Shortly after the existence of quarks was first proposed in 1964, Oscar W. Greenberg introduced the notion of color charge to explain how quarks could coexist inside some hadrons in otherwise identical states and still satisfy the Pauli exclusion principle. The concept turned out to be useful. Quantum chromodynamics has been under development since the 1970s and constitutes an important ingredient in the standard model of particle physics.
One can say that a quark's color can take one of three values: "red", "green", or "blue"; and that an antiquark can take one of three "anticolors", sometimes called "antired", "antigreen" and "antiblue" (occasionally represented as cyan, magenta and yellow, respectively). In the same vein it can be said that gluons are mixtures of two colors: for example red-antigreen, and this constitutes their color charge. QCD considers eight gluons of the possible nine color/anti-color combinations to be unique; see Gluon for the reason of this. Also see Coupling constants for clarification of the following material.
In a quantum field theory the notion of a coupling constant and a charge are different but related. The coupling constant sets the magnitude of the force of interaction; for example, in quantum electrodynamics, the fine structure constant is a coupling constant. The charge in a gauge theory has to do with the way a particle transforms under the gauge symmetry; i.e., its representation under the gauge group. For example, the electron has charge -1 and the positron has charge +1, implying that the gauge transformation has opposite effects on them in some sense. Specifically, if a local gauge transformation φ(x) is applied in electrodynamics, then one finds
In QCD the gauge group is the non-Abelian group SU(3). The running coupling is usually denoted by αs. Each flavor of quark belongs to the fundamental representation (3) and contains a triplet of fields together denoted by ψ. The antiquark field belongs to the complex conjugate representation (3*) and also contains a triplet of fields. We can write
In the simple language introduced previously, the three indices "1", "2" and "3" in the quark triplet above are usually identified with the three colors. The colorful language misses the following point. A gauge transformation in color SU(3) can be written as ψ → Uψ, where U is a 3X3 matrix which belongs to the group SU(3). Thus, after gauge transformation, the new colors are linear combinations of the old colors. In short, the simplified language introduced before is not gauge invariant.
Color charge is conserved, but the book-keeping involved in this is more complicated than just adding up the charges, as is done in quantum electrodynamics. One simple way of doing this is to look at the interaction vertex in QCD and replace it by a color line representation. The meaning is the following. Let ψi represent the i-th component of a quark field (loosely called the i-th color). The color of a gluon is similarly given by a which corresponds to the particular Gell-Mann matrix it is associated with. This matrix has indices i and j. These are the color labels on the gluon. At the interaction vertex one has qi→gij+qj. The color-line representation tracks these indices. Color charge conservation means that the ends of these color-lines must be either in the initial or final state, equivalently, that no lines break in the middle of a diagram.
Since gluons carry color charge, two gluons can also interact. A typical interaction vertex (called the three gluon vertex) for gluons involves g+g→g. This is shown here, along with its color line representation. The color-line diagrams can be restated in terms of conservation laws of color; however, as noted before, this is not a gauge invariant language. Note that in a typical non-Abelian gauge theory the gauge boson carries the charge of the theory, and hence has interactions of this kind; for example, the W boson in the electroweak theory. In the electroweak theory, the W also carries electric charge, and hence interacts with a photon.