Put your hand under the light?
Presumably discrete red and green light sources could be distinguished from a yellow one by measuring the frequency spectrum of the light coming off the paper - two peaks or one? Possibly if you have absorbtion leaving yellow that has a kind of a characteristic spectrum, not sure if it's reliably distinguishable from the spectrum of a yellow light source... would depend on what kind of yellow light source, really.
Aheh. This brings up the complications that I've been struggling with. As I understand it, a yellow pigment actually reflects red and green light as well, which is why when you mix yellow and magenta ink you get red - both pigments reflect red light, but together they cancel out one another's ability to reflect yellow, green, magenta and blue. (My source here is Michael Wilcox.)
So, we get the following problem. Shining a single yellow light source, or two sources - one red, one green - onto a sheet of white paper will make it appear yellow, correct?
Now suppose we're using perfect light sources - the yellow source emits no red or green light, the r+g sources are similarly pure in output.
The paper under the R+G light appears yellow. So does the paper under the Y light. If the light reflected by these surfaces is focussed onto a green sheet of paper, the R+G light will make the paper appear green. But what effect will the Y light have?
As far as I can see, the three possibilities are:
- The paper appears black, as there is no green light to reflect. (This contradicts Wilcox's account of the properties of pigments.)
- The paper appears green; the pigment changes the wavelength of the light it reflects. Like I say, I'm not well educated in optics, but this strikes me as unlikely.
- The paper appears dark yellow (not green in any way); the pigment reflects a small proportion of the yellow light. (This is consistent with Wilcox's account.)
Now, given that Wilcox states that a pure pigment, ie a Magenta that does not reflect Red or Blue, is near impossible to make and would be nearly useless to a painter, it would not surprise me to learn that filters are not perfect - meaning that we are not generally familiar with 'pure' coloured lights. But it strikes me that if the third hypothesis given here is invalid, it undermines a central assumption made by Wilcox.
I do hope I've conveyed my meaning here, I'm not used to articulating this sort of thing.
Noodling around... Laser light can be totally monochromatic, so you could actually take a red laser pointer and do this sort of experiment. Or find a yellow laser.
Pigment wise,I'd agree that "pure" pigment seems unlikely, as in something that only reflects a very narrow wavelength band would probably be very dark anyway since there's only a small amount of incoming light for it to reflect.
For pigments there is the possibility of having an absorption-emission which can shift the wavelength (like glow in the dark stuff, or things that glow under black light - they put that in 'whitening' detergents so that whites look whiter as they're actually emitting extra light absorbed and shifted from the UV spectrum). So there could be some sort of shifting; this would depend on the pigment's chemistry and maybe isn't something you're too likely to get by accident. But possible.
I'd expect for most passive (not wavelength shifting) pigment, the reflective band of wavelengths would be broad enough that you'd get a dark yellow effect.
The red laser is one thing - but it's emitting one of light's primary colours. So a yellow laser would be better, since it's emitting a secondary colour, and the object is to find if this secondary colour is the result of the mixing of two primary colours (like mixing magenta and yellow inks), or if it is an 'elemental' colour (you couldn't separate the yellow and magenta in Burnt Umber, for instance).
there could be some sort of shifting; this would depend on the pigment's chemistry and maybe isn't something you're too likely to get by accident
True - I'm trying to figure out the possibilities in getting it by design.
So from this assume we can assess how monochromatic a yellow filter is by shining it on red and green pigments; if they appear dark yellow, it is very monochromatic, and if they appear red or green, our filter is allowing through a wider range of wavelengths which in combination appear yellow.
It all seems to be coming together for me here, and I think Wilcox is validated. Thanks! I can go back to my drawing now :D
Well, there is a wavelength of light that is percieved as yellow, a particularly narrow band as far as colours go, as well as mixtures of red and green wavelenths being so perceived, if I recall correctly. So you can get a yellow laser that is emitting a single wavelength of light percieved as yellow, as well as filtering a white light to get yellow which may be a broader mix of various red-to-green frequencies.
But your monochramaticity test sounds good; it would distinguish the two.
if you google yellow fluorescence you can get some reports of cyan-to-yellow, under specific circumstances, as possibilities of getting such by design. Although I guess the wavelength shift, is limited from shorter wavelength to longer wavelength, so red-to-yellow would be impossible. Perhaps a two-photon absorbtion one-phot emission would allow for it though.