The energy difference $\Delta E$ is equal to the energy of a photon of wavelength $\lambda=397~\text{nm}$:
\[\Delta E = \hbar \omega = \frac{2\pi \hbar c}{\lambda} \]
For the initial conditions $x(0) = 0.0$, $dx/d\xi(0)=2.0$ the maximum $x_\text{max}=4.20$ and the maximum $(dx/d \xi)_\text{max}=2.87$. For the initial conditions $x(0) = 0.0$, $dx/d\xi(0)=4.0$ the maximum $x_\text{max}=8.34$ and the maximum $(dx/d \xi)_\text{max}=5.87$.
Let's express $\sigma_p$ from $\sigma_x$:
$$\sigma_p = m \sigma_{dx/dt} = m\omega \sigma_{dx/d\xi}/2 = m\omega \sigma_x/2A,$$thus
\[ \sigma_x = \sqrt{\frac{A \hbar}{\omega m}}\]
The energy of the photon in the lab frame is $E_p=\hbar(\omega + \delta \omega)$ and the projection of the momentum onto the $x$-axis is $p_{x} = \hbar(\omega + \delta \omega) /c$. Using the Lorentz transformation, we can obtain the energy of the photon in the rest frame of the ion:
$$ E_p' = \gamma (E_p - vp_{px}) =\gamma \hbar (\omega + \delta \omega_0) \left( 1-\frac{v}{c} \right)= \hbar (\omega + \delta \omega_0) \sqrt{\cfrac{1-\frac{v}{c}}{1+\frac{v}{c}}}$$On the other hand the energy of the photon $E_p'$ in the rest frame of the ion is coincide with the transition eneregy $\hbar \omega$. So we have
$$ \delta \omega_0 = \omega \left( \sqrt{\cfrac{1+\frac{v}{c}}{1-\frac{v}{c}}} - 1 \right). $$
Using the approximation $v/c \ll 1$ we have:
\[ \delta \omega_0 \approx \omega \frac{v}{c} \]
Since a photon carries $\hbar \omega$ energy and its flux density is $\Phi$, we have $I = \hbar \omega \Phi$.
Consider a photon emission in the rest frame of the ion. Let $\theta$ be the angle between the momentum of the emitted photon and the positive direction of the $x$-axis. In this frame all direction of the emission are eqiuprobable, so the probability that $\theta$ will be in the range $[\theta, \theta + d\theta]$ is proportional to the solid angel in equal to the $dw = \cfrac{2\pi \sin \theta d \theta }{4 \pi} = \dfrac{\sin \theta d \theta}{2}$.
Using the Lorentz transformation, we find the momentum projection on the $x$-axis of the emitted photon in the lab frame.
\[ p_{2x} = \cfrac{\gamma \hbar \omega(\cos \theta + \beta)}{c} \]Hence, mean value $\langle p_{2x} \rangle $ is
\[ \langle p_{2x} \rangle = \int\limits_0^\pi \cfrac{\gamma \hbar \omega(\cos \theta + \beta)}{c} \cdot \cfrac{\sin \theta d \theta}{2} = \cfrac{\gamma \beta \hbar \omega}{c}. \]The projection on the $x$-axis of the ion's momentum change $\delta p_x$ is equal to the difference between absorbed and emitted photons:
\[ \delta p_x = \cfrac{\hbar (\omega + \delta \omega)}{c} - \langle p_{2x} \rangle = \cfrac{\hbar \omega}{c} \Big( 1+ \cfrac{\delta \omega}{\omega} - \cfrac{\beta}{\sqrt{1-\beta^2}} \Big) \]According to the approximations $\beta \ll 1$ and $\delta \omega \ll \omega$ we have
\[ \delta p_x \simeq \frac{\hbar \omega}{c}.\]It is following from the symmetry that mean projections of the momentum of emitted photon on the $y$- and $z$-axes are also equal to zero, so $\delta p = \delta p_x$.
Let's assume that the ion has transitioned to the excited state at $t'=0$.
A simple estimation comes from the idea that if $\Gamma$ is a probability of emission per unit time, then at $t' = 1/\Gamma$ the probability of emission is equal to one.
A more precise approach can be applied. Let $P_1(t')$ be the probability that the ion is still in the excited state at $t'$. We also know that the probability of emission in the time span $t' \in [t'; t'+dt']$ (if the photon hasn't been emitted yet) is equal to $\Gamma dt'$. Thus
$$P_1(t'+dt') = P(t') \cdot (1-\Gamma dt').$$
Hence, we obtain a differential equation for $P_1(t')$:
$$ \cfrac{dP_1}{P_1} = - \Gamma dt'.$$With the initial condition $P_1(0) = 1$ we have
$$ P_1(t') = \exp(-\Gamma t') .$$
And finally
$$ \tau_1' = \int_0^{+\infty} t' \cdot \Gamma \exp(-\Gamma t') dt' = \cfrac{1}{\Gamma}.$$
Let's find a mean time $\tau_2'$ between the photon emission and absorption in the rest frame of the ion.
In this frame $\Phi' = \gamma \Phi$ and the energy of the incident photon is $\gamma \hbar (\omega + \delta \omega)\left( 1- v/c\right) = \hbar(\omega + \Delta \omega)$. Thus, the mean lifetime of the ion in the unexcited state is
$$\tau_2' = \cfrac{1}{P\Phi' \sigma} = \cfrac{\left[ \gamma \left(1-\frac{v}{c}\right)(\omega + \delta \omega) -\omega \right]^2 + \Gamma^2}{\Gamma^2 \gamma \Phi \sigma}.$$
The mean time $\tau_2$ between photon emission and absorption in the laboratory frame, taking into account $\beta \ll 1$ and $\delta \omega \ll \omega$, is
$$ \tau_2 = \gamma \tau_2' \approx \cfrac{\Gamma^2 + (\delta \omega - v \omega / c)^2}{\Gamma^2 \Phi \sigma} $$
In accordance with $\Gamma \gg \Phi \sigma$, the force acting onto the ion is
$$ F_x \simeq \cfrac{\delta p_x}{\tau_2} = \frac{I\sigma}{c} \frac{\Gamma^2}{\left(\delta \omega - \omega \frac{v}{c}\right)^2 + \Gamma^2 }.$$
Let's separate the constant and the variable components of the force:
$$ F \simeq \frac{I\sigma}{c} + \frac{I\sigma}{c^2} \frac{2 \Gamma^2 \delta \omega \, \omega v }{(\delta \omega^2 + \Gamma^2)^2} $$
If $F=-kv$, where $k$ is a constant, the force $F$ acts like a viscous friction, so it cools the ion. This case is realized when $\delta \omega < 0$.
The energy difference between the states $4^2S_{1/2}$ and $4^2P_{1/2}$ could be written as the energy of one transition or as the sum of the energies of two transitions:
$$ \Delta E = \cfrac{2\pi \hbar}{\lambda} = \cfrac{2\pi \hbar}{\lambda_1} + \cfrac{2\pi \hbar}{\lambda_2} $$
Thus,
\[\lambda_2 = \frac{\lambda_1 \lambda}{\lambda_1 - \lambda}\]
From the thin lens equation we have
$$ \cfrac{1}{a} + \cfrac{1}{b} = \cfrac{1}{f}, $$so
$$b = \frac{af}{a-f}.$$
The size of the clearest image could be estimated as the size of Airy disk
$$ s = \cfrac{1.22 \lambda f}{D}.$$