Tuesday, November 9, 2004

Portable Nuclear Power Plants (SSTAR: a small, sealed, transportable, autonomous reactor)

Nuclear power has a greatly mixed blessing. On the one hand it offers vast quantities of electricity without the hydrocarbon-style pollutions, and you don't have to ruin a river valley by installing a dam. On the other hand nuclear wastes last for thousands of years, so disposal is a problem, and then what if the nuclear material falls into the wrong hands and they make bombs out of it? Proliferation of nuclear weapons is a serious problem.

The tension with both Iraq and North Korea right now is over their development of nuclear technology. Is it simply for peaceful purposes, providing electricity to run the country? Or are they building bombs? Since they're refusing to abide by the nuclear inspection regimine run by the International Atomic Energy Commission, how can we be certain?

All countries, developing and otherwise require power in order to modernize themselves. A team at Lawrence Livermore National Laboratory has been working on a particular solution to this problem.

SSTAR is a small and relatively lightweight nuclear reactor with which they intend to solve a broad range of issues around the use of nuclear reactors. Here's the feature set:

  • Transportable by ship in a shipping container
  • At 500 tons weight, it can be transported from the port to its destination by heavy truck
  • It is tamper resistant and contains notification capabilities, making it safe to deliver into uncertain territory and have assuredness the nuclear material will not be misdirected
  • The coolant is a liquid metal that will not boil off
  • The design is sealed and meant to remain in operation for 30 years without much intervention
  • Wastes remain within the reactor
  • After 30 years the recipient returns the nuclear reactor for "recycling"

Clearly we have a two-edged sword here. Past nuclear reactors have safety problems, could provide weapons making material to undesirable institutions, and have massive technology and infrastructure requirements to build and maintain. Further the waste disposal problem is very bad, with growing resistance to the transport of nuclear waste around the world (and with good reason).

These problems have resulted in a vehement anti-nuclear crowd that knee-jerkily shouts "NO NUKES" whenever possible. They regularly protest anything remotely nuclear, and just last week a protester in Germany lost his life trying to block a train carrying nuclear waste.

With the current state of nuclear technology, I must agree with them. It is very dangerous stuff, and I am glad that the U.S. has not built new nuclear plants in decades.

However, this LLNL design provides for some very interesting thinking.

What if ... The design is of a size useful to a developing country, and appears to have enough safeguards for deployment nearly anywhere. And it offers "clean" power, in that the immediate output of the reactor is steam that runs a turbine. Due to the high temperature it can also be used in hydrogen production, providing a source of pure hydrogen for fuel cells.

What are the problems with nuclear power and how does this design stack up against them? Here's a few, and it makes the rosy scenario painted by the scientists to be a little less certain. I am comforted somewhat that they envision shipping these as product in 2015, which gives us a few years (5-10 maybe) of design debugging.

Issue Solution Devils Advocate
Proliferation Tamper-"resistant" and includes notification facilities. "Resistant"? I'd rather it be tamper "proof" but how could that be accomplished?

The notification facility is likely some kind of satellite radio system. The article discusses the notification system being triggered by detectors rigged to "identify actions that threaten the security of the reactor".

What if a country gets real "uppity" and successfully wards off inspectors and the "international community" to tamper with and remove the nuclear material for their own nefarious ends?

Waste & Radiation in general The waste stays within the reactor. It gets "recycled" after the 30 year lifespan is up. It has to be returned to the supplier for recycling. What if the supplier goes out of business?

How can we test against leaks over a 30 year lifespan?

What if there's an accident during transport to the destination, or upon return to the supplier? Will the design withstand being dropped while being loaded or unloaded on the cargo ship? What if the cargo ship sinks?

Meltdown Liquid metal "coolant" cannot boil off

Liquid metal is at a low pressure

The design can include a passive shutdown feature

In the article they discuss "corrosiveness" of some formulations of the liquid metal coolant. Will this work over a 30 year lifespan?

Control systems fail all the time, and computer systems always have bugs. How can we trust 30 years of successful operation to a computer software system?