4 ms·

I think that you're right. AFAICT artificial illumination for growing plants makes sense for growing staple foods in deep space if crewed missions ever get out

I think that you're right. AFAICT artificial illumination for growing plants makes sense for growing staple foods in deep space if crewed missions ever get out there, or high-value specialty crops (delicate fruits/greens/herbs, cannabis) near terrestrial demand centers. It's worthwhile to grow tomatoes artificially in Edinburgh in the winter if you have customers who will pay the premium over bred-for-transport-durability tomatoes shipped in from abroad. A really good diode light source might have a wall plug electricity-to-photons efficiency of ~50%, and the best rooftop solar modules achieve a bit over 20% light-to-electricity conversion efficiency. I would really be curious if anyone can provide citations to the effect that sunlight -> PV -> LED light -> plants is more efficient than sunlight -> plants. You'd have to get an enormous benefit from wavelength shifting, enough to offset the ~90% energy losses from sunlight -> PV -> LED light.

Retric 10y ago
We have 46% efficient solar cells that are not currently cost effective. 45% of sunlight is in the useful wavelengths for plants, but the important comparison is their peak effecency with the optimum wavelength of 30% vs there theoretical limit with sunlight of 11% which is a ~3x boost under optimum conditions. Another consideration is leaf peak obsorbtion is around 10,000 lux though trees for example capture light through multiple layers. So, some plants are going to be more efferent at lower light levels. (this is also why trees don't look like a mushroom with one big leaf for maximum light capture.) So, ignoring costs you can come surprisingly close to break even in terms of energy budget. Though we are likely better off going with GMO plants to capture more light than build millions of acres of solar farms just to break even.

Retric 10y ago

We have 46% efficient solar cells that are not currently cost effective. 45% of sunlight is in the useful wavelengths for plants, but the important comparison is their peak effecency with the optimum wavelength of 30% vs there theoretical limit with sunlight of 11% which is a ~3x boost under optimum conditions. Another consideration is leaf peak obsorbtion is around 10,000 lux though trees for example capture light through multiple layers. So, some plants are going to be more efferent at lower light levels. (this is also why trees don't look like a mushroom with one big leaf for maximum light capture.) So, ignoring costs you can come surprisingly close to break even in terms of energy budget. Though we are likely better off going with GMO plants to capture more light than build millions of acres of solar farms just to break even.