The nature of light and the nature of color as it is produced in gemstones was considered briefly. Thus far, the only consideration that has been given to color, insofar as testing instruments are concerned, is the use of the dichroscope to detect double refraction. This assignment is concerned with the spectroscope, which, in qualified hands, may provide positive identification of a gemstone based on close observation of certain parts of the spectrum that the stone selectively absorbs. When used in conjunction with other tests, this instrument may provide conclusive information when examining transparent rough pebbles or stones mounted in such a way that regular instrument tests are impossible.

The spectroscope has developed slowly in its importance in gem testing, for there are few gem varieties that have diagnostic spectra. In recent years it has been used more frequently, because new uses for which it is valuable have been developed. Among the important gemstones ruby, red garnet, some blue and yellow sapphires, jadeite, zircon, alexandrite and emerald have characteristic spectra, but the spectra of synthetic ruby and emerald are indistinguishable from those of their natural counterparts. Its limited application led to concentrate efforts on the development of such tools as dark-field illumination combined with binocular magnification, the polariscope, and the spot method for R.I. determinations.

Elsewhere in the same general period, emphasis was directed to different methods. At this time, serious attention was given to the spectroscope. Those who used it most and and who contributed most to the literature covering its use as a gem-,testing instrument were LW. Anderson and his associates. Studies reported by a prominent American mineralogist, Edgar T. Wherry, in the American Mineralogist in 1929 were followed in the late 1930's and in the 1940's by the British efforts. Interest increased when several new kinds of artificial coloration defied other means of detection.

In 1664, Sir Isaac Newton proved that sunlight is made up of mixed colored light consisting of rays of different refractive ability, and that for each degree of refraction there is a corresponding pure spectral color. To make his discovery, he passed sunlight through a glass prism and noted the characteristic "rainbow" that was projected on a screen.

Two other physicist's should also be mentioned in any discussion of spectroscopy, since their names are used to designate light-wave measurements and to define dispersion. The name of A. J. Angstrom (1814 - 1874) is used to express a unit wavelength of light (one ten-millionth of a millimeter; the Angstrom unit, expressed here as AU.). Joseph Fraunhofer (1787 - 1826) was the first to observe the sun's spectrum with various kinds of glass prisms and to prove the existence of innumerable fine vertical lines, indicating that some of the wavelengths in sunlight are absorbed. To the most prominent of these lines he gave the letters A, B, C, etc.; this was in 1817.

As more effective instruments became available for observing spectra, investigators proved that these dark lines in the sun's spectrum were caused by the absorption of light by the atmosphere surrounding the sun. Later, most of the more prominent lines were correctly ascribed to precise elements in the spectrum. Other workers noted that these elements (e.g., sodium, which is responsible for Fraunhofer's line D), when rendered incandescent in the laboratory, produced not dark absorption lines but bright, or emission, lines in the absence of the sun's atmosphere. The reason is that a gas that radiates a line spectrum will absorb the lines that it radiates at the same temperature. Thus, the emission of incandescent gases at the sun's surface are partially absorbed by the atmosphere.

The difference in refractive index between the B and G lines is taken as the standard for determining the dispersion of gemstones, although the interval between the C and F lines has been adopted as the standard for measuring the dispersion of optical glass.

As explained in the Colored Stone Assignments, the color of most gemstones is caused by an element, usually metallic, that is present either as an impurity or as a necessary component of the chemical makeup. The first ran to note that light passing through a cold solid may be selectively absorbed to the extent that certain lines or bands will appear in the spectrum produced by a spectroscope was David Brewster, when, in 1833, he observed the three bands that are typical of cobalt.

In 1866, Sir Arthur Church became the first man to observe a gemstone through a spectroscope and note the characteristic absorption lines in both zircon and almandite garnet. In a letter describing this discovery he stated,

"You are aware that the orange jacinth, a variety of zircon, is very precious and that the essonite, or cinnamon garnet, is constantly sold for it. Curiously enough, the cinnamon garnet (a lime garnet) has no conspicuous absorption bands, and so the spectroscope may be brought to bear upon the discrimination of these two stones."

Not only did Church note two of the most easily seen absorption spectra, zircon and garnet, but he also recognized the implications of his discovery in gem testing. He did not, however, turn his attention to other stones, or at least he reported no other spectra, probably because of the quality of the instruments available to him and the fact that his light source was almost certainly sunlight, which shows so many absorption bands that they mask all but quite heavy bands in the spectra being studied. The effect of Church's report was that for more than fifty years writers of books on precious stones assumed that no other stones had characteristic spectra. And since other instruments had been perfected that would identify garnet and zircon, the spectroscope was given little attention.

Meanwhile, spectroscopy had become a recognized branch of physics with widely varying applications; for instances through the use of spectroscope with sunlight, both invisible ultraviolet and infrared wavelengths were discovered. In addition, spectroscopes were used to discover helium, thallium and indium; to record a long list of elements known to be in the sun's atmosphere; and to analyze light from the stars. Because spectroscopists often publish the results of their investigations in obscure chemical and other journals, only a truly dedicated gemologist would take the time to research their findings.

Requirements For Gemstone Spectroscopy

The use of a spectroscope for gem testing is somewhat complicated by the varied shape and size of the stones encountered. Spectroscopy has wide application in industry, and almost without exception the instruments used are designed to accommodate specimens that are prepared to a specific dimensions with parallel sides. Gemstones, on the other hand, cannot be altered for testing, and faceting is utilized for the express purpose of preventing the direct transmission of light through the stone. Add to this the characteristics of uneven coloration, dichroism, and surface coatings or dyes and it becomes obvious that it may be difficult to set up a stone to transmit enough light so that a thorough spectroscopic analysis can be made.

There are three requirements for effective gemstone spectroscopy:

  1.  a good spectroscope;
  2. a light source, preferably of variable intensity;
  3.  a convenient means of transmitting the light through, or reflecting it from the stone,

so that it passes into the spectroscope slit without heating the specimen.

  1. Types of Gem-Testing Spectroscopes

    The most convenient and least expensive type of instrument is known as the hand spectroscope (Figure b). It is fairly small and has the following essential parts: (a) an adjustable slit for admitting the light that has passed through the specimen; (b) an adjustable draw tube for focusing on the spectrum produced; (c) either a train of prisms made of alternating crown and flint glass, to produce a dispersion of the light of approximately 10°, or a diffraction grating that produces a spectrum from fine parallel lines engraved on a glass plate; and (d) an aperture for viewing the spectrum. In addition, both prism and diffraction grating hand spectroscopes may have attachments or be designed for other functions. For instance, some produce two spectra, so that a comparison can be made between the spectrum from the specimen and the unaltered spectrum from the light source. Others have a superimposed wavelength scale, so that approximate measurements in Angstrom units can be made. Many other types exist, ranging from very expensive laboratory bench models to diffraction grating hand units that are more toy than instrument. Because of high cost and the difficulty of manipulating the instruments to direct the best light from the specimen into the slit, most of these are impractical for the gemologist.
  2. Light Sources

    Although various kinds of light sources are used for gemstone spectroscopy, the one chosen must fulfill certain functions. First, a sufficient amount of light must pass through or reflect from the specimen to illuminate the slit of the spectroscope adequately. Also, the light should contain no absorption lines nor fluorescent lines to confuse the observations; for this reason, neither sunlight nor fluorescent light is acceptable. Perhaps the best source for the beginner is an ordinary filament-type incandescent lamp. Since this type of light is available in a wide range of wattages and styles, the type chosen depends on how the third requirement listed at the beginning of this section is to be met.

    Illuminating the Specimen

    Regardless of the kind of light selected, provision must be made for transmitting as powerful a beam as possible through the specimen without allowing any extraneous light to reach the slit of the spectroscope. At the same time, direct light must be prevented from reaching the eye, so that it will retain maximum sensitivity. Light reflected from the surface is used only with opaque or semi-opaque specimens such as turquois and some jadeite; otherwise, light should come through the specimen, if it is sufficiently transparent. Following are a number of suggestions, all of which have been used successfully. The advantages and disadvantages of each are indicated.

One of the simplest arrangements is to install a 50-watt bulb in a reflector base similar to that incorporated in the Gemolite or Diamondscope. Over the top of the base is placed a platform with a revolving plate containing holes of various sizes. The stone is placed in a hole of appropriate size and the plate revolved until it is directly over the light. For a number of years models of both the Gemolite and Diamondscope have had iris diaphragms that can be closed down to a small aperture over which the stone may be placed. With light reaching the spectroscope slit coming only from the stone, its spectrum may be studied. This arrangement has two serious drawbacks when used with gemstones that do not have pronounced spectra. First, the light source is not strong enough to illuminate dark- colored stones; secondly, even though the wattage is low, a stone being studied for any length of time will become heated, producing an adverse effect on the absorption lines of some stones and risking damage to the specimen. Stronger lamps only increase the heat problem, of course.

Using a simple 300 to 500 watt slide projector, the Lab has had success in placing the stone table down on black velvet and projecting light through one side of the pavilion, allowing it to be totally reflected from the inside of the table and out the opposite side of the pavilion. Less reflection from the background is encountered if the stone is placed on a small pedestal over a dull-black background. The spectroscope is then held in the hand and the brightest reflections studied.

Using a 150 watt "baby spot" or other strong lamp and condensing the beam with a simple lens, this same arrangement may be used, provided the beam is cooled with a round, flat-bottomed flask of water. If the light is strong enough, this setup may be used to example turquois and other opaque stones whose characteristic spectra may become apparent in light reflected from them. The "cyclospot" a low-voltage "intensity" lamp, may be used for the light source, source, but it is considerably less powerful than a slide projector lamp. It also may be used in conjunction with a simple microscope, as described below, although, again, it is not sufficiently strong for dark-colored stones.

The arrangement suggested by Anderson is to house a 500 Watt projection lamp in an asbestos-lined and ventilated (louvered) box. A slide projector of this or greater wattage is excellent. The light is projected through a flat-bottomed water-filled flask onto the mirror of a microscope. The stone to be studied is placed in the path of the light, the eyepiece of the microscope removed, and the spectroscope placed over the tube. Using a low-power (i.e., 1x to 1½x) objective, the column is adjusted until the tube is filled with an "out-of-focus" colored light from the stone, and the spectrum can then be studied. This arrangement provides the agential power of 500 watts, and the flask of water prevents heat from reaching the stone. Any type of microscope with a low-power objective will suffice for this purpose. This method permits the study of emission lines (seen readily only in ruby and red spinel), but is not ideal for certain observations requiring more direct illumination; for this purpose the spectroscope may be brought to bear directly on the stone without using the microscope tube. An adaptation of the above method consists of a non mineral microscope with a 1½x objective placed horizontally in front of a 500 watt slide projector with the spectroscope on a stand designed for it by the manufacturer. The stone or item of jewelry is held in a pair of Gemolite tweezers and the light is controlled by the iris diaphragm, so that no extraneous light reaches the instrument's slit.

The Gemolite or Diamondscope may be used to analyze the more common stones, by removing one eyepiece and observing the stone held in the light field over the diaphragm with the spectroscope resting on the microscope's tube. Direct observation of the illuminated stone also is possible with these instruments. The stone should be held by the stone holder in a path of direct illumination, rather than placed in the iris opening, because of the adverse heating effect of the lamp. The low power of the lamp in the base (30 watt) limits the usefulness of this method, however.

In laboratories that use a zirconium-arc or carbon-arc lamp, the light either is condensed with a lens and projected through a stone held by a stone holder in such a position that the eyes are protected from the direct glare, or projected through the pavilion and out the other side by reflection from the table. The spectroscope is held either in the hand or attached to an upright stand provided for the purpose.

Mounted stones that are difficult to identify by other means are sometimes identified readily by spectroscope. If a mounting is open, a ring may be placed table down over a light source. A light shield blocks extraneous light but permits light to enter the stone. This is the light that passes through the slit of the spectroscope. If the stone cannot be examined in this manner, a strong beam may be directed down on the surface at about 45° and the reflection thus examined.

The problem with all such arrangements of independent light source and hand or stand-held spectroscope is awkwardness, inefficiency, and usually the lack of results that lead to the incorrect assumption that spectroscopy is not a valuable testing tool.

To overcome this problem, the late Lester B.Senson, Jr., designed a unit incorporating a variable high intensity light source that can be moved to various positions. In addition, a moving arm holds the spectroscope and the wave-length scale is lighted. Transmitted light passes through a prism, to prevent the heating of the specimen. To restrict the size of the spot, an iris diaphragm supplements the focusable light. The light source is mounted on a flexible arm, so that light may be transmitted or reflected from the stone. The light may be turned through 90° in the horizontal, as well as the vertical; in addition, the spectroscope may be moved through a 90° are.


If a prism-type hand spectroscope is directed toward a bright sky, a continuous spectrum will be seen. The slit adjustment knob should always be manipulated with the left hand. When held in this manner, the Beck instrument will show the red on the right and the blue on the left, and the wavelength scale will be clearly illuminated. If the slit is closed gradually, a series of lines crossing the spectrum vertically will begin to appear and will be sharpest just before the slit closes completely and all light is shut out.

When these Fraunhofer lines are at their sharpest, pull the draw tube out about 1/8 inch and note that the lines in the red and the strong line in the yellow (Fraunhofer's D, or sodium, line) become even sharper, while the lines in the blue become sharper if the tube is again pushed in.

The scale marker of instrument (the line just below the scale near 5900 A.U.) should be adjusted so that it coincides exactly with the D line at the focus setting being used. To do this, the lockout must first be loosened slightly and then tightened after setting.

Horizontal lines, which may appear just before the slit is closed completely, are caused by dust on the jaws of the slit. Unless these lines are very obvious, they do no harm; in fact, they are useful in determining whether the slit is open too much when examining light-colored stones. If the lines persist when the slit is opened considerably, the jaws may be cleaned with a wooden toothpick. A small, inexpensive Japanese instrument has a built-in dust cover that prevents the jaws of the slit from becoming dirty. However, this instrument has not proved satisfactory; it is particularly poor for studying absorption lines in the blue part of the spectrum.

If you are using a hand spectroscope, it is best to work out some method of holding it in a steady position. There are several ways in which this can be done. A local chemical-supply firm is sure to have metal stands and test-tube clamps. The clamp may be mounted on the stand and clamped over the instrument, so that it may be directed at any angle from horizontal to vertical. With the spectroscope held steady, it is possible to study a spectrum at length. Stands built expressly for either Beck model are inexpensive.

Another method is to build an inclined trough from a block of wood, with the trough at 30° to 45° from the horizontal. The light source is directed toward the gem at the same angle to the horizontal, directly opposite the slit end of the spectroscope resting in the trough. If a clamp stand is used, the same arrangement is a possibility.

A second, and more often useful, setup is to mount the spectroscope in a vertical or nearly vertical position with the light source beneath it. In this arrangement, the light source is masked with black paper or metal with a hole or holes cut through, so that light passes only through the gem. The light source may be whatever is handy, but a high intensity source is recommended.

If the spectroscope is directed at an ordinary Mae Mazda lamp bulb, a continuous spectrum will be seen, and it will not show any absorption lines. On the other hand, if the light from an ordinary fluorescent lamp is observed, four distinct fluorescent lines in the yellow, green blue and violet will be noted; these are caused by mercury vapor.

It will be noticed that the colors of the spectrum in the prism spectroscope are not equally distributed, since the red end appears compressed and the blue end extended.

The compact unit was designed for convenience and efficiency. The stone is either held in the mechanical stone holder mounted on a post, or laid directly over the opening of the iris diaphragm covering the table of the instrument. The position in which the instrument is used for the majority of tests is with the lamp horizontal in its lowest position, as shown by the uninterrupted lines in the sketch labeled B. In this position, the gemstone is protected from the heat of the intense beam by the beam's passage through the right-angle prism before it is transmitted through the stone. The spectroscope may be moved through a 90° are to the position in which light transmitted through the specimen appears brightest (sketch A). The table and the light arm also may be moved to achieve enough light intensity to permit the spectrum to be studied. The light itself may be varied in intensity from dim to brilliant. If possible, the entire spectrum should be illuminated evenly by choosing a broad beam coming from the stone. Some beams are so small that they appear as small lighted circles that fail to cover even the narrow dimension of the spectrum on an otherwise dark scale. The light beam should be directed at the center of the table, after reflecting through the prism; this should be done with the iris open. The beam should be focused to as small and intense a beam as possible. When the iris is closed down to an opening smaller than the stone, the lamp should be adjusted to center the beam on the opening.

If the stone is opaque or if for some reason the direct transmitted light is not satisfactory, there are other possible positions for the lamp. Usually, light is directed at the pavilion of the at an angle of about 30° to 45° from the horizontal. The beam is reflected from the table and escapes through the opposite side of the pavilion. The spectroscope is directed at the angle at which the reflected beam emerges from the pavilion. This method is helpful for very pale stones, to obtain a greater path within the stone in which absorption can occur. A similar position of the lamp is used if an opaque, such as turquois, is to be studied. The beam is reflected from the surface of the stone into the slit.

If the spectroscope is directed at incandescent light that has been transmitted through a synthetic ruby and focused on the most intense beams with the slit nearly closed and the draw tube pulled out slightly, the typical absorption lines and bands produced by this stone should be seen (see accompanying spectrum diagram). If the light beam has not been scattered during transmission, several absorption lines and a broad absorption band will be visible. In the wavelength spectroscope, this band extends approximately from 6000 to 5000 A.U. In the red, two lines in very close proximity will be seen at about 6942 and 6928 A.U., in addition to a moderately strong line at 6592 and a medium-strong line at 6680 A.U. Beyond the area of broad absorption, there will be a line at 4685 and a very close pair at 4765 and 4750 A.U., with general absorption of the violet. If the light beam has been scattered, a bright fluorescent line, instead of the pair at 6942 and 6928 A.U., will be seen. It is suggested that the beginner experiment with his instrument and lighting setup until the characteristics of the ruby spectrum can be recognized easily and obtained quickly. Pale stones will have correspondingly weaker absorption lines.

Another stone that has a predictable spectrum is almandite garnet. If the stone, selected for study is very dark, the slit will have to be adjusted in order for the instrument to pass enough light to form a satisfactory spectrum. If successful, a striking absorption spectrum will be seen, with the strongest line at 5050 A.U. in the blue-green (see accompanying spectrum diagram). It will be noted that the spectra for ruby and garnet are so dissimilar that a positive separation can be made without additional tests, other than magnification to determine whether the ruby is natural or synthetic.

The wavelength-scale model is not recommended just because of a need to read wavelengths, but because the dispersion (and thus the spread of the spectrum) seems almost ideal for a hand instrument. In actual practice, the more common colored stones do not require precise wavelength readings of the lines; however, there are times when it is essential to know the position of a line within fifty A.U.

When it is necessary to study the blue end of the spectrum in such stones as natural blue sapphire, the use of a blue filter to eliminate all but the blue and blue-green part of the spectrum is recommended. A glass flask containing a solution of approximately seven ounces of copper sulphate dissolved by gentle warming in one pint of water makes an ideal filter. If kept stoppered, it will last indefinitely. Not only are faint lines in the blue end of the spectrum made sharper, but fluorescence lines in ruby, synthetic ruby and red spinel are displayed clearly. Of course, it should not be used if the entire spectrum of an unknown stone is being studied.


Considerable space has been devoted to the use of the spectroscope and to the necessary conditions for its use all of which may seem unduly lengthy. However, since it requires considerably more experimentation than other instruments, even the most proficient spectroscopist may have to try several different arrangements with some stones before the most satisfactory one is found, unless a complete unit designed for the purpose is acquired. The problems caused by different cutting styles and varying degrees of transparency, plus the additional complications caused by mountings and pleochroic character, make such experimentation necessary.

It has been explained previously that certain elements cause the colors in certain gemstones; likewise, the absorption lines and bands in gemstones are attributable to specific elements. It will be recalled that chromium is responsible for the colors of a number of red and green stones; e.g., ruby and synthetic ruby, red spinel, the rare chrome-pyrope garnet, pink topaz, green jadeite, alexandrite and emerald. In combination with iron it is partly responsible for the colors of demantoid garnet and chrome diopside. The absorption lines caused by chromium are mainly fine ones in the red, together with a broad band in the center of the spectrum in the yellow and green. The position, width and intensity of this band largely determines the precise hue of the stone.

Iron is another important color-causing agent in gemstones. There are two categories of iron spectra: that caused by ferric (trivalent) iron and that caused by ferrous (divalent) iron. In the former, the important stones are green and blue sapphire, demantoid garnet, peridot, sinhalite and blue spinel. In addition, it is thought that the spectra of iolite, yellow orthoclase, some jadeite, enstatite, green tourmaline, aquamarine, andalusite and kornerupine are partly due to iron, but whether ferric or ferrous is not known. In general, the fines and bands caused by iron are found in the blue and green part of the spectrum; and, with the exception of the exceedingly sharp line at 5060 A.U. in brown enstatite, the bands are not sharp and clear but tend to be rather broad and fuzzy.

Cobalt, which is used to color synthetic blue spinel and cobalt glass, is responsible for a distinctive series of heavy absorption bands in the yellow, green and blue-green. No natural gem material is colored by this element.

Manganese is responsible for the color in rhodonite, rhodochrosite and spessartite garnet. Lines attributed to it in the spectra of these stones are centered in the blue, violet and even in the ultraviolet; the latter must be studied with other than a hand spectroscope.

Uranium is believed to be the cause of the spectral absorption in zircon.

Didymium, a combination of two rare-earth elements, is responsible for the characteristic spectrum of yellow apatite, and its presence may sometimes be detected in danburite and yellow sphene, color of which is thought to be caused primarily by structural differences rather than tinctorial agents. This idea is bolstered when it is realized that bombarding a pale-yellow to colorless diamond in the cyclotron or atomic pile (deuterons or fast neutrons) may cause the color to change to green and that subsequent heat treatment may further change it to brown or yellow.

In addition to its usefulness in the identification of mineral varieties and species, the spectroscope has more recently become important in detecting two types of treatment: atomic bombardment of diamonds to change the color to green, brown, yellow or pink, and artificial staining of green jadeite and jadeite triplets. Although some natural brown diamonds with green fluorescence will show a pair of absorption lines at 5040 and 4980 A.U., their presence is rare in other colors of untreated diamonds. (Untreated green diamonds examined to date show a strong 5040 but no 4980.) However, the process of bombarding an off-colored yellowish diamond to change the color to green seems to produce these two lines automatically, since they are seen in most such stones. When dark-green treated stones are heated to turn them brown or yellow, another line appears at 5920 A.U. This line has not been reported in natural brown and yellow diamonds; therefore, its presence may be used to identify these stones.

The dye used to date to make white or very light jadeite a more pleasing green is detectable with a spectroscope. Natural fine green jadeite shows three lines in the red portion of the spectrum. The dyed material encountered thus far is characterized by one smudged-appearing band at about 6450 to 6800 A.U. (see accompanying spectrum diagram).

There are some gem materials in which the use of the spectroscope is particularly useful in identification for one or more of several reasons. The line or lines in the blue region of natural blue sapphires speeds the identification of sapphire-set bracelets, necklaces, guard rings or dark solitaires that are difficult or impossible to identify, if only magnification is available. It is also useful with orange sapphires, in that the synthetic shows chromium lines in the red. Natural yellow sapphires from Sri Lanka and Australia show from one to three lines in the blue, in contrast to the synthetic.

Jadeite carvings and cabochons that may prove difficult to identify by ordinary means are identified beyond question by the strong line at 4370 A.U. (usually masked in very dark stones).

Zircons of any color often show a line in the red at 6535 A.U. and some (even colorless stones) show many more, as indicated in the following diagrams. Very low-property green zircon may show only a very faint diffused band, but some green stones have a number of lines.

Almandite and ruby have been mentioned previously.

Many other stones show characteristic spectra, although they are perhaps slightly less often called upon in identification. These include emerald, alexandrite, peridot, demantoid, enstatitc, blue spinel and many others.

The following reproductions are useful to understand the spectrum of different gems:

If no star is used, the spectrum represents one that has been recorded in any gem laboratory, but either rarely seen or too few stones examined to be able to indicate value in testing. Some two and three-star spectra represent very few stones examined in rare species or colors, but the presence of certain essential coloring agents that cause characteristic absorption makes for a safe assumption of diagnostic value.