Batteries and The "Negative Temperature Coefficient of Reactivity"

Batteries

Energy storage costs under Proposition 7. I discussed Ed Ring's analysis of the costs of making the California electrical system 50% renewable by 2025. I mentioned electrical storage which Ring estimated would cost $350,000 per MWh. Ring's estimate struck me as low, because my own check last year had yielded considerably higher figures. Ring's figure was based on a study he did based on sodium-sulphur battery costs. Roger Brown came up with a somewhat higher figure:

NGK Insulators claims that in mass production the price of NAS batteries will drop to U.S. $140/kWh. The round trip efficiency in AC mode is 75%. With a lifetime of 2500 cycles at 100% DOD this adds up to a cost of $140/2500*0.75= $0.075 on top of the cost of primary generation. If private finance capitalism and a rising stock market last forever (or at least during our lifetime, a standard assumption of many posters on this site) then the cost of interest must be added to this number, thus bringing the cost up close to $0.12/kWh.

"Jim from The Energy Blog" reported:

Sorry I missed this announcement on Altair delivering 2 MW of batteries to AES:

January 2, 2008 -- Altair Nanotechnologies, Inc. (Nasdaq: ALTI), a leading manufacturer of safe, high-performance lithium-titanate battery and energy storage products, announced today that it completed on schedule in December the manufacturing of battery packs to be used in a two (2) megawatt energy storage system ordered by the AES Corporation. The $1 million purchase initiated by AES was previously announced in August 2007. Altairnano expects the system to be connected to the grid and tested during the first quarter of 2008.

According to a story in the New York Times, dated September 11, 2007, American Electric Power had contracted to pay NGK Insulators Ltd. of Japan, $27 million for six megawatts of sodium sulfur battery capacity. The price included the cost of substation improvements. The Times story stated that each battery could deliver one megawatt of power for a little more than seven hour. That would yield a cost of $640,000 per MWh of storage capacity. According to a USA Today story, the lifespan of a sodium sulfur battery is expected to run about 15 years.

The "negative temperature coefficient of reactivity"

The term "negative temperature coefficient of reactivity" (NTCR) sounds technical enough to be forbidding to the average person. What it means in a reactor is that as the temperature inside the reactor core goes up, the nuclear reaction decreases. Since increases in the nuclear reaction, increases temperature the nuclear reactions in reactors with NTCR will decrease. Eventually, in a well designed, inherently safe reactor, NTCR will turn the reactor off without any operator intervention.

In a fluid salt reactor the NTCR operates by taking advantages of the rules of nature. As the temperature of the core salts rise, they expand. Thus if the reactor fluid is some variant of the LiF-BeF2-UF4 approach, as core fluid temperature rises the fluid salt mixture expands. Since the size of the core is going to stay the same, some of the salt mixture will be pushed out of the core. that means pushing fissionable uranium out of the core, since the uranium is chemically bonded to the core salt. The less uranium in the core, the less fuel is available, so the nuclear reaction slows down. Eventually before the temperature reaches a dangerous point, so much uranium is forced out of the core that the nuclear reaction will completely stop.

The reactor core can be maintained at this temperature, and indeed some heat will have to be dumped because the nuclear decay of fission products that are dissolved in the core salts will continue to release heat. Decay heat thus must continue to be extracted from the reactor through the heat exchange system to prevent overheating. Thus the salt in the reactor core is held at maximum temperature without any fuel use, and this condition can be maintained for some time.

Liquid salt is an excellent heat storage medium. In fact liquid salts have been proposed as heat storage mediums for solar thermal electrical systems. The heat in the salt can be rapidly and efficiently transfered through a heat exchange system to power generating turbine. The turbines can be generating 100% of their capacity as quickly as they can ramp up to full speed. Thus the LFTR fulfills the basic requirements of peak load electrical generators. Currently natural gas burning turbines, hydro electric systems, and pump storage facilities fulfill the peak generation role. Natural gas systems are cheap to build, but expensive to operate because of the high cost of natural gas. Hydro systems are used to nearly the fullest possible extent already. Pump storage facilities are expensive to build and have limited flexibility, and limited storage capacity. In contrast liquid salt reactors such as the LFTR can be held in reserve, but when needed they can be operated for prolonged periods with very low fuel costs. The question then becomes one of capitol costs. Can the cost of LFTRs be low enough to justify their use as peak power producers?

That question cannot be fully answered until we know what would be the competing peak power production systems, and what would they cost. Some possible peak systems would include thermal solar with heat storage, pump storage, compressed air storage, and natural gas with carbon penalties. Power from the natural gas system would be very expensive, because the power user would be paying in effect a double carbon premium. Thus the lower capitol cost would not be a significant advantage.

Solar thermal with heat storage would be well matched to normal daytime demand, provided its capital costs were competitive. But my power system model is that of Texas, where summer heat waves place extreme peak demands on the power system. Although solar power does have some seasonable flexibility, it is probably not sufficient for the peak demands of Texas summers.

The term "negative temperature coefficient of reactivity" (NTCR) sounds technical enough to be forbidding to the average person. What it means in a reactor is that as the temperature inside the reactor core goes up, the nuclear reaction decreases. Since increases in the nuclear reaction, increases temperature the nuclear reactions in reactors with NTCR will decrease. Eventually, in a well designed, inherently safe reactor, NTCR will turn the reactor off without any operator intervention.

In a fluid salt reactor the NTCR operates by taking advantages of the rules of nature. As the temperature of the core salts rise, they expand. Thus if the reactor fluid is some variant of the LiF-BeF2-UF4 approach, as core fluid temperature rises the fluid salt mixture expands. Since the size of the core is going to stay the same, some of the salt mixture will be pushed out of the core. that means pushing fissionable uranium out of the core, since the uranium is chemically bonded to the core salt. The less uranium in the core, the less fuel is available, so the nuclear reaction slows down. Eventually before the temperature reaches a dangerous point, so much uranium is forced out of the core that the nuclear reaction will completely stop.

The reactor core can be maintained at this temperature, and indeed some heat will have to be dumped because the nuclear decay of fission products that are dissolved in the core salts will continue to release heat. Decay heat thus must continue to be extracted from the reactor through the heat exchange system to prevent overheating. Thus the salt in the reactor core is held at maximum temperature without any fuel use, and this condition can be maintained for some time.

Liquid salt is an excellent heat storage medium. In fact liquid salts have been proposed as heat storage mediums for solar thermal electrical systems. The heat in the salt can be rapidly and efficiently transfered through a heat exchange system to power generating turbine. The turbines can be generating 100% of their capacity as quickly as they can ramp up to full speed. Thus the LFTR fulfills the basic requirements of peak load electrical generators. Currently natural gas burning turbines, hydro electric systems, and pump storage facilities fulfill the peak generation role. Natural gas systems are cheap to build, but expensive to operate because of the high cost of natural gas. Hydro systems are used to nearly the fullest possible extent already. Pump storage facilities are expensive to build and have limited flexibility, and limited storage capacity. And we see that mass storage batteries don't come cheap. In contrast liquid salt reactors such as the LFTR can be held in reserve, but when needed they can be operated for prolonged periods with very low fuel costs. The question then becomes one of capitol costs. Can the cost of LFTRs be low enough to justify their use as peak power producers?

That question cannot be fully answered until we know what would be the competing peak power production systems, and what would they cost. Some possible peak systems would include thermal solar with heat storage, pump storage, compressed air storage, and natural gas with carbon penalties. Power from the natural gas system would be very expensive, because the power user would be paying in effect a double carbon premium. Thus the lower capitol cost would not be a significant advantage.

Solar thermal with heat storage would be well matched to normal daytime demand, provided its capital costs were competitive. But my power system model is that of Texas, where summer heat waves place extreme peak demands on the power system. Although solar power does have some seasonable flexibility, it is probably not sufficient for the peak demands of Texas summers. However, if the cost of a LFTR can be brought in at a low enough cost, the LFTR would be a Jim Dandy load follower-peek power generator.

Let us assume for a moment that we need a system that will assure the delivery of 7 GWh of overnight electricity to run Texas Air Conditioners on summer nights. You can get a battery system for from $3.5 to $4.5 Billion. The batteries will last 15 years. Would it be possible to deliver a LFTR system for $3.5 per GW? If you were a utility which would you prefer?