Fast crystallization is a desired property of the phase change materials, and it is well established that supercooled liquid freezes faster that the non supercooled liquid.
However, is not always the case. It is obvious that when a supercooled melt crystallized portion of the latent heat released will be used to heat up the liquid. So it is like a compromise between how much energy you will harvest and, and how fast it will be released.
One of the applications of phase change energy storage is temperature control i.e. heat is extracted within a fixed temperature range. With supercooling, the objective is compromised, since the temperature of the PCM goes below the nominal phase change temperature range during heat extraction.
Have spent a little time studying supercooling. There are two primary drawbacks, as some others also mentioned. how important they are depends on how you are attempting to use the PCM. But the supercooling will directly impact the temperature margins in the rest of your thermal system, and this could be a significant design problem.
1 - Thermal Protection or temperature suppression applications - if you are attempting to buffer a known thermal transient, and have designed the PCM to absorb most of the heat, you require a 'system reset' prior to the next thermal cycle. If nucleation and recrystallization fails to occur, the next thermal cycle will see energy absorption only through sensible heating, and temperatures can get much higher that allowable. This could lead to critical failure of the system. We did a little work looking at using Erythritol (TM ~ 118°C) for silicon electronics thermal protection, where you'd like to keep temperatures under 125° or so (maybe higher depending on the specific device). If the PCM was not to absorb through latent heating on a later transient, the sensible heating would easily drive temperatures high enough for failure. (in the lab, I've seen Erythritol in small volumes stay liquid almost indefinitely as far down as 50°C, so it's a concern.)
2 - For a temperature control application, you require operation within a certain temperature margin. For air conditioning, this will be dictated by the the nominal temperature where it's integrated into the cooling loop, and the allowable high and low temperatures during PCM absorption and discharge. Supercooling forces the margins to be wider, and may even eliminate otherwise excellent PCM candidates from consideration. Also, to guarantee nucleation, it forces you to implement a colder 'recharge' condition. (i.e., if you have a 50% chance of nucleation at 30°C, but a 90% chance at 20°C, you design the system to run at 10C and have to pay for that extra 10°C somewhere in the system design.) so, it turns into an inefficiency for the system.
Supercooling,makes crystallization under its melting temperature，which results in some inconvenience in application.
"Fast crystallization is a desired property of the phase change materials, and it is well established that supercooled liquid freezes faster that the non supercooled liquid."
Fast crystallization makes the energy releasing fast.And the established fact is that supercooled liquid freezes faster that the non supercooled liquid. However,the crystallization of supercooled liquid can't match the temperature need,and it can't conpensate the drawback induced in utilization.
I agree with the issues mentioned above. I especially want to highlight the issue mentioned by Fabien Rouault concerning the randomness of the phenomenon which makes it difficult (if not impossible) to control the storage system adequately.
I also want to mention that if a significant change in density (and therefore volume) occurs during phase change then the presence of supercooling could cause problems. Even though a container might be sized to leave some additional space for the expansion/contraction of the PCM, the quickness with which this change of volume occurs during supercooling might cause breakage or buckling of the container. This could damage the container and cause leaks in future melting/freezing cycles.
I recommend you to read this paper:
http://ijens.org/Vol_18_I_05/180505-3737-IJMME-IJENS.pdf
Best regards
- Badges
- Science topic