To understand RO desalination of seawater it’s essential to understand osmotic pressure, which is proportional to the salinity.
The osmotic pressure π in bar is roughly 7.5 times the salt concentration in weight %. So, at 3.5 wt% salt - the mean seawater salt concentration in many regions of the world – the osmotic pressure is around about 26 bar. This is the minimum pressure that has to be applied to get any water at all to permeate the RO membrane. If 50% of the water is recovered as permeate, and all of the salt from that permeate is rejected, the retentate salt concentration doubles and so π also doubles to more than 50 bar. For 75% recovery it doubles again to more than 100 bar. For 87.5% conversion it doubles again, to more than 200 bar. And the actual operating pressure is higher than the osmotic pressure because of the various hydraulic resistances in the system: the pipework, the membrane channels, and the membrane transmembrane itself. So, clearly it's not feasible to operate an RO desalination plant at the sort of conversions (more than 95%) of an MBR. In practice 50% recovery is about as much as can be expected.
Now let's take a look at the reverse osmosis process. The seawater is pumped at low pressure to the pre-treatment stage – that’s the microfiltration/ultrafiltration step mentioned earlier. The membranes used for this are similar to the immersed hollow fibre membranes used for an MBR, but they're configured differently and are usually above ground and pumped rather than immersed. The water is then chemically dosed with anti-scalant. This is to suppress the formation of calcium carbonate, and other sparingly-soluble inorganic compounds, within RO module membrane channels – which are only around 1 mm thick.
The water is then pumped at high pressure to the RO membranes. The RO process itself is normally configured in two stages, with each stage containing dozens or hundreds of modules in parallel. As with any membrane process, there’s a permeate stream – which is post-treated to stabilise the pH - and a retentate stream. Since the retentate stream flow rate is significant after this first stage, because the conversion per stage is limited, the retentate is pumped to the second stage where further conversion to permeate takes place. The retentate stream from the second stage is invariably discharged to the sea.
Now, as stated previously, this retentate stream is normally at a pressure of more than 50 bar and represents around half the feed flow. This means that it has a lot of latent kinetic energy. Remember, power is flow rate times pressure: the retentate flow rate is around half that of the feed and the pressure just a few bar lower. So, an energy-recovery turbine is installed on the retentate discharge stream, to capture this energy.
That means that the total energy consumption is given by the energy consumed by pumping up to high pressures to generate permeate, minus the energy recovered from the retentate stream by the turbine. And the balance between these two contributions is determined by the recovery as you'll see in the next exercise.