Large critical current density Josephson $π$ junctions with PdNi barriers

This paper reports the achievement of record-high critical current densities in Nb/Pd$_{89}$Ni$_{11}$/Nb Josephson $\pi$ junctions, demonstrating that the Pd$_{89}$Ni$_{11}$ barrier's perpendicular magnetic anisotropy enables zero-field operation and establishing it as a promising material for superconducting digital logic and qubit applications.

The Gist

Imagine you are trying to build a super-fast, ultra-efficient computer that runs on electricity without generating any heat. To do this, scientists use "superconductors"—materials that let electricity flow perfectly with zero resistance. But to make these computers think and make decisions, they need tiny switches called Josephson junctions.

Think of a Josephson junction as a bridge between two islands of superconducting electricity. Usually, this bridge is made of an insulator (like a wall) that the electricity has to "tunnel" through. However, for the next generation of quantum computers and super-fast logic circuits, scientists want to build bridges out of magnets instead.

Here is the problem: Magnets usually mess up the delicate flow of superconducting electricity. But, if you make the magnet just the right thickness, something magical happens: the electricity flowing across the bridge flips its "phase" by 180 degrees. In the world of superconductors, this is called a $\pi$-junction (Pi-junction).

Think of a normal junction as a bridge where traffic flows forward. A $\pi$-junction is like a bridge where the traffic is forced to flow backward. This "backward flow" is incredibly useful because it acts like a built-in switch. You don't need to plug in a giant magnet to force the switch to flip; the material does it automatically. This is called a "passive $\pi$-shifter."

The Challenge

For years, scientists have been trying to build these magnetic bridges using a specific alloy called PdNi (Palladium-Nickel). It's a great candidate because it naturally wants to magnetize in a specific direction (up and down, rather than side-to-side), which is perfect for these tiny chips.

However, there was a catch: The "traffic" (the critical current) flowing through these PdNi bridges was too weak. It was like trying to drive a Formula 1 car on a dirt path; the engine wasn't powerful enough to handle the job. Previous studies showed these bridges could only carry about 70 kA/cm² of current.

The Breakthrough

In this paper, the team at Texas State University and Michigan State University decided to try again, but with a new recipe and better tools. They built new bridges using Niobium (Nb) superconductors instead of the older materials, and they carefully tuned the thickness of the PdNi magnet layer.

Here is what they discovered:

1. The Sweet Spot: They found that if the magnet layer is exactly 9.4 nanometers thick (that's about 1/10,000th the width of a human hair), the bridge enters that special "backward flow" state ($\pi$-state).
2. Super Power: When they measured the current flowing through this specific bridge, it wasn't just strong; it was massive. They achieved a current density of 410 kA/cm².
   • The Analogy: If the old bridges were like a garden hose, this new bridge is a firehose. It carries nearly 6 times more current than anyone had ever seen in a PdNi-based bridge before.
3. No External Help Needed: Because the PdNi material naturally has "perpendicular magnetic anisotropy" (a fancy way of saying it has a built-in compass pointing straight up), the bridge works perfectly without needing any external magnets to "initialize" it. It just works right out of the box, even in zero magnetic fields.

Why Does This Matter?

Imagine you are building a city of super-fast computers (superconducting digital logic) or a quantum computer (which uses qubits).

• The Old Way: You had to use giant, power-hungry magnets to force the switches to work, or you had to use materials that were too weak to carry the necessary data.
• The New Way: With this new PdNi bridge, you have a switch that is:
  • Strong: It can handle huge amounts of data traffic.
  • Self-Contained: It doesn't need external magnets to flip its state.
  • Robust: It works even if the thickness of the magnet layer varies slightly (like a bridge that doesn't collapse if the concrete is a millimeter off).

The Bottom Line

The researchers have essentially built a "super-bridge" for electricity. By combining the right materials and finding the perfect thickness, they created a component that is much more powerful than anything previously made with this alloy. This brings us one step closer to building superconducting computers that are faster, smaller, and more energy-efficient than anything we have today, and it could be a game-changer for the future of quantum computing.

Technical Summary

1. Problem Statement

Josephson junctions with ferromagnetic barriers (SFS junctions) are critical components for developing energy-efficient superconducting digital logic and specific qubit architectures. A key requirement for these applications is the use of $\pi$-junctions, which act as passive phase shifters to replace applied magnetic flux biases.

However, realizing high-performance $\pi$-junctions faces three major challenges:

1. Low Critical Current Density ($J_c$): Ferromagnetic materials typically have short coherence lengths, leading to exponentially decaying supercurrents. Achieving $J_c$ values high enough to compete with insulating junctions in superconducting loops is difficult.
2. Magnetic Initialization: Many ferromagnetic materials require large in-plane magnetic fields to initialize their magnetic state and produce reproducible $\pi$-junction behavior, which is impractical for integrated circuits and qubits.
3. Damping: For qubit applications, junctions must be underdamped, which is difficult to achieve in all-metallic junctions without specific engineering.

While Palladium-Nickel (PdNi) alloys have shown promise due to intrinsic perpendicular magnetic anisotropy (PMA), previous studies (e.g., Pham et al.) reported relatively low $J_c$ values ($\approx 70$ kA/cm$^2$) in the first $\pi$-state.

2. Methodology

The authors fabricated and characterized Nb/Pd$_{89}$Ni$_{11}$/Nb Josephson junctions to address these limitations.

• Fabrication:

  • Films were deposited via DC sputtering on Si substrates with thermal oxide.
  • Structure: 60 nm Nb (bottom electrode) / varying thickness PdNi barrier / 10 nm Pd capping layer / 150 nm Nb (top electrode).
  • The PdNi barrier was created by co-sputtering separate Pd and Ni targets to achieve an 89:11 atomic ratio.
  • Junctions were patterned into 3 $\mu$m diameter circular devices using photolithography and ion milling.
  • A key fabrication step involved in-situ partial ion milling of the Pd cap before depositing the top Nb electrode to ensure a clean interface.

• Magnetic Characterization:

  • Continuous 35 nm PdNi films and patterned arrays (3.5 $\mu$m dots) were measured using Vibrating Sample Magnetometry (VSM).
  • Measurements included magnetization loops ($M$ vs. $H$) in-plane and out-of-plane at 5 K, and temperature-dependent magnetization ($M$ vs. $T$) to determine the Curie temperature.

• Electrical Transport:

  • Current-Voltage ($I-V$) characteristics were measured in a Physical Property Measurement System (PPMS).
  • Critical current ($I_c$) was measured as a function of applied magnetic field (Fraunhofer pattern) and temperature.
  • Measurements were performed at 5.75 K to mitigate Joule heating from high currents, with temperature dependence extrapolated to lower temperatures.

3. Key Contributions

• Record High $J_c(\pi)$: The study reports a critical current density of 410 kA/cm$^2$ at 4.2 K for a 9.4 nm PdNi barrier. This significantly exceeds previous PdNi-based records (approx. 70 kA/cm$^2$) and approaches values typically found only in elemental ferromagnet (Ni) junctions.
• Zero-Field Operation: The authors demonstrate that the intrinsic Perpendicular Magnetic Anisotropy (PMA) of PdNi allows for robust $\pi$-junction operation without the need for external magnetic initialization fields.
• Interface Transparency Analysis: By comparing their results with NbN-based junctions, the authors attribute the performance boost to the higher interface transparency between Nb and PdNi compared to NbN and PdNi, combined with the longer superconducting coherence length of Nb.
• Scalability Confirmation: They confirmed that the magnetic properties (PMA and coercivity) of PdNi are preserved when patterned down to the micrometer scale (3.5 $\mu$m), validating the material for device integration.

4. Key Results

• Magnetic Properties:

  • Pd$_{89}$Ni$_{11}$ exhibits strong Perpendicular Magnetic Anisotropy (PMA), evidenced by square hysteresis loops in the out-of-plane direction and negligible in-plane hysteresis.
  • Saturation magnetization ($M_s$) is 145 $\pm$ 15 emu/cm$^3$.
  • The Curie temperature ($T_C$) is approximately 170 K.
  • Patterning the film into 3.5 $\mu$m dots did not alter the magnetic response, confirming no significant dipolar coupling or shape anisotropy effects at this scale.

• Transport Properties:

  • Fraunhofer Patterns: The junctions displayed clear Fraunhofer interference patterns, confirming the superconducting nature of the junctions. The patterns were centered at zero field, indicating no trapped flux or need for magnetic initialization.
  • 0-$\pi$ Transition: The critical current density oscillated with barrier thickness, confirming the transition from the 0-state to the $\pi$-state.
  • Peak Performance: For a 9.4 nm PdNi barrier (near the first $\pi$-state peak), the device achieved:
    • $J_c(\pi) = 410$ kA/cm$^2$ at 4.2 K.
    • $J_c > 550$ kA/cm$^2$ at 2 K (zero field).
  • Comparison: These values are roughly 6 times higher than previous PdNi studies (Pham et al.) and significantly higher than extrapolated values from Khaire et al.

• Theoretical Fit:

  • The thickness dependence of $J_c$ was fitted to the standard decay/oscillation model (Eq. 5). The data aligned remarkably well with fitting parameters derived from much thicker (30–100 nm) PdNi junctions by Khaire et al., suggesting the physical mechanisms are consistent across thickness scales.

5. Significance

This work establishes Pd$_{89}$Ni$_{11}$ as a superior barrier material for superconducting digital logic and qubit architectures for the following reasons:

1. High Performance: The combination of high $J_c$ and the ability to operate at zero magnetic field solves the primary bottleneck for using $\pi$-junctions as passive phase shifters.
2. Robustness: The intrinsic PMA ensures reproducible operation without complex magnetic initialization protocols.
3. Tolerance: The relatively long ferromagnetic coherence length in PdNi creates a broader thickness window for the $\pi$-state, making the devices more tolerant to nanometer-scale fabrication variations compared to elemental ferromagnets.
4. Application Potential: These junctions are ideal candidates for Superconducting Digital Logic (e.g., RSFQ circuits) and Flux Qubits, where eliminating external bias fields and maximizing critical currents are essential for scalability and energy efficiency.

In conclusion, the authors have successfully engineered Nb/PdNi/Nb junctions that overcome the historical trade-off between critical current density and magnetic initialization requirements, paving the way for practical superconducting circuits utilizing $\pi$-junctions.