The list of links below is a brief but by no means exclusive or exhaustive list of papers or where to start.
For the SF Sci Hackday and #BASF17, design and implement an experiment and simulation of an optical centrifuge for O2, CO2, CO, and related ionic species.
Create a simulation of an experiment that would show if laser light can separate O2 from CO2 by application of circularly polarized laser light.
Can circularly polarized light without the additional setup of a true optical centrifuge (both LCP and RCP in a single beam, trapping molecules then rotating them) separate O2 from CO2 or CO, or their related species by increasing angular molecular momentum?
Create circularly polarized wavefront from a laser light source, linear polarizer, and 1/4 wave plate. Have circularly polarized light directed through vacuum setup on a gas mix of O2, CO2 and possibly Ar. Create a magnetic field on half of vacuum chamber and lastly measure the output gasses for O2 content. Repeat experiment without laser beam. Which test had more oxygen in one container?
Optical Centrifuge: Use of alternating light waves formed in a circular polarized manner that don't resonate with the target (gas) molecules, deposits energy on to the molecule such that electrons vibrate the molecule in alternating or rotating direction (pending on experiment setup) to raise spin to trap molecules. Trapped molecules can then be moved by magnetic fields easier due to enhanced molecular spin. This technique has been used to separate heterogeneous gasses and also split gasses to their constituent atoms.
How does this compare with IR or microwave radiation operating on resonate wavelengths? IR/microwave would be larger waves, and vibrate more molecules. Per Shannon A. Fiume: My best estimation is that resonate wavelengths would heat the molecule and likely cause some amount of rotation among other random vibration, but not controlled enough to affect the spins to enhance paramagnetic effects. Whichever type of approach (resonate or non-resonate) is most efficient at increasing separation distance of O2 and carbon will ultimately be utilized. Both should be investigated.
A chirped pulsed amplified circular polarized laser light can raise total angular momentum of the atoms, it's likely one could use a laser wavelength in a VUV wavelength (either a resonate or non-resonate wavelength of CO2) to create an optical centrifuge to dissociate CO2, with likely a very low yield. The yield to energy consumed ratio is likely to be too unfavorable to justify the cost to build a unit.
However, it may be possible to use this theory to ad-on method to increase separation distance from newly formed solid carbon and oxygen, both resonant and non-resonate laser wavelengths should be investigated at that time.