Question 11.30: In this problem we will recover the results Section 11.2.1 d...

In this problem we will recover the results Section 11.2.1 directly from the Hamiltonian for a charged particle in an electromagnetic field (Equation 4.188). An electromagnetic wave can be described by the potentials

H=\frac{1}{2 m}( p -q A )^{2}+q \varphi             (4.188).

A =\frac{ E _{0}}{\omega} \sin ( k \cdot r -\omega t), \quad \varphi=0  .

where in order to satisfy Maxwell’s equations, the wave must be transverse \left( E _{0} \cdot k =0\right) and of course travel at the speed of light (\omega=c| k |) .

(a) Find the electric and magnetic fields for this plane wave.

(b) The Hamiltonian may be written as H^{0}+H^{\prime} where H^{0} is the Hamiltonian in the absence of the electromagnetic wave and H^{\prime}  is the perturbation. Show that the perturbation is given by

\hat{H}^{\prime}(t)=\frac{e}{2 i m \omega} e^{i k \cdot r } E _{0} \cdot \hat{ p } e^{-i \omega t}-\frac{e}{2 i m \omega} e^{-i k \cdot r } E _{0} \cdot \hat{ p } e^{i \omega t}       (11.127).

plus a term proportional to E_{0}^{2} that we will ignore. Note: the first term corresponds to absorption and the second to emission.

(c) In the dipole approximation we set e^{i k \cdot r } \approx 1 . With the electromagnetic wave polarized along the z direction, show that the matrix element for absorption is then

V_{b a}=-\frac{\omega_{0}}{\omega} \wp E_{0} .

Compare Equation 11.41. They’re not exactly the same; would the difference effect our calculations in Section 11.2.3 or 11.3? Why or why not? Hint: To turn the matrix element of p into a matrix element of r, you need to prove the following identity: i m\left[\hat{H}^{0}, \hat{ r }\right]=\hbar \hat{ p } .

V_{b a}=-\wp E_{0}                    (11.41).

The Blue Check Mark means that this solution has been answered and checked by an expert. This guarantees that the final answer is accurate.
Learn more on how we answer questions.

(a)

E =-\nabla \varphi-\frac{\partial A }{\partial t}=-\frac{ E _{0}}{\omega} \cos ( k \cdot r -\omega t)(-\omega)= E _{0} \cos ( k \cdot r -\omega t) .

B = \nabla \times A =\frac{1}{\omega} \sin ( k \cdot r -\omega t) \nabla \times E _{ 0 }-\frac{1}{\omega} E _{ 0 } \times \nabla (\sin ( k \cdot r -\omega t)) .

=0-\frac{1}{\omega} E _{0} \times[\cos ( k \cdot r -\omega t) k ]=\frac{ k \times E _{0}}{\omega} \cos ( k \cdot r -\omega t)=\frac{1}{\omega} k \times E .

(note that E_0 is a constant, so its derivatives are zero).

(b) From Eq. 4.188

H=\frac{1}{2 m}( p -q A )^{2}+q \varphi             (4.188).

H=\frac{1}{2 m}( p -q A ) \cdot( p -q A )+q \varphi+V=\frac{p^{2}}{2 m}+V-\frac{q}{2 m}( p \cdot A + A \cdot p )+\frac{q^{2}}{2 m} A^{2} .

=H^{0}-\frac{q}{2 m}( p \cdot A + A \cdot p )+\frac{q^{2}}{2 m} A^{2} .

Now, acting on a test function f(r),

( p \cdot A + A \cdot p ) f=-i \hbar[ \nabla ( A f)+ A \cdot(\nabla f)]=-i \hbar[( \nabla \cdot A ) f+ A \cdot( \nabla f)+ A \cdot( \nabla f)] ,

But

\nabla \cdot A =\frac{1}{\omega} \nabla \cdot\left[ E _{0} \sin ( k \cdot r -\omega t)\right]=\left.\frac{1}{\omega}[( \nabla \cdot E ) 0) \sin ( k \cdot r -\omega t)+ E _{ 0 } \cdot( \nabla \sin ( k \cdot r -\omega t))\right] .

=\frac{1}{\omega}\left[0+ E _{0} \cdot \cos ( k \cdot r -\omega t) k \right]=0 .

\left(\text { since } E _{0} \cdot k =0 .\right)   So (dropping the test function)

( p \cdot A + A \cdot p )=2 A \cdot p .

Ignoring the A² term, then,

H^{\prime}=\frac{e}{m} A \cdot p =\frac{e}{m \omega} \sin ( k \cdot r -\omega t) E _{0} \cdot p =\frac{e}{2 \operatorname{im\omega}} e^{i k \cdot r } E _{0} \cdot p e^{-i \omega t}-\frac{e}{2 i m \omega} e^{-i k \cdot r } E _{0} \cdot p e^{i \omega t} .

(c) Referring to Eq. 11.36, we have

H_{b a}^{\prime}=\frac{V_{b a}}{2} e^{-i \omega t}, \quad H_{a b}^{\prime}=\frac{V_{a b}}{2} e^{i \omega t}                     (11.36).

V_{b a}=\frac{e}{i m \omega} E_{0}\left\langle b\left|p_{z}\right| a\right\rangle .

But (using Eqs. 2.52 and 3.65)

[x, \hat{p}]=i \hbar                    (2.52).

[\hat{A} \hat{B}, \hat{C}]=\hat{A}[\hat{B}, \hat{C}]+[\hat{A}, \hat{C}] \hat{B}                    (3.65).

\left[H^{0}, z\right]=\frac{1}{2 m}\left[p^{2}, z\right]=\frac{1}{2 m}\left[p_{z}^{2}, z\right]=\frac{1}{2 m}\left(p_{z}\left[p_{z}, z\right]+\left[p_{z}, z\right] p_{z}\right)=\frac{-i \hbar}{m} p_{z} .

so (referring to Eqs. 11.18 and 11.40)

\omega_{0} \equiv \frac{E_{b}-E_{a}}{\hbar}                 (11.18).

H_{b a}^{\prime}=-\wp E_{0} \cos (\omega t), \quad \text { where } \wp \equiv q\left\langle\psi_{b}|z| \psi_{a}\right\rangle               (11.40).

V_{b a}=\frac{e}{i m \omega} E_{0} \frac{i m}{\hbar}\left\langle b\left|\left[H^{0}, z\right]\right| a\right\rangle=\frac{e E_{0}}{\hbar \omega}\left\langle b\left|\left(H^{0} z-z H^{0}\right)\right| a\right\rangle=\frac{e E_{0}}{\hbar \omega}\left(E_{b}-E_{a}\right)\langle b|z| a\rangle .

=\frac{e E_{0}}{\hbar \omega} \hbar \omega_{0}\langle b|z| a\rangle=-\frac{\omega_{0}}{\omega} E_{0} \wp .

This differs from Eq. 11.41, which did not include the ratio of the !’s, but it does not affect our conclusions in Sections 11.2.3 or 11.3, because we only considered transitions at resonance, \omega=\omega_{0} (the probability of a transition at other driving frequencies being negligible).

V_{b a}=-\wp E_{0}                    (11.41).

Related Answered Questions