## Wednesday, December 19, 2012 ... /////

Peter Thiel, a favorite venture capitalist of mine, just paid \$300,000 to Louis Michaud, a Canadian inventor on the picture below who plans to build artificial tornadoes – the so-called [atmospheric] vortex engines: Wikipedia, Michaud's web – that may supply us with lots of energy.

This idea surely sounds provoking at first – way too close to a description of a perpetual motion machine – but I am already in a different stage in which I tend to think that this most elementary criticism is unjustified. However, it is still not clear how ambitious a change in the energy sector is being promised here.

The basic underlying mechanism is said to be the same as for a solar chimney. In the troposphere – atmospheric layer between the surface and the tropopause 10 km or so higher – the temperature generally decreases with the altitude. This gradient isn't far from the "adiabatic lapse rate". What is it?

The warm air heated from the Earth's surface at the bottom wants to go up because at the same pressure, it has a lower density than the cooler air. As it goes up, it expands, its pressure decreases, and so does the temperature. The calculation of these changes is enough to see that the temperature decreases by 6.5 °C per kilometer of height. Note that it's possible, albeit a bit counterintuitive, that the atmosphere may sustain a quasi-equilibrium with non-uniform temperatures. There's no paradox here, however: these non-uniformities are powered by the constantly added heat from the Sun and, indirectly, from the Earth's surface.

This behavior stops above the tropopause, in the stratosphere. The stratosphere is stratified – "separated" into horizontal layers that keep their altitude and don't mix. It's possible because the air temperature is actually increasing with the altitude in the stratosphere – because it's being heated by the Sun – so the cooler (and therefore denser, heavier) air at the bottom has no reason to go up.

Back to the troposphere. The mixing of the air that enforces the lapse rate also brings some circulation which is responsible for many kinds of weather phenomena and winds – and tornadoes are actually the most typical ones. The processes and gradients occurring in a tornado are pretty complicated, however, and your humble correspondent isn't able to evaluate all these things.

It's conceivable that the artificial tornadoes could replaces chimneys and increasing their efficiency by dozens of percent. When it comes to this "evolutionary" improvement, I am willing to bet that the existing chimneys are far from optimal, so some tornado-inspired improvement is likely to exist, whatever it is. However, it's plausible that one could do better (and it seems like the folks actually claim that this is the ambition here): the modestly elevated temperatures on the surface could be enough to play the role of the "hot burning coal" in a conventional power plant that powers the flow so if one could extract the mechanical energy from the man-made tornado, it could be energy obtained "almost for free".

It would be nice if someone told me whether this is physically possible at all. And if it is, whether there's an upper limit on the number of tornadoes or the energy per unit time that could be extracted from this truly attractive hypothetical source. ;-) For example, the restriction may arise because the man-made tornado would lower the lapse rate and would stop working when the lapse rate decreased beneath a certain threshold.

Thanks.

I have asked the same question at Physics Stack Exchange and there's already an interesting answer over there.

#### snail feedback (8) :

I think some of the more efficiently and safely fuel converting nuclear fission technologies, that are "waiting in the wings", will provide us with a far more manageable method (I can't see how turbines could be built that could withstand being turned by natural twister!) of keeping us supplied with energy both in the near and far future.

If I have understood correctly, real tornados get most of their energy from condensation (the "dry" vortices in the atmosphere are all weak). A heat source near the ground could trigger the process but sustaining it would probably require a lot of vapour.

Once you've started your tornado, how do you make sure it stays put and doesn't wander off on its own causing bother?

Also, - this is just a quick thought, not fully considered - if this effect is dependent on the adiabatic lapse, then isn't the whole GHG effect put in question?

Peter Thiel needs better technical guidance. All of these schemes try to extract mechanical power from air that is warmed by the sun. This air is abundant but not very hot. Unfortunately, the conversion efficiency is limited to the Carnot efficiency less parasitic (frictional) losses. The swirling vortex only adds additional frictional losses and nothing can be gained by trying to use the natural adiabatic gradient in atmospheric temperature. It is possible to generate a bit of electrical power this way but thermodynamic restraints will render the efficiency to something less than photovoltaics with a likely higher cost per unit area. It would need even more government subsidies than solar power in order to be implemented.

This is a fantasy, Lubos. Or, it may be a scheme to extract money from the public purse. I would not be surprised.

Thanks, Gene, this comment of yours sounds fully compatible with my intuition but I am not able to really "prove it" that the vortices etc. make the efficiency worse rather than better etc.

Peter Thiel is certainly not one of the blood-suckers, Lubos, but Michaud probably is. I can’t be privy to Peter's thinking but 300k is not much money and he may just regard this as a small gamble.

The vortices are not a source of energy, Lubos, and that’s my problem. There are many ways to concentrate the energy from sunlight but you still need a vast area to collect it in the first place and that, necessarily, involves a big capital expenditure. I just can’t imagine anything cheaper that a thin film on a square meter basis. A collection of big chimneys, each having its vertical dimension much larger than its horizontal dimension will be much more costly. If the vertical dimension is not much larger than the horizontal one, cold air will just spill in from the top and ruin everything. Of course, if the horizontal dimension is small, friction will eat up all of the energy.