Relate intracavity dynamics to the measured output
Declare port phases and linewidths, derive reflection, and preserve vacuum noise while connecting cavity population to external photon flux.
On this page
The question for today
What you’ll understand
- Use a consistent sign convention for the Langevin and boundary equations.
- Distinguish dimensionless cavity occupation from a traveling photon flux.
- Compute resonance reflection and critical coupling with internal loss.
A detector outside a cavity measures the outgoing field, not the intracavity photon number directly. The outgoing amplitude contains both prompt incident-field reflection and radiation leaking from the resonator, so interference can cancel the output even when the cavity stores energy. Input–output theory relates those amplitudes while accounting for damping and the vacuum noise required by quantum commutators. We will use kappa as the photon-number decay rate and state the input–output phase convention explicitly.
Normalize a traveling field as a flux amplitude
The cavity annihilation operator is dimensionless and its number operator counts stored photons. A traveling input operator instead has a delta-function time commutator and units of inverse square-root second. Its normally ordered squared amplitude is photon flux. Multiplication by square root of a port decay rate converts that incoming flux amplitude into a cavity drive. Confusing these normalizations produces an incorrect steady photon number or a leaking flux missing a decay-rate factor.
Couple the cavity to an external measured port with energy decay kappa external and to an unobserved internal-loss bath with decay kappa internal. Their sum is the total photon-number decay. In a frame rotating at the drive frequency, define detuning as drive minus cavity frequency. Choose a plus sign for the incoming drive and a minus sign for emitted cavity amplitude in the boundary relation. Other port phase choices are allowed if both equations are transformed consistently.
Solve the mean and then combine amplitudes
For a coherent incident beam, let beta be its mean input amplitude and take the internal bath to have zero mean. The steady cavity amplitude is obtained by dividing the drive by total amplitude damping minus i detuning. In a linear cavity with vacuum baths this gives a coherent intracavity state and occupation equal to the squared amplitude. Its output amplitude is the incident amplitude minus the leaking cavity amplitude, rather than a sum of their separately squared intensities.
On resonance the reflection coefficient is internal loss minus external coupling divided by their sum. A lossless one-sided cavity reflects all incident power with a phase reversal on resonance, despite nonzero stored energy. At equal external and internal rates the coherent reflected amplitude vanishes: incident energy is dissipated through the internal channel. If a second physical output mirror replaces internal absorption, that energy exits another port and must be included explicitly rather than labeled absorbed.
Damping alone cannot preserve a quantum mode
If the cavity operator were merely multiplied by an exponentially decaying factor, its equal-time commutator would decay as well, contradicting the bosonic algebra. The bath inputs supply exactly the noise contribution needed to preserve that commutator. Vacuum noise can therefore be essential even though its normally ordered incoming photon flux is zero. Omitting it may accidentally preserve a classical mean-field solution while making the quantum fluctuations inconsistent.
With the drive removed, stored photon number decays at kappa and mean amplitude at kappa divided by two. Under drive, however, the output flux includes interference with the incident beam and is not generally kappa external times stored occupation. The wave lab can display addition of two coherent amplitudes with a relative phase, but its controls do not set cavity linewidth, detuning, ports, or vacuum noise. It illustrates the boundary interference arithmetic and leaves the actual cavity dynamics to these equations.
Work it through
Balance reflected and internally lost photons
A resonant empty cavity has κex=3.0×10⁶ s⁻¹ and κi=1.0×10⁶ s⁻¹. A coherent input supplies 1.0×10⁶ photons/s. Find intracavity occupation, reflected flux, and internal-loss flux in steady state.
Total photon decay is κ=4.0×10⁶ s⁻¹. On resonance n̄=4κex|β|²/κ²=4×3×1/16=0.75 photons.
The reflection amplitude is (1−3)/(1+3)=−1/2, so reflected power fraction is one quarter. Reflected photon flux is 2.5×10⁵ s⁻¹, including prompt-leakage interference.
Internal loss removes photons at κi n̄=7.5×10⁵ s⁻¹. Adding that to reflected flux returns the full input 1.0×10⁶ s⁻¹, as a stationary energy balance requires.
The cavity stores mean occupation 0.75, reflects 2.5×10⁵ photons/s, and loses 7.5×10⁵ photons/s internally. Leakage intensity alone would give the wrong measured reflection.
The living notebook
Explore the model
Interactive lab · waves
Combine prompt and cavity amplitudes
For independent runs Reset and set BOTH phase 180°, amplitude ratio 1; then Reset and set phase 0°, ratio 1. Predict zero and doubled summed amplitudes. The fixed-frequency sine model illustrates coherent boundary interference; it contains no cavity, port rates, linewidth, or quantum bath noise.
- Sum at x = 0
- 0.866
- Resultant amplitude
- 1.732
- Phase difference
- 60°
Two waves share the same frequency and wave number. Playback advances t in sin(2x − t) + a sin(2x − t + phase). 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 forOutput detection combines amplitudes before squaring, so zero reflection can coexist with stored cavity energy.
Pause & reason
Check your understanding
Your turn
Try it yourself
A resonant cavity has κex=κi=2.0×10⁶ s⁻¹ and input flux 1.0×10⁶ photons/s. Find stored occupation and reflected flux. Then double both rates at fixed input flux and determine the new occupation and photon-number ringdown time.
Take a moment to make a prediction or work it out on paper. Open a hint when you need one.
Give me a hint
At critical coupling reflection vanishes; stored occupation scales inversely with a common scaling of both rates.
Compare with the solution
Initially κ=4.0×10⁶ s⁻¹ and n̄=4κexΦ/κ²=0.5, with zero coherent reflected flux in the ideal model. After doubling both rates κ=8.0×10⁶ s⁻¹ and n̄=0.25. The undriven photon-number ringdown time is 1/κ=125 ns. Critical coupling persists, while the field-amplitude ringdown time is twice the photon-number time.
Keep the idea
Take it with you
- Traveling flux amplitudes and dimensionless cavity operators have different units.
- Port phases and amplitude-decay conventions must remain consistent between equations.
- Vacuum inputs preserve quantum commutators even when their normal-ordered flux vanishes.
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.