Why the fan stops at two hours

The fan on a Branholt column moves 4,200 cubic metres of air an hour. At 420 parts per million there are about 3.4 kilograms of carbon dioxide in that hour of air, and the bed catches a share of it that falls the whole time the fan is running.

At ten minutes the bed is taking almost everything the air offers. At two hours the outlet analyser reads within a few parts per million of the inlet: the air is going through the column and coming out unchanged. Every further minute of fan power is bought and not used.

Where we stop, and why not later

We stop at 120 minutes. By then the bed is at 1.188 millimoles per gram against an equilibrium of 1.25, so we are leaving five per cent of the bed’s capacity on the table.

Chasing that five per cent takes about another hour and a half of fan power, and it delays the regeneration that produces the actual tonne. Cycles a day is the number that sets output, not loading per cycle. Stopping at breakthrough gives us 6.86 cycles a day per column. Running to 99 per cent of equilibrium instead of 95 would give us fewer.

How the minute was picked

Not from the curve. From the meters.

We instrumented one column for eleven weeks with a separate power meter on the fan and a Coriolis meter on the collection line, and looked at grams of carbon dioxide delivered per kilowatt hour as a function of when the fan stopped. The number is flat between 105 and 130 minutes and falls off a cliff after 150. We took 120 because it sits in the flat part and divides the day tidily with the regeneration step.

That is the whole method. It is unremarkable, and it is the reason our cost per tonne is a number we will show you rather than a number we model.