One-Loop Vertex Correction and Electron Anomalous Magnetic Moment (g-2)

Hello! Welcome to our next lesson on Quantum Field Theory.

In the past two lessons, we've systematically addressed the one-loop quantum corrections to the fundamental building blocks of QED. First, we tackled vacuum polarization, which corrected the photon propagator and led to the concept of a running charge. Then, we investigated the electron self-energy, which corrected the electron propagator and forced us to renormalize the electron's mass and field strength.

Today, we will complete our one-loop tour of QED's "primitive divergences" by examining the third and final piece: the correction to the interaction vertex itself. The learning outcome for this lesson is to calculate the one-loop vertex correction and relate it to the electron's anomalous magnetic moment (g-2). This calculation is a historic triumph of QED, yielding a precise, testable prediction that has been verified to astonishing accuracy.

You'll recall from your study of quantum mechanics that the non-relativistic Schrödinger-Pauli theory and the relativistic Dirac equation both predict the electron's gyromagnetic ratio to be exactly $g = 2$ . We will see how quantum field theory introduces a subtle, finite correction to this value, arising directly from the electron's interaction with the quantum vacuum.

1. The Vertex Function and Form Factors

The fundamental interaction in QED is an electron absorbing or emitting a photon. At tree level, this is represented by a simple vertex, to which we assign the Feynman rule $- i e γ^{μ}$ . Quantum corrections "dress" this vertex with loops. The first and simplest correction is the one-loop diagram shown below.

Feynman Diagram and Integral for One-Loop Electron Vertex Function — The one-loop Feynman diagram for the vertex correction. An electron with momentum $p$ interacts with a photon (momentum $q$), and scatters to a final momentum $p'$. The vertex is corrected by a virtual photon loop.

The full vertex, including all corrections, is denoted by $i e Γ^{μ} (p^{'}, p)$ . We are interested in the case where the incoming and outgoing electrons are on-shell, meaning $p^{2} = (p^{'})^{2} = m^{2}$ . The most general Lorentz-covariant structure for this object, when sandwiched between Dirac spinors $u ˉ (p^{'})$ and $u (p)$ , can be written in terms of two scalar functions called form factors, $F_{1} (q^{2})$ and $F_{2} (q^{2})$ , where $q = p^{'} - p$ is the momentum of the external photon.

The decomposition is:

u ˉ (p^{'}) Γ^{μ} (p^{'}, p) u (p) = u ˉ (p^{'}) [F_{1} (q^{2}) γ^{μ} + F_{2} (q^{2}) \frac{i σ ^{μν} q _{ν}}{2 m}] u (p)

where $σ^{μν} = \frac{i}{2} [γ^{μ}, γ^{ν}]$ .

Let's break down the meaning of these form factors:

$F_{1} (q^{2})$ - The Electric Form Factor: This term modifies the original $γ^{μ}$ structure. It can be thought of as a correction to the electron's charge distribution as seen by the probing photon. At $q^{2} = 0$ (a zero-energy photon), the Ward-Takahashi identity (which we will explore in a future lesson) guarantees that $F_{1} (0) = 1$ . This means that for long-wavelength interactions, the electron's charge is exactly $e$ , ensuring charge universality.
$F_{2} (q^{2})$ - The Magnetic Form Factor: This is a new tensor structure that does not exist at tree level. It describes an additional contribution to the electron's interaction with a magnetic field. At tree level, we have $F_{1} (q^{2}) = 1$ and $F_{2} (q^{2}) = 0$ . The one-loop correction will generate a non-zero $F_{2}$ .

The Anomalous Magnetic Moment

The gyromagnetic ratio, $g$ , relates an electron's magnetic dipole moment $μ$ to its spin $S$ via $μ = g (\frac{e}{2 m}) S$ . To see how $g$ relates to the form factors, we can examine the non-relativistic limit of the electron-photon interaction.

The following reading provides an excellent bridge from the familiar Dirac equation to the vertex form factors and their relation to $g$ .

1 Loops and renormalization

To understand the physical significance of the form factors, let's review how the Dirac equation predicts g=2 and how the general vertex function modifies this prediction.

Please read section 1.1.4, 'Vertex correction and anomalous magnetic moment', from the beginning up to (and including) equation (45). Focus on: The derivation showing the Dirac equation leads to g=2 (eqs. 33-39). The general decomposition of the vertex function Γν into F₁(q²) and F₂(q²) (eq. 43). The crucial relations that define the physical meaning of the form factors at zero momentum transfer (eqs. 44 and 45): F₁(0) = Q (the total charge, which is 1 in units of e). F₂(0) = (g-2)/2. This last relation is key: it tells us that the entire deviation of g from the Dirac value is captured by F₂(0). This quantity is often called the electron's anomalous magnetic moment, a_e.

As the reading makes clear, our entire task boils down to calculating $F_{2} (q^{2})$ at one-loop and evaluating it at $q^{2} = 0$ .

2. Calculating the One-Loop Vertex Correction

Now, we will perform the explicit calculation of the one-loop diagram. The calculation is algebraically intensive, but the logic follows the familiar pattern: write down the integral using Feynman rules, introduce Feynman parameters, shift the loop momentum, and evaluate the resulting integral.

Applying the Feynman rules in Feynman gauge gives the expression for the one-loop vertex function, $Γ_{1-loop}^{μ}$ :

i e Γ_{1-loop}^{μ} (p^{'}, p) = \int \frac{d ^{4} k}{( 2 π ) ^{4}} \frac{- i g _{ν λ}}{k ^{2} + i ϵ} (i e γ^{ν}) \frac{i ( \neq p ^{'} - \neq k + m )}{( p ^{'} - k ) ^{2} - m ^{2} + i ϵ} (i e γ^{μ}) \frac{i ( \neq p - \neq k + m )}{( p - k ) ^{2} - m ^{2} + i ϵ} (i e γ^{λ})

The following resource provides a masterclass in evaluating this integral. We will follow its steps closely.

QED Vertex Correction

This document walks through the entire calculation with meticulous detail. I encourage you to follow along with a pen and paper to trace the key algebraic manipulations.

Please read the entire document (pages 1-5). It's dense but very systematic. Here is a roadmap to guide you: Setup (Section 1): The initial integral is set up (eq. 5), and the three propagator denominators are combined using Feynman parameters (eq. 8). The loop momentum is shifted (k -> ℓ), resulting in the denominator [ℓ² - Δ + iε]³ (eq. 14), where Δ = (1-z)²m² - xyq². Numerator Algebra (pp. 4-6): This is the most complex part. The goal is to simplify the numerator N^μ (eq. 6) and express it in terms of γ^μ and iσ^{μν}q_ν. Key techniques used are: D-dimensional gamma matrix identities (eq. 16). Sandwiching the expression between spinors and using the Dirac equation (not{p}-m)u(p)=0 (eq. 21). Applying the Gordon Identity (eq. 25), which you have seen before, to trade momentum terms for gamma matrices. Isolating Form Factors (p. 7): The algebra culminates in separating the numerator into two pieces, N₁ and N₂ (eq. 30), which are directly proportional to the integrands for F₁ and F₂, respectively. Calculating F₂(0) (Section 'Electron's Gyromagnetic Moment'): Here is the payoff. We need to compute F₂(0). The document sets q²=0, which simplifies Δ enormously (eq. 39). Notice the numerator for F₂ (eq. 32) is independent of the loop momentum ℓ. The momentum integral is then performed (eq. 40), and it is UV finite. Finally, the integral over Feynman parameters (x, y, z) is done (eqs. 41-43), yielding the final answer.

The stunning result of this calculation is:

F_{2} (q^{2} = 0) = \frac{α}{2 π}

where $α = e^{2} /4 π$ is the fine-structure constant.

This gives the one-loop prediction for the electron's anomalous magnetic moment:

a_{e} = \frac{g - 2}{2} = F_{2} (0) = \frac{α}{2 π} \approx 0.00116

This result, first derived by Julian Schwinger in 1948, was a landmark achievement for QED.

Test your understanding!

In the previous lessons, the integrals for vacuum polarization and electron self-energy were UV divergent, requiring regularization and the introduction of counterterms. However, the integral for $F_{2} (0)$ turned out to be UV finite. Why is this significant? What does it imply about the nature of the anomalous magnetic moment in QED?

Show answer

The UV finiteness of the $F_{2} (0)$ calculation is profoundly important. It means that the anomalous magnetic moment is not an arbitrary parameter of the theory that needs to be fixed by an experimental measurement. Instead, it is a firm prediction derived from the theory's fundamental principles.

Because the result is finite, we do not need to introduce a new, independent "magnetic moment counterterm" into the Lagrangian. The value $g - 2$ is completely determined by the electron's charge and mass. This predictive power is a hallmark of a successful physical theory. Had the calculation diverged, it might have suggested that $g$ is another fundamental constant of nature that must be renormalized, just like mass and charge. The finiteness proves this is not the case.

Conclusion

Today, we have calculated the final one-loop primitive correction in QED, the vertex correction. This process tied together many of the techniques you have learned and led to a remarkable physical prediction.

Key Takeaways:

The full QED vertex can be described by two form factors, $F_{1} (q^{2})$ and $F_{2} (q^{2})$ .
The form factor $F_{2} (q^{2})$ gives rise to a quantum correction to the electron's magnetic moment. The anomalous magnetic moment is given by $a_{e} = F_{2} (0)$ .
The one-loop calculation for $F_{2} (0)$ is UV finite, meaning $a_{e}$ is a direct, calculable prediction of QED.
The one-loop result, first calculated by Schwinger, is $a_{e} = \frac{α}{2 π}$ .

The agreement between the theoretical prediction and experimental measurement of $g - 2$ is one of the most stringent tests of the Standard Model. Modern calculations have pushed this to five loops in QED, including electroweak and hadronic contributions. The experimental value from Fermilab's Muon g-2 experiment is measured with incredible precision, and the slight tension between the experimental result and the theoretical prediction is a major focus of particle physics research today, possibly hinting at new physics.

Preview of the Next Lesson:
We have now calculated the three fundamental one-loop corrections in QED. We've repeatedly mentioned the Ward-Takahashi identity, particularly its role in ensuring $F_{1} (0) = 1$ . In the next lesson, we will formally state this identity and explore its powerful consequences. It provides a crucial consistency check on our renormalization procedure, proving that the relations between counterterms ( $δ Z_{1} = δ Z_{2}$ ) are not accidental but are enforced by the underlying gauge symmetry of QED. This will finalize our theoretical toolkit for one-loop QED.

Recommended resources

1. The Vertex Function and Form Factors

The Anomalous Magnetic Moment

2. Calculating the One-Loop Vertex Correction

Conclusion