# Is it possible to power a car with chemiluminescence? [closed]

Chemiluminescence means substances that in a reaction emit light.

I wonder if such a reaction would be possible to power an electric car using an photovoltaic panel lighted by the chemical reaction?

• This seems to be one of the least efficient means of powering a car that I can thing of (well, something that runs on proton decay would be worse). – Jon Custer May 30 '18 at 15:18

An electric car needs a minimum of about 10 kW power.

Light from Chemiluminescence is quite weak compared to sunlight, so to be very optimistic I'll use the solar constant ( $1.362\,kWm^{-2}$ ) as the power per unit area available to the photovoltaic cells. Note when you see photos of Chemiluminescence you'll typically see the images taken in completely dark rooms - i.e. no other light source to hide the effect. This is low intensity light.

Even with this source and near perfect efficiency you'd need something like $7.5\,m^2$ of panel to get enough power to match a (low powered) electric car model.

So I would say that this is very unlikely to be practical. A specially designed ultra-lightweight vehicle (which would not be remotely practical) might possible manage it at a low speed, but I'd be very surprised.

For some perspective, an electric bicycle (i.e. a bicycle powered by a small electric motor) might have a 200 Watt engine at minimum. Remember the light source is low intensity and you just won't get enough power.

Note that solar powered cars do exist. These are impressive but read the small print, as they say, as the headline figures can rely on augmentation by battery power.

• The research in the chemilunescence domain has said all it could or is there enough room for improvement? – user38657 May 29 '18 at 9:27
• maibe with a second harmonic generation would work? – user38657 May 29 '18 at 10:47
• if we add something like sum frequency generation to the project would get more energy? – user38657 May 29 '18 at 14:09
• @user38657 You can't get more energy doing that because energy is conserved. – StephenG May 29 '18 at 15:40
• Yes but it depend on the material that is used as a photoovoltaic cell. Thus increasing chances of electron scattering from orbital. – user38657 May 29 '18 at 17:28

Possible? Possibly. Practically? Certainly not.

You have too many steps between the original energy form (Chemical Potential Energy) and the final, desired form (Kinetic Energy in the car). At each transition there is a loss factor (efficiency). It goes something like this:

• Chemical process produces light ($\eta_c$)
• Fraction of light incident on photovoltaics ($\eta_f$)
• Conversion to electricity ($\eta_p$)
• Electric motor efficiency ($\eta_m$)

So your overall efficiency is $\eta = \eta_c \eta_f \eta_p \eta_m$

If you plug in some real-world efficiencies (between 10-30%), you get an overall efficiency of much less than 1%.

In principle, there's no reason why not. That process can produce electricity, so it could power an electric device.

There's a major problem with scale, though. Even a small electric car has a $15~\rm kW$ engine. To put that in perspective, the average solar panel installation for a house produces only $6~\rm kW$, and that covers most of the roof of a house.