Quantum Experiments

Quantum Experiments · III · Control and reconstruction · Chapter 14

Ramsey, Decoherence, and Echo

How can phase evolution reveal coherence and its loss?

In this chapter
  1. Separated oscillatory fields and Hahn spin echo
  2. Transverse decay envelopes and relaxation bounds
  3. Noise spectroscopy and reduced-state interpretation
  4. Interactive experiment
  5. Problems and solutions

14.1 Separated oscillatory fields and Hahn spin echo

14.1.1 Converting free-evolution phase into population with two pulses

A single population measurement cannot read the equatorial phase \(\phi\) of a superposition \((|0\rangle + e^{i\phi}|1\rangle)/\sqrt{2}\) directly because \(P_1 = 1/2\) for every \(\phi\). Norman Ramsey’s method of separated oscillatory fields turns phase accumulation into a population signal using two resonant \(\pi/2\) pulses separated by a free-precession interval \(\tau\). The first \(\pi/2\) pulse rotates \(|0\rangle\) onto the Bloch equator; during the dark interval \(\tau\), a detuning \(\Delta = \omega - \omega_0\) rotates the equatorial state around the \(Z\) axis by angle \(\phi = \Delta\tau\); the second \(\pi/2\) pulse maps the resulting equatorial projection into the readout population \(P_1(\tau)\).

14.1.2 Inhomogeneous frequency spread versus irreversible environment coupling

As the free-evolution time \(\tau\) increases, the amplitude of Ramsey fringes decays with characteristic transverse coherence time \(T_2^*\). That decay combines two physically distinct mechanisms: reversible inhomogeneous broadening—where static or slowly drifting shifts \(\delta\omega\) vary from shot to shot or across an ensemble of atoms—and irreversible homogeneous dephasing (\(T_2\)) caused by high-frequency fluctuations and entangling interactions with an uncontrolled environment, alongside longitudinal energy relaxation (\(T_1\)).

14.1.3 Refocusing static detuning with a midpoint inversion pulse

Erwin Hahn’s spin-echo sequence separates static frequency offsets from genuinely stochastic dephasing by inserting a resonant \(\pi\) inversion pulse at the midpoint \(t = \tau/2\) of the dark interval. A static detuning \(\Delta\) accumulates phase \(+\Delta\tau/2\) during the first half \([0,\tau/2]\); the \(\pi\) pulse flips the equatorial phase angle \(\phi \mapsto -\phi\) by exchanging \(|0\rangle\) and \(|1\rangle\); during the second half \([\tau/2,\tau]\), the same static detuning adds \(+\Delta\tau/2\), bringing the net detuning phase back to zero at \(t = \tau\). Any fringe decay that remains at the echo time \(\tau\) measures dynamic noise that varied between the first and second halves of the trajectory.

Visual animation · Section 14.1

Separated oscillatory fields and Hahn spin echo

Separated oscillatory fields and Hahn spin echoSchematic apparatus; geometry and moving indicators are explanatory, not measured trajectories.
\(|0\rangle\)
\(|1\rangle\)
\(+x\)
\(+y\)
\(\pi/2\) pulse \(\to\) free delay \(\tau\)
accumulate phase \(\Delta\tau\)
Optional midpoint \(\pi\) echo
refocus static detuning
Second \(\pi/2\) + readout
envelope \(W(\tau) = \exp(-\tau/T_2)\)
Section 14.1 Live State:
\(P_0(\tau) = \frac{1}{2}\big[1 + W(\tau)\cos(\Delta\tau)\big], \quad W(\tau) = e^{-\tau/T_2}\)
Stage 1 (14.1.1): Converting free-evolution phase into population with two pReady

Interactive visual model for 14.1 Separated oscillatory fields and Hahn spin echo (Converting free-evolution phase into population with two pulses · Inhomogeneous frequency spread versus irreversible environment coupling · Refocusing static detuning with a midpoint inversion pulse). Use Animate or Step to follow the physical state evolution and quantitative predictions.

14.2 Transverse decay envelopes and relaxation bounds

14.2.1 Exponential visibility decay, longitudinal relaxation, and pure dephasing

In the rotating frame of the drive, free evolution for duration \(\tau\) followed by a second \(\pi/2\) pulse with relative phase \(\phi_{\mathrm{ref}}\) produces the Ramsey population signal

\[P_1(\tau) = \frac{1}{2}\Big[1 + W(\tau)\cos(\Delta\tau + \phi_{\mathrm{ref}})\Big],\]

where \(W(\tau) \in [0,1]\) is the coherence envelope equal to the magnitude of the normalized off-diagonal density-matrix element \(2|\rho_{01}(\tau)|\). For Markovian white-noise fluctuations, \(W(\tau) = \exp(-\tau/T_2)\) decays exponentially, and complete positivity of the quantum channel bounds the transverse dephasing rate \(\Gamma_2 = 1/T_2\) in terms of the longitudinal energy relaxation rate \(\Gamma_1 = 1/T_1\) and the pure dephasing rate \(\Gamma_\phi = 1/T_\phi\):

\[\frac{1}{T_2} = \frac{1}{2T_1} + \frac{1}{T_\phi} \quad\Longrightarrow\quad T_2 \le 2T_1.\]

14.2.2 Gaussian envelopes from static ensemble detuning distributions

When each experimental shot or sub-ensemble member experiences a quasi-static frequency shift \(\delta\omega\) drawn from a zero-mean Gaussian distribution \(p(\delta\omega) = \exp[-\delta\omega^2/(2\sigma_\omega^2)]/\sqrt{2\pi\sigma_\omega^2}\), averaging the phase factor \(e^{i(\Delta + \delta\omega)\tau}\) over \(p(\delta\omega)\) yields a Gaussian envelope

\[W_{\mathrm{inhom}}(\tau) = \int_{-\infty}^{\infty} p(\delta\omega)\cos(\delta\omega\,\tau)\,\mathrm{d}(\delta\omega) = \exp\!\left(-\frac{1}{2}\sigma_\omega^2\tau^2\right) = \exp\!\left[-\left(\frac{\tau}{T_2^*}\right)^2\right],\]

with \(T_2^* = \sqrt{2}/\sigma_\omega\). Combined Markovian and quasi-static noise gives a Voigt-profile envelope \(W(\tau) = \exp(-\tau/T_2 - (\tau/T_2^*)^2)\).

14.2.3 Phase reversal under an ideal echo pulse

With a midpoint \(\pi\) pulse about the \(X\) axis at \(\tau/2\), a static frequency offset \(\Delta + \delta\omega\) evolves the initial superposition \((|0\rangle + |1\rangle)/\sqrt{2}\) through

\[\frac{|0\rangle + e^{i(\Delta+\delta\omega)\tau/2}|1\rangle}{\sqrt{2}} \;\xrightarrow{\;\pi_X\;}\; \frac{e^{i(\Delta+\delta\omega)\tau/2}|0\rangle + |1\rangle}{\sqrt{2}} \;\xrightarrow{\;\tau/2\;}\; e^{i(\Delta+\delta\omega)\tau/2}\frac{|0\rangle + |1\rangle}{\sqrt{2}}.\]

Up to a global phase, every static \(\delta\omega\) refocuses at \(t = \tau\), extending the decay time from the inhomogeneous \(T_2^*\) to the homogeneous Hahn-echo coherence time \(T_{2\mathrm{E}} \le 2T_1\).

Visual animation · Section 14.2

Transverse decay envelopes and relaxation bounds

14.2 Transverse decay envelopes and relaxation boundsInteractive geometric state/phasor dial and quantitative prediction curve for Transverse decay envelopes and relaxation bounds.1. State / Amplitude GeometryAmplitude / Bloch projection
Value \(= \) 0.500
2. Quantitative Law (14.2)
\(P_0(\tau) = \frac{1}{2}\big[1 + W(\tau)\cos(\Delta\tau)\big], \quad W(\tau) = e^{-\tau/T_2}\)

Interactive visual model for 14.2 Transverse decay envelopes and relaxation bounds (Exponential visibility decay, longitudinal relaxation, and pure dephasing · Gaussian envelopes from static ensemble detuning distributions · Phase reversal under an ideal echo pulse). Use Animate or Step to follow the physical state evolution and quantitative predictions.

14.3 Noise spectroscopy and reduced-state interpretation

14.3.1 Fitting envelope shapes and non-Markovian revivals

More generally, for stationary Gaussian phase noise with power spectral density \(S_\omega(\omega)\), the coherence envelope obeys \(W(\tau) = \exp[-\chi(\tau)]\), where the attenuation integral \(\chi(\tau) = \frac{1}{2\pi}\int_0^\infty S_\omega(\omega)\,[F(\omega\tau)/\omega^2]\,\mathrm{d}\omega\) weights the noise spectrum by the sequence filter function \(F(\omega\tau)\). Free induction decay (Ramsey) has \(F_{\mathrm{FID}}(\omega\tau) = 4\sin^2(\omega\tau/2)\), which peaks at \(\omega = 0\) and is maximally sensitive to DC drift, whereas Hahn echo has \(F_{\mathrm{echo}}(\omega\tau) = 8\sin^4(\omega\tau/4)\), which vanishes at \(\omega = 0\) and acts as a bandpass filter centered near \(\omega \sim 2\pi/\tau\). If the qubit couples to a discrete coherent bath mode (such as a nearby nuclear spin), \(W(\tau)\) can exhibit collapse and revival oscillations rather than monotonic decay.

14.3.2 Shrinking Bloch vectors as mixed reduced states

When the fringe visibility \(W(\tau)\) decays toward zero, the reduced qubit density matrix at the end of the free-evolution window approaches

\[\rho(\tau) = \frac{1}{2}\begin{pmatrix} 1 & W(\tau)e^{-i\Delta\tau} \\ W(\tau)e^{i\Delta\tau} & 1 \end{pmatrix}, \qquad \mathrm{Tr}\!\left[\rho(\tau)^2\right] = \frac{1 + W(\tau)^2}{2}.\]

Geometrically, the Bloch vector shrinks radially inward along the equatorial plane from length \(|\vec{r}| = 1\) (pure state on the surface) toward \(|\vec{r}| = 0\) (maximally mixed state at the center of the ball).

14.3.3 Operational scope of coherence times

Because \(T_2^*\), \(T_{2\mathrm{E}}\), and dynamical-decoupling coherence times \(T_2^{(N)}\) correspond to different filter functions \(F(\omega\tau)\) and different ensemble averaging windows, quoting a single number “\(T_2\)” without specifying the pulse sequence and averaging time is incomplete. An operational report always pairs the fitted decay constant with the pulse protocol (\(T_2^*\) for Ramsey, \(T_{2\mathrm{E}}\) for single-pulse Hahn echo) and the exponent \(p\) of the stretched-exponential envelope \(\exp[-(\tau/T_2)^p]\).

Visual animation · Section 14.3

Noise spectroscopy and reduced-state interpretation

14.3 Noise spectroscopy and reduced-state interpretationCompares ideal prediction against finite efficiency/coherence and audits the experimental assumptions for Noise spectroscopy and reduced-state interpretation.1. Ideal Model vs. Imperfect Apparatus
\(\frac{1}{T_2} = \frac{1}{2T_1} + \frac{1}{T_\phi} \;\Longrightarrow\; T_2 \le 2T_1\)
2. Audit Ledger (14.3)
Fidelity / efficiency \(\eta = \) 0.85
14.3.1 CheckFitting envelope shapes and non-Markov14.3.2 CheckShrinking Bloch vectors as mixed reduc14.3.3 ScopeOperational scope of coherence times
Ideal \(= \) 1.00 \(\;\mid\;\) Obs. \(= \) 0.85
Audit status: calibrated

Interactive visual model for 14.3 Noise spectroscopy and reduced-state interpretation (Fitting envelope shapes and non-Markovian revivals · Shrinking Bloch vectors as mixed reduced states · Operational scope of coherence times). Use Animate or Step to follow the physical state evolution and quantitative predictions.

Explore the Apparatus

Adjust the detuning \(\Delta\), the free-evolution time \(\tau\), and the inhomogeneous and homogeneous dephasing rates to compare Ramsey free-induction fringes against the Hahn spin-echo envelope. Observe how switching the midpoint \(\pi\) pulse on cancels quasi-static frequency offsets and restores coherence up to the irreversible \(T_{2\mathrm{E}}\) ceiling.

Interactive model · Idealized predictions

A clock made from a superposition

A clock made from a superpositionSchematic apparatus; geometry and moving indicators are explanatory, not measured trajectories.
\(|0\rangle\)
\(|1\rangle\)
\(+x\)
\(+y\)
\(\pi/2\) pulse \(\to\) free delay \(\tau\)
accumulate phase \(\Delta\tau\)
Optional midpoint \(\pi\) echo
refocus static detuning
Second \(\pi/2\) + readout
envelope \(W(\tau) = \exp(-\tau/T_2)\)

On a small screen, swipe the diagram to inspect its components. Moving marks illustrate the model; they are not measured trajectories.

How can phase evolution reveal coherence and its loss? The text and worked examples remain available without JavaScript. Animation starts only when requested; Step gives a motion-free way to follow the model.

Worked Check

Problem: A qubit has energy relaxation time \(T_1 = 40\,\mu\text{s}\), Markovian pure dephasing time \(T_\phi = 80\,\mu\text{s}\), and Gaussian shot-to-shot frequency jitter that yields \(T_2^* = 2.5\,\mu\text{s}\) in a Ramsey scan. What homogeneous coherence time \(T_{2\mathrm{E}}\) is measured when a Hahn echo \(\pi\) pulse removes the static jitter, and what is the echo visibility at \(\tau = 16\,\mu\text{s}\)?

Solution:

1. Combine the longitudinal relaxation contribution \(1/(2T_1)\) and the Markovian pure dephasing rate \(1/T_\phi\):

\[\frac{1}{T_{2\mathrm{E}}} = \frac{1}{2(40\,\mu\text{s})} + \frac{1}{80\,\mu\text{s}} = \frac{1}{80\,\mu\text{s}} + \frac{1}{80\,\mu\text{s}} = \frac{1}{40\,\mu\text{s}} \quad\Longrightarrow\quad T_{2\mathrm{E}} = 40\,\mu\text{s}.\]

2. At \(\tau = 16\,\mu\text{s}\), the unrefocused Ramsey contrast would be \(\exp[-(16/2.5)^2] \approx 1.5\times 10^{-18} \approx 0\), whereas the Hahn echo visibility is

\[W_{\mathrm{echo}}(16\,\mu\text{s}) = \exp\!\left(-\frac{16\,\mu\text{s}}{40\,\mu\text{s}}\right) = e^{-0.40} \approx 0.670.\]

Summary

Practice Problems

Problem 14.1: Extracting detuning from Ramsey fringe spacing

In a Ramsey experiment with zero phase advance on the second \(\pi/2\) pulse, consecutive maxima of \(P_1(\tau)\) occur at \(\tau_k = k \times 0.40\,\mu\text{s}\) for integer \(k\). What is the magnitude of the detuning \(|\Delta|/(2\pi)\) between the drive and the qubit transition?

Hint

One full fringe period \(\Delta\tau_{\mathrm{period}}\) corresponds to a phase advance \(|\Delta|\,\Delta\tau_{\mathrm{period}} = 2\pi\).

Solution

The fringe period is \(\Delta\tau_{\mathrm{period}} = 0.40\,\mu\text{s}\), so \(|\Delta|/(2\pi) = 1/\Delta\tau_{\mathrm{period}} = 1/(0.40\,\mu\text{s}) = 2.5\text{ MHz}\).

Problem 14.2: Testing the \(T_2 \le 2T_1\) bound

An experimental calibration table reports \(T_1 = 15\,\mu\text{s}\) and \(T_{2\mathrm{E}} = 38\,\mu\text{s}\) for the same two-level transition under Markovian relaxation. Explain why these two numbers are mutually inconsistent and identify the tightest upper bound on \(T_{2\mathrm{E}}\) compatible with \(T_1 = 15\,\mu\text{s}\).

Hint

Check the sign of \(1/T_\phi\) in \(1/T_2 = 1/(2T_1) + 1/T_\phi\).

Solution

Since the pure dephasing rate \(1/T_\phi\) cannot be negative, \(1/T_{2\mathrm{E}} \ge 1/(2T_1)\), which requires \(T_{2\mathrm{E}} \le 2T_1 = 30\,\mu\text{s}\). Reporting \(T_{2\mathrm{E}} = 38\,\mu\text{s}\) indicates either a misfit baseline or that \(T_1\) drifted upward between the two measurements.

Problem 14.3: Purity from Ramsey fringe contrast

At free-evolution delay \(\tau = 5.0\,\mu\text{s}\), a Ramsey phase scan over the second pulse’s phase \(\phi_{\mathrm{ref}} \in [0, 2\pi)\) yields \(P_{1,\max} = 0.85\) and \(P_{1,\min} = 0.15\). Assuming equal populations \(\rho_{00} = \rho_{11} = 1/2\) prior to the second pulse, find the fringe contrast \(W(\tau)\) and the purity \(\mathrm{Tr}(\rho^2)\) of the qubit state at time \(\tau\).

Hint

Use \(W(\tau) = P_{1,\max} - P_{1,\min}\) and \(\mathrm{Tr}(\rho^2) = (1 + W^2)/2\).

Solution

The coherence amplitude is \(W(\tau) = 0.85 - 0.15 = 0.70\). With \(\rho_{00} = \rho_{11} = 1/2\) and \(|\rho_{01}| = W/2 = 0.35\), the purity is \(\mathrm{Tr}(\rho^2) = (1 + 0.70^2)/2 = 1.49/2 = 0.745\).

Key Takeaways

Sources and Further Reading

Search Quantum Experiments

Search all 24 chapters.