[Paper Review] Titanium Nitride Film on Sapphire Substrate with Low Dielectric Loss for Superconducting Qubits

Hao Deng, Zhijun Song+20 more|arXiv (Cornell University)|May 7, 2022

Physics of Superconductivity and Magnetism26 citations

TL;DR

The paper demonstrates TiN on sapphire with ultra-low substrate-metal dielectric loss, enabling transmon qubits with T1 up to ~300 μs and Q ~8×10^6, and provides quantified loss tangents for SM and junction regions.

ABSTRACT

Dielectric loss is one of the major decoherence sources of superconducting qubits. Contemporary high-coherence superconducting qubits are formed by material systems mostly consisting of superconducting films on substrate with low dielectric loss, where the loss mainly originates from the surfaces and interfaces. Among the multiple candidates for material systems, a combination of titanium nitride (TiN) film and sapphire substrate has good potential because of its chemical stability against oxidization, and high quality at interfaces. In this work, we report a TiN film deposited onto sapphire substrate achieving low dielectric loss at the material interface. Through the systematic characterizations of a series of transmon qubits fabricated with identical batches of TiN base layers, but different geometries of qubit shunting capacitors with various participation ratios of the material interface, we quantitatively extract the loss tangent value at the substrate-metal interface smaller than $8.9 imes 10^{-4}$ in 1-nm disordered layer. By optimizing the interface participation ratio of the transmon qubit, we reproducibly achieve qubit lifetimes of up to 300 $μ$s and quality factors approaching 8 million. We demonstrate that TiN film on sapphire substrate is an ideal material system for high-coherence superconducting qubits. Our analyses further suggest that the interface dielectric loss around the Josephson junction part of the circuit could be the dominant limitation of lifetimes for state-of-the-art transmon qubits.

Motivation & Objective

• Motivate reducing dielectric loss in superconducting qubits to achieve longer coherence times.
• Quantitatively characterize dielectric loss at the substrate-metal (SM) interface for TiN on sapphire.
• Extract intrinsic loss tangents for SM and junction regions to identify dominant decoherence sources.
• Demonstrate reproducible, high-coherence transmon qubits by engineering SM interface participation.

Proposed method

• Fabricate TiN base layers on c-plane sapphire with 100 nm TiN and 1 nm-scale disordered SM layer inferred from TEM.
• Pattern shunting capacitors with varied geometries to span P_SM over >2 orders of magnitude and extract loss tangents.
• Characterize qubit T1 and extract Q, correcting for Purcell effects from the readout cavity.
• Model qubit loss as Q^{-1}=P_SM tan(delta_SM)+P_J tan(delta_J) and fit to data to obtain tan(delta_SM) and tan(delta_J).
• Perform TEM to estimate disordered layer thicknesses at SM (~1 nm) and MA (~5.5 nm) interfaces, and infer dielectric constants used in PR calculations.
• Use 2D interdigital and dumbbell geometries, plus 3D dumbbell qubits, to isolate interface contributions.

Experimental results

Research questions

• RQ1 What is the dielectric loss tangent at the TiN/sapphire SM interface for TiN-based qubits?
• RQ2 How does the SM interface participation P_SM affect qubit Q and T1 across designs?
• RQ3 What are the loss tangents for the Josephson junction related interfaces (tan(delta_J)) and their contribution to lifetime limits?
• RQ4 Is the TiN/sapphire material system competitive with the best low-loss platforms for high-coherence qubits?
• RQ5 What is the dominant remaining loss mechanism in state-of-the-art transmon qubits when SM loss is minimized?

Key findings

• tan(delta_SM) is less than 8.9×10^-4 for a 1-nm disordered SM layer.
• tan(delta_J) is approximately 3.5×10^-3 for the Josephson junction region.
• Quibit lifetimes up to 300 μs and Q approaching 8×10^6 were reproducibly achieved by optimizing P_SM.
• A simple SM-only loss model underpredicts Q at low P_SM, indicating an additional P_SM–independent loss (Q0) likely linked to junction-region dissipation.
• After including P_J and tan(delta_J), the data align with a two-loss-channel model, supporting the quantified loss tangents.
• TEM confirms a ~1 nm disordered layer at SM and ~5.5 nm at MA in the junction region, informing PR estimates.