Would cryodesalination be cheaper than current methods?

[PraAnan]

Hey everyone,

So I came across a company which is working on desalinating water by freezing it and was wondering what everyone here thought about it:
http://cryodesalination.com/

A similar project was also being done by students of UC San Diego:
https://www.aicheprojects.org/uploads/9/8/6/9/9869628/cds_proposal__1_.pdf

There are also a number of research papers about this method when you search for it on google.

I noticed that they mention that cryodesalination could be more efficient than reverse osmosis. How much energy is required to turn 1 liter of water which is 20C into ice?

While doing some initial research reverse osmosis deslination seems to use about 4-6 watt hour to produce one liter of water while cryodesalination would use a minimum of 29 watt hour per liter.

First you need to cool down the water to zero. It takes about 4200 J per liter/degree. So to cool the water down you need to extract 4200 * 20 = 84 kJ. Then you need to turn the water into ice. It takes 333 kJ to solidify 1 l of water. So in total you need to extract 417 KJ of energy. You'll need to use some kind of a refrigerator to do it. The coefficient of performance of a good chiller could be about 4 which means you need 4 times less energy to transfer the amount of energy required.

So we get about 105KJ of energy spent (or 29Wh).

Am I missing something completely obvious / using wrong numbers or is cryo desalination way more energy intensive than reverse osmosis?

Thanks for reading.

 

[mfb]

Use a heat exchanger to transfer heat from your salt water to the ice, cooling (and freezing) it while melting the ice. Freezing the salt water has to happen at a lower temperature than melting the resulting salt-free ice, and there is a theoretical minimum energy input (given by the energy needed to separate H2O and salt), but you can make it more efficient than just freezing salt water and then putting it somewhere to thaw.

 

[BvU]

Here's a few theses on freeze crystallization:

https://repository.tudelft.nl/islan...23-3d40-41e7-a71b-c91a8fcb1702/datastream/OBJ
https://research.tue.nl/en/publications/the-eutectic-crystallization-of-nacl2h2o-and-ice
https://repository.tudelft.nl/islan...480-a8ab-ddf7a3a7afea/datastream/OBJ/download
https://repository.tudelft.nl/islan...e9-43d3-8b6f-541947cc8b13/datastream/OBJ/view

lots of practical challenges, in particular to avoid the whole lot from freezing up and clogging
A fun research area !

 

[zbikraw]

The technique is relatively slow and major energetic expenses might be heat (in "form of cold") losses. You can use "directional freezing" by contacting reactor wall with something "very cold", mainly "dry ice". As in any partitioning with small partition coefficient, you can expect "really fresh" or "potable" water after repeating freeze-thaw many times and water yield is small. For less demanding applications, as preparing substrate for distillation or ultrafiltration, the method can work well.