Unresolved Problems

I have made this page to document all the current unresolved problems that readers have put forward.  Problems are not a bad thing!  These panels will never reach the market unless countless problems are solved.  This means identifying them and finding solutions.  Will you help me?

• Freezing Temperatures  
• User TekArt over at solarpaneltalk brings up a good point:  "Here in Maine a drain back system might turn on when the TOP sensor on the collector in a traditional copper collector gets substantially hotter than the storage tank. But with a plastic collector how do you measure that temperature? Without some thermal mass up there what gets hot?"
• This is actually one of my biggest concern.  Related is my concern that the mixture of particles at the bottom would be a big lump of frozen "particle mud", so to speak. My solution here (the only one I have thought of so far) is to use residual heat stored in the heat exchanger tank coupled with some light sensors on the panel. The system would be able to tell if there is enough light hitting the panel to warrant starting, and it would also know how much heat was in the tank. It would use this heat to prime itself, melting the particle mud and getting everything started. [note that this requires one of the channels on the panel to be a bypass that would let the warm water flow around and over the particles.
• It appears that polypropylene will expand/bend and perhaps prevent damage due to freezing water.  However, there are a number of other things that could go wrong (the filters, for example). 
• Particle Degradation
• First, you may want to read an article I posted regarding how the variability in the size distribution affects the particle distribution in the liquid. When the panel is first manufactured, the particle size distribution should be as low-variance as economically practical.  This low-variance favors the particles in achieving a uniform distribution.  As the particle break down, smaller particle fragments will be generated.  Smaller particles feel a higher force due to flow (drag) and are selectivity pushed to the top of the panel.  If the size distribution were too great, the small particles could plug the exit and reduce the flow to a point where the larger particles will stay close to the bottom.  To avoid this, the filter can be made to a size that allows the smaller fragments to pass.  A larger and system-wide filter can then capture these.  The result is that the particles will be consumed at some rate, but the size distribution will remain within bounds.
• I suspect that the particles will degrade over time, but I have no real experience to know how much or how long it will take.  Any readers out there who have experience with this?
• Panel Wear due to Particle Abrasive Action
• I would expect wear on the panels.  The density and abrasive action of the particles will be higher on the bottom, so the wear will probably be highest there.  
• The particles will undoubtedly scratch the inner wall of the plastic.  However, the refractive index of water and the plastic will be similar (1.3 for water and 1.4 for polypropylene), which mostly mitigates problems having to do with lower optical transmittance due to scratches.  
• One thing to keep in mind is that the plastic is not hard and could very well be more resilient then a harder material such as glass.  
• The particles are extremely fine.  It could be possible that the particles would actually polish the inner surface.  If this is the case, I need to know the rate that this occurs.
• I currently have no information here. This could be a big problem or no problem.  
• Failure from Pressure at Elevated Temperatures
• Pressure in the panel can be lowered considerably.  All that is needed is to increase the channel width.  Given an applied pressure, the particle size is chosen so that its sink rate is just a little faster then the flow within the panel.  Extremely hot stagnation temperatures are eliminated due the design.  I would add a temperature sensor so that the controller could turn the flow off if the temperature exceeded, for example, 60C (for residential hot water or heating applications)    
• Particle Oxidation
• I have no data on the oxidation of the particles, nor really any experience here.  I used silicon carbonate, but there are many possibilities.  I have little doubt that a particle could be found that would not oxidize.  I used aquarium sand in my first attempt, for example.  A quick search on Wikipedia yields "The natural resistance to oxidation exhibited by silicon carbide....". 
• My first prototype used aquarium sand.  I suspect that it was stable in water for extended time.
• Any comments from readers would be appreciated here.