greatitself.
Continue
research34 min studyLesson 482 in Physics

Magneto-optical traps and loading

Combine position-selective scattering with velocity damping and extract loading and loss parameters without mistaking fluorescence for an uncalibrated atom count.

Before you begin
On this page

The question for today

What you’ll understand

  • Explain the restoring role of a magnetic gradient and circular polarization.
  • Solve a one-body loading equation and identify its measured parameters.
  • Distinguish density-dependent losses and fluorescence calibration from a simple exponential model.

Optical molasses damps velocity but does not necessarily return atoms to a chosen position. A magneto-optical trap adds a magnetic-field gradient and selected light polarizations so displaced atoms scatter more strongly from a beam pushing them toward the center. It is a driven dissipative trap, usually the first collection stage before conservative trapping. Its atom number reflects a balance of capture and loss, making a loading curve both useful and easy to overinterpret.

Position shifts the resonant transition

In a simple transition picture, the quadrupole magnetic field changes sign across the trap center and Zeeman shifts magnetic sublevels. Opposing circularly polarized beams address different transitions. Choose their handedness together with the field gradient so the inward-pushing beam is closer to resonance for an atom displaced from the center. Red detuning supplies velocity damping as well. The polarizations cannot be specified independently of the atomic transition and field orientation.

Near the center at low speed, expand the mean scattering force into a restoring position term and a damping velocity term. That linear expression defines a spring coefficient but does not make the MOT a globally conservative potential: scattering, pumping, and diffusion continually exchange energy with light. Capture from a hot vapor occurs outside the linear regime and depends on beam size, detuning, intensity, and slowing distance. A good small-amplitude spring constant does not automatically imply a large capture velocity.

Fx≃−κx−αvx,δωZ≃μeffb′xℏF_x\simeq-\kappa x-\alpha v_x,\qquad \delta\omega_Z\simeq\frac{\mu_{\mathrm{eff}}b^{\prime}x}{\hbar}
κ is a local spring coefficient in N/m, α a damping coefficient in kg/s, b′ a field gradient in T/m, and μeff the relevant differential magnetic moment in J/T. The Zeeman formula is a local linear shift; a full multilevel MOT needs its transition strengths and pumping.↔ Scroll wide equations horizontally.

A loading curve separates rate and timescale

Suppose incoming capturable atoms arrive at a constant rate and background collisions remove each trapped atom with a constant probability per unit time. The number then approaches a steady value exponentially from an empty initial trap. The initial slope measures capture rate; the late plateau divided by the initial slope determines the lifetime in this ideal model. A large plateau can result from better capture, smaller loss, or both, so its value alone cannot distinguish those improvements.

Measure an independent decay after turning off the source if loading and holding backgrounds may differ. A dispenser can raise useful atomic flux and harmful hot-atom collision rates simultaneously. Its loading time constant therefore need not equal the lifetime after source turnoff. Vacuum pressure inference also requires collision cross sections and velocity distributions, rather than a universal conversion from seconds to pressure. State the source and laser conditions attached to each fitted parameter.

N˙=R−γN,N(t)=Rγ(1−e−γt),N(0)=0\dot N=R-\gamma N,\qquad N(t)=\frac R\gamma(1-e^{-\gamma t}),\qquad N(0)=0
R is capture rate in atoms/s, γ is one-body loss rate in s⁻¹, N is an atom count, and R/γ is the steady count. This solution requires constant coefficients, negligible density-dependent loss, and initially empty trapping volume.↔ Scroll wide equations horizontally.

Cloud shape and detection change the fit

At higher density, light-assisted two-body collisions can introduce a loss proportional to the integral of density squared. Writing this as a coefficient times number squared assumes a known effective volume. Define that volume from the measured density profile rather than substituting the geometric beam volume. If the cloud expands with number because of multiple scattering or repulsion, the effective volume can change during loading and a constant quadratic coefficient may fail.

Fluorescence power is proportional to atom number only after the single-atom scattering rate, collection solid angle, detector gain, and transmission are calibrated. Saturation, detuning, polarization, and reabsorption can change this proportionality. Comparing loading and decay traces at fixed optical settings reduces ambiguity but does not eliminate it. A statistically convincing exponential fit is evidence for that rate model over the measured interval; it does not prove that two-body losses or density evolution are absent at every setting.

N˙=R−γN−β∫n2d3r=R−γN−βVeffN2,Veff=N2∫n2d3r\dot N=R-\gamma N-\beta\int n^2d^3r=R-\gamma N-\frac\beta{V_{\mathrm{eff}}}N^2,\qquad V_{\mathrm{eff}}=\frac{N^2}{\int n^2d^3r}
n is in m⁻³, β is a two-body atom-loss coefficient in m³/s with event multiplicity included, and Veff is in m³. Using constant β/Veff assumes the relevant density shape and volume do not change appreciably.↔ Scroll wide equations horizontally.

Work it through

Read a low-density loading curve

An initially empty MOT has constant capture rate R = 10⁷ atoms/s and one-body loss rate γ = 0.50 s⁻¹, with negligible pair loss. Find its steady number, loading time constant, and number at t = 2 ln 2 seconds.

  1. The steady number is N∞=R/γ=107/0.50=2.0×107N_\infty=R/\gamma=10^7/0.50=2.0\times10^7 atoms.

  2. The one-body loading time constant is 1/γ=2.0 s1/\gamma=2.0\,\mathrm s. At the specified time, γt = ln 2, so exp(−γt) = 1/2.

  3. The count is N(t)=N∞(1−1/2)=1.0×107N(t)=N_\infty(1-1/2)=1.0\times10^7 atoms. The initial slope remains R, not γ or the plateau value.

The result

The steady count is 2 × 10⁷ atoms, the time constant is 2 s, and the specified time gives 10⁷ atoms.

The living notebook

Explore the model

Interactive lab

Interactive lab · markov

A gain–loss balance has a stationary fraction

Reset and set BOTH Transition p · 0 → 1 to 0.2 and Transition q · 1 → 0 to 0.2; predict equal stationary probabilities. Then Reset and set BOTH p to 0.4 and q to 0.2; predict stationary state-one probability 2/3 before Play. This exact two-state discrete probability model illustrates competing rates only; it is not a MOT atom-count model, a sampled loading trace, or a two-body-loss simulation.

A gain–loss balance has a stationary fractionExact finite-step probabilities for a two-state chain starting in state 0, not sampled trajectories. Playback advances two transitions per second. When p = q = 0 every distribution is stationary; when p = q = 1 the chain alternates forever. A single zero transition creates an absorbing state. Current readings: P(state 1) at step 0: 0%; Stationary P(state 1): 60%; Chain behavior: Monotone convergence. All parameters are adjustable with the labeled controls below.Two-state Markov distribution after 0 steps: state zero probability 1, state one probability 0.Initial distribution (1, 0) · step 0p = 0.3q = 0.201100%0%1−p1−qP(0)P(1)Monotone convergence
P(state 0)P(state 1)Stationary distribution
P(state 1) at step 0
0%
Stationary P(state 1)
60%
Chain behavior
Monotone convergence
t = 0 s

Exact finite-step probabilities for a two-state chain starting in state 0, not sampled trajectories. Playback advances two transitions per second. When p = q = 0 every distribution is stationary; when p = q = 1 the chain alternates forever. A single zero transition creates an absorbing state. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forA stationary population reflects both entrance and exit processes, whose individual values require additional information.

Pause & reason

Check your understanding

In an initially empty constant-rate one-body MOT model, what does the initial slope dN/dt at t = 0 equal?

Your turn

Try it yourself

A constant-volume MOT model has R = 3 × 10⁶ atoms/s, γ = 0.20 s⁻¹, and β/Veff = 10⁻⁸ s⁻¹ per atom. Verify that N = 10⁷ is a steady solution and find the other algebraic root. Which root is physical?

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

Insert the proposed positive count into both loss terms, then factor the resulting quadratic.

Compare with the solution

At 10⁷ atoms, one-body loss is 0.20(107)=2×1060.20(10^7)=2\times10^6 atoms/s and pair loss is 10−8(107)2=10610^{-8}(10^7)^2=10^6 atoms/s. They sum to R. The steady quadratic is N2+2×107N−3×1014=0N^2+2\times10^7N-3\times10^{14}=0, whose roots are 10⁷ and −3 × 10⁷. Only the nonnegative root is a physical atom count. This result assumes the effective volume remains fixed.

Keep the idea

Take it with you

  • A MOT combines Zeeman-sensitive position restoration with optical velocity damping.
  • Initial slope, plateau, and approach time distinguish capture and loss under a declared rate model.
  • Density profiles and photon-detection calibration are required for reliable number and loss estimates.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.