Direct evidence for the absence of coupling between shear strain and superconductivity in Sr2RuO4

By directly applying three types of shear strain to Sr2RuO4 single crystals and observing negligible changes in the superconducting transition temperature, this study provides evidence that shear strain does not couple to superconductivity, supporting a one-component order parameter model while highlighting its inability to fully explain other experimental anomalies.

The Gist

Imagine a crystal of Sr₂RuO₄ (a special superconducting material) as a tiny, perfectly organized dance floor. For decades, physicists have been arguing about the "dance moves" the electrons perform when they become superconductors.

The big question was: Do the electrons dance as a solo act (one component), or do they dance as a synchronized pair (two components)?

Here is the story of how this paper settled a major part of that argument, using a creative mix of stretching, squeezing, and high-tech photography.

The Great Debate: The "Shear" Mystery

In the world of superconductors, scientists can learn a lot by poking and prodding the material.

• The Solo Theory: Some experiments suggested the electrons dance alone. If you push them one way, they shouldn't react much to a specific type of "sliding" motion (called shear strain).
• The Pair Theory: Other experiments, specifically using sound waves (ultrasound), suggested the electrons dance in pairs. If this were true, sliding the crystal layers past each other (shear strain) should act like a strong magnet, drastically changing the temperature at which the material becomes superconducting ($T_c$).

It was like hearing two different stories about a magic trick. One group said, "If you slide the stage, the magician disappears!" The other said, "Sliding the stage does nothing."

The New Experiment: The "Piezo-Push"

To solve this, the researchers built a custom machine. Imagine gluing a thin slice of the crystal onto a special ceramic tile (a piezoelectric device). When you apply electricity to this tile, it physically twists and slides, like a hand sliding a deck of cards.

1. The Setup: They glued the crystal to the tile and put it inside a super-cold fridge.
2. The Camera: Instead of guessing how much the crystal was twisting, they used a high-powered microscope and a computer program (like a digital "spot the difference" game) to watch the crystal move pixel by pixel. This let them measure the exact amount of "sliding" (shear strain) happening on the surface.
3. The Test: They applied three different types of sliding motions to the crystal while carefully measuring its superconducting temperature ($T_c$).

The Result: The "Silent" Crystal

Here is the surprising twist: The crystal didn't care.

No matter how much they slid the layers of the crystal (up to a significant amount), the temperature at which it became superconducting did not change. The change was so tiny (less than 10 thousandths of a degree) that it was effectively zero.

The Analogy:
Imagine you are trying to test if a rubber band is made of two intertwined strands or just one. You pull it sideways.

• If it were two strands, the sideways pull would make it snap or change shape immediately.
• If it is one strand, it might just wiggle a little and stay the same.

In this experiment, the "rubber band" (the superconductor) didn't budge. This strongly suggests the electrons are not dancing as a two-component pair. It points toward a one-component model.

The Plot Twist: The Mystery Remains

However, the story isn't a simple "Case Closed."

The paper admits a confusing contradiction:

• Our new test: Says "No coupling to shear strain" (Supports the one-component theory).
• Old ultrasound tests: Said "Huge coupling to shear strain" (Supports the two-component theory).

The authors point out that if the electrons were truly a one-component pair, they should explain other weird behaviors seen in the past, like the material breaking time-reversal symmetry (acting like a tiny magnet) and forming specific "domains." A simple one-component model struggles to explain those other facts.

The Conclusion

The researchers have delivered a very strong piece of evidence: Shear strain does not affect the superconducting temperature of Sr₂RuO₄.

This rules out many popular theories that claimed the electrons were dancing in a complex, two-part routine. However, because this result clashes with other famous experiments (the ultrasound ones), the full mystery of what kind of "dance" the electrons are actually doing is still unsolved. The paper suggests we need a new, more exotic explanation that fits all the clues, not just the ones about sliding.

In short: They tried to slide the crystal to see if it would change its superconducting nature. It didn't. This breaks some theories, but the full puzzle of the material's identity is still waiting to be solved.

Technical Summary

Technical Summary: Direct Evidence for the Absence of Coupling between Shear Strain and Superconductivity in Sr$_2$RuO$_4$

Problem Statement
The superconducting (SC) symmetry of Sr$_2$RuO$_4$ has remained a subject of intense debate for decades. A critical contradiction has emerged between two classes of experimental evidence. On one hand, shear-mode ultrasound experiments have observed a jump in the elastic modulus $c_{66}$ at the superconducting transition temperature ($T_c$), implying a strong coupling between shear strain and the SC order parameter (OP). This observation supports a two-component OP model (e.g., chiral or nematic states with 2D irreducible representations). On the other hand, uniaxial pressure experiments have suggested a one-component OP, and specific heat measurements under uniaxial strain have failed to observe the splitting between $T_c$ and the time-reversal symmetry breaking (TRSB) phase predicted by chiral models. Resolving the discrepancy between the ultrasound results (which suggest strong shear coupling) and uniaxial stress results (which suggest weak or no coupling) is essential for determining the nature of the superconducting state.

Methodology
To address this controversy, the authors developed a novel experimental approach to directly apply static shear strains to single crystals of Sr$_2$RuO$_4$ and measure the resulting changes in $T_c$ with high precision.

• Sample Preparation: Thin Sr$_2$RuO$_4$ crystals (approx. 30 $\mu$m thick) were rigidly glued onto the active surface of shear piezoelectric devices. The samples were selected for sharp superconducting transitions ($T_c \approx 1.5$ K) to ensure the absence of metallic-ruthenium inclusions.
• Strain Application and Quantification: Three distinct shear strain modes were applied corresponding to different symmetry channels:
  1. $\epsilon_{xy}$ (Shear along [100], $B_{2g}$ symmetry).
  2. $\epsilon_{x'y'}^{110}$ (Diagonal shear along [110], $B_{1g}$ symmetry).
  3. $\epsilon_{zx}$ (Shear along [001], $E_g$ symmetry).
     Unlike previous ultrasound experiments that used high-frequency oscillating strains, this setup applied static strains across the full dynamic range of the piezoelectric actuators.
• Strain Characterization: The magnitude of the applied strain was directly quantified using optical imaging and digital image correlation (DIC) down to 30 K. This non-contact method allowed the authors to measure in-plane displacements and calculate the effective shear strain transferred to the bulk sample, minimizing uncertainties associated with glue-mediated strain transfer.
• $T_c$ Measurement: The superconducting transition was monitored using mutual-inductance AC susceptibility measurements. The real ($\chi'$) and imaginary ($\chi''$) parts of the susceptibility were measured as a function of temperature and applied voltage. $T_c$ was determined with a resolution better than 10 mK, using the drop in $\chi'$ as the primary criterion.

Key Results

• Absence of Shear Coupling: The study found no detectable change in $T_c$ under the application of any of the three shear strain modes. The variations in $T_c$ were smaller than the experimental resolution limit of $\pm 6$ mK %$^{-1}$ for linear slopes and $\pm 60$ mK %$^{-2}$ for quadratic slopes.
• Symmetry Analysis: Consistent with $C_{4z}$ rotational symmetry, the data showed no kink (linear dependence) or quadratic trend around zero strain. This lack of response contradicts the expectation for a two-component OP, which should exhibit a kink in $T_c$ under $B_{2g}$ or $B_{1g}$ shear strains.
• Control Experiment: To validate the sensitivity of their technique, the authors performed a control experiment using a tensile piezo device. This experiment successfully detected a clear, quadratic dependence of $T_c$ on uniaxial tensile strain, confirming that the setup is capable of resolving small $T_c$ variations.
• Comparison with Thermodynamics: The authors compared their direct measurements with estimates derived from the Ehrenfest relations using jumps in elastic moduli (from ultrasound) and specific heat. The Ehrenfest relations, when applied to the ultrasound jump in $c_{66}$, predict a significant linear kink in $T_c$ (slope $\sim 70-750$ mK %$^{-1}$). The authors' direct measurement of $|\partial T_c / \partial \epsilon_{xy}| \le 6$ mK %$^{-1}$ is orders of magnitude smaller than these predictions, revealing a severe inconsistency between the ultrasound data and static strain data.

Significance and Claims
The paper claims to provide direct evidence that shear strain has little to no coupling to the superconductivity in Sr$_2$RuO$_4$. This result seriously challenges the interpretation of the SC state as a two-component order parameter, which is the primary theoretical framework supported by the ultrasound $c_{66}$ jump.

The authors conclude that their results are consistent with a one-component order parameter model. However, they explicitly note that a one-component model cannot simultaneously explain other established experimental facts, such as time-reversal symmetry breaking (TRSB), the existence of superconducting domains, and horizontal line nodes. Consequently, the paper argues that the current understanding of Sr$_2$RuO$_4$ requires alternative interpretations that can reconcile the absence of shear coupling with the presence of TRSB and other phenomena. The authors emphasize that the discrepancy between ultrasound jumps and static strain responses remains an unresolved issue that suggests the ultrasound response may contain contributions not solely due to coupling with the SC order parameter.

The study also highlights the utility of the developed optical strain quantification technique for investigating other superconductors where multi-component order parameters are suspected, such as UPt$_3$.