That just shows that the actual problem in a base with self-sustaining food production will be the inverse - not enough CO2 to produce food instead of too much

That just shows that the actual problem in a base with self-sustaining food production will be the inverse - not enough CO2 to produce food instead of too much of it to breath. To solve this you need initial biomass (or CO2) and then after the harvest you burn or otherwise decompose everything you didn't eat to close the cycle and get your CO2 (and some of the energy) back. The easiest way to achieve that initial CO2 concentration would probably be to bring about ~3.5 tonnes of coal per person, burn it with oxygen mined from lunar soil generating energy for your initial base-building (24 MWh of heat about 40% of which you can convert to electric energy) and capturing the resulting 8 tonnes of CO2 for agriculture. Of course you'd bring reasonably pure carbon, not the cheap dirty stuff we dig from the ground with heavy metals and sulphur in it. Also catalyst would be needed to avoid producing carbon monoxide. Another concern is how much volume of air we need in these tubes to keep the atmosphere breathable and suitable for enough CO2 to supply plants. Assume the tube is a cylinder 150m in radius and 10 meter in height (height is horizontal here ;) ). That's 700 000 m3 of air which should weight about 857 500 kg. At 410.28 ppmv CO2 in air we can calculate that 0.0623240117 % of the air mass should be CO2, which is ~535 kg in that 10-meter section of the tube. This is a long way off the 8 tonnes of CO2 needed to grow the plants, but they don't want all of that CO2 at once anyway - so we store the waste plant matter from harvest and burn it slowly over the year to keep CO2 concentration in air breathable. We could also separate the atmospheres for plants and for people to optimize CO2 ppm for each, but that makes math more complicated :)