Chapter 14 — Beta Decay: The Weak Interaction in the Nucleus

"I have done a terrible thing. I have postulated a particle that cannot be detected." — Wolfgang Pauli, on his neutrino hypothesis (1930)

Chapter Overview

In alpha decay (Chapter 13), a nucleus emits a tightly bound $^4$He cluster. The physics is electromagnetic (the Coulomb barrier) and quantum mechanical (tunneling), but the strong force holds center stage. Beta decay is fundamentally different. It is the manifestation of the weak interaction inside the nucleus — the only fundamental force that changes particle identity, converting neutrons to protons or protons to neutrons.

Beta decay is the most common form of radioactive decay. Of the roughly 3,300 known nuclides, the vast majority are unstable against beta decay — it is the mechanism by which nuclei move toward the valley of stability on the chart of nuclides. It is also the process that powers the neutron-rich ejecta of neutron star mergers (Chapter 23), drives the pp chain in the Sun (Chapter 22), and produces the neutrinos detected by every neutrino observatory on Earth.

This chapter develops the physics of beta decay in its full richness:

• Section 14.1 introduces the three modes of beta decay and their energetics.
• Section 14.2 tells the story of the continuous beta spectrum and Pauli's bold hypothesis of a neutral, nearly massless particle — the neutrino.
• Section 14.3 develops Fermi's theory of beta decay (1934), deriving the allowed beta spectrum shape from Fermi's golden rule.
• Section 14.4 classifies transitions by their selection rules into Fermi and Gamow-Teller types, and into allowed and forbidden categories.
• Section 14.5 introduces the Kurie plot and the ft-value formalism for systematizing beta decay data.
• Section 14.6 tells the dramatic story of the Wu experiment (1957) and the fall of parity symmetry.
• Section 14.7 explores double beta decay — the rarest observed nuclear process — and the search for its neutrinoless variant, one of the most important open questions in physics.

🏃 Fast Track: If you are comfortable with Fermi's golden rule (Chapter 5), begin at Section 14.3. The essential derivation is in Sections 14.3–14.5. The Wu experiment (Section 14.6) and double beta decay (Section 14.7) are conceptually self-contained.

🔬 Deep Dive: The classification of beta transitions by log ft (Section 14.5) is heavily used in nuclear spectroscopy. The connection to the modern electroweak theory (Section 14.3.5) places Fermi's theory in its broader context.

14.1 The Three Modes of Beta Decay

14.1.1 Beta-Minus Decay ($\beta^-$)

In $\beta^-$ decay, a neutron inside the nucleus converts to a proton, emitting an electron and an electron antineutrino:

$$n \to p + e^- + \bar{\nu}_e$$

At the nuclear level:

$$^A_Z X \to ^A_{Z+1} Y + e^- + \bar{\nu}_e$$

The mass number $A$ is unchanged — beta decay is an isobaric transition. The daughter nucleus has one more proton and one fewer neutron.

Energetics. The Q-value for $\beta^-$ decay, expressed in terms of atomic masses (which include the electron masses), is:

$$Q_{\beta^-} = \left[M(^A_Z X) - M(^A_{Z+1} Y)\right]c^2$$

The atomic mass of the parent includes $Z$ electrons; the daughter includes $Z+1$ electrons, and one electron is emitted. These $Z+1$ electrons cancel, so the Q-value depends only on the difference of atomic masses. Beta-minus decay is energetically allowed whenever $Q_{\beta^-} > 0$, i.e., whenever the parent atom is heavier than the daughter atom.

💡 Key Point: Using atomic masses automatically accounts for the electron rest mass. This is a standard convention in nuclear physics — always check whether a Q-value formula uses atomic or nuclear masses.

Example: The decay of tritium:

$$^3_1\text{H} \to\, ^3_2\text{He} + e^- + \bar{\nu}_e$$

Using the 2020 Atomic Mass Evaluation (AME2020): $M(^3\text{H}) = 3.016049\,\text{u}$, $M(^3\text{He}) = 3.016029\,\text{u}$.

$$Q = (3.016049 - 3.016029) \times 931.494\,\text{MeV/u} = 0.01863\,\text{MeV} = 18.63\,\text{keV}$$

This tiny Q-value makes tritium beta decay uniquely sensitive to the neutrino mass — a fact we will exploit in Section 14.7.

14.1.2 Beta-Plus Decay ($\beta^+$)

In $\beta^+$ decay, a proton converts to a neutron, emitting a positron and an electron neutrino:

$$p \to n + e^+ + \nu_e$$

At the nuclear level:

$$^A_Z X \to ^A_{Z-1} Y + e^+ + \nu_e$$

Energetics. In terms of atomic masses:

$$Q_{\beta^+} = \left[M(^A_Z X) - M(^A_{Z-1} Y) - 2m_e\right]c^2$$

The extra $2m_e c^2 = 1.022\,\text{MeV}$ arises because the daughter atom has one fewer electron than needed to be neutral, and a positron is created. Beta-plus decay therefore requires the parent atom to be heavier than the daughter atom by at least $2m_e c^2$.

⚠️ Common Error: Students sometimes forget the $2m_e$ term in $\beta^+$ Q-values. This is the single most common mistake in beta decay energetics problems.

Example: The decay of $^{18}$F (the workhorse isotope of PET imaging, introduced in Chapter 1):

$$^{18}_9\text{F} \to\, ^{18}_8\text{O} + e^+ + \nu_e$$

$M(^{18}\text{F}) = 18.000938\,\text{u}$, $M(^{18}\text{O}) = 17.999161\,\text{u}$.

$$Q = (18.000938 - 17.999161 - 2 \times 0.000549) \times 931.494 = 0.634\,\text{MeV}$$

The positron annihilates with an electron in the surrounding tissue, producing two 511 keV gamma rays emitted back-to-back — the signal detected by PET scanners (Chapter 27).

14.1.3 Electron Capture (EC)

In electron capture, a proton in the nucleus absorbs an inner-shell (usually K-shell) atomic electron:

$$p + e^- \to n + \nu_e$$

At the nuclear level:

$$^A_Z X + e^- \to ^A_{Z-1} Y + \nu_e$$

Energetics:

$$Q_\text{EC} = \left[M(^A_Z X) - M(^A_{Z-1} Y)\right]c^2 - B_n$$

where $B_n$ is the binding energy of the captured electron. Since $B_n$ is typically a few keV (and at most $\sim$100 keV for the heaviest elements), EC is energetically allowed whenever the parent atom is heavier than the daughter atom — a less stringent condition than $\beta^+$ decay. In fact, EC is always energetically allowed whenever $\beta^+$ is allowed, but EC can also occur when $\beta^+$ is forbidden (when $0 < Q < 2m_ec^2$).

Signature: EC produces no charged particle in the final state — the only nuclear emission is the monoenergetic neutrino. The experimental signature is the characteristic X-ray or Auger electron emitted when the vacancy in the inner electron shell is filled.

Example: $^{7}$Be decays exclusively by EC (the Q-value of 0.862 MeV is below the $\beta^+$ threshold of $2m_ec^2 = 1.022$ MeV):

$$^7_4\text{Be} + e^- \to\, ^7_3\text{Li} + \nu_e$$

This process is critical in solar physics — the $^7$Be neutrino line at 0.862 MeV (or 0.384 MeV, depending on whether $^7$Li is left in the ground state or first excited state) is one of the key solar neutrino signals detected by the Borexino experiment.

14.1.4 Competition Between Modes

For a given isobar, the three modes compete. The situation is summarized by:

For heavy nuclei, EC competes with $\beta^+$ even when both are allowed, because the EC rate is enhanced by the high electron density at the nucleus (which scales as $Z^3$ for K-shell electrons). In the lightest nuclei, the EC rate is suppressed and $\beta^+$ dominates.

14.2 The Continuous Beta Spectrum and the Neutrino Hypothesis

14.2.1 The Crisis: A Continuous Spectrum

If beta decay were a two-body process ($^A X \to\, ^A Y + e^-$), the electron would be emitted with a unique kinetic energy determined by the Q-value and conservation of momentum — just as the alpha particle in alpha decay has a well-defined energy (Chapter 13). But experimentally, the electron energy spectrum is continuous, extending from zero to a maximum energy $T_\text{max} = Q - T_\text{recoil} \approx Q$ (since the nuclear recoil is negligible).

This was first observed by James Chadwick in 1914 and confirmed definitively by Charles Drummond Ellis and William Wooster in a landmark calorimetric experiment in 1927. Ellis and Wooster measured the total energy deposited by beta particles from $^{210}$Bi (then called RaE) in a calorimeter and found an average energy of $0.344 \pm 0.004$ MeV — far below the maximum energy of 1.161 MeV. The beta particles genuinely had a continuous distribution of energies; it was not an instrumental artifact.

This was a crisis. Energy appeared to not be conserved. Niels Bohr seriously entertained the idea that energy conservation might be only statistical in quantum mechanics — valid on average but violated in individual events. This was a radical proposal, and most physicists found it deeply unsettling.

The angular momentum problem was equally severe. Consider the $\beta^-$ decay of $^{14}$C ($0^+ \to 1^+$). The initial nucleus has $J = 0$. If only an electron (spin 1/2) is emitted, angular momentum conservation requires $J_f + s_e = J_i$, giving $J_f = 1/2$ — but the daughter $^{14}$N has $J = 1$ (integer). No combination of the electron spin and orbital angular momentum (both half-integer and integer, respectively) can account for the integer spin change. A second spin-1/2 particle — the neutrino — resolves this: the electron and neutrino spins can couple to 0 or 1, and their orbital angular momenta provide the remaining quantum numbers. This spin-statistics argument was one of Pauli's key motivations.

📊 Numerical Contrast: Compare the beta spectrum with alpha decay. In $^{210}$Po alpha decay, the alpha particle is emitted with a single energy $T_\alpha = 5.304\,\text{MeV}$ (monoenergetic, two-body decay). In $^{210}$Bi beta decay, the electrons have a continuous spectrum from 0 to $Q = 1.161\,\text{MeV}$, with a mean energy of only $\langle T_e \rangle = 0.389\,\text{MeV}$ — one-third of the endpoint. The "missing" energy in each decay event is carried by the antineutrino.

14.2.2 Pauli's Desperate Remedy

On December 4, 1930, Wolfgang Pauli wrote his famous letter to the "Radioactive Ladies and Gentlemen" gathered at a physics meeting in Tubingen (which he could not attend because of a dance in Zurich — a detail that humanizes one of physics' most brilliant minds). In this letter, Pauli proposed:

"I have hit upon a desperate remedy to save the... law of conservation of energy. Namely, the possibility that in the nuclei there could exist electrically neutral particles, which I will call neutrons, that have spin 1/2... The continuous beta spectrum would then become understandable by the assumption that in beta decay, in addition to the electron, a neutron is emitted in such a way that the sum of the energies of the neutron and the electron is constant."

Pauli's "neutron" was renamed the neutrino ("little neutral one") by Enrico Fermi in 1933, after Chadwick's discovery of the actual neutron in 1932. The key properties Pauli required:

1. Electrically neutral — otherwise it would have been detected already.
2. Spin 1/2 — to conserve angular momentum in the three-body decay.
3. Very small mass — the endpoint of the beta spectrum was consistent with zero neutrino mass (or at most a very small one).
4. Weakly interacting — otherwise it would have been observed in absorption experiments.

With a third particle, beta decay becomes a three-body process, and the electron energy is no longer uniquely determined. The electron and neutrino share the available energy $Q$ continuously, producing the observed spectrum. The maximum electron energy occurs when the neutrino carries away zero kinetic energy (and minimal momentum), while the minimum electron energy approaches zero when the neutrino carries nearly all the energy.

📜 Historical Note: Pauli was deeply uncomfortable with his own proposal. His famous remark about postulating a particle that cannot be detected reflected a genuine concern. He even offered to bet a case of champagne that the neutrino would never be observed. He would have lost that bet in 1956.

14.2.3 Experimental Confirmation: Reines and Cowan

Frederick Reines and Clyde Cowan detected the (anti)neutrino in 1956 at the Savannah River nuclear reactor in South Carolina. Their experiment exploited the inverse beta decay reaction:

$$\bar{\nu}_e + p \to n + e^+$$

The positron annihilates immediately, producing two 511 keV gamma rays (the "prompt" signal). The neutron thermalizes and is captured by cadmium (added to the liquid scintillator), producing a gamma cascade about $5\,\mu\text{s}$ later (the "delayed" signal). This coincidence signature — two 511 keV gammas followed microseconds later by a gamma cascade — provided a nearly background-free detection of the neutrino.

The measured cross section was $\sigma = (11 \pm 2.6) \times 10^{-44}\,\text{cm}^2$, in agreement with theoretical predictions based on Fermi's theory. The neutrino was real. Reines received the Nobel Prize in Physics in 1995 (Cowan had died in 1974).

💡 Scale: A cross section of $10^{-43}\,\text{cm}^2$ means that a neutrino can pass through roughly one light-year of solid lead before being absorbed. The neutrino interacts only through the weak force (and gravity, negligibly), making it the most penetrating particle known.

14.3 Fermi's Theory of Beta Decay

14.3.1 The Physical Picture

In 1934, Enrico Fermi published his theory of beta decay — one of the most influential papers in the history of physics. Fermi treated beta decay as a point interaction in which four fermions meet at a single spacetime point: the initial neutron (or proton) and the three final-state particles. This was directly analogous to the quantum theory of electromagnetic radiation, where a photon is created at the point of an electronic transition — except here, an electron-neutrino pair is created instead of a photon.

At the quark level, $\beta^-$ decay is:

$$d \to u + e^- + \bar{\nu}_e$$

mediated by the emission and decay of a virtual $W^-$ boson. But at the energy scales of nuclear beta decay ($Q \lesssim 20\,\text{MeV}$), the $W$ boson mass ($M_W = 80.4\,\text{GeV}$) is enormous compared to the energy transfer. The $W$ propagator reduces to a constant, and the interaction is effectively a four-fermion contact interaction — exactly Fermi's original picture. This is the low-energy limit of the electroweak theory, valid for $Q \ll M_W c^2$.

14.3.2 Fermi's Golden Rule Applied to Beta Decay

From Chapter 5, Fermi's golden rule gives the transition rate:

$$\lambda = \frac{2\pi}{\hbar} |\langle f | H_\beta | i \rangle|^2 \rho(E_f)$$

where $H_\beta$ is the weak interaction Hamiltonian, $|i\rangle$ and $|f\rangle$ are the initial and final states, and $\rho(E_f)$ is the density of final states.

We need to evaluate three ingredients:

1. The matrix element $|\langle f | H_\beta | i \rangle|^2$
2. The density of states $\rho(E_f)$ for the three-body final state
3. The Coulomb correction for the effect of the nuclear charge on the emitted electron

14.3.3 The Matrix Element

The initial state is the parent nucleus $|i\rangle = |\Psi_i\rangle$. The final state is $|f\rangle = |\Psi_f\rangle |e^-\rangle |\bar{\nu}_e\rangle$ — the daughter nucleus, the electron, and the antineutrino.

Fermi's interaction Hamiltonian has the form:

$$H_\beta = \frac{G_F}{\sqrt{2}} \int \left[\bar{\psi}_p \hat{O} \psi_n\right] \left[\bar{\psi}_e \hat{O}' \psi_\nu\right] d^3r$$

where $G_F$ is the Fermi coupling constant and $\hat{O}$, $\hat{O}'$ are operators that specify the Lorentz structure of the interaction. The current value of the Fermi coupling constant, determined from muon decay, is:

$$G_F / (\hbar c)^3 = 1.1664 \times 10^{-5}\,\text{GeV}^{-2}$$

This tiny coupling constant — compared to the electromagnetic fine structure constant $\alpha \approx 1/137$ or the strong coupling $\alpha_s \sim 1$ — explains why the weak interaction is "weak" and why beta decay half-lives are long compared to electromagnetic (gamma) transitions.

The allowed approximation. The electron and neutrino wavefunctions inside the nucleus can be expanded in plane waves:

$$\psi_e(\mathbf{r}) = \frac{1}{\sqrt{V}} e^{i\mathbf{k}_e \cdot \mathbf{r}}, \qquad \psi_\nu(\mathbf{r}) = \frac{1}{\sqrt{V}} e^{i\mathbf{k}_\nu \cdot \mathbf{r}}$$

For an allowed transition, the wavelengths of the electron and neutrino ($\lambda_e, \lambda_\nu \sim 10^2 - 10^3\,\text{fm}$) are much larger than the nuclear radius ($R \sim \text{few}\,\text{fm}$), so:

$$e^{i\mathbf{k} \cdot \mathbf{r}} \approx 1 + i\mathbf{k} \cdot \mathbf{r} + \cdots$$

Keeping only the leading term ($e^{i\mathbf{k}\cdot\mathbf{r}} \approx 1$) gives the allowed approximation — the leptons are emitted in $s$-waves ($l = 0$) with respect to the nucleus. Higher-order terms give forbidden transitions.

In the allowed approximation, the matrix element factorizes:

$$|\mathcal{M}|^2 = \frac{G_F^2}{2V^2} |M_{fi}|^2$$

where $M_{fi}$ is the nuclear matrix element:

$$M_{fi} = \langle \Psi_f | \hat{O}_\text{nuclear} | \Psi_i \rangle$$

The nuclear operator $\hat{O}_\text{nuclear}$ depends on the type of transition (Fermi or Gamow-Teller), as we discuss in Section 14.4.

14.3.4 Deriving the Allowed Beta Spectrum Shape

This is the central derivation of the chapter. We seek $N(T_e)\,dT_e$, the number of electrons emitted per unit time with kinetic energy between $T_e$ and $T_e + dT_e$.

Step 1: Density of states. The final state has three particles: the daughter nucleus (which recoils but whose kinetic energy is negligible), the electron, and the neutrino. The number of electron states in a volume $V$ with momentum between $p_e$ and $p_e + dp_e$ in solid angle $d\Omega_e$ is:

$$dn_e = \frac{V}{(2\pi\hbar)^3} p_e^2 \, dp_e \, d\Omega_e$$

Similarly for the neutrino:

$$dn_\nu = \frac{V}{(2\pi\hbar)^3} p_\nu^2 \, dp_\nu \, d\Omega_\nu$$

Step 2: Energy conservation. For a massless neutrino (an excellent approximation for the present discussion — neutrino masses are at most $\sim 1$ eV):

$$Q = T_e + E_\nu = T_e + p_\nu c$$

where $T_e$ is the electron kinetic energy and $E_\nu = p_\nu c$ is the neutrino energy. Therefore:

$$p_\nu = \frac{Q - T_e}{c}, \qquad dp_\nu = \frac{dT_e}{c} \quad (\text{for fixed } T_e)$$

Wait — we want the spectrum as a function of electron energy, so we eliminate $p_\nu$ using energy conservation:

$$p_\nu = \frac{(Q - T_e)}{c}$$

Step 3: Transition rate. Integrating over the neutrino solid angle ($4\pi$) and electron solid angle ($4\pi$), and using Fermi's golden rule:

$$dN(T_e) = \frac{2\pi}{\hbar} \frac{G_F^2}{2V^2} |M_{fi}|^2 \cdot \frac{V}{(2\pi\hbar)^3} p_e^2 dp_e \cdot 4\pi \cdot \frac{V}{(2\pi\hbar)^3} p_\nu^2 dp_\nu \cdot 4\pi$$

The factors of $V$ cancel (as they must for a physical observable). Converting $dp_e$ and $dp_\nu$ to $dT_e$ using relativistic kinematics:

For the electron: $E_e = T_e + m_e c^2$, $E_e^2 = (p_e c)^2 + (m_e c^2)^2$, so:

$$p_e c = \sqrt{T_e(T_e + 2m_e c^2)}, \qquad p_e \, dp_e = \frac{E_e}{c^2} dT_e$$

For the neutrino ($m_\nu \approx 0$): $p_\nu = E_\nu / c = (Q - T_e)/c$.

Collecting all factors, the allowed beta spectrum shape is:

$$\boxed{N(T_e) = C \cdot F(Z', T_e) \cdot p_e \cdot E_e \cdot (Q - T_e)^2}$$

where:

• $C = \frac{G_F^2 |M_{fi}|^2}{2\pi^3 \hbar^7 c^5}$ is a constant containing the nuclear matrix element
• $p_e = \frac{1}{c}\sqrt{T_e(T_e + 2m_ec^2)}$ is the electron momentum
• $E_e = T_e + m_e c^2$ is the total electron energy
• $(Q - T_e)^2 = E_\nu^2$ is the neutrino phase space factor (for $m_\nu = 0$)
• $F(Z', T_e)$ is the Fermi function — the Coulomb correction

💡 Physical Interpretation: The spectrum shape has a transparent physical origin. The factor $p_e E_e$ counts electron states (momentum-space density of states times the Jacobian from momentum to energy). The factor $(Q - T_e)^2$ counts neutrino states — when the electron takes most of the energy, few states are available for the neutrino, and vice versa. The spectrum vanishes at $T_e = 0$ (because $p_e \to 0$, so there are no electron states) and at $T_e = Q$ (because $(Q - T_e)^2 \to 0$, so there are no neutrino states).

14.3.5 The Fermi Function

The derivation above treated the electron as a free particle. But the emitted electron (or positron) moves in the Coulomb field of the daughter nucleus (charge $Z'e$). For $\beta^-$ decay, the Coulomb attraction pulls the electron toward the nucleus, enhancing the emission probability for low-energy electrons. For $\beta^+$ decay, the Coulomb repulsion suppresses low-energy positrons.

The Fermi function $F(Z', T_e)$ corrects for this effect. It is defined as the ratio of the electron probability density at the nuclear surface with the Coulomb field to the density without it:

$$F(Z', T_e) = \frac{|\psi_e(R)|^2_\text{Coulomb}}{|\psi_e(R)|^2_\text{free}}$$

For a point-charge nucleus, the exact result involves the gamma function:

$$F(Z', T_e) = 2(1 + \gamma_0)\left(\frac{2p_e R}{\hbar}\right)^{2(\gamma_0 - 1)} e^{\pi\eta} \frac{|\Gamma(\gamma_0 + i\eta)|^2}{[\Gamma(2\gamma_0 + 1)]^2}$$

where $\gamma_0 = \sqrt{1 - (\alpha Z')^2}$, $\eta = \pm \alpha Z' E_e / (p_e c)$ (positive for $\beta^-$, negative for $\beta^+$), and $\alpha \approx 1/137$ is the fine structure constant.

For practical calculations, an excellent non-relativistic approximation (valid for $T_e \gg$ a few keV and not-too-heavy nuclei) is:

$$F(Z', T_e) \approx \frac{2\pi\eta}{1 - e^{-2\pi\eta}}$$

where $\eta = \pm \alpha Z' / \beta_e$ and $\beta_e = v_e/c = p_e c / E_e$ is the electron velocity in units of $c$. This is the formula implemented in the chapter's Python code.

The effect of the Fermi function on the spectrum is dramatic:

• $\beta^-$ decay: $F > 1$ for all energies, with the enhancement largest at low $T_e$ (where the electron moves slowly and the Coulomb attraction is strongest). For heavy nuclei ($Z' \sim 80$), $F$ can exceed 10 at low energies, producing a pronounced skewing of the spectrum toward low energies.
• $\beta^+$ decay: $F < 1$ for all energies, suppressing the low-energy end of the spectrum.

📊 Worked Example — Fermi Function for $^{60}$Co:

For $^{60}$Co $\beta^-$ decay ($Z' = 28$), compute $F$ at $T_e = 50\,\text{keV}$:

1. Total energy: $E_e = 50 + 511 = 561\,\text{keV}$
2. Momentum: $p_e c = \sqrt{50 \times (50 + 1022)} = \sqrt{53{,}600} = 231.5\,\text{keV}$
3. Velocity: $\beta_e = p_e c / E_e = 231.5 / 561 = 0.4126$
4. Sommerfeld parameter: $\eta = \alpha Z' / \beta_e = (28/137) / 0.4126 = 0.495$
5. $2\pi\eta = 3.112$
6. $F = 2\pi\eta / (1 - e^{-2\pi\eta}) = 3.112 / (1 - 0.0442) = 3.26$

The Coulomb field of the daughter nickel nucleus more than triples the electron emission probability at this energy. At $T_e = 300\,\text{keV}$, the same calculation gives $\beta_e = 0.798$, $\eta = 0.256$, and $F = 1.76$ — still a significant enhancement, but much reduced at higher electron velocities where the Coulomb effect is weaker. These results are confirmed by the Python code in beta_spectrum.py.

14.3.6 Connection to the Standard Model

Fermi's 1934 theory assumed a vector (V) interaction — the same Lorentz structure as electromagnetism. We now know from the full electroweak theory (Glashow, Weinberg, Salam, 1967-1968) that the weak interaction in beta decay has the structure V$-$A (vector minus axial vector). The V interaction is responsible for Fermi transitions and the A interaction for Gamow-Teller transitions (Section 14.4).

The Fermi coupling constant $G_F$ is related to the fundamental parameters of the Standard Model:

$$\frac{G_F}{\sqrt{2}} = \frac{g_W^2}{8 M_W^2 c^4}$$

where $g_W$ is the weak coupling constant and $M_W = 80.4\,\text{GeV}/c^2$ is the $W$ boson mass. This relation makes explicit why the weak interaction appears "weak" at low energies: it is not that the fundamental coupling $g_W$ is particularly small ($g_W \approx 0.65$, comparable to the electromagnetic coupling), but that the $W$ boson is heavy. The effective low-energy coupling $G_F \propto g_W^2 / M_W^2$ is suppressed by two powers of the large $W$ mass.

14.4 Selection Rules: Fermi and Gamow-Teller Transitions

14.4.1 The Nuclear Matrix Element

The nuclear matrix element depends on the operator $\hat{O}_\text{nuclear}$ that acts on the nuclear wavefunctions. The V$-$A structure of the weak interaction gives two types of transitions.

Fermi transitions arise from the vector (V) part of the interaction. The nuclear operator is the isospin raising/lowering operator:

$$\hat{O}_F = \sum_{k=1}^{A} \hat{\tau}_\pm(k)$$

This operator changes a neutron to a proton (or vice versa) without affecting the nucleon's spin or spatial wavefunction. The nuclear matrix element is:

$$|M_F|^2 = |\langle \Psi_f | \sum_k \hat{\tau}_\pm(k) | \Psi_i \rangle|^2$$

Selection rules for Fermi transitions:

• $\Delta J = 0$ (no change in total angular momentum of the nucleus)
• $\Delta \pi = +$ (no parity change)
• $\Delta T = 0$, $\Delta T_z = \pm 1$ (isospin selection rules in the isospin-symmetric limit)

💡 Physical Intuition: In a Fermi transition, only the nucleon's isospin projection changes ($n \to p$ or $p \to n$). The spin and spatial wavefunctions are untouched, so the nuclear spin $J$ cannot change. It is the simplest possible nuclear rearrangement.

Gamow-Teller transitions arise from the axial-vector (A) part of the interaction. The nuclear operator is:

$$\hat{O}_{GT} = \sum_{k=1}^{A} \hat{\boldsymbol{\sigma}}(k) \hat{\tau}_\pm(k)$$

where $\hat{\boldsymbol{\sigma}}(k)$ is the Pauli spin operator for nucleon $k$. This operator flips both the isospin and the spin of a nucleon.

Selection rules for Gamow-Teller transitions:

• $\Delta J = 0, \pm 1$ (but not $0 \to 0$)
• $\Delta \pi = +$ (no parity change)

The exclusion of $0 \to 0$ for Gamow-Teller is critical: a $0^+ \to 0^+$ transition can only proceed via the Fermi operator. This makes superallowed $0^+ \to 0^+$ transitions pure Fermi transitions — the cleanest probe of the vector coupling constant.

14.4.2 Allowed vs. Forbidden Transitions

Allowed transitions ($l = 0$): the electron-neutrino pair carries zero orbital angular momentum. Both Fermi and Gamow-Teller selections apply. The spectrum shape is given by the formula derived in Section 14.3.

First forbidden transitions ($l = 1$): the first-order term in the plane-wave expansion contributes. The matrix elements involve operators like $\mathbf{r}\hat{\tau}_\pm$ (for the parity-changing vector current) or $[\mathbf{r} \times \boldsymbol{\sigma}]\hat{\tau}_\pm$ (for the parity-changing axial current). Selection rules:

• $\Delta J = 0, \pm 1, \pm 2$
• $\Delta \pi = -$ (parity changes)

The additional factor of $|\mathbf{k} \cdot \mathbf{r}|^2 \sim (R/\lambda)^2 \sim 10^{-4}$ to $10^{-2}$ suppresses forbidden transitions relative to allowed ones by several orders of magnitude in rate.

Second forbidden ($l = 2$): $\Delta J = 2, 3$; $\Delta \pi = +$. Suppressed by another factor of $(R/\lambda)^2$.

Higher forbidden transitions continue the pattern. Each additional unit of $l$ suppresses the rate by $\sim (R/\lambda)^2$ and changes the parity by $(-)^l$.

🔬 Extreme Example: The beta decay of $^{115}$In ($9/2^+ \to 1/2^-$, $\Delta J = 4$, $\Delta \pi = -$) is a fourth-forbidden unique transition with a half-life of $4.4 \times 10^{14}$ years — roughly $10^4$ times the age of the universe. It has been measured.

A special subclass are the unique forbidden transitions, where $\Delta J = l + 1$ (the maximum angular momentum change for a given order of forbiddenness). These transitions proceed through a single matrix element, making their spectrum shapes theoretically clean. The spectrum shape for an $n$th-forbidden unique transition is modified from the allowed shape by an additional factor:

$$N_n(T_e) \propto p_e^{2n} \cdot p_\nu^{2n} \cdot N_\text{allowed}(T_e)$$

This distortion is observable in the Kurie plot: a forbidden unique transition produces a characteristically curved Kurie plot. The deviation from linearity is itself a diagnostic of the degree of forbiddenness.

Example — $^{40}$K: Potassium-40 ($J^\pi = 4^-, T_{1/2} = 1.248 \times 10^9\,\text{yr}$) is one of the most geologically important beta emitters. Its $\beta^-$ branch ($89.3\%$) goes to $^{40}$Ca ($J^\pi = 0^+$), with $\Delta J = 4$ and $\Delta\pi = -$. This is a third-forbidden unique transition ($l = 3$, $\Delta J = l + 1 = 4$). The EC branch ($10.7\%$) goes to $^{40}$Ar ($J^\pi = 0^+$), also third-forbidden. The potassium-argon dating method in geology relies on the accumulation of $^{40}$Ar from the EC branch in potassium-bearing minerals.

14.4.3 Mixed Transitions

When the selection rules allow both Fermi and Gamow-Teller contributions ($\Delta J = 0$ or $\pm 1$, $\Delta\pi = +$, and neither initial nor final state has $J=0$ if $\Delta J = 0$ is involved), the total rate is the incoherent sum:

$$|M_{fi}|^2 = g_V^2 |M_F|^2 + g_A^2 |M_{GT}|^2$$

where $g_V = 1$ (conserved vector current, CVC) and $g_A \approx -1.276$ (experimentally determined from neutron beta decay). The ratio $\rho^2 = g_A^2 |M_{GT}|^2 / (g_V^2 |M_F|^2)$ characterizes the mixing.

The value $|g_A/g_V| = 1.2754 \pm 0.0013$ (from the Particle Data Group, 2024) is one of the most precisely measured quantities in weak interaction physics. Its deviation from unity arises from strong-interaction effects (QCD corrections to the axial current inside the nucleon) — a beautiful example of the interplay between the weak and strong forces.

14.5 The Kurie Plot and ft-Values

14.5.1 The Kurie Plot

The allowed beta spectrum (Section 14.3) can be rearranged to define the Kurie function:

$$K(T_e) = \sqrt{\frac{N(T_e)}{F(Z', T_e) \cdot p_e \cdot E_e}}$$

If the transition is allowed and $m_\nu = 0$, then:

$$K(T_e) = C' \cdot (Q - T_e)$$

A plot of $K(T_e)$ versus $T_e$ — the Kurie plot (also called a Fermi-Kurie plot) — should be a straight line that intersects the $T_e$ axis at $T_e = Q$. This provides:

1. A test of the allowed spectrum shape. Deviations from linearity indicate forbidden transitions, atomic final-state effects, or instrumental distortions.
2. A precise determination of $Q$. The x-intercept gives the endpoint energy.
3. Sensitivity to neutrino mass. If $m_\nu > 0$, the spectrum near the endpoint is modified:

$$(Q - T_e)^2 \to (Q - T_e)\sqrt{(Q - T_e)^2 - m_\nu^2 c^4}$$

This produces a deviation from linearity near the endpoint, with the Kurie plot curving downward and intersecting the axis at $T_e = Q - m_\nu c^2$. The KATRIN experiment (Section 14.7) exploits this to measure $m_\nu$.

📊 Worked Example — Tritium Kurie Plot: Tritium ($^3$H $\to$ $^3$He) has $Q = 18.6\,\text{keV}$ and is a superallowed mixed transition. The Kurie plot is beautifully linear over almost the entire spectrum, with the endpoint clearly determining $Q$. The tiny deviation from linearity near the endpoint constrains the neutrino mass to $m_\nu < 0.45\,\text{eV}$ (KATRIN, 2024).

Practical considerations: In real experiments, several effects can distort the Kurie plot and must be corrected:

• Detector resolution. Finite energy resolution smears the sharp endpoint, making the Kurie plot curve upward near $Q$. This is particularly problematic for neutrino mass measurements, where the effect of $m_\nu$ is a subtle downward curvature in exactly the same region.
• Source thickness effects. If the source is not infinitely thin, electrons lose energy escaping the source material, distorting the low-energy part of the spectrum. Correction requires knowledge of the source geometry and electron stopping power (Chapter 16).
• Atomic final-state effects. The sudden change in nuclear charge ($Z \to Z \pm 1$) leaves the atomic electron cloud in an excited state. For tritium decay, the daughter $^3$He$^+$ ion can be left in various electronic states, each contributing a spectrum with a slightly different endpoint. These final-state distributions must be computed quantum mechanically and folded into the Kurie analysis.
• Radiative corrections. Virtual and real photon emission modifies the spectrum shape at the percent level. These QED corrections are well-calculated and must be included in precision analyses.

14.5.2 The ft-Value

The total decay rate for an allowed transition is obtained by integrating $N(T_e)$ over all electron energies:

$$\lambda = \frac{1}{\tau} = \frac{G_F^2 |M_{fi}|^2}{2\pi^3 \hbar^7 c^5} \int_0^{Q} F(Z', T_e) \, p_e \, E_e \, (Q - T_e)^2 \, dT_e$$

The integral over the spectrum shape is called the Fermi integral $f(Z', Q)$:

$$f(Z', Q) = \frac{1}{(m_e c)^3 (m_e c^2)^2} \int_0^{Q} F(Z', T_e) \, p_e \, E_e \, (Q - T_e)^2 \, dT_e$$

(the prefactor makes $f$ dimensionless). Then:

$$\lambda = \frac{m_e^5 c^4 G_F^2}{2\pi^3 \hbar^7} |M_{fi}|^2 \cdot f(Z', Q)$$

Defining the partial half-life $t = t_{1/2} / \text{BR}$ (where BR is the branching ratio for the specific transition), the ft-value is:

$$\boxed{ft = \frac{2\pi^3 \hbar^7 \ln 2}{m_e^5 c^4 G_F^2 |M_{fi}|^2} = \frac{K}{|M_{fi}|^2}}$$

where $K = 2\pi^3 \hbar^7 \ln 2 / (m_e^5 c^4 G_F^2) = 6144.2 \pm 1.6\,\text{s}$ (using the current best values of the constants).

The ft-value depends only on the nuclear matrix element (and fundamental constants). It is independent of the Q-value and $Z$ — those dependences are absorbed into the Fermi integral $f$. The ft-value is therefore the correct quantity for comparing the "intrinsic strength" of different beta transitions.

14.5.3 The log ft Classification

In practice, the comparative half-life $\log_{10}(ft)$ (with $ft$ in seconds) is used to classify transitions:

💡 Key Insight: The neutron itself undergoes beta decay: $n \to p + e^- + \bar{\nu}_e$ with $\tau = 878.4 \pm 0.5\,\text{s}$ and $Q = 0.782\,\text{MeV}$. This is a superallowed mixed (Fermi + Gamow-Teller) transition with $\log ft = 3.07$. Free neutron decay is one of the most precisely studied beta decays and provides the most direct measurement of $g_A / g_V$.

📊 Worked Example — Classifying a Transition by log ft:

The beta decay of $^{60}$Co ($J^\pi = 5^+$) to $^{60}$Ni* ($J^\pi = 4^+$) has $Q = 0.318\,\text{MeV}$, $t_{1/2} = 5.271\,\text{yr} = 1.663 \times 10^8\,\text{s}$, and $\text{BR} = 99.88\%$.

Step 1: Partial half-life: $t = t_{1/2}/\text{BR} = 1.663 \times 10^8 / 0.9988 = 1.665 \times 10^8\,\text{s}$.

Step 2: The Fermi integral for $Z' = 28$, $Q = 0.318\,\text{MeV}$ (computed numerically or from tables) is $f = 1.076$.

Step 3: $ft = 1.076 \times 1.665 \times 10^8 = 1.79 \times 10^8\,\text{s}$.

Step 4: $\log ft = \log_{10}(1.79 \times 10^8) = 8.25$.

Classification: $\log ft = 8.25$ falls in the first-forbidden range ($6-9$). But wait — the selection rules give $\Delta J = 1$, $\Delta\pi = +$, which is allowed Gamow-Teller. Why is the log ft so high? The answer is that the $^{60}$Co $\to$ $^{60}$Ni transition involves a change in the nuclear configuration that makes the GT matrix element small (the initial and final nuclear wavefunctions have poor overlap). This illustrates that log ft classifies the nuclear matrix element, not just the selection rule — a transition can be "allowed" by selection rules but have a small matrix element, giving a large log ft similar to forbidden transitions. Such transitions are called $l$-forbidden or hindered allowed*.

14.5.4 Superallowed $0^+ \to 0^+$ Transitions

The $0^+ \to 0^+$ superallowed transitions are the gold standard of beta decay physics. Since the Gamow-Teller operator cannot connect two $0^+$ states, these are pure Fermi transitions with:

$$|M_F|^2 = |N - Z| = 2 \quad (\text{for } T=1 \text{ analog transitions})$$

where the last equality holds for transitions between members of the same isospin triplet. The ft-values for all superallowed $0^+ \to 0^+$ transitions should therefore be identical (after small radiative and isospin-breaking corrections), and they are:

$$\mathcal{F}t = ft(1 + \delta_R')(1 + \delta_\text{NS} - \delta_C) = 3072.24 \pm 0.72\,\text{s}$$

This world-average $\mathcal{F}t$ value, compiled from 15 precisely measured superallowed transitions ($^{10}$C, $^{14}$O, $^{22}$Mg, $^{26}$Al$^m$, $^{34}$Cl, $^{34}$Ar, $^{38}$K$^m$, $^{42}$Sc, $^{46}$V, $^{50}$Mn, $^{54}$Co, $^{62}$Ga, $^{66}$As, $^{70}$Br, $^{74}$Rb), provides:

1. The most precise determination of $G_V$ (and hence $V_{ud}$): $|V_{ud}|^2 = K / (2\mathcal{F}t G_F^2)$, yielding $|V_{ud}| = 0.97373 \pm 0.00031$.
2. A stringent test of the CVC hypothesis (the vector coupling is not renormalized by the strong interaction).
3. A test of CKM unitarity: $|V_{ud}|^2 + |V_{us}|^2 + |V_{ub}|^2 = 1$ — currently satisfied to $\sim 0.1\%$.

🔬 Precision Frontier: The superallowed beta decay program is one of the most precise programs in all of nuclear physics. The experimental ft-values are known to $\sim 0.03\%$. The theoretical corrections ($\delta_R'$, $\delta_\text{NS}$, $\delta_C$) — radiative corrections and isospin symmetry breaking — are the limiting factor and the subject of active research.

14.6 Parity Violation: The Wu Experiment

14.6.1 The $\theta$-$\tau$ Puzzle

By the mid-1950s, two particles — called $\theta^+$ and $\tau^+$ — appeared to be identical in mass, lifetime, and charge, but decayed into final states of opposite parity: $\theta^+ \to \pi^+\pi^0$ (even parity) and $\tau^+ \to \pi^+\pi^+\pi^-$ (odd parity). If parity were conserved, these had to be different particles. But their identical properties made this hard to accept.

In 1956, Tsung-Dao Lee and Chen-Ning Yang published a landmark paper analyzing all existing experimental evidence for parity conservation in the weak interaction. They found, to the astonishment of the physics community, that there was no experimental evidence for parity conservation in weak decays. Parity had been tested extensively in electromagnetic and strong interactions, but everyone had simply assumed it held for the weak interaction as well. Lee and Yang proposed specific experiments to test it.

14.6.2 The Experiment

Chien-Shiung Wu, an experimental physicist at Columbia University and the world's leading expert on beta decay, immediately recognized the significance of Lee and Yang's paper and moved swiftly to perform the definitive test. Working with the National Bureau of Standards low-temperature group (Ernest Ambler, Raymond Hayward, Dale Hoppes, and Ralph Hudson), she designed an experiment using $^{60}$Co.

The setup: $^{60}$Co nuclei ($J^\pi = 5^+$) were embedded in a cerium magnesium nitrate crystal and cooled to 0.01 K in a magnetic field. At this temperature, the thermal energy $k_BT \approx 10^{-6}\,\text{eV}$ is much less than the nuclear magnetic interaction energy, so the $^{60}$Co nuclear spins are highly polarized — predominantly aligned along the magnetic field direction.

$^{60}$Co undergoes beta decay:

$$^{60}_{27}\text{Co} \to\, ^{60}_{28}\text{Ni}^* + e^- + \bar{\nu}_e$$

This is predominantly a Gamow-Teller transition ($5^+ \to 4^+$, $\Delta J = 1$, $\Delta\pi = +$) with $Q = 0.318\,\text{MeV}$.

The test: If parity is conserved, the angular distribution of emitted electrons must be symmetric with respect to the nuclear spin direction. That is, the number of electrons emitted parallel to the spin ($\hat{J}$) must equal the number emitted antiparallel to it. In mathematical terms, the distribution is:

$$W(\theta) = 1 + \alpha \frac{\langle J \rangle}{J} \cos\theta$$

where $\theta$ is the angle between the electron momentum and the nuclear spin, and $\alpha$ is an asymmetry parameter. Parity conservation requires $\alpha = 0$.

The parity argument: Under a parity transformation $\hat{P}$: the nuclear spin $\mathbf{J}$ (an axial vector) is unchanged, but the electron momentum $\mathbf{p}$ (a polar vector) reverses. The quantity $\mathbf{J} \cdot \mathbf{p}$ is therefore a pseudoscalar — it changes sign under parity. If parity is conserved, the expectation value of any pseudoscalar must vanish, requiring $\alpha = 0$.

14.6.3 The Result

Wu's experiment, published in January 1957, found a large asymmetry: more electrons were emitted opposite to the nuclear spin direction than along it, with $\alpha \approx -1$ (consistent with the V$-$A theory prediction of $\alpha = -v/c$ for the electron). Parity was violated — not slightly, but maximally.

The asymmetry disappeared as the sample warmed up and the nuclear polarization was lost, confirming that the effect was genuine and not an instrumental artifact.

📜 Historical Impact: The result was described by I.I. Rabi as leaving physics "a shambles." Wolfgang Pauli, who had bet that parity was conserved, famously remarked: "I cannot believe that God is a weak left-hander." Richard Feynman lost $50 in a bet. The discovery fundamentally changed our understanding of the weak interaction and led directly to the V$-$A theory of the weak force.

Lee and Yang received the 1957 Nobel Prize in Physics — one of the fastest Nobel awards in history, given in the same year as the experimental confirmation. Wu, despite having performed the crucial experiment, was not included — a decision widely regarded as one of the most egregious omissions in Nobel history.

14.6.4 Consequences

The Wu experiment and its immediate successors established that:

1. The weak interaction violates parity maximally. In the V$-$A theory, only left-handed particles and right-handed antiparticles participate in the charged-current weak interaction. This is built into the Standard Model.

2. The neutrino has a definite helicity. The Goldhaber experiment (1958) measured the helicity of the neutrino emitted in EC of $^{152}$Eu and found it to be left-handed ($h = -1$). The antineutrino is right-handed ($h = +1$).

3. C symmetry (charge conjugation) is also violated. The combined operation CP (charge conjugation times parity) was initially thought to be conserved, but CP violation was discovered in 1964 in the kaon system (Cronin and Fitch) and is now understood as a fundamental feature of the Standard Model.

4. Beta decay angular correlations — between the electron momentum, neutrino momentum, and nuclear spin — contain rich information about the structure of the weak interaction. Precision measurements of these correlations continue to test the Standard Model and search for new physics.

14.7 Double Beta Decay

14.7.1 Two-Neutrino Double Beta Decay ($2\nu\beta\beta$)

Some even-even nuclei ($J^\pi = 0^+$) cannot undergo ordinary beta decay because the neighboring odd-odd isobar is heavier (due to the pairing energy term in the SEMF, Chapter 4). However, they can decay to the next even-even isobar, which is lighter, by emitting two electrons and two antineutrinos simultaneously:

$$(Z, A) \to (Z+2, A) + 2e^- + 2\bar{\nu}_e$$

This is a second-order weak process — two simultaneous beta decays. Its rate is proportional to $G_F^4$, making it extraordinarily slow. Typical half-lives are $10^{18} - 10^{24}$ years.

Observation: Two-neutrino double beta decay ($2\nu\beta\beta$) has been observed in 11 nuclei, including:

These are among the longest half-lives ever measured — and measuring them required heroic experiments with ultra-low backgrounds and massive detectors.

14.7.2 Neutrinoless Double Beta Decay ($0\nu\beta\beta$)

If the neutrino is a Majorana particle — that is, if the neutrino is its own antiparticle ($\nu = \bar{\nu}$) — then a fundamentally different process becomes possible:

$$(Z, A) \to (Z+2, A) + 2e^-$$

No neutrinos are emitted. The virtual neutrino emitted at one vertex is absorbed at the other, which is possible only if $\nu = \bar{\nu}$ (Majorana condition) and if the neutrino has nonzero mass (to allow helicity flip at the propagator).

Why this matters: The observation of $0\nu\beta\beta$ decay would:

1. Prove that lepton number is violated ($\Delta L = 2$). This is not merely a symmetry violation — it would be the first observation of a process that violates any of the accidental symmetries of the Standard Model.
2. Establish that the neutrino is a Majorana particle, resolving one of the fundamental questions in particle physics.
3. Provide a measurement of the effective Majorana mass $\langle m_{\beta\beta} \rangle = |\sum_i U_{ei}^2 m_i|$, where the sum runs over mass eigenstates and $U_{ei}$ are elements of the neutrino mixing matrix.
4. Support leptogenesis as a possible explanation for the baryon asymmetry of the universe.

The experimental signature is distinctive: in $2\nu\beta\beta$, the two electrons share the available energy with two neutrinos, producing a continuous summed-energy spectrum. In $0\nu\beta\beta$, the two electrons carry the full $Q$-value, producing a sharp peak at $E = Q$ in the summed electron energy spectrum. The experimental challenge is detecting this monoenergetic peak above backgrounds.

14.7.3 Current Experimental Searches

The search for $0\nu\beta\beta$ is one of the most active experimental programs in nuclear and particle physics. The current generation of experiments includes:

GERDA/LEGEND (Gran Sasso, Italy): Uses bare $^{76}$Ge detectors immersed in liquid argon. GERDA set the strongest limit on $^{76}$Ge: $T_{1/2}^{0\nu} > 1.8 \times 10^{26}$ years (90% CL). LEGEND-200 began data-taking in 2023 with 200 kg of enriched $^{76}$Ge, and LEGEND-1000 (1 ton) is planned for the late 2020s.

KamLAND-Zen (Kamioka, Japan): Dissolves $^{136}$Xe in liquid scintillator inside the KamLAND detector. Current limit: $T_{1/2}^{0\nu} > 2.3 \times 10^{26}$ years. KamLAND-Zen 800 (with 750 kg of enriched Xe) is operating.

CUORE (Gran Sasso, Italy): Uses 988 TeO$_2$ bolometers containing $^{130}$Te, operated at 10 mK. Current limit: $T_{1/2}^{0\nu} > 2.2 \times 10^{25}$ years. The successor CUPID will use $^{100}$Mo-enriched Li$_2$MoO$_4$ scintillating bolometers.

nEXO (proposed): A 5-tonne liquid $^{136}$Xe time projection chamber planned for SNOLAB (Canada). Projected sensitivity: $T_{1/2}^{0\nu} > 1.35 \times 10^{28}$ years, covering the entire inverted mass ordering region.

NEXT (Canfranc, Spain): A high-pressure gaseous $^{136}$Xe TPC exploiting the topological signature of two-electron events. NEXT-100 (100 kg) is under construction.

The current best limits on the effective Majorana mass from these experiments are:

$$\langle m_{\beta\beta} \rangle < 36 - 156\,\text{meV}$$

(the range reflects uncertainties in the nuclear matrix elements). The next generation of experiments aims to reach $\langle m_{\beta\beta} \rangle \sim 10 - 20\,\text{meV}$, which would cover the entire parameter space predicted by the inverted mass ordering of neutrino masses.

⚠️ The Nuclear Matrix Element Problem: The sensitivity of $0\nu\beta\beta$ experiments to $\langle m_{\beta\beta} \rangle$ depends on the nuclear matrix element (NME) of the $0\nu\beta\beta$ transition, which must be calculated theoretically. Different nuclear structure models (shell model, QRPA, interacting boson model, energy density functional methods, ab initio calculations) give NMEs that differ by factors of 2-3. Reducing this theoretical uncertainty is one of the most important challenges in nuclear theory.

14.7.4 Complementary Neutrino Mass Measurements

The search for $0\nu\beta\beta$ is complementary to other approaches to neutrino mass:

• KATRIN (Karlsruhe, Germany): Measures the tritium beta spectrum endpoint with an electrostatic spectrometer. The 2024 result gives $m_\nu < 0.45\,\text{eV}$ (90% CL), with a design sensitivity of $0.2\,\text{eV}$.
• Cosmological constraints: The Planck satellite combined with baryon acoustic oscillation data gives $\sum m_\nu < 0.12\,\text{eV}$ (95% CL), though this depends on the cosmological model.
• Project 8 (proposed): Measures the cyclotron radiation emitted by individual tritium beta electrons in a magnetic field (cyclotron radiation emission spectroscopy, CRES). Aims for sensitivity below $40\,\text{meV}$.

The relationship between these measurements is:

• KATRIN measures $m_\beta = \sqrt{\sum |U_{ei}|^2 m_i^2}$ (the kinematic mass)
• $0\nu\beta\beta$ measures $\langle m_{\beta\beta} \rangle = |\sum U_{ei}^2 m_i|$ (the Majorana mass, includes phases)
• Cosmology measures $\sum m_i$ (the total mass)

These three quantities probe different combinations of neutrino masses and mixing parameters, and together they can determine the mass hierarchy and the Majorana phases.

14.8 Summary and Connections

Key Results

1. Three modes of beta decay — $\beta^-$, $\beta^+$, and EC — are manifestations of the same weak interaction process: the conversion of a $d$ quark to a $u$ quark (or vice versa) via $W$ boson exchange.

2. The allowed beta spectrum has the shape $N(T_e) \propto F(Z', T_e) \cdot p_e \cdot E_e \cdot (Q - T_e)^2$, derived from Fermi's golden rule with phase space counting for the three-body final state.

3. Fermi transitions ($\Delta J = 0$, $\Delta\pi = +$) and Gamow-Teller transitions ($\Delta J = 0, \pm 1$ not $0 \to 0$, $\Delta\pi = +$) are the two types of allowed transitions, corresponding to the V and A parts of the weak interaction.

4. The Kurie plot linearizes the spectrum to extract $Q$ and test the allowed shape. The ft-value removes the $Q$ and $Z$ dependences to give a measure of the intrinsic nuclear matrix element.

5. Parity is maximally violated in beta decay — demonstrated by the Wu experiment (1957). Only left-handed particles and right-handed antiparticles participate in charged-current weak interactions.

6. Double beta decay ($2\nu\beta\beta$) is the rarest observed nuclear process. The search for its neutrinoless variant ($0\nu\beta\beta$) would prove the neutrino is a Majorana particle and is one of the most important experimental programs in physics.

Looking Forward

• Chapter 15 (Gamma Decay): We will see how the electromagnetic interaction de-excites nuclei — the selection rules for gamma transitions complement those for beta decay, and together they determine how excited nuclear states populated in beta decay return to the ground state.
• Chapter 22 (Stellar Nucleosynthesis): Beta decay is the rate-limiting step in the pp chain and CNO cycle. The weak interaction's slowness is literally why the Sun shines for billions of years rather than burning out in seconds.
• Chapter 23 (r-Process): In the rapid neutron capture process, beta decay rates determine the path through the chart of nuclides and the final abundance pattern.
• Chapter 31 (Fundamental Symmetries): Precision beta decay experiments — angular correlations, ft-values, neutron lifetime — test the Standard Model at the precision frontier.

🔗 The Big Picture: Beta decay is where nuclear physics meets particle physics. Fermi's theory of beta decay (1934) was the first successful quantum field theory of the weak interaction. The Wu experiment (1957) shattered the assumption of parity symmetry. The search for neutrinoless double beta decay may reveal whether the neutrino is fundamentally different from all other fermions. From a humble nuclear transition to the deepest questions about the nature of matter — that is the reach of beta decay physics.