Does the Lovins/passivehaus building construction theme render the concept of thermal storage systems for regulating the intermittency of renewable energy sources obsolete? In order to examine this question we would need to first compare the capital investment of each system. However, since neither is deployed in an quantity, this isn't really possible. The issue that can be analyzed then is the ancillary value of thermal storage to a renewable electricity grid versus the efficiency gains of the passivehaus concept.
Passive Home Concept
For those of you who have never come across the Amory Lovins schtik or the German Passivehaus building standards, it has been shown that residential or commercial structures can be built with practically no heating requirements. This is done through construction that is properly insulated and sealed with minimal air exchange and uses passive heating (or cooling, depending on climate) strategies. The passivehaus benefits from entropy in terms of home heating. Every electronic device is essentially a resistance heater in addition to its functional purpose, and every person an 80-100 W thermal source.
Scale of Thermal Storage
The most likely medium for thermal storage is water due to its low cost, heat capacity, and the fact that it is liquid and hence is easy to transfer heat with it.
The specific heat capacity of water is 4.184 KJ kg-1 K-1. The heat of fusion − the energy required to change the phase of water from solid ice to liquid water − is 334 KJ kg-1 or the equivalent of almost 80 K. Relative to 20 °C, ice stores a greater amount of cooling power than boiling water.
Consider data on space heating and air conditioning for the USA in 2001. I'll use the worst case: Northeast homes for heating and Southwest for cooling. I am going to ignore hot water heating even though it's significant because it's not relevant to the argument in the end. The average Northeast home uses 63 mmbtu/year for space heating. This works out to an average of about 0.18 GJ/day or 0.365 GJ/day during the peak heating season assuming some sinusoidal distribution. This is the equivalent of 87,000 kg K/day of water; if we store the water at 80 °C to heat the home at 20 °C then we need approximately 1.5 tonnes of water, or 1.5 cubic meters worth (nearly 400 gallons).
On the cooling front, the average Southwest annual electricity consumption is 4,000 kWh/year. If we use an average coefficient of performance of 3.o then the actual cooling supplied is 0.12 GJ/day or a estimated peak of 0.235 GJ/day. If, again, we assume the house is kept at 20 °C then 2/3 of a tonne of ice is required. Consider that 1-2.5 tons (of ice) are common ratings for a centralized air conditioning systems.
Taken over North America (say 100 million homes), these seem like significant numbers. 44 mmbtu of natural gas at $10/mmbtu is $44 billion dollars a year and the production of ~67 Megatonnes of CO2 . 2,300 kWh of electricity for cooling is a total of $18.4 billion dollars (at $0.08/kWh) per year and assuming coal power (at 900 g/kWh), 207 Megatonnes of CO2.
The North American GDP is about $12 trillion per year, so residential heating and cooling alone constitute 0.5 % of that. Scale-wise, there is plenty of potential for passivehaus or thermal storage systems. But can they be friends?
Heat Pump Efficiency
The coefficient of performance is how much heat is moved for a given amount of work (electricity in this case). Commercially air conditioners in the USA are rated based on a Energy Efficiency Rating (EER) which is the square of the COP between 80 F and 95 F (or about 300 K and 308 K). There is also a Seasonal EER (SEER) which is a different (more relaxed) standard. The theoretical COP for this temperature range is 37, but in reality most systems are in the range of around 3.5. The ultimate theoretical coefficient of performance of a heat pump is given by:
COP = TH / (TH - TC) = TH / ΔT
Herein lies a problem. As the temperature difference a heat pump has to cross increases its efficiency decreases. Normally an air conditioner only has to work across 8 K or so. However, if we want to use it to make ice, from a night-time temperature of 24 °C, then the ultimate efficiency of the system will only be a third normal. For a real-world system the drop in efficiency on a percentage basis will not be so precipitous but it will still be disadvantaged trying to make ice.
Overall it's a tough sell for thermal storage as a means of handling renewables intermittency. As we've seen, thermal storage sets efficiency against grid regulation. Generally, when we have schemes with competing criteria they fail to be economically attractive. Witness my investigation into solar thermal cooling. In that case there was a competition between the efficiency of the solar thermal collector and absorption chiller on the basis of temperature. Here we have competition on pure power. Yes, we can store off-peak power, but the effective round-trip efficiency is going to be unimpressive simply due to the drop of in performance of the heat pump.
There is still the possibility to run numerous appliances on a deferrable basis. Heat, air-conditioning and refrigeration all only need to maintain a given (if narrow) temperature range so with good insulation they should be able to run on relatively short duty cycles. Other appliances, such as the dishwasher or combination washer/dryer can be scheduled.
Chicken or Egg?
One problem with pumping efficient homes is that houses last for such a long time. One often heard meme in the peak oil world is that car fleets take too long to be replaced. Houses can be renovated, cars can't. Still if you are one of those people who think suburbia is evil (as opposed to just soulless) then the choice of whether to concentrate on improving the efficiency of houses or cars presents quite a quandary. From my point of view, I'm more concerned overall with climate change and more localized pollution of the air and water. If people want to live in rows of identical pink stucco houses... enjoy.