[Physltest] [Phys-L] Re: Color (was LED mini-flashlight price break)

["Frohne, Vickie" <VFrohne@BEN.EDU>]

Interesting experiments, John!  BTW, the green is a bit nicer if one adds a large amount of white frosting to the dye. Even so, the color does tend toward the yellowish side of green.

Thanks to all on the list who contributed to this great discussion! I look forward to future installments.

Vickie Frohne

-----Original Message-----
From: Forum for Physics Educators [mailto:PHYS-L@list1.ucc.nau.edu]On
Behalf Of John Denker
Sent: Monday, January 31, 2005 3:49 PM
To: PHYS-L@LISTS.NAU.EDU
Subject: Re: Color (was LED mini-flashlight price break)


Regarding the statement

> "RYB works very nicely for kindergarten paints and food dyes.
> These paints and dyes ARE red and blue, NOT mis-named magenta
> or cyan. Kids and cooks easily make orange with RY, green with
> YB, and purple with RB."

Well, if kids and cooks can do this "easily", then
they must be much smarter than the people who make
inkjet printers ... or they have much lower standards.

Here's an easy experiment that everybody can do.  Get
some food coloring.  It's cheap and widely available.
Sometimes the store keeps it near the flour, sugar,
and baking supplies, and sometimes near the spices.

Put a drop of "yellow" in a petri dish.  If you
haven't got a petri dish, you can use a custard
cup or anything with a reasonably flat clear
bottom.  Add water gradually.  Tilt the dish
so you can see a thin layer as well as a thick
layer.

When I do this using Schilling (aka McCormick)
yellow dye, what I observe can be summarized as:

    yellow plus yellow makes orange    [1]

    orange plus orange makes red       [2]

Yeah, I know those equations are heretical.  They
disagree with the color theory that you learned in
third grade.  But do the experiment already!  Tell
us what you see.

Also, those results are entirely consistent with
intelligent post-third-grade color theory.  I
haven't got any spectroscopic data on the dye,
but my observations suggest that first of all it
has very high absorption cross section in the
blue parts of the spectrum ... and that's why
it appears yellow in thin layers.  Secondly, it
has moderately high cross section in the middle
wavelengths, so in moderately thick layers only
the orange survives.  Thirdly, it has very little
cross section at the red end of the spectrum, so
in thick layers it looks red, not black.

If you can express the observed results using
any kind of color wheel, I'm gonna be verrry
impressed.

Also:  Although I've heretofore used the term
"subtractive" mixing of inks (in contrast to
additive mixing of lights) I now realize that
subtractive is the wrong word.  Really it's
multiplicative.  If you run any particular
monochromatic light through any given gel, the
intensity gets multiplied by a certain factor,
call it X.  If you put a second gel of the
same kind in series, you get a second factor
of X.  The fact that X depends on wavelength
makes things complicated.

Mixing lights is an additive process, and to a
first approximation it is linear, as addition
should be.  Mixing inks is nowhere near being a
linear process, so it really shouldn't be called
subtractive.  (Subtraction is linear.)


While we've got the dye-bottles out, let's do some
more experiments:  I observe that the "blue" dye
makes a nice cyan shade in thin layers, and makes
a nice deep "true blue" in thick layers.  That is:

      cyan plus cyan makes blue      [3]

Again this is easy to explain in terms of grown-up
color theory ... but not easy to explain in terms
of the third-grade color wheel.

There are many other cheap & easy experiments you
can do with food coloring.  For instance, if you mix
the so-called yellow with the so-called blue, you
can make an approximation of green.  We can agree
that it is green in the sense that the hue is
green and nothing but green ... but it is not a
very nice green.  It is halfway between green and
black.  In thin layers it looks greenish, but in
thick layers it is black.  (This is in contrast
with the red and the blue, which retain their
brightness over a much wider range of coverages.)

This behavior is entirely consistent with the story
I told to explain equations [1], [2], and [3].
_______________________________________________
Phys-L mailing list
Phys-L@electron.physics.buffalo.edu
https://www.physics.buffalo.edu/mailman/listinfo/phys-l