Magnetic resonance: three ways to notice a transition

Rabi’s signal was a beam that failed to refocus

The 1938 experiment sent molecules through deflecting fields that could refocus a selected beam onto a detector. Between them, a weak oscillating magnetic field was applied transverse to a much larger uniform static field. A resonant change of magnetic state changed the later deflection, reducing the refocused detector signal.

The published example used lithium chloride and scanned the static field while holding the radiofrequency fixed. Its trace was beam intensity against magnet current, not an image of a precessing nucleus. The authors reported sharp resonances while reserving final magnetic-moment values until recalibration of the uniform field.

Read the evidence: I. I. Rabi, J. R. Zacharias, S. Millman and P. Kusch (1938), A New Method of Measuring Nuclear Magnetic Moment, Phys. Rev. 53, 318 — original scan

The shared reference is a frequency–field relation

For an isolated spin in a uniform magnetic flux density, the Larmor angular-frequency magnitude is the magnitude of the gyromagnetic ratio times the field. The second equality converts angular frequency to cycles per second. The gyromagnetic ratio here uses radians per second per tesla, and taking its magnitude removes the precession-sign convention.

This relation supplies a resonance center in the simplest case. It does not specify a beam dip, an absorbed power or a pickup voltage. Couplings, shielding and local fields can shift or split a real spectrum, while finite interaction time, field variation and drive strength alter its width.

\omega_{\mathrm L}=|\gamma|B_0=2\pi\nu_{\mathrm L}

Read the evidence: I. I. Rabi, J. R. Zacharias, S. Millman and P. Kusch (1938), A New Method of Measuring Nuclear Magnetic Moment, Phys. Rev. 53, 318 — original scan E. M. Purcell, H. C. Torrey and R. V. Pound (1946), Resonance Absorption by Nuclear Magnetic Moments in a Solid, Phys. Rev. 69, 37–38 F. Bloch, W. W. Hansen and M. Packard (1946), Nuclear Induction, Phys. Rev. 69, 127

Purcell, Torrey and Pound detected absorption in a solid

The 1946 Purcell–Torrey–Pound report placed proton-containing paraffin in a radiofrequency resonator. Sweeping the static field produced a sharp change in the balanced circuit output, interpreted as absorption by nuclear magnetic moments. A large bulk sample replaced a dilute moving molecular beam.

At room temperature, the two spin populations differ only slightly. Net absorption requires that imbalance because absorption and stimulated emission compete. Too strong a radiofrequency field can reduce it through saturation unless relaxation restores the population difference. A resonance is therefore a response of a prepared, relaxing ensemble.

Read the evidence: E. M. Purcell, H. C. Torrey and R. V. Pound (1946), Resonance Absorption by Nuclear Magnetic Moments in a Solid, Phys. Rev. 69, 37–38

Bloch, Hansen and Packard detected induction

Bloch, Hansen and Packard used water at room temperature. A transverse driving field set the bulk nuclear magnetization into motion, and a pickup coil along another axis detected an induced voltage. Their short 1946 letter compared the resonance field at two drive frequencies with previously measured proton magnetic data.

The original observation used an applied oscillating field; it should not be silently replaced by a later pulsed free-induction experiment. Absorption and induction probe related spin dynamics through different circuit observables. They do not amount to tracking each proton separately.

Read the evidence: F. Bloch, W. W. Hansen and M. Packard (1946), Nuclear Induction, Phys. Rev. 69, 127

A narrow line is not yet a complete measurement

A magnetic-moment inference needs both a frequency scale and the actual field seen by the sample. Field inhomogeneity, chemical surroundings, power shifts and calibration can matter even when the trace looks sharp. A reproducible peak tests repeatability before it establishes an absolute value.

The 1944 Rabi award and the 1952 Bloch–Purcell award recognize related advances with distinct methods and teams. Neither early bulk report provides spatial encoding for magnetic-resonance imaging. Producing an image needs additional gradients, acquisition and reconstruction, which this reading does not claim to implement.

Continue to control and refocusing

The Rabi-pulse Lab follows an ideal two-state drive and a chosen readout; it does not model the 1938 lithium-chloride beam and refocusing detector. The spin-echo Lab isolates how a controlled pulse can reverse some phase dispersion; it does not reproduce the early resonator or pickup-coil circuit.

Use them to ask a more precise question than whether resonance exists: which state was prepared, which dynamics occurred and which detector converted them into a record? The Ramsey article extends the interrogation time, while the optical-pumping article asks how light can prepare populations and reveal a radiofrequency transition.