(S-4) The Many Colors of Sunlight

The Spectrum

The colors of the rainbow are the "basic spectrum" from which all the light we see is composed. Although these colors merge smoothly, they are sometimes divided into red, orange, yellow, green, blue, indigo and violet (and other names). Just as various musical sounds contain the tones of the basic scale (often combinations of tones, e.g. chords), so any colored light is made up of its "spectral components. "

Isaac Newton showed that not only can a triangular prism separate a beam of sunlight into rainbow colors (that had already been known), but also that, when a second prism brings the different colors together again, white light is once more obtained. Therefore white light is a combination of all the rainbow colors, and the prism separates its colors because the angle by which a beam of light is bent, when it enters glass, differs from one color to the next.

    [For the same reason, a simple glass lens brings different colors to a focus at different distances. In Newton's time, if an astronomer focused a telescope to give (say) a sharp yellow image of a star, that image would be surrounded by unfocused patches of red and green. Newton thought the problem was insoluble, and proceeded to invent a new kind of telescope, based not on lenses but on concave mirrors, which reflect all colors equally. In later times optics were created which focused all colors together, using a combination of several lenses made of different kinds of glass, and these are nowadays found in cameras, projectors and small telescopes. However, all big modern telescopes follow Newton's idea and use mirrors.]

Perceived color

Even with the rainbow explained, the puzzle of color still baffled scientists. For instance, children experimenting with crayons usually find that a combination of yellow and blue produces green. Is green then a basic color as the rainbow suggests, or is it the combination of two other such colors?

The riddle was solved around 1860 by James Clerk Maxwell (pictured on the left), the brilliant Scottish physicist who also gave us the basic equations of electricity, the ones that predicted electromagnetic waves (see Section S-5). Maxwell showed, while still a student, that two kinds of color existed, depending on whether it was perceived by an instrument or by the human eye:

  1. "Spectral color, " i.e. the colors of the rainbow and their combinations. The amount which each part of the rainbow spectrum contributes to a beam of light can be determined by splitting the beam with a prism.

  2. "Perceived color" reported by the human eye to the brain.

An instrument using prisms ("spectrograph") will reveal that the green color in the rainbow and the green formed by (yellow + blue) are not the same. However, the human eye cannot tell the difference.

Our eye contains three kinds of light-sensitive cells, each sensing a different band of colors--one band centered in the red, one in the yellow and one in the blue. When we see green, the blue-sensitive and yellow sensitive cells are both stimulated; but our eyes cannot tell whether that happens because we see both these colors mixed together, or because we see just one color (rainbow green) that is halfway between those two color bands.

Any color which we see--including brown, olive-green and others absent in the rainbow--is an impression our brain conveys as it combines signals from these 3 color bands. Color-blind persons do not have some types of eye cells, and their world lacks certain colors, or even (for those with only one kind of cell) any color at all.

That is why color TV and color printers can be based on the three "primary colors" red, yellow and blue. These devices do not in any way reproduce the true spectral color of the objects they show, but they are still capable of representing any color our eyes can see.

The Spectrum

Any color discussed from now on will be a spectral color. Two kinds of color distributions are important in nature:

(1) In light emitted from solids, liquids or extensive bodies of dense gas such as the Sun, the colors are distributed continuously. Their exact distribution ("black body spectrum") depends on the temperature at which it is produced--a warm hand radiates mostly in the infra-red, a glowing bar of iron is cherry-red, a lightbulb filament is bright yellow, and sunlight is white-hot.

    [Also of this type is the distribution of microwave radiation left over from the "big bang" when the universe apparently began, a radiation observed by NASA's COBE satellite, the Cosmic Background Explorer. When the observed COBE spectrum was first shown before a meeting of astronomers, it caused a great stir. Observed values generally show some experimental error, but here they were so close to the predicted theoretical curve that the first impression of the viewers was that the presenters had drawn the curve first and then placed their points on top of it.]

  Spectra of slected elements,
  © Donald E. Klipstein

(2) The colors of light emitted by individual atoms or molecules in a rarefied gas are not distributed continuously, but are concentrated in narrow ranges of the spectrum. The colors are characteristic of the type of atom or molecule emitting them, just as the tone of any tuning fork is characteristic of its size, thickness and metal. These narrow ranges are known as spectral lines, because in most spectrographs light enters through a narrow slit, so that each emission appears as a line in the resulting image.

For instance, it is well known that flames--in a fireplace, campfire or burning building--are orange-yellow. A spectrograph will reveal that the color comes from two closely spaced spectral lines, characteristic of sodium, which radiates its light even in the moderate heat of a fire. Wood and most other fuels (but not natural gas, which burns blue) contain small amount of table salt (NaCl), and even a trace amount adds color the flame.

Street lights may contain a small amount of mercury vapor, which emits a bluish light, but no red. Because its coverage of the rainbow spectrum is incomplete, colors seen by such light often appear unnatural. Fluorescent lightbulbs also contain mercury (a spectroscope will show mercury "lines"), but to create softer and more pleasant light (and to put the UV light, usually wasted, to good use), they have a fluorescent coating inside the glass, which absorbs the harsh mercury colors (including UV) and re-radiates them in a more even distribution of color. Neon lights operate in a similar way, with small amounts of other gases producing appropriate colors. Some streetlights also contain sodium vapor, and these can be recognized by their orange-yellow color.

The Wave Nature of Light

Prisms and slits can be used to filter light, leaving only the "monochromatic" light of a single, well defined spectral color. Studies with such light have shown that light propagates like a wave. Its wavelength, the distance from crest to crest, is rather tiny, typically 0.5 micrometers or microns (millionths of a meter).

    [We postpone addressing the question "crest of what? " Early physicists did not know the answer, either--they just knew that when two crests overlapped, the light was brighter, while when crest met "valley" (crest in the opposite direction), the waves cancelled each other, giving a darkening.]

The wavelength determines the extent to which a wave can be confined to certain locations. Because light waves are so short, we can also visualize a light wave limited to a well-defined beam. However, outlines begin to blur when we look at small objects through a powerful microscope, magnifying several thousand times, because light waves cannot define details smaller than their wavelength. That is where electron microscopes become useful, using not light but beams of electrons.

A variety of instruments allow physicists to actually measure the wavelength of light. The one most likely to be used by students is a diffraction grating, a plate ruled with fine parallel grooves, with a constant distance between each one and the next. Inexpensive plastic gratings are available, pressed from a metal grating and mounted on cardboard frames like photographic slides. The incoming wave resonates with the spacing between the grooves and some of it is deflected, by an angle which depends on the wavelength, and knowing the angle and the spacing allows the wavelength to be calculated. Thus gratings can separate a beam of light into its colors the way prisms do, and they are often used in spectroscopes.

    [Lit from the side with a reflecting surface behind them, gratings will shimmer in many colors, making them a popular item of costume jewelry. The same process is responsible for the shimmering of laser disks used in recording music and computer data, which also contain many narrow parallel grooves.]


19th century scientists, in particular Robert Bunsen (1811-99) and Gustav Kirchoff (1824-87), observed and catalogued the spectra of many substances. That provided a tool for analyzing the composition of metals and other substances, still widely used.

The Sun, too, emits spectral lines. The ones noted first were dark lines (named Fraunhofer lines after their discoverer), suggesting increased absorption of light, not increased emission. Cool atoms absorb the same wavelengths as the ones they emit when hot--for instance, light from a filament bulb, shining through a tube with mercury vapor too cool to emit light, will develop dark lines at the same wavelengths as those emitted by hot mercury vapor. In the case of sunlight, it turned out that the absorbtion occured not in the Earth's atmosphere (as one might have guessed) but in the Sun's.

In addition, however, sunlight also contains many bright emission lines, characteristic of hydrogen, calcium and other elements. One yellow line, discovered in 1868, was first identified as the yellow line of sodium, but it did not have the proper frequency and did not fit the spectrum of any other known substance. The British astronomer Norman Lockyer finally proposed that here was a new substance, unknown on Earth, and he was right: "helium" (from helios, the Sun) was identified in terrestrial material by William Ramsay in 1895 and was later isolated by him.

Next Stop: (S-5) Waves and Photons

Back to Index

Author and curator: David P. Stern
Last updated 20 August 1999