I have had the privilege of explaining Pass-through distillation to many clever and accomplished engineers during the past few years. One of them said suddenly “Aha! I see what this is! It is like mechanical vapour recompression (MVR) without the compressor”. Frankly, I didn’t understand his comment at the time, but now that I have grasped what he meant I find it very apt.

Mechanical Vapor Recompression (MVR)

The diagram below illustrates the MVR process: vapour from the evaporator vessel is brought to a higher pressure by means of a mechanical compressor. At higher pressure, a fluid changes state at higher temperature. The higher pressure vapour condenses in the heating coils of the evaporator because its temperature is higher than the boiling liquid in the vessel.


Notice that no external heat is being applied to the system. The latent heat demanded by the boiling liquid is fully supplied by the latent heat of the condensing vapour; the two quantities are nearly identical. Mechanical energy is being added to the system through the work of the compressor, but this amount is very small by comparison to the latent heat.The theoretical thermal efficiency of MVR can exceed that of a 100 effect evaporator. Even with the inefficiencies inherent in real equipment, efficiencies equivalent to 30 effect evaporators are reported.

Now let’s look at a SAM in operation. The green feed stream (assume water for our present purposes) enters the left-hand chamber at its boiling point falls over metal bars at a slightly higher temperature. Heat is imparted to the liquid causing evaporation. The vapour is ducted to the bottom of the right hand chamber. Induced by the pull of the vacuum train, the vapours rise counter-current to falling absorbent fluid, a brine solution, and become a part of the brine. This absorption involves a phase change; the vapour becomes liquid and its latent heat is released into the brine. But that heat is immediately conveyed by the bars to the evaporator chamber. ( Note: the “bars” are actually heat pipes which have very low thermal resistance)



Like MVR, the evaporation has taken place without the application of any externally applied heat. In both cases, the heat applied to the evaporator manifests itself as the mass flow rate of the vapour times its enthalpy of evaporation. In both cases the vapour is turned back into a liquid at a higher temperature that of the evaporator, and means have been provided to move that heat back into the evaporator. So my friend’s comparison has a great deal of merit.

But our first diagram shows an MVR process in its entirety while the second diagram shows only a SAM, which is half of the pass-through distillation process. The other half is the brine desorber, which restores the diluted brine from the bottom of the SAM’s right hand chamber into a stream of pure water and the stream of concentrated brine which enters the same chamber at the top. The desorption process is a multiple effect distillation (MED). A three effect desorber is likely the practical choice, perhaps a four effect.

So MVR is simpler and more efficient than PTD. But it has some serious drawbacks that PTD overcomes.

First, it involves a large piece of mechanical equipment: a fan, blower, or compressor. Aside from their high capital cost, these pieces of equipment have reliability issues.

Second, MVR is not well suited for low pressure / low temperature operation. The compressor is already large when designed for operation at 1 atm. It must be twice as large for operation at 1/2 atm, because compressors are volumetric devices. By contrast, PTD can operate at 1 atm or at 1/30 atm without much change to equipment. Yes, it has a vacuum pump which must be upsized for lower absolute pressures, but since it only handles non-condensible gases, it is not a major piece of equipment at any pressure.

Third, practical compressors for MVR have modest compression ratios and consequently the delta T across the heating coil is small – on the order of 5C. This means the surface area must be large to compensate. No problem here – the large energy savings ensure rapid payback despite the high capital cost associated with large equipment.  The real problem is that MVR can only be applied to fluids that boil and condense at roughly the same temperature. A few degrees of boiling point elevation can make the use of MVR infeasible. Pass-through distillation does not have this limitation. Concentrated Lithium Bromide brine can absorb water vapour that was generated 40 Celsius degrees cooler than itself. With that kind of delta T available, feed streams with high boiling point elevation are no problem, and smaller equipment can accomplish the same job.

So are the two processes, MVR and PTD, really similar? I would have to say “no”. Although both offer energy savings through the recycling of latent heat, they are quite distinct.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>