# Why is distillation not a viable way to separate ammonia from water?

Since the steam pressure of ammonia is higher than that of water, I would expect distillation to be a reasonable way of separating a mixture of both.

However, in industrial applications known to me there's always a step of stripping the ammonia chemically from the vapor. Why is this necessary?

• Whilst I can't answer the question, I can confirm that the water-ammonia system is not azeotropic. Do you have any idea of how much ammonia ends up being scrubbed? Apr 30 '12 at 11:14
• I'm thinking about 5-10kg$NH_4$/t in the input, "pretty much all" of it will be scrubbed (no exact figures). The curve you show for ammonia looks qualitatively like the curves in my textbook examples for distillation.
– mart
Apr 30 '12 at 11:58
• What industrial process is this? Distillation is expensive. May 1 '12 at 1:26
• More often than not, in industry, something is done with economics in mind. In this case it appears that for the concentrations and volumes, a chemical strip is the more economically viable option. An exercise for you might be to check that. May 2 '12 at 12:10
• Ammonia stripping from digestate of a Biogas plant, the stripped digestate is to be used to dilute the feedstock for the plant, ammonia has to be stripped because of the inhibition of the biological process.
– mart
May 3 '12 at 12:53

Having embarrassed myself with my first answer, let me try this again. Ammonia and water hydrogen bond exceptionally well. In fact, the $\ce{O-H...N}$ value is 29 kJ/mol while the $\ce{O-H...O}$ value is 21 kJ/mol. (Ref) So, if there is any water around, ammonia would be happy to become solvated as they recondense.

Also, ammonia is a weak base and reacts with water (which is not usually the case when you're doing a distillation). $$\ce{NH3 + H2O <=> NH4+ + OH-}$$ So, if there's any water around as the ammonia is distilling and they recombine, they would also react which would make it difficult to remove the last of the water. (Some reaction would occur because the lowest energy point of the equilibrium is not at reactants or products.)

So, as noted in the comments, distillations to prepare dry ammonia use a scrubber to remove the water.

• (pure) dissolution is always some kind of physical reaction - a change of physical state, combined with a change in energy. In this sense the reaction (while very different to ethanol for example) is integral in any distillation. If it were not for ammonia-water interaction the mixture would be inseparable by distillation! (it would follow raoult's law). Apr 22 '14 at 8:48

I keep hoping someone will jump in and answer this question to save me from thinking about it. Odds and ends from researching this question that might help someone else answer it.

• Clusters of water and ammomia, such as $\ce{NH3(H2O)20}$ as well as various other ratios, are known and have been investigated. I didn't find any references that reported them in vapor phase in distillation columns.
• But, there is a reference that reports that water clusters of varying sizes are found in steam during distillation, so it seems possible that water-ammonia complexes could exist during distillation since hydrogen bonding is strong between water and ammonia in solution. (I did see reports suggesting their existence as atmospheric pollutants.)
• Clusters involving both positive and negative ion complexes of ammonia and water have been reported. (Such as $\ce{(H2O)20NH4+}$ (Ref) or $\ce{(H2O)p(NH3)q-}$ (Ref)
• It's difficult for me to tell from the abstracts (or the papers) whether these species are important in regular laboratory conditions, since the researchers are normally looking at them by MS.
• A single $\ce{H2O}$ bonded with $\ce{NH4+}$ forms a very strong hydrogen bond 92.5 kJ/m.
• Why doesn't ammonia form an azeotrope with water? Of the common solvents that can form effective hydrogen bonds with water, only methanol and ammonia don't form azeotropes. (ref)

The comments for the question mention stripping ammonia from a water in a digestate of a Biogas plant. Ammonia can be removed from waste streams by steam or air stripping after adding NaOH to suppress formation of ammonium ions. Reports suggest that this is fairly efficient (>99% removal). Perhaps the remaining ammonia in the effluent could come from the water-ammonia clusters that are not separated in the distillation. There also seem to be other options for treating the waste stream discussed on this forum.