Effective Potential and the Centrifugal Barrier

One substitution finishes what separation started. The radial equation of the previous article is almost a one-dimensional Schrödinger problem — spoiled only by a first-derivative term and an unfamiliar $r^2\,dr$ measure. Write $u(r)=rR(r)$ and both blemishes vanish at once: $u$ obeys a textbook 1D equation on the half-line, with the ordinary norm $\int|u|^2dr$ , and every tool of the one-dimensional chapters — node counting, shooting, matching craft — imports wholesale. The price of the disguise is one new term, and it is the article's protagonist. The potential the 1D problem sees is not $V(r)$ but $V_{\text{eff}}(r)=V(r)+\hbar^2\ell(\ell+1)/2mr^2$ : the true attraction plus a repulsive centrifugal barrier that dies as $1/r^2$ and diverges at the origin for every $\ell\ge1$ . That barrier is quantum mechanics' version of the classical angular-momentum term $L^2/2mr^2$ — with $\ell(\ell+1)$ replacing $\ell^2$ , an exact quantum fingerprint, not a typo — and it redraws every landscape it enters. For hydrogen it hollows out the Coulomb well into an $\ell$ -dependent trough whose minimum sits at $2a_0$ and bottoms out at $-6.80$ eV for $\ell=1$ (computed, not quoted); for nuclei it walls off low-energy neutrons of $\ell\ge1$ , making low-energy capture an s-wave monopoly; in Maria Goeppert Mayer's shell model, the same landscape's level ordering spells out the magic numbers. The article owns the reduction (every algebra step symbolic-verified), the barrier landscape with its crossovers and well minima for power-law potentials, the honesty card for the dangerous $1/r^2$ borderline — the fall-to-center threshold, derived rather than folklore — and the s-wave-dominance doctrine that low-energy physics runs on.

Key takeaways

The substitution $u=rR$ converts the radial equation exactly into 1D form: $-\dfrac{\hbar^2}{2m}u''+V_{\text{eff}}\,u=Eu$ , with $V_{\text{eff}}=V(r)+\dfrac{\ell(\ell+1)\hbar^2}{2mr^2}$ — the first-derivative term cancels identically (symbolically verified).
The norm converts too: $\int_0^\infty|R|^2r^2dr=\int_0^\infty|u|^2dr$ — $u$ is the honest 1D wavefunction; the half-line + $u(0)=0$ behave like a hard wall at the origin (admissibility doctrine owned by the next serial).
The barrier is exact quantum: $\ell(\ell+1)=\langle\hat{L}^2\rangle/\hbar^2$ , not the classical $\ell^2$ — s-states feel no barrier at all.
Hydrogen's $\ell=1$ landscape: barrier beats Coulomb inside $r=a_0$ ; the effective well bottoms at $r_{\min}=2a_0$ , depth $-6.80$ eV $=-\mathrm{Ry}/2$ (computed; crossover at $\ell(\ell+1)a_0/2$ ).
Power-law wells $V=-\alpha/r^n$ : a minimum exists for $n<2$ ; $n=2$ is marginal — attractive $\beta/r^2$ with $\beta>\hbar^2(2\ell+1)^2/8m$ has no ground state (fall to center, threshold derived).
Low-energy doctrine: a 1 MeV neutron meets a 1.15 MeV barrier at a nucleus' surface for $\ell=1$ (3.45 MeV for $\ell=2$ , computed) — capture and scattering are s-wave; penetration scales as $(kR)^{2\ell+1}$ (stated; scattering chapter's law).
Key idea: 3D radial motion is 1D motion against a wall, in a potential that carries an $\ell$ -dependent centrifugal rent — the landscape $V_{\text{eff}}$ decides who binds, who enters, and in what order levels fill.

The core insight

The substitution $u=rR$ maps the radial problem exactly onto a one-dimensional Schrödinger equation on the half-line with $V_{\text{eff}}(r)=V(r)+\hbar^2\ell(\ell+1)/2mr^2$ : the repulsive centrifugal term — exact eigenvalue $\ell(\ell+1)$ , not classical $\ell^2$ — reshapes every central potential into an $\ell$ -family of 1D landscapes.

Where this lives today

Ultracold-atom physics lives below every centrifugal barrier at once: at microkelvin energies even the van-der-Waals-scale $\ell=1$ barrier (millikelvin-high) is insurmountable, so all collisions are s-wave and a single scattering length characterizes each species pair — the simplification on which Feshbach-resonance control, BEC production, and cold molecular chemistry are built. Nuclear astrophysics prices every stellar reaction with the same tolls: centrifugal-plus-Coulomb barrier penetration factors set which capture channels burn in stars. And in halo-nucleus research, one-neutron halos form preferentially in s- and p-orbitals precisely because higher $\ell$ 's barrier confines the tail — the landscape of this article, photographed at rare-isotope beamlines.

The physical picture

Draw the landscape an electron actually experiences in hydrogen, one $\ell$ at a time. The s-electron sees the bare Coulomb funnel — bottomless at the origin. The p-electron sees the funnel plus a $1/r^2$ wall: the sum is a trough, floored at finite depth, its minimum pushed out to $2a_0$ . The d-electron's wall is three times higher; its trough is shallower and farther still. One potential, an $\ell$ -indexed family of landscapes — and every qualitative fact of atomic structure ("d-states avoid the core", "only s-electrons have contact hyperfine structure") is visible in the picture before any equation is solved.

Figure

One Coulomb well, three effective landscapes: the centrifugal rent rises with ℓ

The second picture is the reduction itself. In the $u$ -variable, three-dimensional radial motion is one-dimensional motion on a half-line: the origin acts as an impenetrable wall (the condition $u(0)=0$ , stated here with its owner named below), the far side is open, and the particle moves in $V_{\text{eff}}$ . Every 1D instinct fires correctly: count nodes for level ordering (The Node Theorem's grammar), classify bound versus scattering by the far field (Bound versus Scattering States), shoot and match for eigenvalues.

Figure

The u = rR disguise: radial 3D motion becomes 1D motion against a wall

The reduction

Start from the radial equation owned by the previous article (cited, never re-derived) and substitute $R=u/r$ :

-\frac{\hbar^2}{2m}\,\frac{d^2u}{dr^2}\;+\;\underbrace{\Big[V(r)+\frac{\ell(\ell+1)\hbar^2}{2mr^2}\Big]}_{V_{\text{eff}}(r)}\,u \;=\; E\,u \qquad (1)

where: this is a bona fide one-dimensional Schrödinger equation on $0<r<\infty$ — the same operator the whole ONED chapter studied — and the reduction is exact: no approximation entered. The disappearance of the first-derivative term is an algebraic identity, verified symbolically in this article's verifier, and derived honestly below because the cancellation looks like luck until you see the geometry behind it.

Derivation AFrom the radial equation to the effective 1D problem

We perform the $u=rR$ reduction with every step shown, and explain both payoffs — the vanished derivative term and the converted norm — as one geometric fact. The strategy is direct substitution plus one integration-by-parts insight; the input radial equation is The Radial Equation's, consumed.

Step 1. Compute the derivative block on $R=u/r$ : $\;\dfrac{dR}{dr}=\dfrac{u'}{r}-\dfrac{u}{r^2}$ , so $r^2\dfrac{dR}{dr}=ru'-u$ and $\dfrac{d}{dr}\big(r^2R'\big)=u'+ru''-u'=r\,u''$ .

Step 2. Insert: the radial kinetic operator becomes $-\dfrac{\hbar^2}{2m}\,\dfrac{1}{r^2}\cdot r\,u''=-\dfrac{\hbar^2}{2m}\,\dfrac{u''}{r}$ . Every remaining term of the radial equation carries $R=u/r$ ; multiply through by $r$ and equation (1) appears with no residue — the cancellation of $u'$ is exact (the verifier confirms the identity symbolically for general $V$ , $\ell$ ). Why the cancellation is geometry, not luck: the radial part of the 3D Laplacian is self-adjoint with respect to the weight $r^2dr$ ; the map $R\mapsto rR$ is precisely the unitary that converts the weighted space $L^2(r^2dr)$ to the flat space $L^2(dr)$ . A first-derivative term is the signature of a non-flat weight — transform the weight away and the term must go with it. The same trick converts every radial Laplacian in any dimension (with $r^{(d-1)/2}$ in $d$ dimensions).

Step 3. The norm converts with the same stroke: $\int_0^\infty|R|^2r^2\,dr=\int_0^\infty|u|^2\,dr$ (verified on a test profile) — so $|u(r)|^2\,dr$ is the honest radial probability, and $u$ , not $R$ , is the function whose square you should plot. The radial-density craft ( $P(r)=r^2|R|^2=|u|^2$ , most-probable radii, moments) belongs to Radial Probability Distribution and Expectation Values, named here once.

Step 4. Boundary inheritance. The origin-entrance $R\sim r^\ell$ of the previous article becomes $u\sim r^{\ell+1}$ : every physical $u$ vanishes at the origin — including $\ell=0$ , where $u(0)=0$ is exactly what kills the delta-carrying $1/r$ solution. Operationally the half-line problem behaves as if a hard wall stood at $r=0$ : the proof that $u(0)=0$ is the correct admissibility condition (normalizability plus hermiticity of the Hamiltonian on the half-line domain) is owned by Boundary Conditions and Normalization of Radial Wavefunctions, the next serial — this article uses the condition with its owner named. Why the wall picture earns its keep: with it, the entire 1D toolkit imports with one caveat. A symmetric 1D well's odd solutions vanish at the center — so the s-wave spectrum of any central well is exactly the odd-parity half of the corresponding 1D well's spectrum. Node counting, shooting stability, the ground state's nodelessness (Sturm grammar of "The Node Theorem", recalled) — all apply verbatim to $u$ . *Verified: reduction identity and norm identity symbolic; entrance exponent $\ell+1$ inherited from the previous article's verified indicial analysis (article verifier).*

Read $V_{\text{eff}}$ 's new term with the care it deserves. It is the quantum rotational energy at lever arm $r$ — and its numerator is the exact eigenvalue $\langle\hat{L}^2\rangle=\ell(\ell+1)\hbar^2$ (the Casimir card of Angular Momentum Commutation Relations, cited), not the classical $L_z^2$ -style $\ell^2\hbar^2$ :

V_{\text{cf}}(r)\;=\;\frac{\ell(\ell+1)\hbar^2}{2mr^2}\;>\;\frac{(\ell\hbar)^2}{2mr^2} \qquad (2)

where: the excess — the "+1" — is the transverse quantum spread of the angular momentum vector, which can never point purely along one axis. A classical orbit with the same maximal projection feels a strictly smaller barrier; the quantum particle pays rent on components it cannot eliminate. For $\ell=0$ the rent is exactly zero: s-waves alone see the naked potential, the structural root of every "only s-states touch the core" fact in atomic and nuclear physics.

The barrier landscape

For attractive power-law potentials the competition between $-\alpha/r^n$ and the barrier is fully solvable, and its verdicts organize a remarkable amount of physics.

Go deeper · Derivation B · 2 minWhere the effective well lives: minimum and depth for power-law attraction

Derivation BWhere the effective well lives: minimum and depth for power-law attraction

We locate the extremum of $V_{\text{eff}}=-\alpha/r^n+\hbar^2\ell(\ell+1)/2mr^2$ for general $n$ , classify when a genuine well exists, and evaluate the hydrogen case exactly. The strategy is elementary calculus made honest by dimensional bookkeeping — and by the marginal case $n=2$ , which the calculus flags by refusing to answer.

Step 1. Differentiate and solve: $V_{\text{eff}}'=n\alpha/r^{n+1}-\hbar^2\ell(\ell+1)/mr^3=0$ gives $r_{\min}^{\,2-n}=\dfrac{\hbar^2\ell(\ell+1)}{mn\alpha}$ — a positive solution exists for any $n\neq2$ ; whether it is a minimum (a well) or a maximum (a hump) depends on which term wins at short distance.

Step 2. Classify by $n$ . For $n<2$ the barrier's $1/r^2$ beats the attraction's $1/r^n$ as $r\to0$ : $V_{\text{eff}}\to+\infty$ at the origin, $\to0^-$ at infinity, and the extremum is a truebottomed well of finite depth. For $n>2$ the attraction wins at short range: $V_{\text{eff}}\to-\infty$ at the origin and the extremum is a barrier summit guarding an abyss — the fall-to-center configuration.

Step 3. Evaluate the physical case $n=1$ (Coulomb, $\alpha=e^2/4\pi\varepsilon_0$ ): $r_{\min}=\dfrac{\ell(\ell+1)\hbar^2}{m\alpha}=\ell(\ell+1)\,a_0$ , and the depth $V_{\text{eff}}(r_{\min})=-\dfrac{m\alpha^2}{2\hbar^2\ell(\ell+1)}=-\dfrac{\mathrm{Ry}}{\ell(\ell+1)}\cdot\ldots$ — computed exactly: for $\ell=1$ , $r_{\min}=2a_0$ and depth $=-\tfrac{\mathrm{Ry}}{2}=-6.80$ eV; for $\ell=2$ , $r_{\min}=6a_0$ , depth $-\tfrac{\mathrm{Ry}}{18}=-0.76$ eV. The crossover where barrier equals $|V|$ sits at $r=\ell(\ell+1)a_0/2$ ( $a_0$ for $\ell{=}1$ , $3a_0$ for $\ell{=}2$ — computed): inside it, the landscape is net-repulsive. Why the ℓ=1 well bottom at $-\mathrm{Ry}/2$ is a preview: hydrogen's $2p$ level sits at $-\mathrm{Ry}/4=-3.40$ eV — above the well floor, as a ground-state-of-its-landscape must be, with the difference paid to radial zero-point motion. The exact spectrum (and why all $\ell$ share the $-\mathrm{Ry}/n^2$ ladder — the accidental degeneracy) belongs to the hydrogen serials ahead; the landscape explains here why no $\ell=1$ state can bind below $-6.80$ eV, and why " $\ell=1$ starts at $n=2$ " is geometrically reasonable.

Step 4. The marginal case $n=2$ deserves its own verdict. For $V=-\beta/r^2$ the two terms merge into a single $1/r^2$ coefficient $\hbar^2\ell(\ell+1)/2m-\beta$ ; no length scale survives, so no minimum exists at any finite $r$ . The indicial analysis (previous article's, applied to the merged coefficient) gives $u\sim r^s$ with $s=\tfrac12\pm\sqrt{\tfrac14+\ell(\ell+1)-2m\beta/\hbar^2}$ : real roots — normal behavior — until the discriminant crosses zero at

\beta_c=\frac{\hbar^2(2\ell+1)^2}{8m}

(derived symbolically, using $\tfrac14+\ell(\ell+1)=\tfrac{(2\ell+1)^2}{4}$ ). Beyond $\beta_c$ the roots turn complex: $u\sim r^{1/2}\cos(\nu\ln r+\phi)$ oscillates infinitely often approaching the origin — infinitely many nodes, hence (Sturm grammar) states below any energy: no ground state. This is the fall to center: the spectrum is unbounded below and the problem as posed is incomplete (its rigorous repair — self-adjoint extensions, a short-distance cutoff — is the rigged-Hilbert-space/domain territory of the formalism chapter, one honesty line, no machinery). For s-waves ( $\ell=0$ ) the threshold is $\beta_c=\hbar^2/8m$ : even an arbitrarily weak $1/r^2$ attraction is dangerous only past a finite, computable strength. Verified: $r_{\min}$ general- $n$ symbolic; $n=1$ minimum $\ell(\ell+1)a_0$ and depth $-m\alpha^2/2\hbar^2\ell(\ell+1)$ exact; hydrogen $\ell=1$ : $2a_0$ , $-6.8028$ eV computed (a spec-sketch value of $-3.40$ eV was corrected by this computation — draft catch); crossovers $a_0$ , $3a_0$ ; threshold $\beta_c=\hbar^2(2\ell+1)^2/8m$ symbolic (article verifier).

The barrier's most consequential verdict is about entry. A particle approaching from outside with energy $E$ must climb $V_{\text{cf}}$ to reach the interaction region. At a nuclear surface ( $R\approx6$ fm, $A\approx125$ ), the toll booth charges (computed): $1.15$ MeV for $\ell=1$ , $3.45$ MeV for $\ell=2$ . A thermal or even 1-MeV neutron simply cannot pay for $\ell\ge1$ : low-energy reactions are s-wave monopolies. Quantitatively, tunneling through the barrier suppresses the $\ell$ -wave contribution by the penetration factor $\sim(kR)^{2\ell+1}$ — stated here as the doctrine's number, owned with its derivation by the scattering chapter (threshold laws). This staircase of suppressions is why "the s-wave scattering length" alone summarizes low-energy physics from neutron capture to ultracold-atom collisions.

Limiting cases

case	result	meaning
$\ell=0$	$V_{\text{eff}}=V$ : no barrier	s-waves see the naked potential; only they reach the core freely
$n<2$ power-law attraction	finite-depth well at $r_{\min}=[\hbar^2\ell(\ell+1)/mn\alpha]^{1/(2-n)}$	barrier always wins the origin; binding is a finite negotiation
$n=2$ , $\beta<\beta_c$	scale-free, normal indicial behavior	marginal but safe — no well, no catastrophe
$n=2$ , $\beta>\beta_c=\hbar^2(2\ell+1)^2/8m$	fall to center: log-periodic $u$ , no ground state	quantum mechanics' escape from Coulomb collapse fails for $1/r^2$
$n>2$	barrier summit guarding an abyss	short-range-dominated attraction: needs a cutoff to make sense
$r\to\infty$	$V_{\text{cf}}\to0$ faster than Coulomb	far field forgets $\ell$ ; tails are $\ell$ -blind (previous article's exit)

Common trapThe centrifugal term uses ℓ(ℓ+1), and the barrier is not a force anyone exerts

Two standing confusions. First, the numerator: it is $\ell(\ell+1)\hbar^2$ — the exact $\hat{L}^2$ eigenvalue — not $\ell^2\hbar^2$ . The difference is not pedantry: for $\ell=1$ it is a factor of 2 in the barrier, and every "semiclassical" substitution $\ell(\ell+1)\to(\ell+\tfrac12)^2$ you will meet later (WKB territory) exists precisely to manage this gap. Second, the ontology: nothing pushes the particle outward. The centrifugal term is kinetic energy in disguise — the rotational share of $\hat{p}^2/2m$ , relocated into the potential column by the separation bookkeeping. It appears repulsive in the radial problem because angular momentum conservation forces rotational cost to rise as $r$ falls. Treating it as a real force leads to genuine errors — e.g., expecting it to do work on the particle or to appear in 3D force balances.

Worked examples

Example 1 — the hydrogen ℓ=1 landscape, end to end. Map the effective potential a $2p$ electron lives in.

Step 1. Crossover: barrier equals $|V_{\text{Coul}}|$ at $r=\ell(\ell+1)a_0/2=a_0$ (computed). Inside one Bohr radius, a p-electron's landscape is net repulsive — the core is walled off.

Step 2. Well minimum: $r_{\min}=2a_0$ , depth $V_{\text{eff}}(2a_0)=-6.80$ eV (computed exactly as $-\mathrm{Ry}/2$ ). The trough is both farther out and eleven times shallower than the s-electron's landscape at the same radius comparison point — geometry, before any eigenvalue is computed.

Step 3. Sanity-check against the known spectrum: the lowest $\ell=1$ level ( $2p$ , $-3.40$ eV) sits above the floor by $3.40$ eV of radial zero-point energy, and below zero — inside the trough, as it must be. Every deeper level is $\ell=0$ 's privilege. Answer: repulsive inside $a_0$ , minimum $-6.80$ eV at $2a_0$ ; the landscape predicts "p-states start at $n=2$ " before hydrogen is solved.

Example 2 — why low-energy neutrons are s-wave customers only. A 1 MeV neutron approaches a mid-mass nucleus ( $R\approx6$ fm). Which partial waves participate?

Step 1. Barrier tolls at the surface (computed): $V_{\text{cf}}(R)=\dfrac{\ell(\ell+1)\hbar^2}{2m_nR^2}=1.15$ MeV for $\ell=1$ , $3.45$ MeV for $\ell=2$ .

Step 2. Compare: at $E=1$ MeV the $\ell=1$ channel is classically forbidden at the surface (toll exceeds budget by 15%), $\ell=2$ hopelessly so. Entry proceeds only by tunneling, suppressed as $(kR)^{2\ell+1}$ — with $kR\approx1.3$ at this energy the suppression is mild, but drop to 1 keV and $kR\approx0.04$ : the $\ell=1$ channel is down by $\sim(0.04)^3\approx6\times10^{-5}$ .

Step 3. Doctrine: thermal and epithermal neutron physics — capture, moderation, the entire reactor bookkeeping — is s-wave physics, and cross sections inherit the famous $1/v$ behavior of barrier-free s-wave capture (stated; the threshold-law derivation is the scattering chapter's). The same staircase, in Mayer's shell model, sets which high- $\ell$ orbitals hug the nuclear surface and how the magic-number level ordering emerges from the $V_{\text{eff}}$ landscape — the portrait band's story. Answer: $\ell=0$ only, with $\ell\ge1$ suppressed by computed MeV-scale tolls and $(kR)^{2\ell+1}$ tunneling factors.

Origins · 1 minMaria Goeppert Mayer and how this was discovered

Maria Goeppert Mayer

1906–1972 · Göttingen, Columbia, Chicago, San Diego

Mayer learned quantum mechanics in Born's Göttingen and then spent nearly twenty years as a "volunteer professor" — brilliant, published, and unpaid at Johns Hopkins, Columbia, and Chicago, because universities would not hire a professor's wife. The work that made her immortal is effective-potential physics: the nuclear shell model of 1949. The puzzle was the magic numbers — nuclei with 2, 8, 20, 28, 50, 82, 126 protons or neutrons are anomalously stable — and the answer was the level ordering of a 3D well's $V_{\text{eff}}$ landscape: high- $\ell$ orbitals, pressed toward the surface by exactly this article's centrifugal term, plus one decisive ingredient. That ingredient arrived in a corridor conversation when Fermi asked, "Is there any indication of spin-orbit coupling?" — she answered overnight, the strong nuclear spin-orbit force split the high- $\ell$ levels, and the magic numbers fell out in order. (The spin-orbit physics itself belongs to the nuclear and fine-structure articles; here it is one sentence of credit, shared with Hans Jensen, with whom she split the 1963 Nobel — the second woman ever to win it in physics.) Her textbook with Jensen made "the $1g_{9/2}$ orbital" the working vocabulary of a field.

History

1913

— Bohr's planetary model implicitly wrestles with rotational energy in atoms

1926

— Schrödinger's hydrogen paper carries the first quantum radial equation with its

\ell(\ell+1)

term

1935

— Bethe's nuclear reviews systematize barrier penetration and s-wave dominance at low energy

1948-49

— Goeppert Mayer reads the magic numbers out of the effective-potential level ordering; spin-orbit completes the shell model

1963

— Nobel Prize to Goeppert Mayer and Jensen for the shell model

Going further

Griffiths & Schroeter §4.1 — the $u=rR$ reduction and the effective potential at the source register.
Sakurai & Napolitano §3.7; Shankar Ch. 13 — the landscape and its consequences.
Cohen-Tannoudji Vol. I, Complement A $_{VII}$ — the careful treatment of the origin condition.
Landau & Lifshitz §35 — the fall-to-center analysis this article's threshold derivation follows in spirit.

Provenance note: Equations cross-checked against Griffiths Introduction to QM 3rd ed. §4.1 (radial equation, $u=rR$ substitution, $V_{\mathrm{eff}}$ ), Sakurai & Napolitano Modern QM 2nd ed. §3.7, Shankar 2nd ed. Ch. 13, and Cohen-Tannoudji Vol. I Complement A_VII. A verifier should confirm: (1) the $u=rR$ derivation reproduces $u''/r$ for the Laplacian radial part; (2) the centrifugal term is $\hbar^2\ell(\ell+1)/2mr^2$ from the $\mathbf{L}^2$ eigenvalue; (3) the indicial roots $s=\ell+1,-\ell$ ; (4) the effective-minimum formula for $V=-A/r^n$ . No numerical constants beyond standard symbols are asserted. — Items (1), (2), (4) are executed in this article's verify script (the reduction and minimum formulas symbolic; hydrogen and neutron numbers computed); item (3) is verified in The Radial Equation's script (in the $R$ -variable, with the $u$ -variable off-by-one noted there). The admissibility proof for $u(0)=0$ is deferred BY NAME to Boundary Conditions and Normalization of Radial Wavefunctions; the $(kR)^{2\ell+1}$ threshold law to the scattering chapter.

Is u(r) or R(r) the "real" wavefunction?

The full wavefunction is $\psi=R\,Y_\ell^m$ — that is what lives in 3D space. But for every radial question, $u=rR$ is the honest object: $|u|^2dr$ is the radial probability, $u$ obeys the clean 1D equation, and $u$ 's nodes are the countable ones. $R$ is what you reconstruct when you need the 3D amplitude (e.g. the value at the origin for contact interactions, where $R(0)=u'(0)$ matters for s-states).

If the centrifugal term is really kinetic energy, why put it in the potential column at all?

Because the separation fixed $\ell$ , and at fixed $\ell$ the rotational energy is a known function of $r$ alone — indistinguishable, for the radial motion, from a potential. This is the same accounting classical mechanics does for orbits (the effective potential of the Kepler problem) and it is exactly as legitimate: energy conservation along the radial direction includes the rotational share. What it is not is a force field in 3D — no third party pushes outward.

Why does ℓ(ℓ+1) → (ℓ+½)² appear in semiclassical treatments?

Because the 1D WKB machinery, applied blindly to the half-line with $\ell(\ell+1)$ , misses the origin's singular character and quantizes incorrectly; Langer showed that the substitution $\ell(\ell+1)\to(\ell+\tfrac12)^2$ (equivalently a change of variable mapping the half-line to the full line) repairs it. Note $(\ell+\tfrac12)^2=\ell(\ell+1)+\tfrac14$ — the correction is exactly the $\tfrac14$ that appeared in this article's fall-to-center discriminant. The WKB story belongs to the approximation chapters; the $\tfrac14$ 's origin is already visible here.

Does the centrifugal barrier ever help bind anything?

Directly, no — it is purely repulsive. Indirectly, yes, in two ways. It creates shape resonances: a well-plus-barrier landscape can trap positive-energy states behind the barrier, long-lived quasi-bound levels that dominate low-energy cross sections (scattering chapter). And it protects weakly bound halo states from short-range decay channels by walling off the interior — some exotic nuclei owe their existence to the barrier's gatekeeping.

What happens to the landscape picture for the Coulomb problem's degenerate spectrum?

The landscape explains the floor of each $\ell$ -family but not hydrogen's signature coincidence — that $2s$ and $2p$ , with visibly different landscapes, land on exactly the same energy. That degeneracy is accidental (not forced by rotational symmetry) and traces to a hidden conserved vector special to $1/r$ ; its story is the SO(4)/Runge-Lenz pair of articles later in this chapter. The landscape gives inequalities and orderings; the exact coincidences need the hidden symmetry.

Is the fall-to-center threshold physically realized anywhere?

Yes — most cleanly in the interaction of an electron with a polar molecule: the charge-dipole potential falls off exactly as $1/r^2$ , and molecules with dipole moment above a critical value (the Fermi-Teller critical dipole, ~1.63 D) bind an electron while weaker ones cannot — the $\beta_c$ threshold as a laboratory fact. Three-body cold-atom physics realizes the supercritical regime: the log-periodic solutions become the Efimov effect's geometric tower of states, a direct experimental echo of the $r^{1/2}\cos(\nu\ln r)$ behavior derived here.

Why is u(0) = 0 treated as a wall rather than derived here?

Discipline of ownership: the statement is used here (it is what makes the 1D import honest), but its proof is a boundary-condition theorem — normalizability does not by itself force it, and the honest argument runs through hermiticity of the Hamiltonian on the half-line domain plus the delta-function analysis of the previous article. That proof, with the full admissibility doctrine (including what happens for singular potentials where the condition changes), is exactly the next serial's mandate.