Cherenkov radiation

Cherenkov radiation is electromagnetic radiation from a charged particle exceeding the medium's phase velocity: $c/n(\omega )<v<c$. Here $n(\omega )$ is the positive refractive index at angular frequency $\omega$. The threshold uses phase velocity, not group velocity.^[ref]

The beauty of Cherenkov radiation is that uniform motion can generate a sharply directed electromagnetic wave: the medium's response along successive portions of the track adds with matching wave phase on a cone. The moving electric charge drives the medium's electric polarization; the resulting radiation depends on the collective refractive index. This differs from luminescence, where excitation of material is followed by light emission. Both mechanisms can contribute to the same irradiated sample. A visible blue glow alone therefore identifies neither Cherenkov radiation nor a particular radioactive source.

The Cherenkov cone angle turns a measured direction into a constraint on particle speed when the refractive index is known. The Frank-Tamm formula instead predicts the number of emitted photons in a specified wavelength interval. These are complementary observables: the finite-band Cherenkov yield model below shows why both depend on optical dispersion and on which wavelengths are collected. This makes Cherenkov radiation useful for particle detectors, with photon production, transport and detection treated separately.

For a charge moving at $v=\beta c$, the Cherenkov radiation photon direction makes angle ${\theta }_C$ to the trajectory with $\cos {\theta }_C=1/(n\beta )$. The full cone of photon directions has opening $2{\theta }_C$. The corresponding conical wavefront is perpendicular to those directions and has half-angle $\mu =\pi /2-{\theta }_C$, satisfying $\sin \mu =1/(n\beta )$. Therefore a full opening angle $\alpha$ gives $n=1/[\beta \cos (\alpha /2)]$ for the photon-direction cone, but $n=1/[\beta \sin (\alpha /2)]$ for the wavefront cone. These are different geometrical definitions, not competing approximations to the same angle.

In his 1958 lecture, Pavel Cherenkov recalls the 1934 reports by himself and Sergey Vavilov: radium gamma rays illuminated solutions and their solvents. Weak light was compared using dark-adapted human vision and threshold photometry. Heating and added fluorescence quenching agents left the solvent component and its polarization of an electromagnetic wave unchanged; parallel aqueous fluorescent controls visibly quenched. Its electric vibration direction was parallel to the exciting beam, unlike the compared polarized fluorescence. Further experiments attributed the emission to fast Electrons produced by Compton scattering, rather than directly to gamma rays. These comparisons supported a mechanism distinct from ordinary luminescence.^[ref]

The evidential step is the controlled comparison, rather than the mere presence of light. A weak glow could include fluorescence from impurities; suppressing a known fluorescent control tests whether the intervention actually works. Visual threshold comparisons also require a stable observer and viewing conditions. These early controls do not themselves calibrate absolute photon yield or supply a single-Electron angular scan. Its polarization of an electromagnetic wave comparison also does not by itself measure the Cherenkov cone angle of an individual Electron.

Ilya Frank and Igor Tamm supplied the theory in 1937; Ilya Frank also investigated nuclear physics experimentally and later directed neutron research.^[ref] Sergey Vavilov, an optics and luminescence researcher and collaborator in the discovery, died in 1951. The 1958 Nobel Prize in Physics went to Pavel Cherenkov, Ilya Frank and Igor Tamm; Sergey Vavilov was not a recipient.^[ref]

Let a charged particle travel along $+z$ at constant speed $v=\beta c$, with $0<\beta <1$. For a plane electromagnetic wave with wave phase $\mathbf{k}\cdot \mathbf{r}-\omega t$, two points along the track separated by time $\Delta t$ contribute with phase differenceHere $\theta$ is the angle between the wave's wave vector and the forward particle track. Constructive addition along a long uniform track requires $kv\cos \theta =\omega$. ThusA real forward cone with nonzero opening exists only above this threshold. At $\beta n=1$ the cone closes and the ideal Frank-Tamm formula gives zero yield. For $\beta n<1$ there is no allowed angle, rather than an imaginary observable cone. The requirement is phase matching; substituting the group refractive index would describe a different condition.

There is a second geometric derivation when $n$ is constant. In time $\Delta t$, the charge advances $OP=v\Delta t$, while an earlier spherical phase front has radius $OQ=c\Delta t/n$. Its tangent envelope passes through the current position $P$. The radius $OQ$ is normal to the envelope, so the right triangle gives $\cos \theta =OQ/OP$. Rotating about the track makes a cone. Its surface half-angle $\eta$ is complementary to the emission angle $\theta$, not equal to it.

In a dispersive medium, apply the phase matching derivation separately at each angular frequency. The circle construction uses one fixed phase speed; it does not determine the envelope of a broadband wave packet. In particular, group velocity controls packet propagation under its usual narrow-band assumptions, and the simple complementary-angle construction need not describe a dispersive pulse front. A finite track also broadens the ideal directional selection: adding contributions over duration $T$ gives a factor proportional to $\sin (\Delta T/2)/\Delta$, with $\Delta =kv\cos \theta -\omega$, rather than an infinitely sharp direction.

For constant particle speed and electric charge $q=ze$, in a transparent, linear, nonmagnetic medium with isotropy, the Frank-Tamm formula gives the produced photon density:Set it to zero below threshold. Here $e$ is the positive elementary charge, $x$ is track length, $\lambda =2\pi c/\omega$ is vacuum wavelength, and $\alpha$ is the fine-structure constant. With SI lengths in metres, the density has units ${\mathrm{m}}^{-2}$; multiplying by $dxd\lambda$ gives a dimensionless expected count. The explicit SI convention isThese conventions agree.^[ref]

To check the convention, start from $E=\hslash \omega =hc/\lambda$ and $d^{2}N/(dxdE)=\alpha z^2{\sin }^2\theta /(\hslash c)$. Changing variables uses the positive density Jacobian $|dE/d\lambda |=hc/{\lambda }^2$; since $h/\hslash =2\pi$, the coefficient becomes $2\pi \alpha z^2$. No factor of the material refractive index belongs in this Jacobian: $\lambda$ is not the in-medium wavelength. Multiplication by $\hslash \omega$ gives the corresponding energy loss density in this SI nonmagnetic convention:

For an original constant-index example, stipulate $n=1.50$, $\beta =0.90$, $z=1$ and $\alpha =1/137.036$. Integration from $400$ to $700\mathrm{nm}$ givesThus a stipulated $1\mathrm{cm}$ track produces about $222$ photons in this band in the ideal model. This is an emitted count before absorption, collection losses and detector response, not a brightness measurement or a measured water yield.

Choose an original transparent-band model, defined only for $400\le \lambda \le 700\mathrm{nm}$:This chosen refractive index is not a fit to water, glass or the historical liquids. It decreases from $1.3425$ to approximately $1.300408$, so the threshold $1/n$ increases across the band. For $\beta =0.75$, solving $n({\lambda }_{\ast })=1/\beta$ givesOnly $400\le \lambda <{\lambda }_{\ast }$ emits. Faster particles with $\beta =0.78$ and $0.90$ emit throughout this chosen band. The Cherenkov cone angle and the Frank-Tamm formula are both evaluated with the same $n(\lambda )$.

For a $1\mathrm{cm}$ track the chosen model predicts about $0.589$, $27.40$ and $142.85$ emitted photons for $\beta =0.75$, $0.78$ and $0.90$, respectively. An expected count below one is meaningful over repeated equivalent tracks; it is not a fractional photon observed on one track. The near-threshold example illustrates why extending a constant-index approximation across an entire optical band can give the wrong emitting interval.

Finite bandwidth is essential. The $1/{\lambda }^2$ factor cannot be extrapolated down to arbitrarily small wavelength: the chosen material model stops at its stated band. Real optical dispersion changes the threshold and direction, and appreciable absorption requires a complex material response beyond the simple transparent-medium expression. A detected count must also include transport, collection geometry and the detector's wavelength-dependent response. Even a correct produced spectrum therefore does not, on its own, predict perceived colour or observed glow intensity.