Origine de la couleur dans les gemmes

Parmi les critères d'appréciation des gemmes, la couleur occupe une place de choix. Cependant chaque personne distingue les couleurs avec différentes nuances, et les couleurs peuvent changer avec la nature de l'éclairage. La compréhension des mécanismes colorant les minéraux et plus particulièrement les gemmes est d'un intérêt fondamental dans la détection des traitements subis par celles-ci.
L'équipe de gemmologues nantais travaille sur ce sujet depuis de nombreuses années afin d'établir un catalogue des causes de la couleur dans chacune des gemmes. Pour plus de détails, consultez nos articles à ce sujet.

Tout d'abord, une gemme peut avoir une couleur intrinsèque due à un des éléments constituant la charpente du cristal : c'est le cas du péridot (Fe,Mg)2SiO4, coloré en vert par le fer (Fe2+). La gemme est alors idiochromatique. Cependant la majorité des gemmes sont intrinsèquement incolores : les variétés colorées sont dites allochromatiques car leur couleur est due à la présence d'impuretés ou de défauts, ou d'une combinaison des deux. C'est le cas par exmeple du rubis, Al2O3, coloré par des impuretés de chrome.

Aussi, beaucoup de gemmes sont colorées par plusieurs processus à la fois. Un exemple courant mais pourtant peu connu est le grenat rouge almandin. Un almandin "pur", coloré uniquement par le fer (Fe2+ en coordination cubique déformée dans la structure du grenat ,voir plus bas) donne en fait une couleur pourpre, qui est la couleur de la rhodolite d'Afrique de l'Est. C'est seulement lorsque cette première origine de la couleur est combinée à une deuxième que la couleur rouge apparaît: il faut un transfert de charge Fe2+-Ti4+ qui donne par lui même une couleur jaune à orange aux grenats. La superposition du pourpre et du jaune résulte en une teinte rouge, classiquement associée aux grenats.

Cette page propose un aperçu des différentes causes de la couleur dans les gemmes.

1. Isolated metallic ions impurities

Metals belonging to the firstseries of transition elements are good chromophores: Ti, V, Cr, Mn, Fe, Co, Ni, Cu, as well as rare-earth elements. When incorporated in crystal structures as positive ions, and in low concentration (typically less than 0.1%), they may absorb some part of the visible ligth, giving coloration to the stone.

Several parameters affect the color:

a· Nature of the ion

Each ion is able of a color panel, and not a single color. The famous green color of emerald is due to Cr3+. Hence, when viewing an emerald-color gemstone, one may believe its color is due to chromium. However, V3+also gives an amerald green color to beryl, when Mn3+ gives it a red color (red beryl) and Fe3+colors it in yellow (heliodore).

b. ion valence

Absorption due to an isolated ion depends on its valence, or its charge, that is, the number of electron withdrawn compared to the neutral element.In beryl, iron gives the blue color to aquamarine when divalent (Fe2+) and the yellow color to heliodore when trivalent (Fe3+). 

c· Ligands nature

Ligands are elements linked to chromophore ions. For example, Co2+ induces a blue color to spinel because linked to  O2- , and a green color to sphalerite because linked to S2-.

d· Coordination

Coordinationindicates the number of atoms in the immediate vicinity of the coloring ion.
- coordination IV, tetrahedral: 4 atoms forming a tetrahedron.
- coordination VI, octaherdal: 6 atoms forming an octahedron.
- coordination VIII, distorded cube: 8 atoms forming nearly a cube.

For example, Mn2+ linked to O2- causes different colors depending on its coordination.
- rhodochrosite, octahedral coordination octaédrique : pink to red.
- willemite, tetrahedral coordination: yellow
- spessartite, cubique coordination: orange.

e· Crystal field

A good example is ruby (red) found in zoïsite (green) at the Longido deposit in Tanzania. These two minerals are colored by Cr3+. However, distances between chromium atoms and surrounding oxygen atoms are not the same in ruby and zoisite, henve different colors.

2. Color centers

Color centers are structural defects (such as vacancies, defects caught by impurities, etc). This is a general term that commonly refers to defects created by irradiation, either natural or artificial.

A vacancy is a lacking atom. Several ways to generate it: at least some vacancies are always created during crystal growth; Irradiation is a good way to produce abundant vacancies; deformation of the crystal lattice. Curiously, a single neutral vacancy gives a green color to diamond, when a cluster of vacancies gives a brown color.



Neutral carbon vacancy in diamond is called GR1 (for General Radiation 1). this is just one carbon atom missin, without residual electric charge. The carbon-carbon bond is strong, so that a pretty high irradiation energy is needed to create this vacancy.














Brown color of diamond around the Penthièvre diamond (Chantilly Castel, Oise, France) is due to vacancy clusters, groups of about 40 to 60 carbon vacancies. Such defect is formed by plastic deformation of diamond.

Irradiation, in a general manner,can generate numerous color centers, most of the time by extracting an electron. This is the case for example of smocky quartz: irradiation extracts an electron to an O2- itself directly linked to an aluminum atom (Al), that commonly substitutes for a silicium atom in the SiO2 quartz structure. This creates a defects that absorbs in the ultraviolet region, and extends into the visible part of the spectrum, hence the brown color (see picture below).




Often, a color centers counts a structural defect (often a vacancy) linked to a chemical impurity. This impurity "sticks" the vacancy when the new defect is stable and will not tend to dissociate. Pink fluorite, for example, is colored by a defect combining an yttrium atom (a rare-earth element) and a vacancy.

Irradiation followed by heating of diamond may induce numerous defects simultaneously (N2, N3, H3, H4, cavancy…). This is commercially usedto modify the appearance of diamonds of less desirable initial color.







3. Atom groups

a. Intensification of Fe3+ absorptions

When grouped to form polymers, ions interact and intensify transitions, hence a deeper color. This is particularly true for minerals containing a coloring element in high concentration. Among them, iron is the most common. Trivalent iron is magnetic. It behaves as a "magnet" at the atomic scale, and can influence other "magnets" around. This is the way Fe3+ ions interact.

b. Charge tansfers
Some colors are due to an electron circulating beween several atoms in the same mineral. "transfer" refers to the movement and "charge" is for the electron, the unit for electric charge. Charge transfers are very efficient as they strongly absorb light (large band). However, absorption can be very strong in one direction of the crystal, and very weak to null in an other: absorptions are strongly directionnal and often induce a strong pleochroism. Several types of charge transfer:
· O2- + ion of transition element (or "metal")
    - O2- – Fe3+ : maximum of absorption in the UV giving a yellow  to orange color (yellow beryl yellow sapphre, citrine quartz).
    - O2- – Cr6+ : chromate ion CrO42- of orange color (absorption in the blue-violet) in crocoite (Pb chromate), vanadinite (chromate), wulfenite.
    - O2- – Mn5+ : dark blue apatite.
    - O2- – Fe4+ : amethyst.
· Metal + O2- + Metal
    - Fe2+ – O – Ti4+ : blue sapphire, brown micas, red to green andalusite, yellow to brown andradite.
    - Mn2+ – O – Ti4+ : yellow to green-yellow tourmaline.
· Intervalence charge transfer
    - Fe2+ – O – Fe3+ : very common, commonly gives a blue color such in blue amphibole (pietersite), cordierite, magmatic blue sapphire, lazulite...
c. Pairs transitions Fe3+ – Fe3+
Magnetic coupling in yellow sapphires, dravite from Chipala, Zambia.

d. Charge transfer without metallic ions
S3- (vivid blue), S2- : lapis-lazuli, sodalite, afghanite, feldspathoids.

Organic materials: color of amber is due to electrons delocalization on the carbon chains.

4. Colors explained by the Bands Theory

A band is an energy range with homogeneous properties. The valence band corresponds to a bond between two atoms, electrons "maintaining" the molecule. The band gap (or "gap") is the band without electrons. The conduction band is usually devoid of electrons, but these can "jump" from the valence band to the conduction band in case of energy input. Several types of gems are distinguished according to the depth of the gap:

- Isolators (thermal and electrical):
This is the case of most gems. The gap is over 3 eV, hence all the visible light is transmitted.The gap for diamond is 5.5 eV, and that of codundum is 8 eV. As visible light is not absorpbed, all these gems are colorless when pure.

- Intrinsically colored semi-conductors:
The gap corresponds to a visible light. All light higher in energy is absorbed, all light of lesser energy is transmitted. This is the case of cuprite (intrinsically red), proustite, native sulphur, orpiment, realgar.

- Opaque semi-conductors:
The gap is lower than the energy of red light. This is the case of native silicium, gremanium...

- Metals:
There is no gap.

5. Colors due to an optical phenomenon

a. Interferences
rainbow-garnet-smallProduces a series of non spectral colors. Interferences are due to lamellar structure salternating two different materrials (rainbow obsidian, rainbow moonstone, rainbow garnet...).

b. Diffraction
opale-diffraction-lumiere-visible-smallDiffraction is an interference phenomenon due to a very high number of superimposed layers. Diffraction produces pure spectral colors (those of the rainbow). this is the origi of play-of-color in precious opal.


c. Scattering
calcedoine-bleue-diffusion-smallInclusions of material different from the host crystal can scatter light: this is Mie scattering. Scattering depends on the size, shape and orientation of inclusions. A case extensively studied at Nantes is opalescence: white light is scattered by opal, giving it a "bluish" tint by reflexion and an "orangey" tint by transmission.

d. Colored inclusions
chrysoprase-inclusions-nickeliferes-smallAlmost all varieties of poorly crystallized silica, such as opal and chalcedony, are colored by colored inclusions, sometimes extremely small. For example, chrysoprase (green variety of chalcedony) is colored in green by nano-inclusions of nickeliferous phyllosilicates.

Additional information