This is not the case with PV, where construction and operation does not involve non-critical steps.
In my PV breeder analysis, after looking at batteries, metal hydride storage, etc., my best net energy conclusion was to use PV electrolysis of water, low pressure storage of H2 in large inflatable cylindrical structures, and fuel cells for H2 back to electricity.
Electrolyzers and fuel cells have a very high specific power density in kW/kg, and much of the technology for natural gas handling and storage is applicable to low pressure H2, where hydrogen embrittlement is not a problem.
You may be familiar with the very large, cylindrical natural gas storage units that are composed of a frame of steel posts, tied together at the top--kind of Stonehedge-looking. A hinged and sealed bladder composed of reinforcing horizontal steel plate strips with a rubberoid liner contains the gas. The bladder rides up and down on tracks inside the posts, depending on their fullness level. The weight of the structure on the bladder would apply more or less constant pressure to the gas supplied to the natural gas network.
The energy storage density of these units reaches an economy of scale. This is because the energy embodied in the fabrication of these units--the skin of the structure--goes as the square of the linear dimension, whereas the energy stored in the form of natural gas --or H2--goes as the cube of the linear dimension. The cube quickly overwhelms the square at a certain linear dimension to achieve an attractive kWh/kg specific-energy storage-density, certainly as compared to batteries, flywheels and the like.
(Note: this allows PV to function in an electrical grid 24/7. It does not address the problem of transportation energy. But clearly in this scenario, H2 energy could be tapped off and used in onboard H2-fuel cells; or the output electricity used in EVs (electric vehicles) or conventional plug-in hybrids).
Although there is always the irrational Hindenberg phobia with H2, modern analysis has shown that the Hindenberg catastrophe was caused by lightning igniting the aluminum impregnated fabric skin--similar to modern solid rocket propellant--not the H2. Also, pure H2 burns up straight up in the air rather than explodes: the small number of Hindenberg deaths were caused by people being crushed in the collapsing debris.
Right now, I would guess that folks in Fukashima would rather have had solar PV with mass H2 storage than uranium/plutonium nuclear.
But in the near term crunch, I say we should push for thorium nuclear for near term baseload. It remains to be seen whether the thorium technology is truly as safe as reported. On paper the thorium cycle looks promising, so much so that the Chinese government has committed themselves to its development—and retention of the intellectual property. The Indian government is also developing thorium reactors. The basic technology was developed decades ago at the US Oak Ridge National Lab. (Does this technology evaporation from the US to other countries sound familiar?)
Thorium nuclear energy produces no bomb fuel; is stable running (liquid salt cooled) with a natural criticality shutdown; little waste--actually eats waste in a breeder cycle. There is no CO2 produced for folks to get all bunched about that. And, very importantly, saves the fossil fuel stocks to produce plastic substrate for PV thin films. By my back of the envelope calculations, such PV film needs to be produced in the hundreds of GWp per year to meet the needed world grid capacity of 4000 GWeff (2006 statistics), or on the order of 24,000 GWp if totally PV with storage. A 24/7 thorium nuclear baseload will gradually and eventually be supplanted with greener storage technologies, like a PV/hydrogen/fuel cell breeder technology. Although far more plentiful than uranium and using its waste fission products, even thorium will run out.
A 4 to 5% PV module replacement per year (20-25 year lifetime) means around 1000 GWp needs yearly replacement, worldwide. To put things into perspective, at say, 10% efficiency, 100 GWp represents 1 billion square meters, or 1000 square km, of PV per year. 1,000 GWp/year means 10,000 square km, or 3,861 square miles, per year. That's a lot of area, and a lot of substrate made of recyclable plastic, from hydrocarbons, or bio-material, or both.
Also, a lot of PV thin film is made from tellurium, indium and gallium material that does not exist in sufficient quantity. Better to realize this inherent scarcity earlier rather than later, and say goodbye soon to CdTe and CIGS that won't meet 1,000 GWp/year expectations.
Derrick Grimmer © 2011
