Skip to content

Semi-Classical Theory of Radiation

CSIR NET Physical Sciences Part C: atom-field interaction, dipole approximation, absorption and stimulated emission, Einstein A and B, Rabi oscillations.

  • 18 min read
  • Updated
On this page · 15 sections
  1. Prerequisites
  2. The Hook: Einstein’s Coefficients Before Quantum Mechanics
  3. The Interaction Hamiltonian
  4. Transition Rates
  5. Einstein’s A and B Coefficients
  6. Selection Rules
  7. Beyond First Order: Rabi Oscillations
  8. What the Semi-Classical Theory Cannot Do
  9. Memory Tricks
  10. Don’t Confuse
  11. Common Mistakes
  12. Practice
  13. Quick Revision Table
  14. Key Takeaways
  15. What’s Next

Prerequisites

The Hook: Einstein’s Coefficients Before Quantum Mechanics

The Interaction Hamiltonian

Minimal coupling

An electron (charge \( q = -e \)) in an atom with potential \( V(r) \), in a radiation field described by \( \mathbf A(\mathbf r, t) \) in Coulomb gauge (\( \nabla\cdot\mathbf A = 0 \), scalar potential zero for the free field), has

$$\hat H = \frac{(\hat{\mathbf p} - q\mathbf A)^2}{2m} + V(r) = \hat H_0 - \frac{q}{m}\mathbf A\cdot\hat{\mathbf p} + \frac{q^2A^2}{2m}$$

(In Coulomb gauge \( \hat{\mathbf p}\cdot\mathbf A = \mathbf A\cdot\hat{\mathbf p} \).) For ordinary light intensities the \( A^2 \) term is second order and is dropped. The perturbation is

$$\boxed{\hat H'(t) = -\frac{q}{m}\mathbf A\cdot\hat{\mathbf p} = \frac{e}{m}\mathbf A\cdot\hat{\mathbf p}}$$

For a monochromatic plane wave \( \mathbf A = A_0\hat{\boldsymbol\epsilon}\cos(\mathbf k\cdot\mathbf r - \omega t) \), the electric field is \( \mathbf E = -\partial\mathbf A/\partial t \), with amplitude \( E_0 = \omega A_0 \).

The electric dipole approximation

Atomic wavefunctions extend over \( a \sim a_0 \approx 0.05 \) nm, while visible light has \( \lambda \sim 500 \) nm. So \( ka \sim 10^{-3} \) and

$$e^{i\mathbf k\cdot\mathbf r} = 1 + i\mathbf k\cdot\mathbf r + \cdots \approx 1$$

The field is effectively uniform across the atom. Two equivalent forms of the coupling follow:

  • Velocity form: \( \hat H' = \frac{e}{m}A_0\cos\omega t\;\hat{\boldsymbol\epsilon}\cdot\hat{\mathbf p} \).
  • Length form: \( \hat H' = -\hat{\mathbf d}\cdot\mathbf E(t) \), with dipole operator \( \hat{\mathbf d} = q\hat{\mathbf r} = -e\hat{\mathbf r} \).

They give the same matrix elements between energy eigenstates because \( [\hat H_0, \hat{\mathbf r}] = -\frac{i\hbar}{m}\hat{\mathbf p} \), so

$$\boxed{\langle f\rvert\hat{\mathbf p}\lvert i\rangle = im\,\omega_{fi}\,\langle f\rvert\hat{\mathbf r}\lvert i\rangle, \qquad \omega_{fi} = \frac{E_f - E_i}{\hbar}}$$

Every rate below therefore depends on the dipole matrix element \( \mathbf d_{fi} = q\langle f\rvert\hat{\mathbf r}\lvert i\rangle \).

Transition Rates

Monochromatic light: first-order result

Write \( \hat H' = \hat V\cos\omega t \) with \( V_{fi} = -q E_0\,\hat{\boldsymbol\epsilon}\cdot\langle f\rvert\hat{\mathbf r}\lvert i\rangle \). First-order time-dependent perturbation theory gives the amplitude as the sum of two terms, one with denominator \( \omega_{fi} - \omega \) and one with \( \omega_{fi} + \omega \). Near resonance one term dominates (this is the rotating-wave approximation), and

$$\boxed{P_{i\to f}(t) = \frac{\lvert V_{fi}\rvert^2}{\hbar^2}\,\frac{\sin^2\left[(\omega_{fi} - \omega)t/2\right]}{(\omega_{fi} - \omega)^2}}$$
  • \( \omega \approx +\omega_{fi} \) (\( E_f > E_i \)): absorption.
  • \( \omega \approx -\omega_{fi} \) (\( E_f < E_i \)): stimulated emission. Since \( \lvert V_{fi}\rvert = \lvert V_{if}\rvert \), the rate is the same as for absorption between the same pair of states.

At exact resonance \( P = \lvert V_{fi}\rvert^2t^2/4\hbar^2 \) grows as \( t^2 \), which cannot continue; the full two-level treatment below (Rabi oscillations) takes over.

Incoherent broadband light

Real light (a lamp, black-body radiation) has a spread of frequencies with random phases. Replace \( E_0^2 \) by the spectral energy density: the time-averaged energy density is \( u = \frac{1}{2}\epsilon_0E_0^2 \), so \( E_0^2 \to \frac{2}{\epsilon_0}\rho(\omega)\,d\omega \). Integrate over \( \omega \): the function \( \sin^2(\Delta t/2)/\Delta^2 \) has area \( \pi t/2 \), so \( P \propto t \) and a constant rate emerges, as in Fermi’s golden rule. For isotropic, unpolarised light, average \( \lvert\hat{\boldsymbol\epsilon}\cdot\mathbf d_{fi}\rvert^2 \) over directions to get \( \frac{1}{3}\lvert\mathbf d_{fi}\rvert^2 \). The result:

$$\boxed{R_{i\to f} = \frac{\pi}{3\epsilon_0\hbar^2}\,\lvert\mathbf d_{fi}\rvert^2\,\rho(\omega_{fi})}$$

with \( \rho(\omega) \) the energy density per unit angular frequency. Writing \( R = B\rho \) identifies

$$B_{ab} = B_{ba} = \frac{\pi\lvert\mathbf d_{ab}\rvert^2}{3\epsilon_0\hbar^2}$$

for a lower state \( a \) and an upper state \( b \), both non-degenerate.

Einstein’s A and B Coefficients

The rate equations

For populations \( N_a \) (lower) and \( N_b \) (upper) in radiation of spectral density \( \rho(\omega) \):

$$\frac{dN_b}{dt} = -A\,N_b - B_{ba}\rho\,N_b + B_{ab}\rho\,N_a$$

The three terms are spontaneous emission, stimulated emission and absorption.

The thermodynamic argument

In equilibrium at temperature \( T \), \( dN_b/dt = 0 \) and \( N_b/N_a = e^{-\hbar\omega/kT} \) (non-degenerate levels). Solving for \( \rho \):

$$\rho(\omega) = \frac{A/B_{ba}}{(B_{ab}/B_{ba})e^{\hbar\omega/kT} - 1}$$

This must equal the Planck formula \( \rho(\omega) = \frac{\hbar\omega^3}{\pi^2c^3}\frac{1}{e^{\hbar\omega/kT} - 1} \) at every temperature. Therefore

$$\boxed{B_{ab} = B_{ba}, \qquad \frac{A}{B_{ba}} = \frac{\hbar\omega^3}{\pi^2c^3}}$$

With degenerate levels the first becomes \( g_aB_{ab} = g_bB_{ba} \). (If \( \rho \) is defined per unit ordinary frequency \( \nu \), the ratio reads \( A/B = 8\pi h\nu^3/c^3 \); check which convention a question uses.)

The spontaneous emission rate

Combining with the semi-classical \( B \):

$$\boxed{A = \frac{\omega^3\lvert\mathbf d_{ba}\rvert^2}{3\pi\epsilon_0\hbar c^3}}$$

summed over all lower states \( a \) for the total decay rate of \( b \). The lifetime is \( \tau = 1/\sum_aA_{b\to a} \).

Key features:

  • \( A \propto \omega^3\lvert d\rvert^2 \). At the same dipole moment, an ultraviolet transition decays far faster than a microwave one. Doubling \( \omega \) shortens the lifetime eightfold.
  • Ratio of spontaneous to stimulated emission in thermal radiation: \( \frac{A}{B\rho} = e^{\hbar\omega/kT} - 1 \). For visible light at room temperature this is enormous, so spontaneous emission dominates; in the microwave and radio range stimulated emission dominates. The crossover \( \hbar\omega = kT\ln2 \) lies near 4.3 THz at 300 K.
  • Population inversion. In a two-level system pumped by light, absorption and stimulated emission rates are equal per atom, so pumping alone can at best equalise \( N_a \) and \( N_b \). Lasers need three or four levels.

Lifetime and natural linewidth

An upper level that decays at rate \( \Gamma = 1/\tau \) has an amplitude \( \propto e^{-\Gamma t/2} \). Its Fourier transform gives a Lorentzian line with full width at half maximum \( \Gamma \) in angular frequency, \( \Delta\nu = \frac{1}{2\pi\tau} \): the natural linewidth, consistent with \( \Delta E\,\tau \sim \hbar \).

For hydrogen \( 2p \to 1s \) (Lyman-\( \alpha \)): \( \lvert\langle100\rvert z\lvert210\rangle\rvert = \frac{128\sqrt2}{243}a_0 = 0.745a_0 \), \( \hbar\omega = 10.2 \) eV. The formula gives \( A = 6.3\times10^8 \) s\( ^{-1} \), \( \tau = 1.6 \) ns and a natural width \( \Gamma/2\pi \approx 100 \) MHz. A rough rule: an allowed optical transition with \( d \sim ea_0 \) has a lifetime of tens of nanoseconds (at 500 nm, \( d = ea_0 \) gives about 60 ns).

graph LR
    A["Upper level b"] -->|"spontaneous: A"| B["Lower level a"]
    A -->|"stimulated: B rho"| B
    B -->|"absorption: B rho"| A
    C["Thermal equilibrium + Planck law"] --> D["B_ab = B_ba ; A/B = hbar omega^3 / pi^2 c^3"]

Selection Rules

A transition is electric-dipole (E1) allowed only if \( \langle f\rvert\hat{\mathbf r}\lvert i\rangle \neq 0 \). Because \( \hat{\mathbf r} \) is odd under parity and is an \( l = 1 \) object (\( z \propto Y_1^0 \), \( x \pm iy \propto Y_1^{\pm1} \)):

QuantityRuleOrigin
Paritymust change\( \hat{\mathbf r} \) is odd
\( l \) (one electron)\( \Delta l = \pm1 \)\( \hat{\mathbf r} \) carries \( l = 1 \), plus parity
\( m_l \)\( \Delta m = 0 \) (\( \pi \), field along \( z \)); \( \pm1 \) (\( \sigma^\pm \), circular)\( z \) and \( x \pm iy \)
Spin\( \Delta s = 0 \)\( \hat{\mathbf r} \) does not act on spin
\( j \)\( \Delta j = 0, \pm1 \), not \( 0 \to 0 \)vector coupling

Consequences that recur in questions:

  • \( 2s \) of hydrogen is metastable. \( 2s \to 1s \) has \( \Delta l = 0 \) and no parity change, and the only lower state is \( 1s \). It decays mainly by two-photon emission, with a lifetime of about a tenth of a second, some \( 10^8 \) times longer than \( 2p \).
  • Higher multipoles. Expanding \( e^{i\mathbf k\cdot\mathbf r} \) one order further gives magnetic-dipole (M1) and electric-quadrupole (E2) transitions. Their rates are smaller than E1 by factors of order \( (ka)^2 \) for E2 and \( \alpha^2 \) (that is, \( (v/c)^2 \)) for M1: several orders of magnitude for optical lines. E2 allows \( \Delta l = 0, \pm2 \) with no parity change. The hydrogen 21 cm hyperfine line is an M1 transition, which is part of why it is so slow.
  • Oscillator strength. \( f_{fi} = \frac{2m\omega_{fi}}{3\hbar}\lvert\langle f\rvert\hat{\mathbf r}\lvert i\rangle\rvert^2 \) (summed over degenerate final states) is dimensionless; the Thomas–Reiche–Kuhn sum rule gives \( \sum_ff_{fi} = 1 \) for one electron. For hydrogen \( 1s \to 2p \), \( f = 0.416 \): this single line carries about 42% of the total.

Beyond First Order: Rabi Oscillations

For a strong, coherent field near resonance with one pair of levels, solve the two-level problem exactly (still in the rotating-wave approximation). Define the Rabi frequency and detuning

$$\Omega = \frac{\lvert V_{fi}\rvert}{\hbar} = \frac{\lvert\mathbf d_{fi}\cdot\mathbf E_0\rvert}{\hbar}, \qquad \Delta = \omega - \omega_{fi}$$

Starting in the lower state,

$$\boxed{P_{i\to f}(t) = \frac{\Omega^2}{\Omega^2 + \Delta^2}\,\sin^2\left(\frac{\sqrt{\Omega^2 + \Delta^2}}{2}\,t\right)}$$
  • On resonance the population swings completely between the two levels: \( P = \sin^2(\Omega t/2) \). A pulse with \( \Omega t = \pi \) (a \( \pi \) pulse) inverts the population; a \( \pi/2 \) pulse makes an equal superposition.
  • Off resonance the oscillation is faster but incomplete: maximum \( \Omega^2/(\Omega^2 + \Delta^2) \). At \( \Delta = \Omega \) the maximum is \( \frac{1}{2} \).
  • For \( \Omega t \ll 1 \) the formula reduces to the first-order result above. First-order perturbation theory is valid only while the upper-state population stays small.

Numbers: with \( d = ea_0 \) and \( E_0 = 10^5 \) V/m, \( \Omega = ea_0E_0/\hbar = 8.0\times10^9 \) s\( ^{-1} \), so a \( \pi \) pulse lasts about 0.39 ns.

What the Semi-Classical Theory Cannot Do

  • Spontaneous emission. With no applied field, \( \hat H' = 0 \) and an excited eigenstate of \( \hat H_0 \) is stationary forever. Spontaneous emission needs the quantised field (the vacuum fluctuations), or Einstein’s thermodynamic argument as a shortcut.
  • The Lamb shift, which comes from the same vacuum fluctuations.
  • Surprisingly, it can explain the main features of the photoelectric effect: a threshold frequency (resonance with the continuum) and a rate proportional to intensity. That effect alone does not prove the existence of photons.

Memory Tricks

Don’t Confuse

ThisvsThat
Stimulated emission: photon in the same mode as the fieldSpontaneous emission: any direction, needs field quantisation
\( A/B = \hbar\omega^3/\pi^2c^3 \) (per unit \( \omega \))\( A/B = 8\pi h\nu^3/c^3 \) (per unit \( \nu \))
Natural width \( \Gamma = 1/\tau \) (angular)\( \Delta\nu = 1/2\pi\tau \) (ordinary frequency)
E1: parity changes, \( \Delta l = \pm1 \)E2: no parity change, \( \Delta l = 0, \pm2 \)
Rabi frequency \( \Omega \propto E_0 \propto \sqrt{I} \)Incoherent rate \( R \propto \rho \propto I \)
Dipole approximation: \( ka \ll 1 \)Weak-field approximation: drop \( A^2 \), first-order perturbation

Common Mistakes

Practice

All questions below are original practice questions written for this page. They are not taken from any CSIR NET paper. Use \( hc = 1240 \) eV nm, \( k_B = 8.617\times10^{-5} \) eV/K, \( \hbar = 6.582\times10^{-16} \) eV s, \( a_0 = 0.0529 \) nm.

Level 1: Foundation

  1. Is \( 2s \to 1s \) in hydrogen an electric-dipole allowed transition?
  2. How does the ratio \( A/B \) depend on the transition frequency?
  3. State the condition for the electric dipole approximation in terms of wavelength and atomic size.
Answers
  1. No. \( \Delta l = 0 \) and parity does not change.
  2. \( A/B \propto \omega^3 \).
  3. \( \lambda \gg a \), i.e. \( ka \ll 1 \), so the field is uniform across the atom.

Level 2: Part B Style

Practice Q1. For a transition at 600 nm in thermal radiation at 3000 K, the ratio of the spontaneous to the stimulated emission rate is closest to:

(a) 8 (b) 300 (c) \( 3\times10^3 \) (d) \( 3\times10^6 \)

Answer: (c)

\( \hbar\omega = 1240/600 = 2.067 \) eV; \( kT = 8.617\times10^{-5}\times3000 = 0.2585 \) eV; \( \hbar\omega/kT = 7.99 \). Ratio \( = e^{7.99} - 1 \approx 2.96\times10^3 \). Option (a) is the exponent itself.

Practice Q2. Which of the following one-electron transitions in hydrogen is electric-dipole allowed?

(a) \( 3d \to 1s \) (b) \( 2s \to 1s \) (c) \( 3d \to 2p \) (d) \( 3p \to 2p \)

Answer: (c)

Only \( 3d \to 2p \) has \( \Delta l = \pm1 \) (here \( -1 \)). (a) has \( \Delta l = -2 \) (an E2 transition); (b) and (d) have \( \Delta l = 0 \).

Practice Q3. Two transitions have equal dipole matrix elements, but the frequency of the second is twice that of the first. The ratio of the spontaneous lifetime of the second upper level to that of the first (each with only that decay channel) is:

(a) 1/2 (b) 1/4 (c) 1/8 (d) 8

Answer: (c) 1/8

\( A \propto \omega^3\lvert d\rvert^2 \), so doubling \( \omega \) multiplies \( A \) by 8 and divides \( \tau \) by 8. Option (d) inverts the ratio.

Practice Q4. A two-level atom with dipole matrix element \( ea_0 \) (parallel to the field) is driven on resonance by a coherent field of amplitude \( 10^5 \) V/m. The duration of a \( \pi \) pulse is about:

(a) 0.04 ns (b) 0.39 ns (c) 0.78 ns (d) 3.9 ns

Answer: (b) 0.39 ns

\( \hbar\Omega = ea_0E_0 = (0.0529\times10^{-9}\ \text{m})(10^5\ \text{V/m})\,e = 5.29\times10^{-6} \) eV. \( \Omega = 5.29\times10^{-6}/6.582\times10^{-16} = 8.04\times10^9 \) s\( ^{-1} \). \( t_\pi = \pi/\Omega = 0.39 \) ns. Option (c) is a \( 2\pi \) pulse, which returns the atom to the ground state.

Level 3: Part C Style

Practice Q5. For the hydrogen \( 2p\,(m = 0) \to 1s \) transition, take \( \lvert\langle100\rvert z\lvert210\rangle\rvert = 0.745a_0 \) and \( \hbar\omega = 10.2 \) eV, and use \( A = \frac{\omega^3\lvert d\rvert^2}{3\pi\epsilon_0\hbar c^3} \). Consider:

  1. The emitted wavelength is about 122 nm.
  2. \( A \approx 6\times10^8 \) s\( ^{-1} \), so the lifetime is about 1.6 ns.
  3. The natural linewidth \( \Delta\nu \) is about 100 MHz.
  4. The radiation is emitted mainly through the electric-quadrupole channel.

Which are correct?

(a) 1 and 2 only (b) 1, 2 and 3 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4

Answer: (b) 1, 2 and 3 only
  1. \( \lambda = 1240/10.2 = 121.6 \) nm. True.
  2. \( \omega = 10.2/6.582\times10^{-16} = 1.55\times10^{16} \) s\( ^{-1} \); \( d = 0.745\times1.602\times10^{-19}\times0.0529\times10^{-9} = 6.31\times10^{-30} \) C m. Then \( A = \frac{(1.55\times10^{16})^3(6.31\times10^{-30})^2}{3\pi(8.854\times10^{-12})(1.055\times10^{-34})(3.00\times10^8)^3} \approx 6.3\times10^8 \) s\( ^{-1} \), so \( \tau = 1.6 \) ns. True.
  3. \( \Delta\nu = \frac{A}{2\pi} = \frac{6.3\times10^8}{2\pi} \approx 1.0\times10^8 \) Hz. True.
  4. \( \Delta l = 1 \) with a parity change: it is a pure E1 transition. False.

Practice Q6. A two-level atom starts in its lower state and is driven by a coherent field with Rabi frequency \( \Omega \) and detuning \( \Delta \). Consider:

  1. For \( \Delta = \Omega \), the maximum upper-state probability is \( \frac{1}{2} \).
  2. For \( \Delta = \Omega \), the population oscillates at angular frequency \( \sqrt2\,\Omega \).
  3. For \( \Delta = \sqrt3\,\Omega \), the maximum upper-state probability is \( \frac{1}{4} \).
  4. First-order perturbation theory gives the correct probability at all times for \( \Delta = 0 \).

Which are correct?

(a) 1 and 3 only (b) 1, 2 and 3 only (c) 2 and 4 only (d) 1, 2, 3 and 4

Answer: (b) 1, 2 and 3 only

\( P = \frac{\Omega^2}{\Omega'^2}\sin^2(\Omega't/2) \) with \( \Omega' = \sqrt{\Omega^2 + \Delta^2} \).

  1. \( \Omega^2/2\Omega^2 = \frac{1}{2} \). True.
  2. \( \Omega' = \sqrt2\,\Omega \); the probability \( \sin^2(\Omega't/2) = \frac{1}{2}(1 - \cos\Omega't) \) oscillates at \( \Omega' \). True.
  3. \( \Omega^2/(\Omega^2 + 3\Omega^2) = \frac{1}{4} \). True.
  4. On resonance first order gives \( (\Omega t/2)^2 \), which exceeds 1 for \( \Omega t > 2 \); the exact result is \( \sin^2(\Omega t/2) \). False.

Practice Q7. Which statements about Einstein coefficients are correct?

  1. For non-degenerate levels, the absorption and stimulated-emission coefficients are equal.
  2. Optical pumping of a closed two-level system cannot produce a steady-state population inversion.
  3. For the 21 cm line (\( \lambda = 0.21 \) m) in radiation at 3 K, stimulated emission is faster than spontaneous emission.
  4. The ratio \( A/B \) is independent of the transition frequency.

(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4

Answer: (a) 1, 2 and 3 only
  1. From detailed balance with Planck’s law. True.
  2. Equal \( B \) coefficients mean pumping can at most equalise the populations. True.
  3. \( h\nu = 1240\ \text{eV nm}/(2.1\times10^8\ \text{nm}) = 5.9\times10^{-6} \) eV; \( kT = 2.59\times10^{-4} \) eV; \( h\nu/kT = 0.0228 \). \( A/B\rho = e^{0.0228} - 1 = 0.023 \), so stimulated emission is about 43 times faster. True.
  4. \( A/B \propto \omega^3 \). False.

Quick Revision Table

IdeaFormula or fact
Interaction\( H' = -\frac{q}{m}\mathbf A\cdot\mathbf p \) (Coulomb gauge, \( A^2 \) dropped)
Dipole approximation\( e^{i\mathbf k\cdot\mathbf r} \approx 1 \), \( ka \sim 10^{-3} \); \( H' = -\mathbf d\cdot\mathbf E \)
p to r\( \langle f\rvert\mathbf p\lvert i\rangle = im\omega_{fi}\langle f\rvert\mathbf r\lvert i\rangle \)
Monochromatic\( P = \frac{\lvert V_{fi}\rvert^2}{\hbar^2}\frac{\sin^2[(\omega_{fi} - \omega)t/2]}{(\omega_{fi} - \omega)^2} \)
Broadband rate\( R = \frac{\pi}{3\epsilon_0\hbar^2}\lvert d\rvert^2\rho(\omega) \)
Einstein relations\( B_{ab} = B_{ba} \) (\( g_aB_{ab} = g_bB_{ba} \)); \( A/B = \hbar\omega^3/\pi^2c^3 \)
Spontaneous rate\( A = \frac{\omega^3\lvert d\rvert^2}{3\pi\epsilon_0\hbar c^3} \)
Spontaneous/stimulated\( e^{\hbar\omega/kT} - 1 \)
Natural width\( \Gamma = 1/\tau \); Lorentzian; \( \Delta\nu = 1/2\pi\tau \)
Hydrogen 2p\( \tau = 1.6 \) ns, \( \Delta\nu \approx 100 \) MHz
E1 rulesparity changes; \( \Delta l = \pm1 \); \( \Delta m = 0, \pm1 \); \( \Delta s = 0 \)
Oscillator strength\( f = \frac{2m\omega}{3\hbar}\lvert r_{fi}\rvert^2 \); \( \sum f = 1 \); H \( 1s\to2p \): 0.416
Rabi\( P = \frac{\Omega^2}{\Omega^2 + \Delta^2}\sin^2\left(\frac{\sqrt{\Omega^2 + \Delta^2}}{2}t\right) \)

Key Takeaways

What’s Next

Advertisement
Advertisement