iSWAP 3: crossmons and coupler

Two crossmons 600 µm apart with a tunable coupler above their inner arms. A through line arches over the chip, and the two qubits’ 2 mm meanders and the coupler’s 1.5 mm meander end open next to it. It is example 3 of the Qiskit Metal iSWAP-gate notebook (after Phys. Rev. X 11, 021058, Fig. 1; Apache-2.0, © IBM and contributors). The Quantcore rebuild is distributed under GPL-3.0.

Build this one to learn how a flux-tunable pair with a tunable coupler is wired, with open readout meanders beside a through line, four control lines and declared couplings. It fits inside the standard 7.5 mm Quantcore die.

The whole iswap3 chip as the v2 notebook builds it

At a glance

Notebook

examples/designs/metal/iswap3_xmon_tc.ipynb (an older one-cell version is kept as iswap3_xmon_tc.v1.ipynb)

Source

Qiskit Metal Full-Physical-Design-of-iSWAP-Gates.ipynb, example 3, © IBM and contributors, Apache-2.0

Chip

7.5 × 7.5 mm Quantcore die (the source’s 4 × 3 mm area centred in it), CPW 10 / 6 µm

On the chip

2 crossmons with Quantcore SQUIDs · 1 tunable coupler · 3 open meanders (2 mm, 2 mm, 1.5 mm) · 1 through line · 4 control lines · 6 bond pads

Builds today

yes (layout built on 2026-11-19 in 9 s)

How the notebook builds it

The layout segments of iswap3, one picture each
  1. Die. The 7.5 mm die with the QCT frame (dicing marks, alignment grids, chip boundary, device ID).

  2. Bond pads. Six launchpads slid to the 100 µm pad-edge rule (the picture zooms on one pad).

  3. Through line. The east-west line arching over the chip, between its two end pads.

  4. Qubits. Two flux-tunable crossmons with Quantcore SQUIDs.

  5. Readout meanders. One 2 mm open meander from each qubit’s claw to an open end beside the through line.

  6. Coupling. The tunable coupler above the inner arms, its 1.5 mm meander, and the declared couplings.

  7. Control lines. Four straight lines to the north and south pads, two crossing over the arch.

  8. Whole chip. Sim groups for the FEM jobs, the content centred in the die, the design saved.

LOM estimates

Numbers are from a run of this notebook on 2026-11-19 (Setup, Design, Check and the LOM cells, 9 s): QuCat normal modes with QuTiP dressed values.

group

qubit

f_q (GHz)

α (MHz)

resonator

f_r (GHz)

κ (MHz)

χ (MHz)

Q1+coupler_right

Q1

5.846

-203

coupler_right

28.59

0

-0.00573

Q2+coupler_left

Q2

5.846

-203

coupler_left

28.59

0

-0.00573

TC+coupler_mid

TC

7.899

-394

coupler_mid

39.7

0

-0.847

group

pair

ZZ (kHz)

neighbours

hybridised

Q1+TC+Q2

Q1-Q2

–

yes

yes

Q1+TC+Q2

Q1-TC

–

yes

yes

Q1+TC+Q2

Q2-TC

–

yes

yes

LOM plot of iswap3_xmon_tc

These are lumped-model estimates from component formulas and calibration, not FEM results. The LOM page says how far to trust them.

Build spec

The rest of this page is the detailed spec the notebook was built from: every component, port, route and length.

The metal: references in this spec build as written, through the metal: wrapper (ports are the Metal pin names, parameters the Metal options). The Quantcore and KQC substitutions are optional.

Source: Quantum Metal docs/circuit-examples/F.Small-quantum-chips/Full-Physical-Design-of-iSWAP-Gates.ipynb, Example 3, after Phys. Rev. X 11, 021058 (2021), Fig. 1. The same qubit and coupler pair is in D.Qubit-couplers/31-TwoCrossmonsTunableCoupler. Licence Apache-2.0. The chip has two TransmonCrossFL crossmons 600 µm apart, with a TunableCoupler01 (an interdigitated C-shaped island with a SQUID and a flux line) above their inner arms. An east–west through line arches over the top of the chip. Two 2 mm qubit readout meanders and a 1.5 mm coupler meander end open next to that line. There are 6 pads. The chip is 4000 × 3000 µm, centred, so Designer = Metal + (2000, 1500). CPW 10/6 µm and Metal defaults (cross_width 20, cross_length 200 per arm, cross_gap 20; claw 30/10/6, ground_spacing 5). A first build uses these substitutions: Xmon for TransmonCrossFL, Xmon(flux_port) for TunableCoupler01, hanger end="open" for the OpenToGround ends.

Components

instance

catalog ref

key params (µm)

at (x, y)

rot

notes

Q1

Xmon(cross_length=400, arm_width=20, gap=20, claws=["n","e","w"], flux_port=True) (metal:TransmonCrossFL)

Metal pads b (N, connector_type 1 = gap coupler), c (E, claw); fl_options default

(2700, 1500)

0

Metal at (2300, 1500); moved 400 µm E to make room for the coupler cell. Metal pins b (2300, 1781) N, c (2581, 1500) E, flux (2300, 1241.5) S

Q2

Xmon(... claws=["n","e","w"]), no flux line

same

(1300, 1500)

180

Metal at (1700, 1500) (moved 400 µm W), orientation 180: b S (1700, 1219), c W (1419, 1500); its flux line is not routed in the source

TC

Xmon(cross_length=200, arm_width=10, gap=10, claws=["w","e"], flux_port=True) (metal:TunableCoupler01)

Metal c_width 400, l_width 20, l_gap 10, a_height 100, fl 5/3/10

(2000, 1500)

180

Metal at (2000, 1650). Metal pins Control (2038, 1784) N, Flux (1886, 1777.5) N; after rot 180 the flux port faces N

T_west, T_east

BondPad(pad_width=80, pad_height=80, pad_gap=58, taper_length=122)

Metal default pad

(525, 1500), (3475, 1500)

270, 90

through line

T_NE, T_NW

BondPad

same

(2500, 2475), (1500, 2475)

180

Q1 drive (b), TC flux

T_SW, T_SE

BondPad

same

(1500, 525), (2500, 525)

0

Q2 drive (b), Q1 flux

Routes

route

from → to

type, length

waypoints / hanger

notes

tline

T_east.a → T_west.a

cpw ≈ 4750–4960

nodes [(3225, 1500), (3225, 2000), (2871, 2354), (1121, 2354), (768, 2000), (768, 1500)]

Metal RouteMeander with jogs R, 45°, 45°, 45°, 45°, R (all 0.5 mm, middle 1.75 mm). The Metal path overshoots T_west and doubles back; use the nodes instead

cpw_1

T_SE.a → Q1.flux

cpw ≈ 800

—

lead 250/250

cpw_2

T_SW.a → Q2.claw_n

cpw ≈ 780

—

Metal Q2.b (gap coupler) → drive line; claw stands in for it

cpw_3

T_NE.a → Q1.claw_n

cpw ≈ 780

—

Metal Q1.b

cpw_4

T_NW.a → TC.flux

cpw ≈ 930

—

coupler flux bias

k1, k2

Q1.claw_w → TC.claw_w; Q2.claw_w → TC.claw_e

cpw ≈ 200

—

short capacitive links that replace the interdigitated overlap

coupler_right

Q1.claw_e → tline.hanger(s≈400, length=100, side="left", end="open")

meander 2000

—

Metal open2 at (3100, 1500), 125 µm from the tline’s x = 3225 vertical run; spacing 90, fillet 40

coupler_left

Q2.claw_e → tline.hanger(s≈3650, length=100, side="left", end="open")

meander 2000

—

Metal open3 at (900, 1500), beside the x = 768 vertical run

coupler_mid

(TC readout) → tline.hanger(s≈2000, side="right", end="open")

meander 1500

—

Metal: TC.Control → open1 (2250, 2300), 54 µm below the y = 2354 run. Optional, because the Xmon stand-in has no free claw

Launchers, pads, test structures, markers, labels

6 default Metal pads: W/E for the through line, NE/SW for the qubit drives, SE for Q1 flux, NW for coupler flux. There are no markers.

LOM node structure

  • Q1 and Q2 are fixed transmons; Q1 has a flux line in the layout. TC is a tunable coupler island with a SQUID biased from T_NW.

  • The coupling chain is Q1, C, TC, C, Q2, plus the direct Q1–Q2 arm capacitance.

  • The readout resonators (coupler_right/left, and the coupler’s coupler_mid) have C_g at the claw and open ends capacitively coupled to the through line, which runs between the 50 Ω ports T_east and T_west.

Targets (if stated)

None. The 2 mm open-open lines would resonate near 30 GHz, so the lengths are illustrative only.

Simplifications for a first rebuild

  1. Use realistic readout lengths: make each a λ/4 hanger with end="short", meander about 4500 µm.

  2. Drop coupler_mid until the coupler cell has a readout claw (or use KQC Swissmon for TC: cplr0/cplr2 to the qubits, cplr1 to the readout, flux port).

  3. The qubit spacing is widened from 600 to 1400 µm so a coupler cell and the k1/k2 links fit. Keep this if metal:TunableCoupler01 is not used.

Quantcore PDK rebuild

The notebook is a Quantcore PDK rebuild of the source, not a replica: the netlist, LOM and imported components are kept, and the process is Quantcore’s.

Notebook: examples/designs/metal/iswap3_xmon_tc.ipynb.

  • Chip: the Quantcore die frame (L dicing markers 1000/0, 3 × 3 alignment grids 1001/0, chip boundary 100/0, device ID where an edge is free; any brand text is QCT), on the 7.5 × 7.5 mm Quantcore die; the source’s 4 × 3 mm chip area is centred in it (box=(-1750, -2250, 5750, 5250)), so every coordinate stays the source’s.

  • Stack: sapphire 650 µm, εr 10.25 (the PDK default stack, used by the LOM estimates and every FEM export); Metal’s DesignPlanar default is silicon (εr 11.45, 750 µm), so the LOM capacitances, resonator frequencies and FEM jobs here are for sapphire rather than the source’s substrate.

  • Junctions: Quantcore Nb/AlOx/Nb trilayer junctions (junction_type from layouts.registry.trilayer_junction): fingers on 2/0 + 4/0, spacer 50/0 with the 51–53/0 gradient, sized on the PDK junction rule (Nb Δ = 1.4 meV, I_c·R_n = πΔ/2e, R_n·A 13.5 kΩ·µm² and C_J 50 fF/µm² assumed); the two TransmonCrossFL (flux lines) and the TunableCoupler01 each get a PDK SQUID of two 0.25 × 0.6 µm junctions → R_n = 45 kΩ in parallel, L_J = 6.73 nH at zero flux, C_J = 15 fF (in the 40–50 kΩ window), in place of the assumed 10 nH + 2 fF. Expected f_q ≈ 5.8 GHz (an estimate), the coupler ≈ 7.9 GHz at zero flux. The two qubits stay identical.

  • Kept: the topology and every Metal component with the source’s options (through the metal: wrapper), positions and orientations, every route’s end pins and total length, the sim groups (the three open meanders 1.5 / 2.0 / 2.0 mm, the arched through line, the two airbridges, the d.couple estimates).