Chapter 3 — The Nuclear Force: What Holds the Nucleus Together

"What is the force between two nucleons? This question, first posed in 1932, is still not fully answered — but the progress we have made is one of the great achievements of twentieth-century physics." — Hans Bethe, Nobel Lecture (1967)

In Chapter 1, we established that nuclei exist — that protons and neutrons are bound together in a compact object roughly $10^{-15}$ m across. In Chapter 2, we catalogued the systematic properties of nuclei: their sizes, masses, spins, and electromagnetic moments. But we left the deepest question unanswered: what is the force that holds the nucleus together?

The question is not trivial. Protons carry positive electric charge, and the Coulomb repulsion between two protons separated by 1 fm is enormous:

$$V_{\text{Coul}} = \frac{e^2}{4\pi\epsilon_0 r} \approx \frac{1.44 \text{ MeV} \cdot \text{fm}}{1 \text{ fm}} = 1.44 \text{ MeV}$$

For a heavy nucleus like $^{208}$Pb with 82 protons, the total Coulomb energy is hundreds of MeV. Something must overpower this repulsion, and that something is the nuclear force — the residual strong interaction between nucleons. This chapter traces our understanding of this force from its phenomenological properties through the elegant meson exchange picture to the modern effective field theory framework that connects the nuclear force to the underlying physics of quarks and gluons.

The story begins with experiments.

3.1 Properties of the Nuclear Force

Before writing down any mathematical model, we should establish what experimental evidence tells us about the nucleon-nucleon interaction. The properties that any successful nuclear force model must reproduce are as follows.

Short Range

The nuclear force acts over distances comparable to the size of the nucleon itself, roughly 1--2 fm. The evidence for this is overwhelming:

1. Nuclear density saturation. As established in Chapter 1, the nuclear radius follows $R = r_0 A^{1/3}$ with $r_0 \approx 1.2$ fm. This means nuclear volume is proportional to the number of nucleons: $V \propto A$. If every nucleon interacted with every other nucleon (as in a $1/r$ force like gravity or electromagnetism), the binding energy per nucleon would grow linearly with $A$. Instead, $B/A$ saturates at about 8.5 MeV for $A \gtrsim 12$. Each nucleon interacts only with its nearest neighbors — the hallmark of a short-range force.

2. The binding energy curve. The near-constancy of $B/A$ across most of the periodic table means the nuclear force saturates. This is analogous to nearest-neighbor interactions in a liquid — and indeed, the liquid drop model of the nucleus (Chapter 4) succeeds precisely because the nuclear force is short-ranged. We can make this quantitative. If each nucleon interacts with $n$ nearest neighbors with an average pair energy $\epsilon$, then $B \approx n\epsilon A/2$ and $B/A \approx n\epsilon/2 \approx 8.5$ MeV. Taking $\epsilon \approx 2$--$3$ MeV per pair (a plausible value given the balance between attraction and kinetic energy), we need $n \sim 6$--$8$ neighbors. In nuclear matter at normal density ($\rho_0 = 0.16$ fm$^{-3}$), the number of neighbors within a sphere of radius $R_f$ is $n \sim (4/3)\pi R_f^3 \rho_0$. Setting $n \sim 6$--$8$ gives $R_f \sim 2.0$--$2.2$ fm, consistent with the independently measured range of the nuclear force.

3. Scattering experiments. Nucleon-nucleon scattering at low energies is well described by an interaction that vanishes beyond about 2--3 fm. The effective range parameter extracted from low-energy $S$-wave scattering (Section 3.2) is $r_0 \approx 1.75$--$2.8$ fm, providing a direct, model-independent measure of the force range.

4. Comparison to other forces. To appreciate how unusual the nuclear force is, consider the hierarchy of length scales. The Bohr radius of hydrogen is $a_0 = 0.529 \times 10^5$ fm — five orders of magnitude larger than the nuclear force range. The electromagnetic force between a proton and electron at nuclear distances ($\sim 1$ fm) is about $1.44$ MeV, comparable to nuclear energies. But the electromagnetic force extends to infinity, while the nuclear force essentially vanishes beyond 3 fm. This dramatic difference in range — not in strength — is what separates nuclear physics from atomic physics.

Strongly Attractive at Intermediate Range, Repulsive at Short Range

The nuclear force is not uniformly attractive. At intermediate distances (1--2 fm), it is strongly attractive — this is what binds nuclei. But at very short distances ($r \lesssim 0.5$ fm), the force becomes strongly repulsive. This repulsive core (or "hard core") prevents nucleons from collapsing onto one another and is essential for nuclear stability and the saturation of nuclear density.

The evidence comes from high-energy nucleon-nucleon scattering. At incident kinetic energies above about 300 MeV, the $S$-wave phase shift for proton-proton scattering passes through zero and becomes negative, indicating a repulsive interaction at the short distances probed by these energies. We can estimate the size of the repulsive core using classical kinematics: at $E_{\text{lab}} = 300$ MeV, the center-of-mass kinetic energy is $T_{\text{cm}} = E_{\text{lab}}/2 = 150$ MeV, and the classical distance of closest approach (where the kinetic energy equals the potential energy) is $r_{\text{min}} \lesssim 0.5$ fm.

The repulsive core has profound consequences:

• It provides the pressure that prevents nuclear matter from collapsing to a point — the same physics that supports neutron stars against gravitational collapse (Chapter 25).
• Together with the Pauli exclusion principle, it ensures that nuclear matter has a well-defined equilibrium density ($\rho_0 \approx 0.16$ fm$^{-3}$).
• It creates a minimum interparticle spacing that, combined with the attractive intermediate-range force, determines the nuclear compressibility $K \approx 230$ MeV — a quantity measurable through giant monopole resonances in nuclei.

The physical origin of the repulsive core lies in the quark substructure of the nucleon. At short distances, the quark wavefunctions of two nucleons overlap, and the Pauli exclusion principle for quarks (which are spin-1/2, color-triplet fermions) creates an effective repulsion. In the meson exchange picture (Section 3.4), the repulsive core is modeled by the exchange of heavy vector mesons, particularly the $\omega$ meson.

Charge Independence and Charge Symmetry

The nuclear force is approximately charge independent: the strong interaction between any two nucleons depends on their total isospin state, not on whether they are protons or neutrons individually. More precisely:

• Charge symmetry means the $pp$ force equals the $nn$ force (after removing electromagnetic effects). This is tested by comparing mirror nuclei — pairs like $^3$H and $^3$He that differ only by interchanging protons and neutrons. After correcting for the Coulomb energy, their binding energies are nearly identical: $B(^3\text{H}) = 8.482$ MeV versus $B(^3\text{He}) = 7.718$ MeV, a difference of 0.764 MeV attributable almost entirely to the Coulomb interaction.

• Charge independence is the stronger statement that $V_{pp} = V_{nn} = V_{np}$ in the same isospin state. This is tested by comparing scattering lengths (Section 3.2). The $nn$ and $pp$ scattering lengths in the $^1S_0$ channel (after Coulomb correction) are $a_{nn} \approx -18.9$ fm and $a_{pp} \approx -17.3$ fm, while the $np$ scattering length in the same isospin $T=1$ channel is $a_{np}(T=1) \approx -23.7$ fm. The $pp$ and $nn$ values agree within a few percent (charge symmetry), while the $np$ value differs more substantially — charge independence is approximate, not exact.

Spin Dependence

The nuclear force depends on the spin state of the nucleon pair. The most dramatic evidence is the deuteron itself. The neutron-proton system with total spin $S = 1$ (triplet state) is bound — this is the deuteron. The $S = 0$ (singlet state) is not bound. The $np$ scattering lengths in these two channels are:

$$a_s(^1S_0) = -23.7 \text{ fm} \quad (\text{singlet, unbound})$$ $$a_t(^3S_1) = +5.42 \text{ fm} \quad (\text{triplet, bound})$$

The sign convention is that a positive scattering length indicates a bound state exists in that channel, while a large negative scattering length indicates a nearly bound state — a virtual state. The singlet $np$ system is tantalizingly close to binding but does not quite make it.

The Tensor Force

The nuclear force includes a tensor component — a non-central force whose operator structure is analogous to the dipole-dipole interaction in electromagnetism:

$$V_T(r) = V_T(r) \, S_{12}$$

where the tensor operator is:

$$S_{12} = \frac{3(\boldsymbol{\sigma}_1 \cdot \hat{r})(\boldsymbol{\sigma}_2 \cdot \hat{r}) - \boldsymbol{\sigma}_1 \cdot \boldsymbol{\sigma}_2}{1}$$

Here $\boldsymbol{\sigma}_1$ and $\boldsymbol{\sigma}_2$ are the Pauli spin operators for the two nucleons and $\hat{r}$ is the unit vector connecting them. The tensor force mixes orbital angular momentum states: in the deuteron, it mixes the dominant $L = 0$ ($S$-wave) component with an $L = 2$ ($D$-wave) admixture. The evidence for this is the deuteron's nonzero electric quadrupole moment, $Q_d = 0.2860$ fm$^2$. A pure $S$-state has spherical symmetry and $Q = 0$; the measured quadrupole moment requires roughly 4--7% $D$-state probability in the deuteron ground state.

Spin-Orbit Force

The nuclear force also contains a spin-orbit component of the form:

$$V_{LS}(r) = V_{LS}(r) \, \mathbf{L} \cdot \mathbf{S}$$

where $\mathbf{L}$ is the relative orbital angular momentum and $\mathbf{S}$ is the total spin. This term is crucial: the nuclear spin-orbit force, when folded into the nuclear mean field, produces the strong spin-orbit splitting that explains the magic numbers (Chapter 6). Its strength and sign were deduced from $P$-wave nucleon-nucleon scattering phase shifts and from polarization measurements.

🔄 Check Your Understanding: Why does the near-constancy of $B/A$ across the periodic table tell us the nuclear force must be short-ranged? What would $B/A$ look like if the nuclear force were long-ranged like the Coulomb force?

3.2 Evidence from Nucleon-Nucleon Scattering

The most direct experimental window into the nuclear force is nucleon-nucleon scattering: firing one nucleon at another and measuring the angular distribution and energy dependence of the scattered particles. The theoretical framework for extracting force information from scattering data is partial wave analysis.

Partial Wave Analysis

In the center-of-mass frame, the scattering of two nucleons can be decomposed into partial waves labeled by the orbital angular momentum quantum number $L$ and the total angular momentum $J$. For identical fermions (pp or nn), the Pauli exclusion principle requires that the total wavefunction be antisymmetric under particle exchange. Since the spatial wavefunction has parity $(-1)^L$, the spin wavefunction has exchange symmetry $(-1)^{S+1}$, and the isospin wavefunction has exchange symmetry $(-1)^{T+1}$, we require:

$$(-1)^L \cdot (-1)^{S+1} \cdot (-1)^{T+1} = -1$$

which gives $(-1)^{L+S+T} = -1$, or equivalently $L + S + T = \text{odd}$.

For the $pp$ system ($T = 1$), this means $L + S$ must be even: $^1S_0$, $^3P_{0,1,2}$, $^1D_2$, $^3F_{2,3,4}$, .... For the $np$ system, both $T = 0$ and $T = 1$ are accessible, which is why $np$ scattering provides more information than $pp$ scattering alone.

The Phase Shift

For each partial wave, the effect of the nuclear force is encoded in a single real number at each energy: the phase shift $\delta_L(E)$. The scattering amplitude in terms of phase shifts is:

$$f(\theta) = \frac{1}{k} \sum_L (2L+1) e^{i\delta_L} \sin\delta_L \; P_L(\cos\theta)$$

and the differential cross section is $d\sigma/d\Omega = |f(\theta)|^2$.

At low energies (below about 10 MeV in the center of mass), only $S$-waves contribute significantly. As the energy increases, higher partial waves ($P$, $D$, $F$, ...) become important. The phase shift carries a definite physical meaning:

• $\delta_L > 0$: the interaction is attractive in that partial wave (the wavefunction is "pulled in" relative to the free-particle case).
• $\delta_L < 0$: the interaction is repulsive.
• $\delta_L = \pi/2$: a resonance exists at that energy.

Scattering Lengths and Effective Ranges

At very low energies, $S$-wave scattering is parametrized by just two numbers: the scattering length $a$ and the effective range $r_0$, defined through the effective range expansion:

$$k \cot\delta_0(k) = -\frac{1}{a} + \frac{1}{2} r_0 k^2 + \mathcal{O}(k^4)$$

where $k$ is the center-of-mass momentum. The scattering length and effective range for the key channels are:

Several features are striking:

1. The singlet $np$ scattering length is anomalously large in magnitude ($-23.7$ fm) compared to the range of the force ($\sim 2$ fm). This is the hallmark of a near-threshold state — the singlet $np$ system is very nearly bound. If the nuclear force were about 2% stronger, this channel would have a bound state, and di-neutrons and di-protons would exist. The consequences for nuclear physics, stellar nucleosynthesis, and the existence of the chemical elements would be profound.

2. The triplet scattering length is positive, confirming the existence of the bound state (deuteron) in that channel.

3. The effective ranges are all approximately 2.7--2.8 fm in the singlet channel and 1.75 fm in the triplet channel, consistent with a force range set by the pion Compton wavelength $\hbar/(m_\pi c) \approx 1.4$ fm.

Phase Shift Phenomenology

The experimental $NN$ phase shifts, extracted from decades of scattering experiments at laboratories worldwide, are the gold standard against which every nuclear force model is tested. The Nijmegen group and the Granada group have performed exhaustive partial wave analyses of the world $NN$ scattering database ($pp$ and $np$), comprising over 6,000 data points below 350 MeV.

The partial wave analysis database represents one of the most painstaking experimental programs in all of physics. The data come from proton-proton and neutron-proton scattering experiments performed at dozens of laboratories over half a century: Los Alamos, TRIUMF, PSI, Saclay, LAMPF, Uppsala, and many others. Each experiment measured cross sections, analyzing powers (polarization asymmetries), spin correlation coefficients, and other observables at specific energies and angles. The Nijmegen analysis (Stoks et al., 1993) used 1,787 $pp$ and 2,514 $np$ data points; the more recent Granada analysis (Navarro Perez et al., 2013) included over 6,700 data points.

Key features of the experimental phase shifts include:

• $^1S_0$ channel: Large positive phase shift at low energy ($\delta_0 \approx 62^\circ$ at $E_{\text{lab}} = 1$ MeV for $pp$), decreasing and eventually passing through zero at $E_{\text{lab}} \approx 260$ MeV, then becoming negative (repulsive core dominates at short distances). The initial large positive value reflects the strong attraction, while the sign change is the most direct evidence for the repulsive core. At $E_{\text{lab}} = 350$ MeV, $\delta_0 \approx -20^\circ$.

• $^3S_1$ channel: Even larger positive phase shift at low energy ($\delta_0 \approx 147^\circ$ at $E_{\text{lab}} = 1$ MeV), reflecting the stronger attraction in this channel (which binds the deuteron). The phase shift decreases with energy but remains positive up to the highest energies analyzed, never reaching the sign change seen in $^1S_0$.

• $^3P_2$ channel: Attractive, with a significant positive phase shift peaking around $\delta \approx 25^\circ$ at 200 MeV. This is the most attractive $P$-wave channel and plays an important role in nuclear structure, particularly in $^3P_2$ neutron-neutron pairing, which is relevant for superfluidity in neutron stars (Chapter 25).

• $^1P_1$ channel: Repulsive (negative phase shift, $\delta \approx -10^\circ$ at 100 MeV), as expected from the spin-isospin structure.

• $^3P_0$ channel: Attractive at low energy ($\delta > 0$ below $\sim 200$ MeV), then repulsive. This channel is important for $^3P_0$ pairing in dense nuclear matter.

• $^3S_1$-$^3D_1$ mixing parameter ($\epsilon_1$): Nonzero (reaching $\epsilon_1 \approx 2^\circ$ at low energy, increasing to $\sim 6^\circ$ at 300 MeV), confirming the presence of the tensor force. Without a tensor interaction, this mixing parameter would be exactly zero.

The totality of the phase shift data provides an essentially complete characterization of the on-shell nucleon-nucleon interaction below the pion production threshold. Any model of the nuclear force must reproduce these phase shifts; a model that achieves $\chi^2/\text{datum} \approx 1$ for the full database is considered "high precision."

🔄 Check Your Understanding: Why does the very large magnitude of the singlet $np$ scattering length ($-23.7$ fm) indicate a near-threshold state? What is the physical significance of the sign difference between the singlet and triplet scattering lengths?

3.3 The Deuteron: Nature's Simplest Nucleus

The deuteron — the bound state of one proton and one neutron — is to nuclear physics what the hydrogen atom is to atomic physics: the simplest system in which the fundamental interaction can be studied. But where the hydrogen atom has an infinite tower of excited states (the Balmer series, the Lyman series, ...), the deuteron has no excited states. It is just barely bound. This fact, more than any other, shapes our understanding of the nuclear force.

Experimental Properties

The deuteron's properties are measured with extraordinary precision:

These numbers are all informative:

• $B_d = 2.225$ MeV is tiny compared to the depth of the nuclear potential well ($\sim 40$ MeV). The deuteron is barely bound.
• $J^\pi = 1^+$ means $J = 1$ (total angular momentum), positive parity. Since parity is $(-1)^L$, this requires $L = 0$ or $L = 2$ (or higher even $L$). The ground state is predominantly $L = 0$ ($S$-wave) with a small $L = 2$ ($D$-wave) admixture.
• $T = 0$ means the deuteron is in the isospin-singlet state: $|np\rangle - |pn\rangle)/\sqrt{2}$. This is consistent with the fact that no bound $pp$ or $nn$ state exists (those would require $T = 1$).
• $\mu_d = 0.857$ $\mu_N$ is close to but not equal to the sum of the proton and neutron magnetic moments: $\mu_p + \mu_n = 2.793 - 1.913 = 0.880$ $\mu_N$. The small discrepancy is due to the $D$-state admixture and meson exchange currents.
• $Q_d = 0.286$ fm$^2$ is nonzero, proving the deuteron is not spherically symmetric and therefore cannot be a pure $S$-state. This is the smoking gun for the tensor force.

Analytical Solution: Square Well Approximation

We now solve for the deuteron bound state analytically, using a simplified spherical square well potential. This calculation is instructive because it shows, with minimal mathematical machinery, why the deuteron is barely bound and why its wavefunction extends far beyond the range of the nuclear force.

The model. Consider the neutron-proton system in the center-of-mass frame. For the $S$-wave ($L = 0$) component (which dominates), the radial Schrodinger equation for $u(r) = r R(r)$ is:

$$-\frac{\hbar^2}{2\mu} \frac{d^2 u}{dr^2} + V(r) \, u(r) = E \, u(r)$$

where $\mu = m_N/2 \approx 469.5$ MeV/$c^2$ is the reduced mass of the two-nucleon system (with $m_N \approx 939$ MeV/$c^2$), and we model the nuclear potential as a square well:

$$V(r) = \begin{cases} -V_0 & r \leq R \\ 0 & r > R \end{cases}$$

with $R \approx 2.1$ fm (the range of the nuclear force) and $V_0 > 0$ (the depth of the potential).

The bound state has $E = -B_d$ where $B_d = 2.225$ MeV.

Region I ($r \leq R$): Inside the well. The Schrodinger equation becomes:

$$\frac{d^2 u}{dr^2} + K^2 u = 0$$

where

$$K^2 = \frac{2\mu(V_0 - B_d)}{\hbar^2}$$

The solution regular at the origin (since $u(0) = 0$ for $L = 0$) is:

$$u_I(r) = A \sin(Kr) \quad (r \leq R)$$

Region II ($r > R$): Outside the well. With $V = 0$ and $E = -B_d < 0$:

$$\frac{d^2 u}{dr^2} - \kappa^2 u = 0$$

where

$$\kappa^2 = \frac{2\mu B_d}{\hbar^2}$$

The normalizable solution (decaying at infinity) is:

$$u_{II}(r) = C e^{-\kappa r} \quad (r > R)$$

Numerical values of the wavevectors. Using $\hbar c = 197.3$ MeV$\cdot$fm:

$$\kappa = \frac{\sqrt{2\mu B_d}}{\hbar} = \frac{\sqrt{2 \times 469.5 \times 2.225}}{197.3} \text{ fm}^{-1} = \frac{\sqrt{2089.3}}{197.3} \text{ fm}^{-1} = \frac{45.71}{197.3} \text{ fm}^{-1} \approx 0.2316 \text{ fm}^{-1}$$

The corresponding decay length is $1/\kappa \approx 4.32$ fm — more than twice the range $R$ of the potential. The deuteron wavefunction extends far beyond the potential well. This is a direct consequence of the small binding energy.

Matching conditions. At $r = R$, both $u$ and $du/dr$ must be continuous:

$$A \sin(KR) = C e^{-\kappa R}$$

$$AK \cos(KR) = -C\kappa e^{-\kappa R}$$

Dividing the second equation by the first:

$$K \cot(KR) = -\kappa$$

This is the transcendental eigenvalue equation for the bound state. Given $R$ and $B_d$ (and hence $\kappa$), this equation determines $V_0$ (and hence $K$).

Solving for the well depth. We can rewrite the eigenvalue equation as:

$$KR \cot(KR) = -\kappa R$$

With $\kappa R = 0.2316 \times 2.1 = 0.4864$, the right side is $-0.4864$. The equation $x \cot x = -0.4864$ (where $x = KR$) has its first solution near $x \approx 1.83$ (which can be verified numerically). Therefore:

$$K = \frac{1.83}{2.1} \approx 0.871 \text{ fm}^{-1}$$

and

$$V_0 = \frac{\hbar^2 K^2}{2\mu} + B_d = \frac{(197.3)^2 \times (0.871)^2}{2 \times 469.5} + 2.225 \approx 31.5 + 2.225 \approx 33.7 \text{ MeV}$$

Note that $KR \approx 1.83$ is only slightly above $\pi/2 \approx 1.57$. If $KR$ were exactly $\pi/2$, the cotangent would be zero and $\kappa = 0$, meaning $B_d = 0$ — the state would be at the threshold of binding. The fact that $KR$ is only marginally above $\pi/2$ quantifies what we mean when we say the deuteron is barely bound. A modest reduction in the potential depth $V_0$ would push the state above threshold and unbind the deuteron.

The condition for a bound state in a square well is $KR > \pi/2$, or equivalently:

$$V_0 > \frac{\pi^2 \hbar^2}{8\mu R^2} \approx \frac{\pi^2 \times (197.3)^2}{8 \times 469.5 \times (2.1)^2} \approx 23.2 \text{ MeV}$$

The actual well depth ($\sim 34$ MeV) exceeds this critical value by only about 45%. Compare this to the hydrogen atom, where the lowest bound state is far below the ionization threshold. The nuclear force is just barely strong enough to bind one neutron to one proton.

Why no excited states? A second bound state would require $KR > 3\pi/2$, meaning:

$$V_0 > \frac{9\pi^2 \hbar^2}{8\mu R^2} \approx 209 \text{ MeV}$$

This is far above the actual well depth, confirming that the deuteron has no excited states.

The Wavefunction and RMS Radius

The full (normalized) $S$-wave wavefunction is:

$$u(r) = \begin{cases} A \sin(Kr) & r \leq R \\ A \sin(KR) \, e^{\kappa(R-r)} & r > R \end{cases}$$

where we have used the matching condition to express $C$ in terms of $A$. The normalization condition is:

$$\int_0^\infty |u(r)|^2 \, dr = 1$$

The probability of finding the nucleon outside the well is:

$$P_{\text{outside}} = \int_R^\infty |u(r)|^2 dr = A^2 \sin^2(KR) \int_R^\infty e^{2\kappa(R-r)} dr = \frac{A^2 \sin^2(KR)}{2\kappa}$$

Numerical evaluation gives $P_{\text{outside}} \approx 0.70$ — roughly 70% of the time, the nucleons are separated by more than the range of the force. The deuteron is a loosely bound, spatially extended object. This is why its RMS charge radius ($1.97$ fm) substantially exceeds the force range.

The RMS matter radius is:

$$\langle r^2 \rangle = \int_0^\infty r^2 |u(r)|^2 dr$$

In the limit where the binding energy is small ($\kappa R \ll 1$), the wavefunction outside the well dominates, and:

$$\sqrt{\langle r^2 \rangle} \approx \frac{1}{2\kappa} \approx \frac{4.32}{2} \approx 2.16 \text{ fm}$$

The factor of 2 converts from the relative coordinate to the single-particle radius; the result is within about 10% of the measured value, which is satisfactory for such a simplified potential. The full square-well calculation, including the interior contribution, gives closer agreement.

Normalization and the Asymptotic Normalization Constant

The normalization integral can be evaluated analytically for the square well:

$$\int_0^\infty |u(r)|^2 dr = A^2 \left[ \frac{R}{2} - \frac{\sin(2KR)}{4K} + \frac{\sin^2(KR)}{2\kappa} \right] = 1$$

The quantity $A_S = A\sin(KR)e^{\kappa R}$ is the asymptotic normalization constant of the $S$-wave component. It determines the amplitude of the wavefunction at large distances and is directly measurable through low-energy neutron-proton capture cross sections and deuteron breakup reactions. The experimental value is $A_S = 0.8845 \pm 0.0008$ fm$^{-1/2}$, and it provides an important constraint on nuclear force models independent of the phase shifts.

For realistic potentials that include the $D$-wave component, the asymptotic wavefunction has two parts:

$$u(r) \to A_S \frac{e^{-\kappa r}}{r} + A_D \frac{e^{-\kappa r}}{r}\left(1 + \frac{3}{\kappa r} + \frac{3}{(\kappa r)^2}\right) P_2(\cos\theta)$$

The ratio $\eta = A_D/A_S = 0.0256 \pm 0.0004$ is a precisely measured quantity that constrains the strength of the tensor force.

Limitations of the Square Well Model

The square well calculation captures the essential physics — barely bound, no excited states, extended wavefunction — but it has quantitative limitations that should be acknowledged:

1. The sharp boundary at $r = R$ is unphysical. Realistic nuclear potentials have a smooth surface (like the Woods-Saxon form), which changes the interior wavefunction shape.

2. No $D$-state. The square well with $L = 0$ gives a spherically symmetric wavefunction and $Q_d = 0$. Reproducing the quadrupole moment requires including the tensor force, which is beyond the scope of a single-partial-wave calculation.

3. The well parameters ($V_0$, $R$) are correlated. The binding energy constrains only the combination $V_0 R^2$ (approximately). Deeper, narrower wells and shallower, wider wells can give the same binding energy but different wavefunctions. The scattering data (particularly the effective range) break this degeneracy.

4. No repulsive core. The square well is purely attractive. Adding a hard core at small $r$ modifies the interior wavefunction and is needed for a realistic description of $NN$ scattering at higher energies.

Despite these limitations, the square well deuteron remains one of the most instructive calculations in all of nuclear physics. It demonstrates, with minimal mathematics, the essential features of nuclear binding.

📜 Historical Context: Why Does the Deuteron Matter?

The deuteron was first studied systematically in the 1930s. Its binding energy was measured by Chadwick and Goldhaber in 1934 via photodisintegration: $\gamma + d \to p + n$. The threshold photon energy (2.225 MeV) gave a direct measurement of $B_d$. The deuteron's spin ($J = 1$) was established by molecular spectroscopy of HD. The quadrupole moment, measured by Kellogg, Rabi, Ramsey, and Zacharias in 1939 using molecular beam magnetic resonance, was the first evidence for the tensor force. These measurements — all before 1940 — established the basic program of nuclear force physics that continues to this day.

The deuteron also plays a central role in Big Bang nucleosynthesis (Chapter 24). The reaction $p + n \to d + \gamma$ is the first step in building up the light elements in the early universe. The deuteron's small binding energy means it is easily photodissociated by the intense radiation field of the early universe — the "deuterium bottleneck" delays nucleosynthesis until the universe has cooled sufficiently ($T \lesssim 0.07$ MeV, corresponding to about 3 minutes after the Big Bang). If the deuteron were more tightly bound, nucleosynthesis would begin earlier, more neutrons would be captured into $^4$He, and the primordial helium abundance would be higher. If the deuteron did not exist (i.e., the nuclear force were too weak to bind it), no nucleosynthesis beyond hydrogen would occur at all. The deuteron's marginal binding thus has cosmic consequences.

🔄 Check Your Understanding: In the square well model, why does the deuteron have no excited states? If the well depth were doubled, would a second bound state appear?

3.4 Meson Exchange and the Yukawa Potential

Yukawa's Prediction

In 1935, Hideki Yukawa proposed a revolutionary idea: the nuclear force is mediated by the exchange of a massive particle, just as the electromagnetic force is mediated by the exchange of massless photons. Yukawa's reasoning was as follows.

The electromagnetic interaction between two charges is mediated by the photon, a massless boson. The potential between two static charges is the Coulomb potential, $V(r) \propto 1/r$, which has infinite range. What if the mediating particle had mass $m$? Yukawa showed that the potential would then have the form:

$$V(r) = -g^2 \frac{e^{-\mu r}}{r}$$

where $g$ is the coupling constant and $\mu = mc/\hbar$ is the inverse Compton wavelength of the exchanged particle.

Derivation. The Yukawa potential can be derived from the Klein-Gordon equation for a massive scalar field $\phi$ coupled to a static point source $\rho(\mathbf{r}) = \delta^3(\mathbf{r})$:

$$(\nabla^2 - \mu^2) \phi(\mathbf{r}) = -g \, \delta^3(\mathbf{r})$$

This is the screened Poisson equation. To solve it, take the Fourier transform:

$$\tilde{\phi}(\mathbf{q}) = \frac{g}{q^2 + \mu^2}$$

The inverse Fourier transform gives:

$$\phi(r) = \frac{g}{(2\pi)^3} \int \frac{e^{i\mathbf{q}\cdot\mathbf{r}}}{q^2 + \mu^2} \, d^3q$$

Evaluating in spherical coordinates (choosing $\mathbf{r}$ along the $z$-axis):

$$\phi(r) = \frac{g}{(2\pi)^2} \int_0^\infty \frac{q^2 \, dq}{q^2 + \mu^2} \int_{-1}^{+1} e^{iqr\cos\theta} \, d(\cos\theta)$$

The angular integral gives $2\sin(qr)/(qr)$, so:

$$\phi(r) = \frac{g}{2\pi^2 r} \int_0^\infty \frac{q \sin(qr)}{q^2 + \mu^2} \, dq$$

This integral is evaluated by contour methods (or looked up in standard tables). Extending to the complex $q$-plane and closing in the upper half-plane, the pole at $q = i\mu$ gives:

$$\int_0^\infty \frac{q \sin(qr)}{q^2 + \mu^2} \, dq = \frac{\pi}{2} e^{-\mu r}$$

Therefore:

$$\phi(r) = \frac{g}{4\pi} \frac{e^{-\mu r}}{r}$$

The potential energy between two nucleons, each coupled to this field with strength $g$, is:

$$\boxed{V_{\text{Yukawa}}(r) = -\frac{g^2}{4\pi} \frac{e^{-\mu r}}{r}}$$

This is the Yukawa potential. Its key features are:

1. Range: The exponential factor $e^{-\mu r}$ causes the potential to fall off much faster than $1/r$. The effective range is $R_{\text{eff}} \sim 1/\mu = \hbar/(mc)$. For the nuclear force, $R_{\text{eff}} \sim 1\text{--}2$ fm.

2. Limiting case: When $m \to 0$ ($\mu \to 0$), the Yukawa potential reduces to the Coulomb potential $V \propto 1/r$. The photon is the massless limit of a Yukawa mediator.

3. Prediction of the meson mass: Yukawa estimated the range of the nuclear force as $R \sim 2$ fm, giving:

$$m = \frac{\hbar}{Rc} \approx \frac{197.3 \text{ MeV}\cdot\text{fm}}{2 \text{ fm} \times c} \approx 100 \text{ MeV}/c^2$$

This predicted a particle intermediate in mass between the electron (0.511 MeV/$c^2$) and the nucleon (939 MeV/$c^2$). Yukawa called it the "meson" (from the Greek mesos, meaning "middle").

The Discovery of the Pion

The history of the meson's discovery is a tale of false starts and eventual triumph.

📜 Historical Context: From Yukawa to the Pion

In 1936, Anderson and Neddermeyer discovered a particle with mass $\sim 106$ MeV/$c^2$ in cosmic rays. Initially identified as Yukawa's meson, this particle (now called the muon, $\mu$) turned out to interact only weakly with nuclei — it was a lepton, not a hadron. The puzzle was expressed memorably by I. I. Rabi: "Who ordered that?"

The true Yukawa particle — the pion ($\pi$) — was discovered in 1947 by Cecil Powell, Cesare Lattes, and Giuseppe Occhialini using photographic emulsions exposed to cosmic rays at high altitude on the Pic du Midi in the Pyrenees. They observed the decay chain $\pi \to \mu \to e$ and measured the pion mass as approximately 140 MeV/$c^2$. Powell received the Nobel Prize in 1950.

The charged pions ($\pi^\pm$, mass 139.6 MeV/$c^2$) and the neutral pion ($\pi^0$, mass 135.0 MeV/$c^2$) form an isospin triplet. The pion Compton wavelength is $\hbar/(m_\pi c) = 197.3/139.6 = 1.41$ fm, in excellent agreement with the observed range of the nuclear force.

One-Pion Exchange Potential (OPEP)

When the pion's spin, parity, and isospin ($J^\pi = 0^-$, $T = 1$) are properly accounted for, the one-pion exchange potential between two nucleons takes the form:

$$V_{\text{OPEP}}(r) = \frac{f_{\pi NN}^2}{4\pi} \frac{m_\pi c^2}{3} (\boldsymbol{\tau}_1 \cdot \boldsymbol{\tau}_2) \left[ (\boldsymbol{\sigma}_1 \cdot \boldsymbol{\sigma}_2) \frac{e^{-x}}{x} + S_{12} \left(1 + \frac{3}{x} + \frac{3}{x^2}\right) \frac{e^{-x}}{x} \right]$$

where $x = m_\pi c \, r / \hbar = r/(1.41 \text{ fm})$, $f_{\pi NN}^2/(4\pi) \approx 0.08$ is the pion-nucleon coupling constant, $\boldsymbol{\tau}_{1,2}$ are isospin operators, $\boldsymbol{\sigma}_{1,2}$ are spin operators, and $S_{12}$ is the tensor operator defined in Section 3.1.

This potential has several important features:

1. The $\boldsymbol{\tau}_1 \cdot \boldsymbol{\tau}_2$ factor enforces the isospin dependence of pion exchange. In the deuteron ($T = 0$), $\langle \boldsymbol{\tau}_1 \cdot \boldsymbol{\tau}_2 \rangle = -3$, so the OPEP is attractive (for the dominant spin-triplet tensor configuration). In the $T = 1$ channels, $\langle \boldsymbol{\tau}_1 \cdot \boldsymbol{\tau}_2 \rangle = +1$, and the OPEP is repulsive in the spin-triplet configuration.

2. The $\boldsymbol{\sigma}_1 \cdot \boldsymbol{\sigma}_2$ term provides the central spin-spin interaction. This gives different potentials in the spin-singlet ($\langle \boldsymbol{\sigma}_1 \cdot \boldsymbol{\sigma}_2 \rangle = -3$) and spin-triplet ($\langle \boldsymbol{\sigma}_1 \cdot \boldsymbol{\sigma}_2 \rangle = +1$) channels.

3. The $S_{12}$ term is the tensor force from pion exchange. This is the dominant source of the tensor interaction in the nuclear force and is responsible for the deuteron's quadrupole moment. The pion's pseudoscalar nature ($J^\pi = 0^-$) is what generates this tensor structure: a scalar meson exchange would produce only a central force.

4. At large distances ($r \gtrsim 2$ fm), the OPEP is the dominant contribution to the nuclear force. This is one of the best-established features of nuclear physics: the long-range part of the nucleon-nucleon interaction is one-pion exchange. The experimental evidence is the agreement between the OPEP prediction and the measured $F$-wave and higher partial-wave phase shifts, which are sensitive only to the long-range part of the interaction.

To get a numerical sense of the OPEP strength: at $r = 2$ fm ($x = 1.42$), the central Yukawa factor is $e^{-x}/x = e^{-1.42}/1.42 \approx 0.17$, and the tensor factor $(1 + 3/x + 3/x^2)e^{-x}/x \approx 0.64$. The tensor component is nearly four times stronger than the central component at this distance — this is why the tensor force, not the central force, is the most important feature of pion exchange.

Beyond One-Pion Exchange: Heavier Mesons

At shorter distances ($r \sim 1$ fm), the exchange of heavier mesons contributes:

• The $\sigma$ meson (now understood as correlated two-pion exchange, $f_0(500)$, mass $\sim 400$--$550$ MeV/$c^2$, range $\sim 0.4$ fm): provides the intermediate-range attraction that is essential for nuclear binding. This is a scalar-isoscalar exchange, giving a purely central attractive potential. The $\sigma$ has a controversial history — for decades it was unclear whether it was a genuine resonance or an artifact — but modern analyses establish it as a very broad resonance (width $\Gamma \sim 400$--$700$ MeV) in the two-pion channel. In chiral EFT, the $\sigma$ is not treated as a particle but emerges naturally from the two-pion exchange diagrams at NLO and beyond.

• The $\rho$ meson (vector, $J^\pi = 1^-$, $T = 1$, mass 775 MeV/$c^2$, range $\sim 0.25$ fm): provides a shorter-range contribution with both central and tensor components. The $\rho$ tensor force has the opposite sign to the pion tensor force, partially canceling the OPEP tensor at short distances. This cancellation is important: without it, the tensor force would be too strong, overbinding the deuteron and producing too much $D$-state.

• The $\omega$ meson (vector, $J^\pi = 1^-$, $T = 0$, mass 783 MeV/$c^2$, range $\sim 0.25$ fm): provides the short-range repulsion (the "hard core") of the nuclear force. The $\omega$ exchange gives a strong, repulsive, central potential at short distances. Because the $\omega$ is an isoscalar vector meson, its exchange potential is positive definite and repulsive — this is a consequence of the vector nature of the coupling ($\gamma_\mu$ vertex).

The qualitative picture is therefore: - Long range ($r > 2$ fm): one-pion exchange (tensor + central) - Intermediate range ($1 < r < 2$ fm): two-pion exchange / "$\sigma$" (attraction) - Short range ($r < 1$ fm): $\omega$ and $\rho$ exchange (repulsion)

This hierarchy — long-range pion exchange, intermediate-range attraction, short-range repulsion — is the essential structure of the nuclear force. It produces the characteristic shape of the nuclear potential: a well with depth $\sim 40$--$100$ MeV (depending on the channel and the representation) at $r \sim 1$ fm, rising steeply to large positive values at $r \lesssim 0.5$ fm, and falling exponentially to zero beyond $\sim 3$ fm.

🔄 Check Your Understanding: Explain why the range of a force mediated by a particle of mass $m$ is approximately $\hbar/(mc)$. If a hypothetical meson had mass 500 MeV/$c^2$, what range force would it produce?

3.5 Modern Nuclear Potentials

The meson exchange picture, while physically illuminating, faces fundamental difficulties. At distances below about 1 fm, the nucleons overlap, and a description in terms of point-like nucleon sources exchanging point-like mesons breaks down. The internal structure of the nucleon — three valence quarks bound by gluons, immersed in a sea of virtual quark-antiquark pairs — becomes relevant. The underlying theory of the strong interaction is quantum chromodynamics (QCD) — the non-Abelian gauge theory of quarks (spin-1/2 fermions carrying color charge) and gluons (spin-1 gauge bosons, also carrying color). QCD is asymptotically free: at very high energies (short distances), the coupling constant is small and perturbation theory applies. This is the domain of deep inelastic scattering and jet physics at CERN. But at nuclear energy scales ($E \lesssim 1$ GeV, $r \gtrsim 0.5$ fm), QCD is non-perturbative — the coupling constant is of order unity, confinement operates, and no analytical solution exists.

Lattice QCD — the numerical solution of QCD on a discrete spacetime lattice — can in principle compute the nuclear force from first principles. The HALQCD (Aoki, Hatsuda, Ishii, et al.) and NPLQCD (Beane, Savage, et al.) collaborations have computed nucleon-nucleon potentials and phase shifts on the lattice, though currently at unphysically heavy pion masses ($m_\pi \sim 300$--$800$ MeV) and with limited precision. As computational power grows and lattice techniques improve, direct QCD calculations of nuclear forces will become increasingly important for validating the chiral EFT framework.

In the meantime, two complementary strategies have emerged: phenomenological potentials fitted directly to scattering data, and systematic effective field theories that connect the nuclear force to QCD.

Phenomenological Potentials

The Woods-Saxon Potential

For mean-field applications (Chapter 6), the most widely used single-particle potential has the Woods-Saxon form:

$$V(r) = -\frac{V_0}{1 + \exp\left(\frac{r - R}{a}\right)}$$

with typical parameters $V_0 \approx 50$ MeV, $R = r_0 A^{1/3}$ with $r_0 \approx 1.25$ fm, and diffuseness $a \approx 0.65$ fm. This is not a nucleon-nucleon potential but rather the average potential felt by a single nucleon in the nuclear medium. It has a flat interior (reflecting nuclear density saturation) and a smooth surface (reflecting the finite range of the nuclear force).

High-Precision NN Potentials

For the bare nucleon-nucleon interaction, several high-precision phenomenological potentials have been constructed by fitting directly to the world $NN$ scattering database. The most prominent are:

The Argonne $v_{18}$ potential (Wiringa, Stoks, and Schiavilla, 1995): - Contains 18 operator components: central, spin-spin, tensor, spin-orbit, and quadratic spin-orbit terms, each in different isospin channels. - Has 40 adjustable parameters fitted to 4,301 $pp$ and $np$ scattering data points below 350 MeV. - Achieves $\chi^2/\text{datum} \approx 1.09$ — a near-perfect fit to the data. - The operator structure is:

$$V_{ij} = \sum_{p=1}^{18} v_p(r_{ij}) \, O_{ij}^p$$

where the operators $O_{ij}^p$ include $1$, $\boldsymbol{\sigma}_i \cdot \boldsymbol{\sigma}_j$, $S_{ij}$, $\mathbf{L}\cdot\mathbf{S}$, $L^2$, $L^2 \boldsymbol{\sigma}_i \cdot \boldsymbol{\sigma}_j$, $(\mathbf{L}\cdot\mathbf{S})^2$, each multiplied by isospin operators $1$, $\boldsymbol{\tau}_i \cdot \boldsymbol{\tau}_j$, plus charge-dependent and charge-asymmetric terms.

The CD-Bonn potential (Machleidt, 2001): - Based on one-boson exchange with form factors, using $\pi$, $\rho$, $\omega$, and two fictitious scalar mesons. - 43 parameters, $\chi^2/\text{datum} \approx 1.01$. - Nonlocal (momentum-dependent), making it computationally more challenging for many-body calculations.

The Reid93 potential and the Nijmegen potentials (Nijm I, Nijm II) round out the "high-precision" set. All achieve $\chi^2/\text{datum} \approx 1$ for the $NN$ database below 350 MeV. They differ in functional form and physical interpretation but agree on the observables — demonstrating that the $NN$ scattering data alone do not uniquely determine the off-shell behavior of the nuclear force.

This off-shell ambiguity is a fundamental issue. Two potentials $V$ and $V'$ are phase-equivalent if they produce identical scattering matrices at all energies. They are related by a unitary transformation: $V' = U V U^\dagger$ for some unitary operator $U$. The on-shell scattering data (cross sections, polarization observables) are invariant under this transformation, but the off-shell $T$-matrix elements — which enter in many-body calculations — are not. This means that two potentials that fit the same $NN$ data equally well can give different predictions for the binding energy of $^3$H, $^4$He, or heavier nuclei. The resolution comes from three-body forces: when the off-shell two-body physics is changed by a unitary transformation, the required three-body force also changes, and the total Hamiltonian ($V_{NN} + V_{3N}$) gives the same physical predictions. This is why it is essential to use consistent two- and three-nucleon forces — a point that chiral EFT handles naturally, since both arise from the same Lagrangian.

Chiral Effective Field Theory

🎓 Advanced Sidebar: Chiral EFT — Connecting the Nuclear Force to QCD

The modern theoretical framework for the nuclear force is chiral effective field theory (chiral EFT), developed primarily by Weinberg (1990, 1991), and implemented at high precision by van Kolck, Epelbaum, Krebs, Meissner, Machleidt, Entem, and others.

The key idea is as follows. At the energy scales relevant to nuclear physics ($E \lesssim 300$ MeV), the relevant degrees of freedom are not quarks and gluons but nucleons and pions. The pion is special: it is the pseudo-Goldstone boson of the spontaneously broken chiral symmetry of QCD — the approximate symmetry of QCD under independent rotations of left-handed and right-handed light quarks. The small but nonzero pion mass ($m_\pi \approx 140$ MeV/$c^2$) is a consequence of the small but nonzero masses of the up and down quarks.

Chiral EFT is the most general theory of nucleons and pions consistent with the symmetries of QCD — particularly chiral symmetry, its pattern of explicit and spontaneous breaking, and the discrete symmetries $C$, $P$, and $T$. It provides a systematic power counting scheme: the nuclear force is expanded in powers of $(Q/\Lambda_\chi)$, where $Q$ is a typical momentum scale (or $m_\pi$) and $\Lambda_\chi \sim 1$ GeV is the chiral symmetry breaking scale. Each order in this expansion introduces new operators with coefficients (low-energy constants, LECs) that encode the short-distance physics and must be determined from experiment or, increasingly, from lattice QCD.

The hierarchy is:

• Leading order (LO): One-pion exchange + two contact terms (the two $S$-wave scattering lengths).
• Next-to-leading order (NLO): Two-pion exchange begins to contribute (triangle and box diagrams). Seven additional contact terms.
• N$^2$LO: Leading three-nucleon force appears (two topologies: two-pion exchange, one-pion-contact, and three-nucleon contact).
• N$^3$LO: Subleading two-pion exchange, leading four-nucleon force, additional operator structures.

The state of the art is N$^3$LO (fourth order) for the two-nucleon force, where the Entem-Machleidt (2003) and Epelbaum-Krebs-Meissner (2015) potentials achieve $\chi^2/\text{datum} \approx 1$ for $NN$ scattering below 300 MeV, rivaling the phenomenological potentials.

The great advantage of chiral EFT is threefold: 1. It is systematically improvable: higher orders give smaller corrections, and error estimates are possible. 2. It naturally generates many-body forces: the three-nucleon force is not an ad hoc addition but appears at a definite order (N$^2$LO) with definite structure. Only two new parameters enter. 3. It is connected to QCD through its symmetry structure: it is not just a fit but a controlled approximation to the underlying theory.

The practical impact of chiral EFT on nuclear structure calculations has been transformative. The combination of chiral two- and three-nucleon forces with modern many-body methods (no-core shell model, coupled cluster theory, quantum Monte Carlo, in-medium similarity renormalization group) has enabled ab initio calculations of nuclear properties — starting from the nuclear force and solving the many-body Schrodinger equation without phenomenological adjustments — up through the medium-mass region ($A \sim 100$) and even into the tin isotopes ($A \sim 130$).

The word "ab initio" here deserves emphasis. For the first time in the history of nuclear physics, it is possible to calculate the binding energy, radius, excitation spectrum, and electromagnetic properties of a nucleus like $^{48}$Ca or $^{78}$Ni starting from the nuclear force, without any adjustable parameters beyond those already determined from $NN$ and $3N$ data. The results agree with experiment at the level of a few percent for binding energies and a few percent for charge radii. This represents the convergence of two programs: the development of high-quality nuclear forces from chiral EFT, and the development of many-body methods capable of handling the computational challenge of the nuclear Schrodinger equation with realistic forces.

A crucial recent development is uncertainty quantification. Because chiral EFT is a systematic expansion, the truncation error at each order can be estimated using Bayesian methods (Furnstahl, Melendez, Phillips, and others, 2015 onward). This means that ab initio nuclear theory now provides not just predictions but predictions with error bars — a qualitative advance over the era of phenomenological potentials, where the theoretical uncertainty was unknown. These error bars propagate to predictions for neutron star properties, nucleosynthesis rates, and other applications where the nuclear force enters as input.

🔄 Check Your Understanding: What is the physical origin of the pion in the context of QCD? Why is the pion lighter than other hadrons, and how does this relate to the range of the nuclear force?

3.6 The Three-Nucleon Force

The Problem: Two-Body Forces Are Not Enough

If the nuclear force were purely a two-body interaction — depending only on the relative positions and momenta of pairs of nucleons — then the nuclear Hamiltonian would be:

$$H = \sum_i T_i + \sum_{i

and once $V_{ij}$ was determined from $NN$ scattering, all nuclear properties would be predicted. For decades, this was the working assumption. It fails.

The evidence is systematic and unambiguous:

1. The triton ($^3$H) binding energy. Using any of the high-precision $NN$ potentials (Argonne $v_{18}$, CD-Bonn, Nijmegen) in an exact three-body calculation (the Faddeev equations yield exact results for $A = 3$), the predicted binding energy of the triton is:

$$B(^3\text{H})_{\text{calc}} \approx 7.62 \text{ MeV} \quad (\text{Argonne } v_{18})$$

compared to the experimental value:

$$B(^3\text{H})_{\text{exp}} = 8.482 \text{ MeV}$$

The underbinding is about 1 MeV — roughly 10%. This is far outside the numerical uncertainty and far outside the uncertainty from the $NN$ fit.

2. The $^3$H/$^3$He binding energy difference. After correcting for the Coulomb interaction, the calculated charge-symmetry-breaking effects do not fully account for the Nolen-Schiffer anomaly in mirror nuclei, suggesting additional isospin-dependent three-body effects.

3. The $A = 10$ region. Calculations of the spectra of $^{10}$B and $^{10}$Be with two-body forces alone produce level orderings that disagree with experiment. The ground state spin of $^{10}$B, for example, is predicted incorrectly.

4. Nuclear matter saturation. Perhaps the most dramatic failure: when nuclear matter (an infinite, uniform system of nucleons at the density found in the interior of heavy nuclei) is calculated with two-body forces alone, the saturation point — the equilibrium density and energy per nucleon — is wrong. The calculated saturation occurs at too high a density and too large a binding energy per nucleon. The empirical saturation point ($\rho_0 \approx 0.16$ fm$^{-3}$, $E/A \approx -16$ MeV) lies on the "Coester band" — a line in the $(\rho, E/A)$ plane that all realistic $NN$ potentials trace out, but none of them pass through the empirical point.

The Three-Nucleon Force (3NF)

The resolution is that the nuclear Hamiltonian contains three-body forces — interactions that depend on the simultaneous positions and quantum numbers of three nucleons and cannot be decomposed into a sum of two-body terms:

$$H = \sum_i T_i + \sum_{i

The leading contribution to the three-nucleon force arises from two mechanisms:

1. The Fujita-Miyazawa (FM) two-pion exchange 3NF (1957). In this process, one nucleon emits a virtual pion, which is absorbed by a second nucleon. But before the second nucleon absorbs the pion, it is virtually excited to a $\Delta(1232)$ resonance — the lowest nucleon excitation, with spin-isospin $J^\pi = 3/2^+$, $T = 3/2$. The $\Delta$ then emits a second pion, which is absorbed by a third nucleon. The net result is an effective three-body interaction:

$$W_{ijk}^{\text{FM}} \propto \sum_{\text{cyclic}} \left\{ \left( \frac{\boldsymbol{\sigma}_i \cdot \mathbf{q}_i}{q_i^2 + m_\pi^2} \right) \left( \frac{\boldsymbol{\sigma}_j \cdot \mathbf{q}_j}{q_j^2 + m_\pi^2} \right) \right\} T_\Delta$$

where $T_\Delta$ contains the isospin algebra of the $\Delta$ intermediate state.

This process cannot be absorbed into a redefinition of the two-body force because it depends on the quantum state of the third nucleon. When nucleon $k$ is present, the virtual $\Delta$ excitation of nucleon $j$ is modified by the Pauli exclusion principle — nucleon $j$ cannot occupy states already filled by nucleon $k$. This is an irreducibly three-body effect.

2. The contact and one-pion-exchange 3NF. In chiral EFT at N$^2$LO, two additional three-nucleon force topologies appear beyond the Fujita-Miyazawa term:

• A one-pion exchange term where one nucleon emits a pion absorbed by a second nucleon, with the third nucleon interacting via a contact (short-range) interaction. This term has one new low-energy constant ($c_D$).
• A pure contact term where all three nucleons interact at a single point. This has one new constant ($c_E$).

Together with the two-pion exchange term (whose strength is determined by the known pion-nucleon coupling), these three topologies constitute the complete leading three-nucleon force. Only two new parameters ($c_D$ and $c_E$) need to be determined from three-body observables — typically the triton binding energy and the $nd$ doublet scattering length.

Impact of the Three-Nucleon Force

When the three-nucleon force (calibrated to reproduce $B(^3$H$)$) is included:

1. The $^4$He binding energy (predicted, not fitted) improves dramatically: from $\sim 24$ MeV (with $NN$ only) to $\sim 28.3$ MeV, in excellent agreement with the experimental value of $28.296$ MeV.

2. The $A = 10$--$12$ spectra are significantly improved. The ground state spin-parity of $^{10}$B is correctly reproduced.

3. Nuclear matter saturation moves toward the empirical point. The combination of chiral $NN$ + 3NF forces gives saturation near $\rho_0 \approx 0.16$ fm$^{-3}$ and $E/A \approx -15$ to $-16$ MeV.

4. The neutron matter equation of state — crucial for neutron star physics (Chapter 25) — is significantly stiffened by three-nucleon forces, producing neutron stars consistent with the observed $\sim 2 M_\odot$ maximum mass.

5. The oxygen drip line. A remarkable result: with two-body forces alone, the neutron-rich isotopes of oxygen are predicted to be bound all the way out to $^{28}$O. The three-nucleon force provides additional repulsion in neutron-rich systems, correctly placing the drip line at $^{24}$O — in agreement with experiment. This prediction, made by Otsuka, Suzuki, Holt, Schwenk, and Akaishi in 2010, was one of the landmark successes of chiral EFT with consistent two- and three-nucleon forces. The physical mechanism is that the 3NF, when averaged over the filled neutron orbits, shifts the single-particle energies of the unfilled orbits upward (less binding), making it energetically unfavorable to add more neutrons beyond $^{24}$O.

6. Light nuclei up to $A = 12$. Quantum Monte Carlo (QMC) calculations by Pieper, Wiringa, and collaborators at Argonne National Laboratory, using the AV18 + Illinois-7 Hamiltonian, reproduce the ground-state energies and excitation spectra of all nuclei from $A = 3$ to $A = 12$ with remarkable accuracy — typically within 1--2% of experiment. The spectra of $^{10}$B and $^{12}$C, including the famous Hoyle state ($0_2^+$ in $^{12}$C at 7.65 MeV, crucial for carbon production in stars), are correctly ordered.

The three-nucleon force is not a small correction. It is an essential component of the nuclear interaction that qualitatively changes the predictions of nuclear theory. Its relative importance grows with density: in the interior of neutron stars ($\rho \sim 2$--$5 \rho_0$), three-nucleon forces may contribute as much to the pressure as two-nucleon forces. Understanding the 3NF at these extreme densities is one of the central open questions in nuclear physics.

📜 Historical Context: From Tucson-Melbourne to Chiral 3NF

The first widely used phenomenological three-nucleon force was the Tucson-Melbourne (TM) potential (Coon and colleagues, 1979), based on two-pion exchange with an intermediate $\Delta$ excitation. The Urbana IX model (Pudliner et al., 1995) combined the Fujita-Miyazawa two-pion exchange term with a phenomenological repulsive term and was used extensively with the Argonne $v_{18}$ $NN$ potential. The combination "AV18 + UIX" became the workhorse of quantum Monte Carlo calculations of light nuclei for over a decade. The Illinois-7 (IL7) model refined this approach. Modern chiral EFT provides a principled, systematically improvable replacement for these models, but the physical insights of the earlier work remain valid.

🔄 Check Your Understanding: Why can't the three-nucleon force be absorbed into a redefinition of the two-body force? What role does the Pauli exclusion principle play in generating irreducible three-body effects?

3.7 Connecting Back: From the Nuclear Force to the Chart of Nuclides

We began this chapter by asking what holds the nucleus together. We can now give a layered answer:

1. The strong force binds nucleons via meson exchange at intermediate range and the residual effects of QCD at short range. The one-pion exchange potential dominates beyond 2 fm; heavier meson exchange and short-range QCD dynamics dominate below 1 fm. The force is not simple: it depends on the distance between nucleons, their relative spin orientation, their relative orbital angular momentum, and their isospin state. It requires at least six independent operator structures (central, spin-spin, tensor, spin-orbit, quadratic-$L$, and quadratic-spin-orbit, each in isoscalar and isovector channels) to describe accurately.

2. The nuclear force saturates. Because of the short range, the repulsive core, and the Pauli exclusion principle, each nucleon interacts primarily with its neighbors. This produces nearly constant nuclear density and nearly constant $B/A$ — the liquid-drop-like behavior that we will exploit in Chapter 4. The saturation mechanism involves a delicate balance: the attractive intermediate-range force pulls nucleons together, while the repulsive core and the Pauli kinetic energy cost prevent them from getting too close. The equilibrium density $\rho_0 \approx 0.16$ fm$^{-3}$ corresponds to an average internucleon spacing of about 1.8 fm — right at the edge of the attractive region.

3. The delicate balance between the nuclear force and the Coulomb force determines the structure of the chart of nuclides. For light nuclei ($Z \lesssim 20$), the most stable configuration has $N \approx Z$ — the nuclear force is charge-independent, and the asymmetry energy (Chapter 4) favors equal numbers of protons and neutrons. For heavy nuclei, the growing Coulomb repulsion among protons tilts the balance toward $N > Z$. This is why $^{208}$Pb has 82 protons but 126 neutrons: the extra 44 neutrons provide nuclear attraction without the Coulomb cost.

The valley of stability on the chart of nuclides — the band of long-lived nuclei running from hydrogen to uranium — is shaped by this competition. On the neutron-rich side, the weak interaction (beta decay) converts excess neutrons to protons. On the proton-rich side, proton emission and electron capture reduce the proton number. The boundaries of nuclear existence (the drip lines) are where the nuclear force can no longer compensate the cost of adding another nucleon. The neutron drip line is far from stability for heavy elements and remains largely unexplored experimentally — the Facility for Rare Isotope Beams (FRIB) at Michigan State University, which began operations in 2022, is specifically designed to reach these limits.

4. Three-nucleon forces are essential for quantitative predictions, particularly for neutron-rich nuclei and nuclear matter. They determine the limits of nuclear existence (drip lines) and the equation of state of dense matter in neutron stars. The 3NF is also crucial for the nuclear symmetry energy — the energy cost of converting protons to neutrons at fixed density — which governs the neutron skin thickness of heavy nuclei and the structure of neutron stars.

5. The nuclear force determines the fate of stars. Every nuclear reaction in stellar interiors — hydrogen burning ($pp$ chain and CNO cycle), helium burning (triple-alpha), carbon and oxygen burning, silicon burning, and the rapid neutron capture process ($r$-process) — depends on the properties of the nuclear force. The cross sections, $Q$-values, and resonance energies that enter astrophysical reaction rates are all consequences of the two- and three-nucleon forces discussed in this chapter. We will return to these connections repeatedly in Part V (Nuclear Astrophysics).

The nuclear force, despite being a "residual" interaction — the nuclear analogue of the van der Waals force between neutral atoms — is responsible for the existence of the chemical elements, the energy source of stars, and the structure of neutron stars. It is arguably the most consequential force in the universe.

🔄 Check Your Understanding: The nuclear force is charge-independent, yet heavy nuclei have many more neutrons than protons. Explain this apparent contradiction. Why does $^{208}$Pb need 126 neutrons but only 82 protons?

In Chapter 4, we will see how far the liquid-drop analogy — inspired by the saturation property of the nuclear force — takes us in understanding nuclear masses. The semi-empirical mass formula captures the gross features but misses the shell effects that arise from the quantum-mechanical single-particle structure. That structure, and its connection to the nuclear force via the mean field, is the subject of Chapters 5 and 6.

Chapter Summary

• The nuclear force is short-range ($\sim 1$--$2$ fm), strongly attractive at intermediate distances, repulsive at very short distances ($r < 0.5$ fm), spin-dependent, and approximately charge-independent. These properties are established by nucleon-nucleon scattering data and the systematics of nuclear binding.

• The force has a complex operator structure requiring at least six types of terms: central, spin-spin, tensor, spin-orbit, quadratic-$L$, and quadratic-spin-orbit, each in isoscalar and isovector channels. The tensor force, arising from pion exchange, is particularly important for nuclear structure.

• Nucleon-nucleon scattering data — parametrized by phase shifts, scattering lengths, and effective ranges — provide the most direct experimental constraint on the nuclear force. The anomalously large singlet $np$ scattering length ($a_s = -23.7$ fm, far exceeding the force range of $\sim 2$ fm) indicates a near-threshold virtual state. If the nuclear force were about 2% stronger, the di-neutron would be bound, with profound consequences for nucleosynthesis.

• The deuteron, the only two-nucleon bound state, is barely bound ($B = 2.225$ MeV, compared to a well depth of $\sim 35$ MeV) with no excited states. Its nonzero quadrupole moment ($Q_d = 0.286$ fm$^2$) proves the existence of the tensor force and the presence of a 4--7% $D$-state admixture. Roughly 70% of the deuteron wavefunction lies outside the range of the nuclear force — a direct consequence of the small binding energy.

• Yukawa's meson exchange theory (1935) predicted a mediating particle of mass $\sim 100$ MeV/$c^2$. The pion ($m_\pi \approx 140$ MeV/$c^2$), discovered in 1947, fulfills this prediction. One-pion exchange (OPEP) dominates the nuclear force at long range and is responsible for the tensor interaction. Heavier mesons ($\sigma$, $\rho$, $\omega$) contribute at shorter distances, providing intermediate-range attraction and short-range repulsion.

• Modern high-precision potentials (Argonne $v_{18}$, CD-Bonn, chiral N$^3$LO) fit the $NN$ scattering database with $\chi^2/\text{datum} \approx 1$. Different potentials that fit the same on-shell data can differ off-shell — the off-shell ambiguity — which is resolved only by considering many-body systems and three-nucleon forces consistently.

• Chiral effective field theory provides a systematic, order-by-order framework for the nuclear force connected to QCD through chiral symmetry. It naturally generates many-body forces at definite orders, enables uncertainty quantification through Bayesian methods, and has made ab initio nuclear structure calculations — from the nuclear force to nuclear properties without adjustable parameters — a reality for nuclei up to $A \sim 130$.

• Two-body forces alone cannot reproduce the binding energies of nuclei beyond $A = 2$. Three-nucleon forces, arising from the excitation of intermediate $\Delta(1232)$ resonances and short-range mechanisms (contact and one-pion-exchange topologies in chiral EFT), are essential. They resolve the triton underbinding, improve spectra of light nuclei, fix nuclear matter saturation, correctly locate the oxygen drip line at $^{24}$O, and stiffen the neutron matter equation of state to support observed $\sim 2 M_\odot$ neutron stars.

• The nuclear force, though a "residual" interaction between composite objects, determines the structure of the chart of nuclides, the energy source of stars, the origin of the chemical elements, and the properties of neutron stars. Understanding it remains one of the central challenges of modern physics.

Next: Chapter 4 — The Semi-Empirical Mass Formula: A Liquid Drop with Quantum Corrections