# Visible wavelengths of air fluorescence data - needed for school demo laser telephone design

I'm seeking information about fluorescence in visible wavelengths (390-700nm) of any of the main constituents of air ( $N_2, O_2, CO_2, H_2O, CO, etc.$) that can be excited with a (hopefully low-cost?, ?COTS, ?diode) laser.

Are there any laser output wavelengths, and or laser types, that can do this? (I guess that they would be with output possibly in the UV range to get a Stokes shift down into the visible range.)

What output power level would be required? (low mW hopefully.)

Would the laser output still be able to be PWM? or OOK? modulated at sound signal frequencies?

Would the modulated signal still be able to be detected, without being interfered by the fluorescence? Is fluorescence decay time going to be a problem?

I'm part of an community science group that's designing and building a laser telephone to take to STEM demonstrations at schools in this International Year of Light 2015, and want the laser to fluoresce any constituents of air, so that the children can see laser's path, as opposed to relying on Rayleigh scattering from dust particles. The device is in a box with a transparent perspex cover, and is usually demonstrated in a classroom with fluorescent room background lighting. I'm guessing the laser's output wavelength/s don't need to be in the visible spectrum.

But if they are visible- and if two or more output or fluorescence wavelengths of the laser are close together, they could beat together, which could add to the "wow and awe" factor of the device.

What wavelength difference would they have to be for the beating to be observable visually?

Where should I be looking for publicly accessible data to answer my questions? I'm not a student at a university, and I don't have access to data sources that a university student or researcher has.

• The brief answer is "no" - you cannot excite air molecules to fluoresce in the visible spectrum with a low power COTS laser – user56903 Jul 30 '15 at 7:51
• Plus wavenumber difference for beating should be in single inverse centimetres range or lower (that would be single tenths of nanometres in wavelength domain in the middle of visual range), and stable over time. Emission lines should be appropriately narrow, too. This would probably mean gas lasers with some sort of stabilization. There may be some femtosecond laser set-ups that would do if you did a lot of tweaking. Then the geometry of beams should be extremely stable (laser-cavity-stable I'd say). All of this would be really expensive. – Jarosław Komar Jul 30 '15 at 18:40