Opening the book…
We now turn to calculating the matrix element \(V_{k\ell }\) which dictates how strongly the electromagnetic wave interacts with the charged particles. For the moment, we will neglect the vector nature of this interaction. Recognizing that potential is real but the matrix elements complex, we start with eq. (\(7.4.12\)), and write the matrix elements for the case of a single bound charged particle as
\[V_{k\ell } = - \frac{q}{\mathrm{i}m\omega } \, \tilde{\mathcal{E}}_{0}\left\langle k \right|\hat{\mathbf{p}}\left| \ell \right\rangle \label{DOTDME01} %DOCeq57 \]
To evaluate the matrix elements of the momentum operator, it is helpful to use the commutator relationship between a particle’s position and Hamiltonian which holds for central potential Hamiltonians of the form eq. (\(7.3.12\)):
\[\left[ \hat{\mathbf{r}},\hat H_{0} \right] = \frac{\mathrm{i}\hbar \hat{\mathbf{p}} }{m} \label{DOTDME02} %DOCeq58 \]
Solving for \(\hat p\) in this expression, we can evaluate the matrix element in eq. (\ref{DOTDME01}) as
\[\begin{aligned}
\left\langle k \right| \hat{\mathbf{p}} \left|\ell \right\rangle &= \frac{m}{\mathrm{i}\hbar } \left\langle k\left| \kern.1em \hat{\mathbf{r}}{\hat H}_{0} - {\hat H}_{0}\hat{\mathbf{r}} \kern.1em \right|\ell \right\rangle \\
& = \frac{m}{\mathrm{i}\hbar }\left( \left\langle k\right| \hat{\mathbf{r}} \left| \ell \right\rangle E_{\ell } - E_{k}\left\langle k\right| \hat{\mathbf{r}} \left| \ell \right\rangle \vphantom{A^2} \kern-.15em \right) \\
& = \mathrm{i}m\omega _{k\ell }\left\langle k\right| \hat{\mathbf{r}} \left|\ell \right\rangle \\
\end{aligned} \label{DOTDME03} %DOCeq59 \]
So, by substituting back into eq. (\ref{DOTDME01}), we have
\[V_{k\ell } = - q \left\langle k\right| {\mathbf{\hat r}} \left|\ell \right\rangle \frac{\omega _{k\ell }}{\omega }\tilde{\mathcal{E}}_{0} \label{DOTDME04} %DOCeq60 \]
For the case of many charged particles eq. (\ref{DOTDME04}) can be written
\[V_{k\ell } = - \frac{\omega _{k\ell }}{\omega }\tilde{\mathcal{E}}_{0}\sum_{j} q_{j}\left\langle k \right| \hat{\mathbf{r}}_{j} \left| \ell \right\rangle \label{DOTDME05} %DOCeq61 \]
In this form, we observe that the matrix elements are related to a sum over the charge and position of the constituent particles; that is, the spatial distribution of charge within a molecule, or dipole moment. Therefore, it is helpful to define the dipole operator
\[\hat {\boldsymbol{\mu} } = \sum_{j} q_{j} \hat{\mathbf{r}}_{j} \label{DOTDME06} %DOCeq62 \]
The dipole operator \(\hat \mu \) is the quantum analog of the classical dipole moment \(\mu\), which describes the distribution of charge density \(\rho \) in the molecule:
\[\mu = \int dr \, r \rho ( r ) \label{DOTDME07} %DOCeq63 \]
The strength of interaction between light and matter in eq. (\ref{DOTDME05})
\[V_{k\ell } = - \frac{\omega_{k\ell }}{\omega } \tilde{\mathcal{E}}_0 \mu_{k \ell } \label{DOTDME08} %DOCeq64 \]
is now proportional to matrix elements in the dipole operator
\[\mu_{k\ell } = \left\langle k\right| \hat{\boldsymbol{\mu}} \left|\ell \right\rangle \label{DOTDME09} %DOCeq65 \]
which is known as the transition dipole moment. The magnitudes of the transition dipole matrix elements dictate whether there can be efficient absorption of light, which requires a change of molecular distribution of charge between \(\left| k \right\rangle\) and \(\left| \ell \right\rangle \). This matrix element is the foundation underlying the spectroscopic selection rules, which describe what light-driven transitions are allowed. We will look into this is some detail in Chapter 10.
We can now add back the vector nature of the light-matter interaction. From eq. (\(7.4.12\)) we see that \(V_{0}\) depends on the alignment of the light’s electric field with the dipole moment, \(\mathcal{E}_0\hat \varepsilon \cdot \hat{\boldsymbol{\mu} } = \boldsymbol{\mathcal{E}}_{0} \cdot \hat{\boldsymbol{\mu} }\), so the corresponding matrix elements are
\[V_{k\ell } = - \frac{\omega _{k\ell }}{\omega }(\boldsymbol{\mathcal{E}}_{0} \cdot \boldsymbol{\mu}_{kl}) \label{DOTDME10} %DOCeq66 \]
Substituting this into the Golden Rule eq. (\(7.4.15\)) leads to an expression for the rate of transitions between quantum states induced by the light field:
\[\begin{aligned}
w_{k\ell } & = \frac{\pi }{2\hbar }\left| \mathcal{E}_0 \right|^{2}\frac{\omega _{k\ell}^{2}}{\omega ^{2}}\left| \boldsymbol{\mu} _{kl} \right|^{2}\left[ \vphantom{A^2} \delta ( E_{k} - E_{\ell } - \hbar \omega ) + ( E_{k} - E_{\ell } + \hbar \omega) \right] \\
& = \frac{\pi }{2\hbar ^{2}}\left| \mathcal{E}_0 \right|^{2}\left| \boldsymbol{\mu}_{kl} \right|^{2} \left[ \vphantom{A^2} \delta ( \omega _{k\ell } - \omega ) + \delta ( \omega _{k\ell } + \omega ) \right] \\
\end{aligned} \label{DOTDME11} %DOCeq67 \]
We observe that the resonance condition \(E_{k} - E_{\ell } = \pm \hbar \omega \) dictates that the frequency ratio \({ \omega _{k\ell } / \omega } = 1\), and therefore this factor can be dropped from the original expression.
Note that the electric dipole interaction potential used to obtain eq. (\ref{DOTDME10}) is essentially the same as the well-known potential energy for a dipole in an electric field:
\[V = - \boldsymbol{\mu} \cdot \boldsymbol{\mathcal{E}} \label{DOTDME12} %DOCeq68 \]
This raises the question, why did we go through all the effort to derive the electric dipole Hamiltonian when we simply could have quantized the classical energy of a dipole in a field? Remember that our original expression for the interaction of a charged particle with an electromagnetic field, eq. (\(7.3.14\)) is valid for light-matter interactions under any conditions, and we made a number of additional assumptions in obtaining the EDH, eq. (\(7.4.9\)). We assumed interactions occur with weak classical resonant electric fields. We neglected magnetic fields, spin, and the gradient or curl of the electric field. Remember also that the classical expression was derived for static electric fields rather than time-varying electromagnetic fields. What we have learned are the conditions under which it is possible to quantize eq. (\ref{DOTDME12}) to obtain electric dipole transitions.
Equation (\ref{DOTDME11}) shows that the rates of absorption and stimulated emission by the matter is related to the square of the electric field amplitude (or intensity), and the modulus squared of the matrix element connecting the initial and final states. For a molecular system, this expression can be summed over all of the states participating in the interaction with the electric field. If the system is initially in the ground state \(\left| g \right\rangle \), this sum is over all excited states \(\left\{ \left| e \right\rangle \right\}\), and only the absorption terms remain. To calculate an absorption spectrum, we have to remember that this reflects the power or rate of energy change by the system, which reflects the rate of transitions from \(g\) to \(e\) and the energy change accompanying that transition \(\hbar \omega _{eg}\):
\[\overline {\mathcal{P}} _{\text{abs}}(\omega _{eg}) = \dot E(\omega _{eg}) = \hbar \omega _{eg}w_{eg} \label{DOTDME13} %DOCeq69 \]
Then, if the intensity of the light is constant with frequency, the absorption spectrum is
\[\overline {\mathcal{P}} _{\text{abs}}(\omega ) = \frac{\pi \omega }{2\hbar }\left| \mathcal{E}_0 \right|^{2}\sum_{e} \left| \boldsymbol{\mu} _{eg} \right|^{2}\delta ( \omega _{eg} - \omega ) \label{DOTDME14} %DOCeq70 \]
This shows that the Golden Rule absorption spectrum is a series of delta function resonances at frequencies equal to the transition energies (\(\omega = \omega _{eg}\)).
What we can see thus far is that the Golden Rule approach outlined here results in a description of two important features of a quantum absorption spectrum: (1) the absorption spectrum identifies resonances that are related to the energy gap between initial and final states, \(E_{k} - E_{\ell } = \pm \hbar \kern.1em \omega \), and (2) the strength of the interaction of the light and matter and therefore the peak height is proportional to the square of the transition dipole matrix element connecting the initial and final state \(\left| \mu _{eg} \kern-.05em \right|^{2} \kern-.4em .\) What is missing at this point is any description of lineshape that we observed in the classical driven oscillator model. We will address this last point after we look at how the transition dipole moment influences absorption spectra in practice.
This page titled 7.5: The Dipole Operator and Transition Dipole Matrix Elements was last modified on Thu, 04 Jun 2026 01:16:32 GMT and is shared under a CC BY-NC-SA 4.0 license and was authored, remixed, and/or curated by Andrei Tokmakoff via source content that was edited to the style and standards of the LibreTexts platform.
The book could not be loaded just now.