Curated map of Zoo entries in the Trapped Ion family.
Entries
| Entry | Type | Status |
|---|---|---|
| cirac-zoller-gate | gate | demonstrated |
| molmer-sorenson-gate | gate | demonstrated |
| shuttling-ion-trap-qubit | infrastructure | demonstrated |
| trapped-ion-qubit | qubit | demonstrated |
| ytterbium-hyperfine-qubit | qubit | demonstrated |
Composition
- gate: 2
- infrastructure: 1
- qubit: 2
Conceptual anchors
- motional-mode-coupling-in-ion-traps is the key evergreen note tying the modality to its entangling-gate layer.
- coherence-time-hierarchy is the right lens for comparing long hyperfine memory times to much faster gate and readout cycles.
Curated synthesis
The most useful way to read this family is as a three-layer stack rather than three unrelated notes.
-
trapped-ion-qubitis the modality baseline- Start there when the question is why ions are still canonical: identical qubits, minute-scale coherence, high-fidelity readout, and shared-mode connectivity.
- It owns the broad platform tradeoff against superconducting and neutral-atom systems.
-
cirac-zoller-gateis the historical mechanism note- It explains the original phonon-bus idea in its most literal form: map spin information into motion, apply a conditional phase, map back.
- Read it when the important question is where trapped-ion entangling gates came from conceptually, not which pulse family dominates today.
-
shuttling-ion-trap-qubitis the scaling architecture note- It exists because the same shared-motion resource that gives all-to-all connectivity also stops scaling cleanly in very long chains.
- QCCD is therefore best understood as an architectural response to motional-mode crowding, not as a new qubit modality.
Modern operational stack
- ytterbium-hyperfine-qubit is the memory-and-readout anchor: clock-state storage, fluorescence detection, and long coherence margins.
- molmer-sorenson-gate is the modern entanglement anchor: virtual-phonon XX interactions, robustness beyond the strict Cirac-Zoller ground-state requirement, and pulse shaping toward 99.9%+ two-qubit fidelity.
- Together they expose the practical systems bottleneck: motional-mode engineering and calibration, not the existence of a clean atomic memory.
Gate-to-architecture decision map
The two gate notes and the shuttling note answer different questions. The first two choose how internal states borrow the phonon bus inside one interaction zone; QCCD chooses how many ions should share that bus before routing replaces longer-range collective motion.
| Route | How motion is used | Closure condition | What it buys | Dominant systems debt | Read it when… |
|---|---|---|---|---|---|
| cirac-zoller-gate | Sequential sideband pulses map a qubit excitation into a real phonon, apply a conditional phase through a second ion and auxiliary level, then map the phonon back | The selected mode must begin near its ground state and end disentangled from both ions | The cleanest literal picture of a phonon as a quantum data bus and the historical route to trapped-ion universality | Ground-state preparation, auxiliary-level control, and sensitivity to heating during a multi-pulse sequence | the question is how the original bus-mediated gate works or why sideband resolution matters |
| molmer-sorenson-gate | A bichromatic spin-dependent force drives a closed trajectory in motional phase space and leaves an effective XX interaction | Every significantly driven mode must return to its starting phase-space point at the gate end; ground-state cooling is not required, but the Lamb-Dicke and mode-closure conditions remain | The modern operational default: direct entanglement, multi-ion reach, and better tolerance of imperfect motional preparation | Spectator-mode coupling, detuning and phase calibration, heating, and pulse-shape complexity as modes crowd | the question is how present-day ion processors execute high-fidelity entangling gates |
| shuttling-ion-trap-qubit | Ions are transported between short-chain memory, interaction, and readout zones; a local MS or related gate still supplies entanglement inside each zone | Transport must preserve internal coherence and deliver a sufficiently cold, well-characterized motional state for the next local gate | Keeps local mode spectra tractable while scaling processor connectivity beyond one long Coulomb crystal | Junction routing, scheduling, transport-induced excitation, recooling, and calibration across many zones | the question has shifted from one gate pulse to machine-scale connectivity and throughput |
This map prevents two common category errors. MS gates are less sensitive to the initial phonon occupation than Cirac-Zoller gates, but they are not independent of motion: residual phase-space displacement still leaves spin and motion entangled. Likewise, QCCD does not replace the phonon-bus gate; it repeatedly reconstructs a small, controllable local bus after transport. Read all three through motional-mode-coupling-in-ion-traps to track where motional complexity moves rather than assuming it disappears.
Coherence evidence ladder
Trapped-ion coherence records are easy to flatten into a single platform number, but they answer different architectural questions. Route a claim by the protection resource it consumes before comparing it with another result.
| Evidence layer | What is being protected | Resource that creates the margin | What the headline does not establish |
|---|---|---|---|
| Passive clock-state encoding | One ion’s internal-state splitting | First-order magnetic-field insensitivity of a hyperfine clock transition, as in ytterbium-hyperfine-qubit | That the same coherence survives laser exposure, transport, repeated measurement, or a full processor duty cycle |
| Dynamically decoupled memory | One stored ion over a long idle interval | A sustained refocusing sequence suppresses residual magnetic and oscillator noise | A bare , or a memory that remains simultaneously available for arbitrary gates |
| Sympathetically cooled memory | Internal-state coherence plus usable motional/readout conditions | A second species removes heating without directly scattering from the data ion; long-memory demonstrations may combine this with dynamical decoupling | That cooling itself suppresses magnetic dephasing, or that the result is a property of the qubit species alone |
| Decoherence-free encoded memory | A logical state distributed across multiple ions | Symmetry cancels collective noise inside a protected subspace; see decoherence-free-subspace | Protection against differential noise, gate faults, leakage, or the physical-qubit overhead of the encoding |
The comparison discipline is therefore: use coherence-time-hierarchy to name the measured timescale, then name the active or encoded resource that produced it. In particular, the ten-minute memory result should be read as a stack—clock-state storage, dynamical decoupling, and sympathetic cooling—not attributed to any one ingredient in isolation. This keeps memory demonstrations informative for architecture without silently turning them into bare-species or processor-wide specifications.
Routing rule: when to stay here versus hand off
- Stay in this MOC for the full modality stack: optical and hyperfine encodings, Cirac-Zoller and Mølmer-Sørensen gates, shared motion, and QCCD scaling.
- Enter ytterbium-hyperfine-qubit or molmer-sorenson-gate directly for the modern Yb + MS implementation slice.
- Hand off to neutral-atom-moc when the comparison shifts from Coulomb-crystal buses to tweezer-rearranged atomic arrays with transient Rydberg interactions.
Scope boundary
This is the authoritative trapped-ion platform map. Species-specific entries are reserved for materially distinct encodings or milestones rather than one page per isotope.