Chemists at Seoul National University's College of Engineering have raised the proportion of the wanted product in an electrochemical reaction from about 16 per cent to about 97 per cent without altering the voltage, the catalyst or the composition of the solution. What they changed was the shape of the vessel and the speed of the liquid inside it.
The variable they were chasing is local pH: the acidity of the thin region of liquid immediately beside an electrode. Protons are produced or consumed at electrode surfaces, and when they are generated or removed faster than they can diffuse away, the acidity of that microscopic layer drifts away from the value measured in the bulk.
In this cell the drift was substantial. The researchers put the local extremes at about pH 11 beside the cathode and about pH 2.5 beside the anode. The overall solution was essentially neutral.
The reaction under study was the Baeyer-Villiger oxidation, which turns ketones into esters or lactones and traditionally calls for peroxyacid oxidants. The team used electricity to activate oxygen gas at an electrode instead; isotope tracing confirmed that the oxygen ending up in the product came from oxygen gas rather than from water.
Their fix was mechanical. They rebuilt the reactor to put the electrodes closer together and to circulate the electrolyte between them, so that the acidic region at one electrode and the alkaline region at the other counterbalance each other. Cutting the gap between electrodes from 20 millimetres to 4 millimetres raised the production rate about 2.2 times, and output per unit time rose about 14 times from the lowest flow rate tested to the highest. Without flow, almost none of the desired product formed at all, at the same voltage and in the same solution.
Under the optimised conditions the group made lactones from 18 different ketones, including epsilon-caprolactone at 82 per cent yield, a precursor to the biodegradable polymer polycaprolactone used in sutures and drug-delivery systems. They also produced epoxides from six alkene starting materials, of the kind used in adhesives, coatings and electronic resins. The process runs at room temperature and ordinary atmospheric pressure and could be paired with renewable electricity, though the researchers say it is not yet ready for widespread industrial use and needs further optimisation and larger-scale testing.
The work appears in the Journal of the American Chemical Society under the title "Local pH Control Promotes Oxygen-Atom Transfer Electrosynthesis", supervised by Professor Jaeyune Ryu, with Seonghyeon Min and Junghoon Lee as co-first authors. It was funded by Seoul National University, the National Research Foundation of Korea and the Institute for Basic Science.
To gauge how widely the effect might matter, the team reviewed more than 600 papers on organic electrosynthesis published since 2010. About 89 per cent of them involved reactions in which both electrons and protons were transferred at one or more electrodes.
"This study transforms 'local pH' near the electrode from a hidden variable into a controllable design principle," Ryu said.
It is worth pausing on what the bulk reading was describing all along. A vessel containing one region at pH 11 and another at pH 2.5 was reported, accurately, as essentially neutral.
That figure is an average of two places, and it describes neither of them. No molecule in the cell experienced pH 7. The number refers to the midpoint between two environments, which is to say a corridor, and the corridor is where nothing was happening.
The literature review sharpens the point considerably. A variable that appears in roughly nine experiments out of every ten, for sixteen years, is not hidden. It is the subject. Organic electrosynthesis has been studying local pH continuously since 2010 without mentioning it, which makes it the most thoroughly replicated unreported result in the field.
The Department of Unnecessary Research proposes that bulk pH be reclassified accordingly: not as a measurement of the solution, but as a rumour about the middle of the tank.
Neither end of the cell has ever been neutral. The only part of it that matched the label was the gap between them.