Processes that determine the color of fire

[RubinLicht]

I never really had a good idea of the mechanisms behind color. why we feel heat, what determines the color of fire, etc. So, today I sat down and did some research. I want to explain the processes here to make sure I got it completely right:

For complete combustion of most hydrocarbons (blue flame, this would be for gaseous hydrocarbons):
the hydrocarbon breaks down into carbon monoxide, however carbon monoxide is only meta-stable, and quickly combusts and turns into CO2. CO2 is formed in an excited state, and releases blue light when its electrons return to a lower energy state.

For incomplete combustion of most hydrocarbons (gaseous hydrocarbons without enough oxygen, and wood):
the hydrocarbon breaks down into carbon monoxide as usual, but there is some residual carbon from this reaction, which appears in the form of soot. The soot is heated up to temperatures that make it emit red-orange-yellow light due to thermal radiation (rather than atomic transitions like before). This overpowers most of the atomic transitions that are still happening, though much more weakly than before.

For odd colored flames like magnesium, sodium, copper, etc.
the heat vaporizes the metal (not sure about this part), and the vaporized metal floats away in an excited state, and when the electrons drop down to a lower energy state, the atom emits a photon of some energy, which determines the color of these flames.

so really, the color of fire is determined by two very different processes, either atomic transitions or thermal radiation. which mechanism is dominant depends on what chemicals are used.

question: When green fire is produced by holding a sample of copper above a yellow flame, the fire is often entire green beyond the point where it contacts the copper. is this because there is more green light being emitted than yellow light? does the soot maybe cool down (and fall out of the red-orange glowing temperature range) after heating up the metal atoms? some other mechanism?Looking forward to any constructive criticism.

 

[Wes Tausend]

Rubin,

I think overall you know more than I, and that was a pretty good presentation.

The difference in color should ultimately be due to temperature and it seems that could 'glow' well beyond a physical particle position, a green particle spectrum frequency glow for instance. So your oxidation reactions must produce some specific temperature frequencies. The atomic transitions and thermal radiation are really the same electrodynamics' thing in a cause-and-effect symmetry.

Our sun glows at about an average 5800 degrees Kelvin at it's surface for example, and our eyes are tuned to peak frequency efficiency at that yellowish-white color. Had we developed under a slightly different star, with a slightly different temperature, then our eyes (our radiation receivers) would be tuned to (resonate at) a slightly different radio frequency. As transmitters, atomic transitions must have a certain 'ring' to them too, a resonation at a certain color.

Wes

 

[sophiecentaur]

the fire is often entire green beyond the point where it contacts the copper. is this because there is more green light being emitted than yellow light?

The light you see from such a flame will be a combination of atomic and thermal emissions. If you look with a simple spectroscope (you really should buy one for about £30 off eBay) you will see that there is a lot of black body spectrum there, too. Our colour vision can be very misleading, as it is when looking at a 'really vivid' rainbow and our subjective assessment of the colour over emphasises those de-saturated colours we see and tends to ignore the background 'white' light.

 

[RubinLicht]

Thanks for all the replies, I'll be taking optics/electrodynamics/thermodynamics courses in the three coming quarters, I'll set a reminder to come back to this in a year :D. I do have a general idea of what both of you are saying, but it seems like I'd appreciate many of the basic ideas you are talking about (cause and effect symmetry in electrodynamics, how exactly eyes are an inaccurate in perceiving reality, etc) after I learn some of these topics with more depth. Cheers!

 

[sophiecentaur]

how exactly eyes are an inaccurate in perceiving reality

The progress of Science took an upwards turn, once they started measuring rather than just looking.

 

[gianeshwar]

Take it further...Recent experiments in Biology of visual perception say that nearly 50 % of what we perceive is made by our brain smartly.

 

[sophiecentaur]

Take it further...Recent experiments in Biology of visual perception say that nearly 50 % of what we perceive is made by our brain smartly.

The whole of our visual perception is a mapping of images in the world into our brains. It's all made by our brain "smartly". There are errors in our perception because of the gross sub-sampling and lack of input information. The inverted 'photographic' image in back of our eye and on the retina is about as far as many people consider about our vision. That's hardly even the start of it.

 

[DrDu]

The green color of a flame when you bring copper into it is due to volatile copper compounds, most prominently halogenides. This is used as a simple test in chemistry on whether a sample contains halogens: Simply dip a copper wire inside and if the flame turns green, there are halogenides inside. Warning: As copper catalyses the fomation of dioxines, this should only be performed ouside or in a fume hood.

 

[diogenesNY]

Of note, when hydrogen (at least in the form of an industrial gas product) burns, the flame is invisible.

In cases in which, say, a hydrogen tank truck overturns on the freeway, punctures and ignites, this causes a harrowing and extraordinarily hazardous situation for firefighters.

diogenesNY

 

[Drakkith]

The inverted 'photographic' image in back of our eye and on the retina is about as far as many people consider about our vision. That's hardly even the start of it.

Ain't that the truth. The optic nerve itself isn't even capable of transmitting the information from every cone and rod cell in the eye simultaneously. There's an immense amount of processing done before the signals even get to the optic nerve. And that's only the start of the process. Vision is a bit like trying to send 60 FPS 1080P motion video over a 28.8k modem. You can do it, but it takes a lot of compression and processing.

 

[sophiecentaur]

You can do it, but it takes a lot of compression and processing.

MPEG is an attempt but it is too 'literal', unlike the brain and the reconstructed image is full of artefacts. The thing about the brain is that it is both salesman and customer and it makes do with all sorts of illusions and artefacts whilst convincing itself that what it is seeing is correct. Also, Shannon's ideas about information capacity , noise and bandwidth are borne out in our perception; you can take a long time to assemble a complete internal image of a scene - much longer than a viewer would be prepared to accept of a moving TV picture - and then your short term memory keeps that part of the image even when you look away.