Stacked Superconducting Qubit Layout With Bump-Bonded Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In stacked quantum computing devices, alignment challenges and energy loss due to interactions with lossy dielectric materials limit qubit density and coherence, particularly in 2D arrays where wiring and control elements are deposited, causing decoherence and reducing the ability to perform complex computations.

Innovation Solution

The use of superconductor bump bonds for electrical and mechanical coupling between chips, with control and readout elements on separate chips, reduces energy loss and decoherence by avoiding lossy dielectric materials, and allows for increased qubit density through 3D integration and improved alignment tolerance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If control and readout elements are deposited on the same chip as qubits in 2D arrays, then qubit density and connectivity are improved, but energy loss and decoherence increase due to interactions with lossy dielectric materials

Engineering Contradiction:
Improvequbit densityVSAvoidenergy loss
Core Design Contradiction:
Quantity of substanceVSLoss of energy

Solution Approach 1:

The device is divided into two separate chips: a first chip containing qubits and a second chip containing control and readout elements. This segmentation eliminates the need for lossy dielectric materials between interacting elements, as superconductor bump bonds are used for coupling instead, thereby reducing energy loss while maintaining high qubit density through 3D integration.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Superconductor bump bonds serve as an intermediary coupling mechanism between the first chip (qubits) and second chip (control/readout elements). This intermediary enables electrical and mechanical coupling without requiring lossy dielectric materials, thus preserving quantum coherence while allowing dense qubit arrangements.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Quantity of substance

If multiple chips are stacked for 3D integration, then alignment precision requirements increase, but qubit coupling strength and density can be improved

Engineering Contradiction:
Improvequbit densityVSAvoidalignment precision
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The patent transitions from 2D planar integration to 3D stacked integration by placing control and readout elements on a separate second chip bonded to the first chip. This dimensional change allows qubits and their control elements to be vertically aligned, improving coupling strength while the superconductor bump bonds provide tolerance for alignment variations during manufacturing.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The use of superconductor bump bonds changes the coupling parameter from requiring precise lateral alignment in 2D to allowing vertical stacking with relaxed alignment tolerances. The bump bonds provide mechanical compliance and electrical connectivity that are less sensitive to alignment errors, enabling higher qubit density without proportionally increasing manufacturing difficulty.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If superconductor bump bonds are used for chip coupling, then energy loss is reduced and coherence is preserved, but manufacturing complexity increases

Engineering Contradiction:
ImprovecoherenceVSAvoidmanufacturing complexity
Core Design Contradiction:
ReliabilityVSEase of manufacture

Solution Approach 1:

By segmenting the device into two separate chips bonded via superconductor bump bonds, the patent achieves reduced energy loss and preserved coherence. The segmentation allows each chip to be optimized independently, with the first chip focusing on qubit quality and the second chip on control electronics, while the bump bond interface provides reliable superconducting coupling.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent employs composite construction with superconductor bump bonds combining mechanical bonding and electrical connectivity functions. This composite approach enables reliable chip coupling with low energy loss, as the superconducting material provides both structural integrity and electrical pathways without introducing dielectric losses.

Inventive Principle:
Principle #40Composite materials

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances qubit coupling strength, increases qubit density from 1D to 2D arrays, and reduces energy loss and decoherence, preserving coherence and enabling more complex quantum computations.

Implementation Method 1

a first chip including a superconducting qubit

Methodology Applied
Scientific EffectSuperconductivity: Superconductivity

Implementation Method 2

the superconducting qubit includes a superconducting quantum interference device (SQUID) region

Methodology Applied
Scientific EffectSuperconducting quantum interference: Interference

Implementation Method 3

control elements on separate chips, reduces energy loss and decoherence

Methodology Applied
Scientific EffectCapacitive coupling: Capacitance

Data Source

PatentUS11937520B2Integrating circuit elements in a stacked quantum computing device
Publication Date: 2024.03.19 GOOGLE LLC
  • US11937520B2 patent drawing
  • US11937520B2 patent drawing
  • US11937520B2 patent drawing

AI summary

A stacked quantum computing device including: a first chip including a superconducting qubit, where the superconducting qubit includes a superconducting quantum interference device (SQUID) region, a control region, and a readout region, and a second chip bonded to the first chip, where the second chip includes a first control element overlapping with the SQUID region, a second control element displaced laterally from the control region and without overlapping the control region, and a readout device overlapping the readout region.