Rotational Transitions and the Rigid Rotor

Radio astronomy leans almost entirely on rotational spectroscopy: molecules in cold interstellar clouds (typically 10–100 K) populate only the lowest few rotational levels, and their rotational transitions fall in the radio/millimeter window that penetrates dust and reaches Earth-based and space telescopes. The identification of interstellar CO, HCN, and dozens of more exotic species (including complex prebiotic molecules) relies on matching observed line frequencies to laboratory-measured rotational constants using exactly the $E_J = BJ(J+1)$, $\Delta J=\pm1$ framework developed here. Atmospheric and environmental monitoring instruments use microwave and millimeter-wave absorption to detect trace polar gases (ClO, HCl, and other polar pollutants or ozone-depletion-relevant species) in the stratosphere, since rotational lines are sharp, essentially collision-free at low pressure, and highly species-specific. Industrial gas analysis and isotope ratio measurements exploit the mass-dependence of $B$ (through $\mu = m_1m_2/(m_1+m_2)$): different isotopologues of the same molecule (e.g., H³⁵Cl versus H³⁷Cl) have measurably different rotational constants, allowing non-destructive isotopic composition analysis by microwave spectroscopy. Precision fundamental-physics tests, including searches for a permanent electron electric dipole moment and tests of parity violation in molecules, use ultra-cold polar diatomic and triatomic molecules prepared in single, well-defined rotational states — a direct descendant of the basic rigid-rotor quantum number bookkeeping developed here. Finally, this entire framework is the necessary launching point for real molecular spectra, where rotation and vibration are coupled: Vibration-Rotation Coupling explains how the rotational constant $B$ becomes vibrational-state-dependent ($B_v$), and Vibration-Rotation Spectra: P, Q, R Branches shows how the same $\Delta J=\pm1$ rule, applied on top of a vibrational transition, produces the characteristic P- and R-branch line patterns seen in infrared gas-phase spectra.

Ask the tutor: How does adding vibration change the rotational constant B, and why does that make the P and R branch lines in an infrared spectrum unevenly spaced instead of a perfect 2B comb?

Picture a diatomic molecule as two point masses, $m_1$ and $m_2$, held at a fixed separation $r_e$ by a stiff, unstretchable rod — the rigid rotor approximation. In the center-of-mass frame this two-body problem reduces to a single particle of reduced mass $\mu = m_1 m_2/(m_1+m_2)$ moving on the surface of a sphere of radius $r_e$. The Schrödinger equation for this system is identical in form to the angular part of the hydrogen-atom problem: the eigenfunctions are the spherical harmonics $Y_J^{M}(\theta,\phi)$, labeled by the rotational quantum number $J = 0, 1, 2, \dots$ and the projection quantum number $M = -J, \dots, +J$. Each energy level $E_J$ is $(2J+1)$-fold degenerate, reflecting the $2J+1$ possible orientations of the angular momentum vector in space for a given magnitude. Crucially, in the absence of an external field, all these $M$ states are degenerate — the molecule has no preferred axis in the lab frame until something (an electric field, or the oscillating field of the radiation itself) picks one out. Now add a permanent electric dipole moment $\vec\mu$ that is fixed along the molecular axis (pointing, say, from the negative to positive end of a polar bond like H–Cl). As the molecule rotates, the direction of $\vec\mu$ in the lab frame sweeps around — this rotating dipole is precisely what can couple to an oscillating electric field of electromagnetic radiation. If there is no permanent dipole (as in N₂, O₂, or CO₂ in its ground vibrational state, by symmetry), the molecule presents no rotating charge asymmetry for the field to grab onto, and pure rotational absorption is forbidden regardless of energy matching. The transition intensity is governed by Fermi's Golden Rule and Transition Rates: the rate is proportional to $|\langle f|\hat\mu|i\rangle|^2$, the squared matrix element of the dipole operator between initial and final rotational states. Because $\hat\mu$ transforms as a vector (rank-1 spherical tensor) under rotations, and the initial/final states are spherical harmonics of definite $J$, the integral is governed by the same angular-momentum coupling rules that appear throughout atomic and molecular spectroscopy — this is the deep reason $\Delta J = \pm 1$ emerges, rather than being an ad hoc rule.

Rigid-rotor energies. Solving the angular Schrödinger equation for a rotating diatomic of moment of inertia $I = \mu r_e^2$ gives

where:

• $\hbar$ is the reduced Planck constant,
• $I = \mu r_e^2$ is the moment of inertia, with $\mu$ the reduced mass and $r_e$ the equilibrium bond length,
• $J$ is the rotational angular momentum quantum number, with degeneracy $g_J = 2J+1$.

Reading: levels are not evenly spaced in energy — the gap $E_{J+1}-E_J$ grows linearly with $J$ — yet, as the derivation below shows, the transitions that are actually allowed produce an evenly spaced spectrum. It is conventional to define the rotational constant in wavenumber units,

where:

• $h$ is Planck's constant, $c$ the speed of light,
• $\tilde E_J = E_J/(hc)$ expresses the energy as a spectroscopic term value in cm⁻¹,
• $B$ packages all the molecule-specific structural information ($\mu$, $r_e$) into a single measurable number.

Reading: $B$ is large for light, short molecules (H₂, HF) and small for heavy, long ones (I₂, ICl) — rotational spacings directly encode molecular size and mass distribution.

Transition dipole moment and selection rule. The rate of absorption between rotational states $|J,M\rangle$ and $|J',M'\rangle$ is governed, via Fermi's Golden Rule, by the square of

where:

• $\mu_0$ is the magnitude of the permanent dipole moment along the molecular axis,
• $\hat n(\theta,\phi)$ is the unit vector along the bond axis, whose Cartesian components are proportional to the $\ell=1$ spherical harmonics,
• $d\Omega = \sin\theta\, d\theta\, d\phi$ is the solid-angle element,
• the integral is purely angular because $\mu_0$ (the radial/bond-length dependence) is factored out as a rigid-rotor constant.

Reading: the whole selection-rule problem reduces to a three-spherical-harmonic overlap integral — exactly the type of integral evaluated using Selection Rules from Symmetry and angular-momentum coupling (the Wigner–Eckart theorem in miniature).

Example 1 — Foundational: energy levels and the first three lines of ¹²C¹⁶O. Carbon monoxide has a well known rotational constant $B = 1.9313\ \text{cm}^{-1}$ (a nice asymmetric, non-round number to keep every step honest). Compute the term values for $J=0,1,2,3$ and the first three absorption line positions.

• $J=0$: $\tilde E_0 = 1.9313(0)(1) = 0\ \text{cm}^{-1}$
• $J=1$: $\tilde E_1 = 1.9313(1)(2) = 3.8626\ \text{cm}^{-1}$
• $J=2$: $\tilde E_2 = 1.9313(2)(3) = 11.588\ \text{cm}^{-1}$
• $J=3$: $\tilde E_3 = 1.9313(3)(4) = 23.176\ \text{cm}^{-1}$

Absorption lines, using $\tilde\nu_{J\to J+1} = 2B(J+1)$:

Check the spacing: $7.7252 - 3.8626 = 3.8626\ \text{cm}^{-1} = 2B$, and $11.5878-7.7252 = 3.8626\ \text{cm}^{-1} = 2B$ — the constant spacing predicted in Step 3 of the second derivation holds exactly, as it must for a rigid rotor by construction. Limiting case check: as $J\to 0$ the ground state has zero rotational energy (no zero-point rotational energy, unlike vibration) — this is a genuine feature of an angular-momentum ladder starting at $J=0$, not an approximation artifact.

Example 2 — Applied: extracting the bond length of ¹H³⁵Cl from a measured microwave spectrum. Suppose a microwave absorption scan of gaseous HCl shows lines at (to five significant figures) $20.794$, $41.588$, and $62.382\ \text{cm}^{-1}$ — an intentionally non-round data set as would come from a real recording. Extract $B$, then $I$, then $r_e$. Step 1 — Confirm even spacing and read off $2B$:

So $2B = 20.794\ \text{cm}^{-1}$, giving

Step 2 — Convert to moment of inertia using $B = h/(8\pi^2 c I)$, so $I = h/(8\pi^2 cB)$. Using $h = 6.6261\times10^{-34}\ \text{J s}$, $c = 2.9979\times10^{10}\ \text{cm/s}$:

Evaluate the denominator: $8\pi^2 = 78.957$; then $78.957 \times 2.9979\times10^{10} = 2.3671\times10^{12}$; then $\times 10.397 = 2.4611\times10^{13}$.

(the units work out to kg·m² once the cm⁻¹ and cm/s are consistently converted; carrying the conversion through gives this standard textbook value region for HCl). Step 3 — Compute reduced mass. For ¹H (1.007825 u) and ³⁵Cl (34.96885 u):

Convert: $\mu = 0.97962 \times 1.66054\times10^{-27}\ \text{kg} = 1.6266\times10^{-27}\ \text{kg}$. Step 4 — Solve for bond length:

This lands close to the accepted HCl bond length near $1.27$–$1.29$ Å, illustrating that a handful of microwave line positions — obtained without touching the sample chemically — pin down a bond length to picometer precision. Limiting case check: if instead the two nuclei had equal mass ($m_1=m_2$), the molecule would be homonuclear, $\mu_0=0$ by symmetry, and this entire spectrum would not exist — consistent with the requirement derived in Step 5 of the first derivation.

Example 3 — Exam/competition level: intensity alternation vanishes, but selection-rule logic still predicts the silent partner. A common trap question: "N₂ has a large bond dipole moment change during vibration according to some naive electronegativity argument — does it show a microwave rotational spectrum?" Justify the answer using the transition dipole integral, and then contrast with the isotopically substituted, chemically identical but not-quite-symmetric case of ¹⁴N¹⁵N. Reasoning: N₂ is homonuclear: by the $D_{\infty h}$ point group, the molecule possesses a center of inversion, and a permanent electric dipole moment (a polar vector) cannot be invariant under inversion unless it is exactly zero. Formally, $\mu_0$ transforms as $\Sigma_u^+$ symmetry, while the totally symmetric ground electronic state is $\Sigma_g^+$; the dipole operator has ungerade parity while the molecule's charge distribution has gerade symmetry, forcing $\mu_0 = 0$ identically (see Selection Rules from Symmetry for the general group-theoretic statement). Step 5 of the first derivation shows that once $\mu_0=0$, the entire transition-moment integral is zero for every $J,J'$ pair — there is no accidental "leakage" pure rotational transition, no matter how large the individual bond polarizabilities are. N₂ has zero pure-rotational microwave absorption spectrum. Now consider $^{14}\text{N}^{15}\text{N}$. Chemically the bonding is unchanged — both nuclei are still nitrogen, same electronic structure, same electron cloud, hence still essentially zero electronic dipole moment to extremely good approximation. Isotopic substitution breaks the mechanical symmetry (unequal nuclear masses shift the center of mass off the geometric bond midpoint) but not the electronic symmetry; the equilibrium electron distribution is still (almost) mirror-symmetric about the original bond center. In practice $^{14}\text{N}^{15}\text{N}$ has an extremely tiny nonzero dipole moment (of order $10^{-3}$ D, arising from small non-Born–Oppenheimer/vibrational effects), so an extraordinarily weak pure rotational spectrum is in principle observable with sensitive enough instrumentation — but the intensity is many orders of magnitude below a normal polar molecule like CO, illustrating that the point-group-based selection rule is a robust zeroth-order truth, with only tiny corrections surviving beyond it. Numerical check on the "even spacing" logic surviving isotopic substitution: the rotational constant itself still changes measurably (since $\mu$ changes with isotopic mass), so if ¹⁴N¹⁵N did show lines, they would still be spaced by $2B'$ with $B' = h/8\pi^2cI'$ using the new reduced mass — the derivation's Step 3 (line spacing = $2B$, independent of $J$) is a universal rigid-rotor result, unaffected by whether the transition happens to be allowed or merely extremely weak.

The rigid-rotor treatment of molecular rotation and its use in interpreting microwave spectra is inseparable from the broader development of quantum theory of angular momentum, but the experimental and instrumental breakthrough that made rotational spectroscopy a practical chemical tool is largely credited to Charles Hard Townes and collaborators in the years following World War II, when radar-derived microwave technology was redirected toward molecular spectroscopy.