Chemists separate water isomers

Nine years ago, Russian researchers sparked controversy when they claimed to have separated water into its two spin isomers. Now, chemists in Israel claim to have performed a similar feat using a different method, and suggest the outcome could deliver highly sensitive nuclear magnetic resonance (NMR) experiments.

Water molecules come in two spin isomers: one ’ortho-water’ with the spins of the constituent hydrogen atoms parallel, and one ’para-water’ with the hydrogen spins anti-parallel. The two isomers have subtly different properties that become important in diverse fields of science. In astrophysics, for example, the ratio between ortho- and para-water is used to determine temperatures in interstellar space, although the data are hard to interpret - partly because scientists have been unable to study either isomer on its own.

In 2002, Vladimir Tikhonov and Alexander Volkov at the Russian Academy of Science claimed to create drops of water enriched with either the ortho or para isomer, lasting for 25 minutes or more, by exploiting their different adsorption properties on a surface.¹ But their experiment proved hard to repeat, and other chemists, such as Hans-Heinrich Limbach of the Free University of Berlin in Germany claimed it would be impossible for the drops to remain in an awkward, one-isomer structure for so long without converting back to a mixture.² 

Gil Alexandrowicz and colleagues at the Technion-Israel Institute of Technology in Haifa may have got around the problems of Tikhonov and Volkov’s experiment, however, with a method based instead on the deflection of water molecules in a magnetic field. Based on the famous 1920s Stern-Gerlach experiment in quantum mechanics, the method involves sending a beam of water vapour through a six-poled magnet, which bends the trajectories of the water molecules depending on their spin projection. The result for Alexandrowicz’s group was a focused beam of ortho-water vapour.

Limbach, who was sceptical of the Tikhonov and Volkov experiment, believes Alexandrowicz’s group may have proved successful in separating ortho-water. However, he claims there will be no practical benefits of the separation, because as soon as the ortho-water vapour is captured in a liquid or on a surface the hydrogen spins will convert to produce a normal mixture. As for applications in NMR - ’that’s completely nonsense,’ Limbach says. 

Alexandrowicz’s group believes that single water isomers will benefit NMR like the two spin isomers of hydrogen - ortho- and para-hydrogen - do currently, improving signal resolutions by up to 100,000 times. ’I believe that such experiments are possible and in fact our group is currently working towards that goal,’ says Alexandrowicz. ’I would however like to emphasise that the experiments [would] involve measuring NMR signals of ultra-thin layers of water molecules. Hence these are of great importance for surface science, but are not directly related to medical NMR uses such as magnetic resonance imaging.’ 

Jon Cartwright