Color Factors in SU(N) Gauge Theory

Hello! In our last lesson, we successfully derived the complete set of Feynman rules for Quantum Chromodynamics. We now have a dictionary that translates the terms in the QCD Lagrangian into propagators for quarks, gluons, and ghosts, and vertices that describe their interactions.

However, a crucial complexity remains. Every interaction vertex in QCD carries a color structure, represented by the SU(3) generators $t^{a}$ or the structure constants $f^{ab c}$ . To calculate a physical quantity like a scattering cross-section, we must sum over the colors of unobserved final-state particles and average over the colors of initial-state particles. This procedure involves performing traces and products of these color matrices. The result of this calculation for a given Feynman diagram is a numerical value called the color factor.

Today, we will master the algebra required to compute these color factors. You will learn the fundamental identities of the SU(N) Lie algebra and apply them to systematically evaluate the color structure of QCD interactions. This skill is essential for performing any perturbative calculation in the theory.

The Algebra of SU(N)

While QCD is an SU(3) gauge theory, it's often more insightful to work with a general SU(N) group and specialize to N=3 only at the end. This approach clarifies the origin of various terms and helps in organizing calculations.

The generators of SU(N) in the defining (or fundamental) representation, where quarks reside, are a set of $N^{2} - 1$ traceless, Hermitian $N \times N$ matrices, which we'll denote as $T^{a}$ . They obey the following fundamental relations:

Lie Algebra Commutation Relation:
$[T^{a}, T^{b}] = i f^{ab c} T^{c}$
where $f^{ab c}$ are the structure constants of the SU(N) group, which are totally anti-symmetric in their indices.
Tracelessness:
$Tr (T^{a}) = 0$
Normalization:
$Tr (T^{a} T^{b}) = T_{F} δ^{ab}$
By convention in QCD, we set the normalization factor $T_{F} = 1/2$ .

The following reading provides a concise but rigorous collection of the key identities we will be using. We'll refer to it throughout the lesson.

Relations among generators in SU(N) representations

Please familiarize yourself with the foundational definitions and identities of SU(N) generators from the following notes by Howard E. Haber.

Read Section 2, "The defining representation of the SU(N) Lie algebra". Focus on understanding Equations (1)-(6). In particular, pay close attention to the derivation of the quadratic Casimir, C_F (Eq. 5), and the Fierz identity (Eq. 6), as we will use these extensively.

Casimir Operators

A key concept in group theory is the Casimir operator, which is an operator constructed from the generators that commutes with all of them. By Schur's lemma, in any irreducible representation (like the one for quarks or gluons), the Casimir operator must be proportional to the identity matrix.

1. The Fundamental Representation (Quarks)

The quadratic Casimir operator for the fundamental representation is $T^{a} T^{a} \equiv \sum_{a = 1}^{N^{2} - 1} T^{a} T^{a}$ . It is proportional to the $N \times N$ identity matrix $1$ :

T^{a} T^{a} = C_{F} 1

As you saw in the reading, by taking the trace of this equation, we can determine the value of $C_{F}$ :
$Tr (T^{a} T^{a}) = Tr (C_{F} 1) ⟹ \sum_{a} Tr (T^{a} T^{a}) = C_{F} Tr (1)$
$\sum_{a} \frac{1}{2} δ^{aa} = C_{F} N ⟹ \frac{1}{2} (N^{2} - 1) = C_{F} N$
$C_{F} = \frac{N ^{2} - 1}{2 N}$
For QCD (SU(3)), this gives the famous factor $C_{F} = (3^{2} - 1) / (2 \cdot 3) = 8/6 = 4/3$ . This factor will appear any time a quark emits and reabsorbs a gluon.

2. The Adjoint Representation (Gluons)

Gluons themselves carry color charge and transform under the adjoint representation of SU(N). The generators of this representation are $(N^{2} - 1) \times (N^{2} - 1)$ matrices, $(F^{a})_{b c} = - i f^{ab c}$ . The quadratic Casimir in this representation is $F^{a} F^{a}$ , and its value is denoted $C_{A}$ :

F^{a} F^{a} = C_{A} I

where $I$ is the $(N^{2} - 1) \times (N^{2} - 1)$ identity matrix. The components of this equation are:
$\sum_{a} (F^{a})_{b d} (F^{a})_{d c} = \sum_{a} (- i f^{ab d}) (- i f^{a d c}) = f^{ab d} f^{a c d} = C_{A} δ_{b c}$
A fundamental identity for SU(N) is $\sum_{a, d} f^{ab d} f^{a c d} = N δ_{b c}$ . Using this, we find:
$C_{A} = N$
For QCD (SU(3)), $C_{A} = 3$ . This factor is characteristic of gluon self-interactions.

A Pictorial Method: Birdtracks

Keeping track of all the color indices ( $i, j, \dots$ for quarks and $a, b, \dots$ for gluons) can quickly become overwhelming. Fortunately, there is a powerful and intuitive graphical method developed by Predrag Cvitanović, often called "birdtracks," which simplifies these calculations immensely.

The core idea is to represent quarks, gluons, and the interaction vertices as simple graphical elements. Connecting lines then corresponds to summing over the shared index.

initiation to the birdtrack pictorial technique

The following resource provides a fantastic introduction to the birdtrack technique. It's a fun and elegant way to handle color algebra.

Please study Section 1.1 ("The basic Lego bricks") and Section 1.2 ("First trivial rules"). This will introduce you to the pictorial notation for the SU(N) generators, structure constants, and Kronecker deltas, and show how simple traces are represented.

Let's summarize the key pictorial rules and see how they work.

Color Factors in QCD — This diagram illustrates the fundamental interactions of QCD and their associated color factors. The quark-gluon vertex involves $T^a$, whose square gives $C_F$. The three-gluon vertex involves $f^{abc}$, whose square gives $C_A$. The gluon splitting to a quark-antiquark pair involves $T^a$, whose trace gives $T_F$.

Test your understanding! Deriving C_F with Birdtracks

In the birdtrack notation, the Fierz identity, $(T^{a})_{i}^{j} (T^{a})_{k}^{l} = \frac{1}{2} (δ_{i}^{l} δ_{k}^{j} - \frac{1}{N} δ_{i}^{j} δ_{k}^{l})$ , looks like this:

Now, try to derive the value of $C_{F}$ . The color factor for a quark emitting and re-absorbing a gluon is $\sum_{a} (T^{a} T^{a})_{i}^{j} = C_{F} δ_{i}^{j}$ . In birdtrack notation, this is:

How can you use the Fierz identity to evaluate the loop on the left and find the value of $C_{F}$ ? (This corresponds to Exercise 3 in the birdtrack notes).

Show answer

To evaluate the loop, we can "close" the Fierz identity diagram by connecting the top-left line to the bottom-left line, and the top-right to the bottom-right. This corresponds to setting $i = j$ and $k = l$ and summing.

Let's evaluate each term on the right-hand side:

The first term becomes $\frac{1}{2} \times (a quark loop) = \frac{1}{2} \times N$ .
The second term becomes $- \frac{1}{2 N} \times (two separate quark loops) = - \frac{1}{2 N} \times N^{2} = - \frac{N}{2}$ .

Combining these, we get $\frac{N}{2} - \frac{1}{2 N}$ . But wait, something is wrong. The Fierz identity connects four points. To get the C_F diagram, we need to contract j with k. Let's re-write the Fierz identity with indices: $(T^{a})_{i}^{j} (T^{a})_{k}^{l}$ . The C_F diagram corresponds to $\sum_{a} (T^{a})_{i}^{k} (T^{a})_{k}^{j}$ . So we set $l = i$ in the index form. This means connecting the bottom-left line to the top-right line.

Applying the Fierz identity $(T^{a})_{i}^{j} (T^{a})_{k}^{l} = \frac{1}{2} (δ_{i}^{l} δ_{k}^{j} - \frac{1}{N} δ_{i}^{j} δ_{k}^{l})$ to the C_F diagram $\sum_{a} (T^{a})_{i}^{k} (T^{a})_{k}^{j}$ :
$\sum_{a} (T^{a})_{i}^{k} (T^{a})_{k}^{j} = (T^{a})_{i}^{k} (T^{a})_{k}^{j}$
Using the Fierz identity on the product of two generators is a bit tricky. A simpler way is to use the identity (T^a)_i^j (T^a)_k^l = ....

Let's try a different approach, directly from the notes. We can rewrite the Fierz identity to isolate the gluon line:

Now, connect the j and k lines. This forms a quark loop, which evaluates to N.

The equation becomes:
$2 \times C_{F} = N - \frac{1}{N} \times 1 = \frac{N ^{2} - 1}{N}$
$C_{F} = \frac{N ^{2} - 1}{2 N}$
This matches our algebraic result! The graphical method makes manipulating these structures much more manageable.

Common Color Factor Calculations

Now let's put these rules into practice by calculating the three most important color factors that appear in one-loop QCD corrections.

initiation to the birdtrack pictorial technique

The birdtrack resource derives several key identities as exercises. Working through them is the best way to learn.

Please read Sections 1.3 ("Fierz identity"), 1.4 ("Lie algebra"), and 1.5 ("Sum up"). Pay close attention to the pictorial rules being derived, especially in Exercises 3, 4, 6, 7, and 10. We will use these results directly.

Quark Self-Energy: A quark emits and re-absorbs a gluon.
- Color Structure: $\sum_{a} T^{a} T^{a}$
- Calculation: As we've seen, this is the definition of the Casimir operator. The color factor is $C_{F}$ .
- Birdtrack:
  
  The result of this diagram is $C_{F}$ times a straight quark line.
Gluon Vacuum Polarization (Quark Loop): A gluon fluctuates into a quark-antiquark pair, which then re-annihilates.
- Color Structure: $Tr (T^{a} T^{b})$
- Calculation: This is our normalization condition. The result is $\frac{1}{2} δ^{ab}$ . The factor $T_{F} = 1/2$ is associated with any closed quark loop with two external gluons.
- Birdtrack:
Gluon Vacuum Polarization (Gluon Loop): A gluon fluctuates into two gluons, which then recombine.
- Color Structure: $\sum_{c, d} f^{a c d} f^{b c d}$
- Calculation: This is related to the trace of the adjoint generators, $Tr (F^{a} F^{b}) = \sum_{c, d} (F^{a})_{c d} (F^{b})_{d c} = \sum_{c, d} (- i f^{a c d}) (- i f^{b d c}) = \sum_{c, d} f^{a c d} f^{c d b}$ . Using the anti-symmetry of $f$ , this is $\sum_{c, d} f^{a c d} f^{b c d}$ . From the Casimir identity, this evaluates to $C_{A} δ^{ab} = N δ^{ab}$ .
- Birdtrack (from Exercise 6 in LINK):

Summary of Key Color Factors

Here is a summary of the most important SU(N) color identities and their specific values for QCD (N=3).

Identity	Description	General SU(N) Value	QCD (SU(3)) Value
$Tr (T^{a} T^{b})$	Normalization / Quark Loop	$\frac{1}{2} δ^{ab}$ ( $T_{F} = \frac{1}{2}$ )	$\frac{1}{2} δ^{ab}$
$\sum_{a} T^{a} T^{a}$	Quark Casimir	$C_{F} = \frac{N ^{2} - 1}{2 N}$	$\frac{4}{3}$
$\sum_{c, d} f^{a c d} f^{b c d}$	Gluon Casimir	$C_{A} = N$	$3$
$\sum_{a} Tr (T^{a} T^{a})$	Trace of Casimir	$C_{F} N = \frac{N ^{2} - 1}{2}$	$4$
$\sum_{a, b, c} f^{ab c} f^{ab c}$	Sum of squares of f	$N (N^{2} - 1)$	$24$
$\sum_{c} T^{a} T^{c} T^{a}$	"Twisted" Casimir	$- \frac{1}{2 N} T^{c}$	$- \frac{1}{6} T^{c}$

Conclusion

In this lesson, you've learned how to manage the color algebra of QCD. We've moved from abstract generator definitions to concrete computational rules.

Key Takeaways:

Physical calculations require averaging over initial state colors and summing over final state colors, leading to color factors.
The algebra of SU(N) generators provides a set of identities to compute these factors.
The key numerical factors are the Casimirs $C_{F} = \frac{N ^{2} - 1}{2 N}$ and $C_{A} = N$ , and the normalization $T_{F} = 1/2$ .
The birdtrack pictorial technique provides a powerful and intuitive method for performing these calculations, avoiding the complexity of index manipulation.

Preview of the Next Lesson:

We are now finally ready to perform a full calculation of a physical process in QCD. In the next lesson, we will combine the Feynman rules from the previous lesson with the color factor calculations from this one to compute the tree-level scattering amplitude for quark-quark scattering ( $q q \to q q$ ). This will be our first complete application of the perturbative QCD framework.

Recommended resources