Sub-Doppler cooling

Real atoms have multiple internal quantum states. Optical pumping and spatially varying electromagnetic polarization can remove kinetic energy by additional mechanisms, producing sub-Doppler cooling below the ideal Doppler cooling limit. In polarization-gradient cooling, state-dependent energy shifts from the Stark effect and delayed optical pumping make atoms repeatedly climb potential energy hills. This mechanism needs the internal-state structure omitted by the independent-beam two-level system above. ^[ref].

William D. Phillips's team tested unexpectedly cold sodium molasses with time-of-flight atom detection. A probe above the cloud initially saw no atoms; deliberately heating them verified that the detector worked. Moving the probe below the cloud produced arrival signals consistent with temperatures below the Doppler cooling limit. Independent release/recapture checks used sharply apertured beams and included gravity; uncertain cloud and probe geometry still limited the inferred temperature. These controls tested the thermometer as well as the cooling theory. ^[ref].

The 1988 experiment loaded sodium optical molasses from a slowed atomic beam. After stopping loading, a shutter extinguished the six cooling beams. Released atoms crossed a lower probe; a photomultiplier recorded fluorescence versus arrival time. Calculations integrated the measured initial cloud distribution and an assumed three-dimensional Maxwell-Boltzmann distribution, including gravity and detection geometry. The minimum was $43\pm 20\mu \mathrm{K}$, versus the $240\mu \mathrm{K}$ Doppler cooling limit.^[ref]

Probe separation, probe intensity, cloud density and shutter timing were varied. Fountain-height, horizontal-spreading and release/recapture measurements also gave below-limit results. Geometry remained an uncertainty; agreement among complementary thermometers made a detector artifact less plausible. These measured released atoms, rather than an in-situ velocity distribution.^[ref]

Steven Chu's account connects the explanation to optical pumping, Stark effect shifts and spatially changing electromagnetic polarization; Claude Cohen-Tannoudji's lecture separates this mechanism from subrecoil schemes. Their lectures and William D. Phillips's describe complementary experimental and theoretical routes.^[ref][ref]

A spatial image after a chosen dark interval is a different readout from continuously recording arrivals at a fixed probe. Here is an original idealised fluorescence imaging model: release a dilute cloud, let it move freely, then apply a short uniform probe pulse. Neglect collisions, residual forces, attenuation and motion during the pulse. Use each image's cloud centre to remove mean drift. Uniform gravity shifts that centre without changing the centred variance.

For each atom let $x(t)=x_0+v_{0}t$ along a horizontal imaging axis. Expanding the square about the expected values gives exactlyNo Gaussian distribution is needed for this variance identity. A jointly Gaussian distribution of initial positions and velocities does make the later density Gaussian. For a separable three-dimensional cloud viewed along $z$, integration along the viewing direction gives the column densityHere $N$ is atom number. Integrating this image along $y$ leaves a one-dimensional Gaussian distribution with standard deviation ${\sigma }_x(t)$. This is a projected density measurement, not a three-dimensional radius or a direct velocity histogram. Under the uniform, optically thin probe assumptions, expected fluorescence is proportional to this column density. A known, independent point spread function with finite variance adds its own variance to the measured width; probe duration and intensity gradients need separate treatment.

For a Maxwell-Boltzmann distribution, $T_x=m{\sigma }_v^2/k_B$, where $m$ is atomic mass and $k_B$ the Boltzmann constant. For a nonthermal distribution this defines a directional second-moment temperature, which need not describe every feature of the distribution. A covariance between position and velocity can initially narrow the cloud even though ${\sigma }_v^2$ is positive. The physical covariance matrix requires $C_{xv}^2\le {\sigma }_{x0}^2{\sigma }_v^2$.

Choose ${\sigma }_{x0}=0.60\mathrm{mm}$, ${\sigma }_v=0.060\mathrm{m}{\mathrm{s}}^{-1}$ and $C_{xv}=-0.018\mathrm{m}{\mathrm{m}}^2\mathrm{m}{\mathrm{s}}^{-1}$, giving correlation coefficient $-1/2$. With $t$ in milliseconds, the numerical variance in square millimetres is $0.36-0.036t+0.0036t^2$. Its minimum occurs at $5\mathrm{ms}$, with standard deviation $\sqrt{0.27}\mathrm{mm}$. These chosen values satisfy the covariance matrix constraint; early contraction is compatible with free motion.

Generate thirteen synthetic variance observations at equally spaced times from $0$ to $30\mathrm{ms}$, adding independent Gaussian noise of standard deviation $0.012\mathrm{m}{\mathrm{m}}^2$. Linear least squares fits all three coefficients in $a+bt+c{t}^2$. The example yields $a\simeq 0.37026\mathrm{m}{\mathrm{m}}^2$, $b\simeq -0.036413\mathrm{m}{\mathrm{m}}^2\mathrm{m}{\mathrm{s}}^{-1}$ and $c\simeq 0.0035820\mathrm{m}{\mathrm{m}}^2\mathrm{m}{\mathrm{s}}^{-2}$. With the chosen approximate sodium mass $m=3.82\times 10^{-26}\mathrm{kg}$, the true $T_x$ is $9.96\mu \mathrm{K}$ and the fitted value is $9.91\mu \mathrm{K}$. This is a fit to generated width data, not historical observations.

If the same data are fitted to $a+c{t}^2$, forcing $C_{xv}=0$, the inferred $T_x$ falls to $6.79\mu \mathrm{K}$. The omitted term is partly absorbed into both fitted coefficients. In noiseless linear least squares with a free intercept, its precise effect on the quadratic coefficient iswhere the subscript denotes equal-weight moments over the chosen observation times, using one consistent unit system. Since the ratio is positive for distinct nonnegative times, negative $C_{xv}$ biases this estimate downwards. An incorrect release-time origin can also generate a linear term, so fitting it does not by itself identify a physical initial covariance. Timing calibration, spatial resolution, residual magnetic fields and probe-induced recoil energy remain separate controls. Neither a cold-looking image nor a single long expansion time identifies all three coefficients.

Polarization-gradient cooling uses spatial variation in electromagnetic polarization together with optical pumping between internal quantum states to cool moving chemical atoms.

Jean Dalibard and Claude Cohen-Tannoudji distinguish cooling by changing elliptical polarization from cooling by rotating linear polarization. The hill picture below concerns the former.^[ref]

Finite optical pumping time lets an atom climb a light-shift hill before transfer into a lower magnetic sublevel. Spontaneous emission dissipates the gained potential energy; climbing alone is conservative.^[ref]

To make the geometry and signs explicit, take an original choice of phase origin for equal opposing waves:The real electric field is $\mathrm{Re}[\mathcal{E}(z)e^{-i{\omega }_{L}t}]$; $\lambda$ is wavelength. The fractions $w_{\pm }$ are squared projections onto the defined circular basis vectors, divided by total intensity. Their sum is one: interference changes electromagnetic polarization despite constant total intensity. At $z=0,\lambda /4,\lambda /2$ the field is respectively ${\mathbf{ϵ}}_{+},{\mathbf{ϵ}}_{-},{\mathbf{ϵ}}_{+}$; halfway between these positions it is linearly polarized. Specifying the vectors avoids an observer-dependent verbal handedness convention.

For an illustrative $J_g=1/2\to J_e=3/2$ angular momentum transition, use the squared Clebsch-Gordan coefficients $1$ and $1/3$ for the strong and weak circular couplings from each lower magnetic sublevel. Interchanging circular components interchanges these strengths. The weighted sums are $w_{+}+w_{-}/3=(2+\cos (2kz))/3$ and $w_{-}+w_{+}/3=(2-\cos (2kz))/3$. Rescaling their common negative shift givesBoth potential energies are negative relative to the unilluminated level. Each hill has height $U_0$, and opposite hills are displaced by $\lambda /4$. An atom starting at the $U_{+}$ minimum at $z=0$ loses $U_0$ of kinetic energy if it reaches $z=\lambda /4$ on that branch. A pump transfer there lowers its potential energy by $U_0$, leaving it at a $U_{-}$ minimum. The next uphill segment ends at $z=\lambda /2$, where the favoured transfer is reversed. Actual transfers are stochastic and need not occur exactly at these extrema.

For one completed climb and pump, conservation of energy gives $\Delta K=-U_0$, while the ground-state potential energy returns to its starting value. Neglecting recoil and treating the scattered photon as the sole outgoing radiation change, $\hslash ({\omega }_{\mathrm{out}}-{\omega }_L)=U_0$. A potential drop is an internal-state reset, not an automatic equal upward kick in kinetic energy. Random emission nevertheless adds momentum fluctuations and recoil energy, so useful cooling requires the dissipated energy to compete with that heating.

The recoil energy $E_R=(\hslash k{)}^2/(2m)$ sets a different scale from $k_B{T}_D=\hslash \Gamma /2$. Conventional Sisyphus cooling reaches the recoil regime when reducing the hill height also weakens cooling; it is not a claim of arbitrarily small temperature at vanishing intensity. Subrecoil techniques need further physics.^[ref]

Consider an original reduced rate equation model using the same potentials. Write $s(z)=\cos (2kz)$ and stipulate positive optical pumping rates $R_{+\to -}={\gamma }_p(1-s)$, $R_{-\to +}={\gamma }_p(1+s)$, with ${\gamma }_p>0$. They pump towards $+$ at $z=0$ and towards $-$ at $z=\lambda /4$, consistently with the diagram. These rates retain the relevant symmetry and delay, while omitting excited-state dynamics, quantum coherence and recoil diffusion.

For populations $p_{+}+p_{-}=1$, put $q=p_{+}-p_{-}$. The rate equations conserve population and give $\dot{q}=2{\gamma }_p(s-q)$. The stationary atom has $q=s$; a moving atom cannot follow it instantaneously. On a prescribed trajectory $z=vt$, the periodic solution isSubstitution into $v{q}^{\prime }=2{\gamma }_p(s-q)$ verifies both coefficients. The mean mechanical force is $F=-p_{+}{U}_{+}^{\prime }-p_{-}{U}_{-}^{\prime }=-U_{0}kq\sin (2kz)$. Averaging over a spatial period, with mean squared sine $1/2$ and mean sine times cosine zero, givesThus the force opposes either sign of velocity and vanishes at rest. A small pumping rate makes the small-velocity friction slope large while narrowing its useful velocity range. Holding $v$ fixed while sending ${\gamma }_p$ to zero makes the force vanish; the apparently divergent zero-velocity slope is not unlimited cooling. Because this model has no momentum diffusion, it predicts no equilibrium temperature. It demonstrates the lag mechanism rather than replacing a multilevel quantum calculation.