Unfortunately, at least in the US, desalination cost is dominated by the high cost of capital, not by the cost of the energy.
For example, the largest desal plant in the US is the Carlsbad plant in California [1]. For our purposes, it's a very nice plant, because the numbers are very round: it cost $1 BN to build. It has an output of 190k cubic meters per day, or 5.7 million m3 per month. The plant was funded with bonds. Let's say you want to build it today. A private person has access to 30y mortgage rates of about 7.7%. But this being a large corporation, and using municipal bonds that are tax exempt, let's say you can borrow at 5.5%. If you plug in a mortgage calculator, that comes to $5.7 MM per month, which is very fortunate, or a very round $1.00 per m3.
The energy use of desalination is in the range of 2.5 to 4.0 kWh per m3 ([2], page 91). Let's say it is 4.0 for the Carlsbad plant (to be conservative; most likely it is lower). The average cost of electricity for US industrial users is $0.0775 [4], so that's about 19 cents per m3.
Recap: $1.00 capex cost per m3, $0.19 electricity cost per m3. Increasing the efficiency of desalination by a factor of 2 would not make desalinated water twice as cheap, but only about 10% cheaper.
There is an asterisk here though: I don't know why, but the cost of electricity in California is much higher than the average cost in the US. It costs 18 cents vs 7.75 cents. So, for the particular case of California, the capex cost still dominates, but not by a factor of 5, only by a factor of 2.
My pithy answer for why cost of electricity is so high in California is because it is regulated on a Cost Plus basis. This means that the incentives are misaligned such that the best way for power companies to increase profits is to increase their input cost.
The cool trick is to use a nano-porous hydrophobic material with some air trapped in the pores. Water evaporates into the air, then re-condenses on the other side, while salt can't pass. The water does not stick inside, because the material (based on silane) is hydrophobic.
The practically important part is that the membrane does not degrade from the chlorine and ozone, the typical water disinfectants.
But will not the separated salt clog the membrane? Separated salt is an issue, e.g. in distillation - you have to remove it to remain decently efficient.
And speaking of efficiency: do energy costs make mass desalinization with this technology feasible?
Salt doesn't pass through RO membranes and these appear to have the same pore size. However, they all foul and you perform maintenance like backwashing and chemical cleaning as the flow rate goes down.
In terms of efficiency, it is cost dependent. Making drinking water is not cost prohibitive at this time but for large scale agriculture it would typically be considered too expensive to irrigate with RO desalinated water. This technology appears to lower the energy requirements but more research and scaling it up is needed.
RO membranes typically include a waste path so they're self-cleaning for their lifecycle, but some manufacturers may have flushing procedures to extend service life.
Saudi Arabia, Kuwait, UAE, Qatar, Bahrain, and Libya are almost completely dependent on RO for water.
The Carlsbad plant in San Diego was $1B and costs $80M/y (@ 40 MW) to operate while producing 56k acre-ft/y (7% of demand), enough for about 400k people or 20k acres of farmland with average crops.
What would cut water use greatly would be the elimination of meat agriculture which depends on crops such as hay, alfalfa, and cow corn. 2/3 of Utah's water goes to irrigating crops to feed cows.
When I've looked into RO for irrigation it seems that cost is a problem but it's tantalizingly close to economic. I think you need to get the cost down by 2/3rds.
However boron is a problem, most RO water has more boron than is healthy for plants. Though I saw one press release talking about a membrane that unexpectedly produces water with low levels of boron.
For example, the largest desal plant in the US is the Carlsbad plant in California [1]. For our purposes, it's a very nice plant, because the numbers are very round: it cost $1 BN to build. It has an output of 190k cubic meters per day, or 5.7 million m3 per month. The plant was funded with bonds. Let's say you want to build it today. A private person has access to 30y mortgage rates of about 7.7%. But this being a large corporation, and using municipal bonds that are tax exempt, let's say you can borrow at 5.5%. If you plug in a mortgage calculator, that comes to $5.7 MM per month, which is very fortunate, or a very round $1.00 per m3.
The energy use of desalination is in the range of 2.5 to 4.0 kWh per m3 ([2], page 91). Let's say it is 4.0 for the Carlsbad plant (to be conservative; most likely it is lower). The average cost of electricity for US industrial users is $0.0775 [4], so that's about 19 cents per m3.
Recap: $1.00 capex cost per m3, $0.19 electricity cost per m3. Increasing the efficiency of desalination by a factor of 2 would not make desalinated water twice as cheap, but only about 10% cheaper.
There is an asterisk here though: I don't know why, but the cost of electricity in California is much higher than the average cost in the US. It costs 18 cents vs 7.75 cents. So, for the particular case of California, the capex cost still dominates, but not by a factor of 5, only by a factor of 2.
[1] https://en.wikipedia.org/wiki/Claude_%22Bud%22_Lewis_Carlsba...
[2] https://www.energy.gov/sites/default/files/2019/09/f66/73355...
[3] https://www.amtaorg.com/wp-content/uploads/07_Membrane_Desal...
[4] https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...