At 4.2 Kelvin, the bulk superconducting state of uranium ditelluride ($UTe_2$) is completely dead. Its electrical resistance is fully restored, its Meissner screening currents have collapsed to zero, and its macroscopic phase coherence has evaporated. Yet, when researchers at the University of Illinois Urbana-Champaign lowered the probe of a cryogenic vector-field scanning tunneling microscope onto the cleaved $(01\bar{1})$ surface of the crystal, they detected the unmistakable spatial oscillations of Cooper pairing lingering in the electronic density of states.
The quantitative results, formally documented by Zhen Zhu, Vidya Madhavan, Eduardo Fradkin, and their collaborators in the Proceedings of the National Academy of Sciences (PNAS), provide the first direct experimental verification of a 20-year-old theoretical prediction: superconducting electron pairs can sustain an ordered, spatially modulated array known as a pair density wave (PDW) at temperatures more than 100 percent above the material's bulk critical transition threshold ($T_c \approx 2.1\text{ K}$).
Fradkin, an Illinois Grainger Engineering physics professor and project co-lead who helped formulate the original 2007 theory of pair density waves, described the manifestation through Lewis Carroll’s famous literary metaphor: "Pair density waves are the Cheshire Cat's grin of superconductivity. They are the vestige that remains once the phase itself has disappeared. In conventional superconductors, Cooper pairs form when the full phase transition occurs, but, in this system, their observation in PDWs above the transition point shows that they are formed beforehand in a different state."
This superconductivity physics discovery provides concrete numerical proof that the fundamental mechanism of electron pairing can completely decouple from the macroscopic onset of zero resistance. Across dozens of spectroscopic datasets acquired at local tunneling setpoints of $V_{\text{bias}} = -20\text{ millivolts}$ and currents of $I = 150\text{ picoamperes}$, the team measured modulated electronic patterns that defied standard Bardeen-Cooper-Schrieffer (BCS) expectations. More crucially, when subjected to directional vector magnetic fields, these vestigial patterns melted along crystallographic orientations at thresholds strictly proportional to the material's highly anisotropic superconducting critical fields ($H_{c2}$), proving that the charge structures are direct descendants of an underlying, lingering pair field.
Decoupling Pair Formation from Macroscopic Coherence
The standard model of low-temperature superconductivity, established by John Bardeen, Leon Cooper, and J. Robert Schrieffer in 1957, operates on a rigid relationship between pairing energy and critical temperature. In a weak-coupling BCS framework, the superconducting energy gap $\Delta_0$ at absolute zero relates to the transition temperature by the universal ratio:
$$\frac{2\Delta_0}{k_B T_c} \approx 3.53$$
In this conventional framework, electron pairing and macroscopic phase coherence occur simultaneously at $T = T_c$. Above $T_c$, thermal fluctuations break the weak phonon-mediated electron bonds, causing both the superconducting gap $\Delta_0$ and the Cooper pair condensate to vanish immediately. The normal state returns as a Fermi liquid of uncorrelated, single-electron quasiparticles.
The measurements on uranium ditelluride break this paradigm down into two distinct thermal regimes. In $UTe_2$, bulk superconductivity sets in at a transition temperature between $T_c = 1.6\text{ K}$ and $2.1\text{ K}$, depending on sample purity and synthesis methods. Below this threshold, electrical resistivity drops below the instrument detection floor ($< 10^{-10}\ \Omega\cdot\text{cm}$), confirming macroscopic dissipationless transport.
However, local density of states (LDOS) mapping conducted at $4.2\text{ K}$—a temperature regime easily reached using standard liquid helium cooling at ambient atmospheric pressure (101.3 kPa)—demonstrates that electron correlations remain locked into periodic spatial modulations. The Fourier transforms of these spatial LDOS maps reveal two distinct families of wavevectors that dictate how the electronic charge organizes itself on the atomic lattice:
[Wavevector Spectrum Across Critical Transitions]
T < Tc (0.3 K to 2.1 K):
Bulk Superconductivity ON ---> qi peaks (Incommensurate CDW) + pi peaks (Induced CDW)
[Macroscopic Phase Coherence + Spatially Modulated Cooper Pairing]
Tc < T < 4.2 K:
Bulk Superconductivity OFF ---> pi peaks vanish; qi peaks PERSIST
[Zero Resistance Lost, but Local Cooper Pair Arrays Remain Intact]The scanning tunneling microscope tracked these wavevectors across a continuous thermal sweep from $300\text{ millikelvin}$ to well above $4.2\text{ Kelvin}$. The first group of peaks, designated by the wavevectors $\mathbf{p}_i$ ($p_1, p_2, p_3$), alongside higher-order harmonics $h_1$ and $h_2$, showed an abrupt collapse as the sample warmed past $T_c \approx 2.1\text{ K}$. At $T \ge 2.1\text{ K}$, the spectral weight of the $\mathbf{p}_i$ peaks dropped below background noise levels.
In contrast, the second group of modulations, designated as the $\mathbf{q}_i$ ($q_1, q_2, q_3$) incommensurate charge density wave (CDW) peaks, showed no thermal discontinuity at $T_c$. The $\mathbf{q}_i$ peaks retained their spatial periodicities and sharp Fourier profiles up through $4.2\text{ K}$, demonstrating that a robust, organized electronic state survives long after zero-resistance transport ceases.
This persistent modulation constitutes the physical Cheshire Cat grin. The cat—the bulk macroscopic phase coherence enabling zero electrical resistance—vanishes at $2.1\text{ K}$. But the grin—the periodic ordering of paired electrons manifested as charge density modulations—lingers at $4.2\text{ K}$.
The Ginzburg-Landau Free Energy Mechanism
To understand why a charge modulation can serve as the fingerprint of a lingering pair state, one must examine the mathematical coupling between superconductivity and charge density waves within a Ginzburg-Landau phenomenological framework.
A pair density wave differs fundamentally from an ordinary superconductor because its pairing order parameter possesses a non-zero center-of-mass momentum $\mathbf{P} = \hbar \mathbf{Q}$. While an ordinary BCS superconductor is characterized by a spatially uniform complex scalar order parameter $\Delta_0(\mathbf{r}) = \Delta_0 e^{i\phi}$, a pair density wave is spatially modulated:
$$\Delta_{\text{PDW}}(\mathbf{r}) = \sum_i \left[ \Delta_{\mathbf{p}_i} e^{i \mathbf{p}_i \cdot \mathbf{r}} + \Delta_{-\mathbf{p}_i} e^{-i \mathbf{p}_i \cdot \mathbf{r}} \right]$$
Because Cooper pairs carry electric charge ($-2e$), any spatial modulation of the pairing amplitude induces a corresponding modulation in the underlying charge density $\rho(\mathbf{r})$. The Ginzburg-Landau free energy density $\mathcal{F}$ contains symmetry-allowed interaction terms linking the uniform superconducting order parameter $\Delta_0$, the pair density wave order parameter $\Delta_{\mathbf{p}}$, and the electronic charge density wave modulations $\rho_{\mathbf{k}}$:
$$\mathcal{F}_{\text{coupling}} = \gamma \sum_i \left[ \Delta_0^ \Delta_{\mathbf{p}_i} \rho_{\mathbf{p}_i}^ + \text{c.c.} \right] + \lambda \sum_{i \neq j} \left[ \Delta_{\mathbf{p}_i} \Delta_{\mathbf{p}_j}^ \rho_{\mathbf{q}_k}^ + \text{c.c.} \right]$$
where $\mathbf{q}_k = \mathbf{p}_i - \mathbf{p}_j$, and $\gamma$ and $\lambda$ represent coupling coefficients dictated by the point-group symmetry of the crystal.
+--------------------------------------------------------+
| Parent Pair Density Wave (PDW) Order Parameter |
| Delta_{p_i} (Spatially Modulated) |
+---------------------------+----------------------------+
|
+-----------------------+-----------------------+
| |
Couples via: Trilinear Term Couples via: Quadratic Term
gamma * Delta_0^* * Delta_{p_i} * rho_{p_i}^* lambda * Delta_{p_i} * Delta_{p_j}^* * rho_{q_k}^*
| |
v v
+-------------------------------+ +-------------------------------+
| pi Charge Modulations | | qi Charge Modulations |
| (Dependent on Delta_0) | | (Independent of Delta_0) |
| Collapses at Tc = 2.1 K | | SURVIVES to T = 4.2 K |
+-------------------------------+ +-------------------------------+The mathematical structure of these two terms explains the experimental data:
- The Trilinear Coupling ($\gamma$-term): The generation of charge density waves at wavevectors $\mathbf{p}_i$ requires the simultaneous presence of both the uniform superconducting order parameter $\Delta_0$ and the PDW order parameter $\Delta_{\mathbf{p}_i}$. When the temperature exceeds $T_c \approx 2.1\text{ K}$, the uniform order parameter vanishes ($\Delta_0 \to 0$). As a direct mathematical consequence, the effective driving field for $\rho_{\mathbf{p}_i}$ drops to zero. This is why the $\mathbf{p}_i$ peaks disappear immediately at $T_c$.
- The Quadratic Intermodulation ($\lambda$-term): The generation of charge density waves at wavevectors $\mathbf{q}_k$ does not depend on $\Delta_0$. Instead, it arises as a second-harmonic composite order parameter generated directly by the interference of two PDW components: $\rho_{\mathbf{q}_k} \propto \Delta_{\mathbf{p}_i} \Delta_{\mathbf{p}_j}^$. This composite CDW order does not require macroscopic phase coherence; it requires only that the local pairing amplitude $|\Delta_{\mathbf{p}}|$ remains non-zero.
Because $\rho_{\mathbf{q}_k}$ scales quadratically with the PDW amplitude, it persists into the normal state as long as the parent pair density wave retains finite amplitude. The observation of the $\mathbf{q}_i$ peaks at $4.2\text{ K}$ confirms that the parent order parameter $\Delta_{\mathbf{p}}$ remains active well above $T_c$.
Anisotropic Vector Field Suppression
In standard materials physics, charge density waves are composed of static charge reorganizations that exhibit minimal coupling to external magnetic fields. Because a CDW carries no net electrical current in its ground state, applying a magnetic field of several teslas typically produces negligible changes in its modulation amplitude.
"A charge density wave is just a collective electronic state that is modulated in space, so there is no reason for them to respond to magnetic fields, let alone be destroyed by them," noted project co-lead Vidya Madhavan. "We took our data to Professor Fradkin and his students, and we predicted that this could happen if there is also a pair density wave in the system. There aren't many possible explanations for something like this, and pair density waves are the best one I can think of."
To test whether the lingering $\mathbf{q}_i$ peaks were truly tied to a vestigial superconducting pair field, the UIUC experimental team applied magnetic fields using a 3D vector magnet while holding the sample at $4.2\text{ K}$. This instrument permitted the precise rotation of magnetic field vectors $\mathbf{B} = (B_a, B_b, B_c)$ along arbitrary crystallographic axes.
The data revealed that the $\mathbf{q}_i$ charge modulations at $4.2\text{ K}$ were progressively suppressed and eventually extinguished by external magnetic fields. Critically, the field thresholds required to destroy the modulations ($H_{\text{sup}}$) exhibited a stark directional anisotropy that mirrored the material's bulk superconducting upper critical field hierarchy ($H_{c2}$):
$$H_{\text{sup}}^{b^} > H_{\text{sup}}^{c^} > H_{\text{sup}}^a$$
The quantitative suppression thresholds measured by the vector-field STM at $4.2\text{ K}$ are detailed below:
| Crystallographic Axis | Suppression Field Range at 4.2 K ($H_{\text{sup}}$) | Bulk Upper Critical Field Limit ($H_{c2}$ at $T \to 0$) | Anisotropy Ratio Relative to $a$-axis ($H / H^a$) |
|---|---|---|---|
| $a$-axis (Easy magnetic axis) | $0.4\text{ T} < H_{\text{sup}}^a < 0.8\text{ T}$ | $\approx 6.0\text{–}7.0\text{ T}$ | $1.0\times$ (Baseline) |
| $c^$-axis* (Hard magnetic axis) | $1.5\text{ T} < H_{\text{sup}}^{c^} < 3.0\text{ T}$ | $\approx 11.0\text{ T}$ | $\approx 3.75\times$ |
| $b^$-axis (Intermediate axis) | $4.0\text{ T} < H_{\text{sup}}^{b^} < 8.0\text{ T}$ | $\approx 35.0\text{ T}$ (Exceeds Pauli limit) | $\approx 10.0\times$ |
At $B_a = 0.4\text{ T}$, the $\mathbf{q}_i$ peaks remained detectable, but by $B_a = 0.8\text{ T}$, peaks $q_1$ and $q_2$ were completely eradicated, leaving only a faint trace of $q_3$. Along the $c^$-axis, the modulations resisted destruction up to $1.5\text{ T}$ before vanishing entirely at $3.0\text{ T}$. Along the $b^$-axis, the charge peaks survived beyond $4.0\text{ T}$, requiring fields approaching $8.0\text{ T}$ to be fully quenched.
This exact $10:3.75:1$ directional ratio matches the directional profile of bulk upper critical fields observed in zero-temperature transport studies ($35\text{ T} : 11\text{ T} : 7\text{ T}$). If the $\mathbf{q}_i$ modulation were an ordinary lattice charge density wave driven by Fermi surface nesting, its response to magnetic fields would be roughly isotropic and negligible at sub-tesla levels. The fact that its destruction directly reflects the anisotropic superconducting upper critical field of $UTe_2$ provides quantitative proof that the charge modulation is driven by a underlying pair density wave.
The vector STM measurements confirmed that magnetic vortices formed within the vestigial pair field introduce topological phase dislocations into the intertwined charge density wave. As the magnetic field increases, the density of these vortex-induced dislocations scales linearly with magnetic flux density:
$$n_v = \frac{B}{\Phi_0}$$
where $\Phi_0 = h / (2e) \approx 2.067 \times 10^{-15}\text{ Wb}$ is the superconducting magnetic flux quantum. As these dislocations multiply, they disrupt the phase coherence of the CDW, melting the spatial pattern into an uncorrelated normal state. The persistence of this field-melting mechanism at $4.2\text{ K}$ establishes that the electrons remain paired into Cooper states that are selectively destroyed by magnetic flux lines, even when zero resistance is absent.
Crystallographic Architecture and Heavy-Fermion Mechanics
The physical setting for this behavior is uranium ditelluride's crystallographic and electronic structure. The compound crystallizes in an orthorhombic, body-centered lattice belonging to the space group $Immm$ (No. 71).
The lattice constants measured by X-ray diffraction establish an anisotropic unit cell:
- $a = 4.16\text{ Å}$
- $b = 6.12\text{ Å}$
- $c = 13.96\text{ Å}$
Inside this lattice, uranium atoms assemble into linear chains running parallel to the crystallographic $a$-axis. These uranium chains are arranged in dimers aligned along the $c$-axis, with each uranium atom coordinated by two crystallographically inequivalent tellurium sites, denoted as $\text{Te}1$ and $\text{Te}2$.
[Orthorhombic Immm Structure of Uranium Ditelluride]
c = 13.96 Å
^
| U---U U---U <-- Uranium Dimers along c-axis
| | | | |
| Te1 Te2 Te1 Te2 <-- Inequivalent Tellurium Sites
| | | | |
| U---U U---U
+-----------------------------> a = 4.16 Å (Linear Chains along Easy Axis)
/
/ b = 6.12 Å (Cleavage Plane (01-1))
vCleaving the crystal exposes the $(01\bar{1})$ crystallographic plane, which presents quasi-one-dimensional chains of tellurium atoms along the $a$-axis. This low-symmetry structural arrangement generates an extreme electronic anisotropy. The uranium $5f$ electrons do not behave as localized magnetic moments; instead, they hybridize strongly with the conduction electrons through the Kondo effect below a characteristic coherence temperature $T_K \approx 50\text{ K}$.
This hybridization produces heavy-fermion quasiparticles characterized by an enormous effective mass:
$$m^* \approx 30\text{ to }40\ m_e$$
where $m_e$ is the bare electron rest mass. This high effective mass compresses the Fermi energy, flattening the electronic bands and amplifying electron-electron correlation energies relative to their kinetic energy.
Because of these correlations, $UTe_2$ does not form standard spin-singlet Cooper pairs, where electron spins point in opposite directions ($S = 0$, antiparallel). Instead, it forms spin-triplet Cooper pairs, where electron spins point in the same direction ($S = 1$, parallel). In a spin-triplet superconductor, the pairing wavefunction must have odd spatial parity to satisfy the Pauli exclusion principle: $\Delta(-\mathbf{k}) = -\Delta(\mathbf{k})$.
This triplet pairing is formally described by a three-dimensional complex vector $\mathbf{d}(\mathbf{k})$:
$$\hat{\Delta}(\mathbf{k}) = i (\mathbf{d}(\mathbf{k}) \cdot \boldsymbol{\sigma}) \sigma_y$$
where $\boldsymbol{\sigma} = (\sigma_x, \sigma_y, \sigma_z)$ are the Pauli spin matrices. Unlike singlet pairs, triplet pairs carry an intrinsic magnetic moment. Consequently, an applied magnetic field cannot simply break the pair via Zeeman energy by flipping one electron’s spin, because the pair's spin configuration can align with the external field vector without destroying the pairing state.
This mechanism explains why $UTe_2$ shatters the conventional Chandrasekhar-Clogston Pauli paramagnetic limit. For a conventional superconductor, the maximum magnetic field that a singlet Cooper pair can survive before the Zeeman energy exceeds the superconducting condensation energy is:
$$H_P = \frac{\Delta_0}{\sqrt{2} \mu_B} \approx 1.86\ T_c\ [\text{Tesla/Kelvin}]$$
For $UTe_2$ with $T_c \approx 1.6\text{–}2.1\text{ K}$, the theoretical Pauli limit is roughly $H_P \approx 3.0\text{–}3.9\text{ T}$. Yet along the $b$-axis, the bulk upper critical field $H_{c2}^b$ exceeds $35\text{ Tesla}$ at ambient pressure—more than 900 percent higher than the Pauli limit. When subjected to ultra-high pulsed magnetic fields, the material even develops a secondary, field-induced re-entrant superconducting phase between $40\text{ T}$ and $65\text{ T}$.
This resilience against magnetic destruction is the hallmark of spin-triplet superconductivity, an elusive state of matter shared in nature only by superfluid Helium-3 ($^3\text{He}$). The recent work confirms that this triplet pairing persists as a modulated spatial wave above the critical temperature, marking a milestone in this superconductivity physics discovery.
Metallurgy and the Reduction of Defect Scattering
The ability to detect these spatial modulations depended directly on crystal growth metallurgy. In early experimental investigations following the 2019 discovery of superconductivity in $UTe_2$, single crystals were synthesized primarily via chemical vapor transport (CVT) using iodine or tellurium tetrachloride as transport agents.
While CVT growth produced superconducting crystals, the samples suffered from high concentrations of uranium vacancies, native tellurium site substitutions, and local lattice strain. In scanning tunneling microscopy, these atomic defects create long-range Friedel oscillations—ripples in the electronic density caused by electrons scattering off impurities.
In Fourier-space maps, this defect-induced scattering manifests as a broad, diffuse halo of noise that obscures weak, incommensurate charge modulations. As Madhavan described: "We couldn't see pair density waves in our earlier data because of material impurities that obscured our data. It would have been like trying to spot a light in a cloud of fog."
[Synthesis and Transport Comparison]
Parameter Chemical Vapor Transport (CVT) Molten Salt Flux Growth
----------------------------------------------------------------------------------------
Transition Temp (Tc) 1.60 K 2.10 K
Residual Resistivity Ratio 30 to 50 > 900
Residual Scattering Rate (Gamma) High (Halos in STM) < 0.05 meV (Atomic Clarity)
Defect-Induced Decoherence Severe Suppressed
PDW Visibility at 4.2 K Undetectable Directly ResolvedTo clear this fog, the team utilized single crystals grown using a molten salt flux technique. High-purity uranium and tellurium elements were dissolved in a eutectic molten flux mixture (such as $\text{NaCl-KCl}$) inside sealed quartz ampoules evacuated to $10^{-6}\text{ Torr}$, then cooled at controlled rates below $0.5\text{ K/hour}$.
The resulting flux-grown crystals yielded a leap in electronic quality:
- Transition Temperature ($T_c$): Increased from $1.60\text{ K}$ up to $2.10\text{ K}$, indicating reduced pair-breaking defect density.
- Residual Resistivity Ratio (RRR): Calculated as the ratio between room-temperature resistivity and residual low-temperature normal-state resistivity:
$$\text{RRR} = \frac{\rho(300\text{ K})}{\rho_0}$$
The RRR jumped from earlier values of $30\text{–}50$ in CVT crystals to greater than $900$ in flux-grown crystals.
- Quasiparticle Mean Free Path ($\ell$): Expanded beyond $100\text{ nanometers}$, exceeding the superconducting coherence length ($\xi_0 \approx 5\text{–}10\text{ nm}$) by more than an order of magnitude, firmly placing the material in the clean limit ($\ell \gg \xi_0$).
- Quasiparticle Scattering Rate ($\Gamma$): Reduced below $0.05\text{ meV}$, suppressing the diffuse background in scanning tunneling spectroscopy and enabling Fourier resolution of the subtle $\mathbf{p}_i$ and $\mathbf{q}_i$ wavevectors.
By eliminating native point defects, the intrinsic electronic order parameters could organize across thousands of unit cells without being pinned or destroyed by local disorder. This level of crystal purity allowed the vector-field STM to isolate the distinct temperature dependences of the $\mathbf{p}_i$ and $\mathbf{q}_i$ modulations.
The Landscape of Intertwined Quantum Orders
The observation that a pair density wave outlives uniform superconductivity alters how condensed matter physicists model intertwined electronic orders. For decades, competing phases in quantum materials were viewed as mutually antagonistic. In this traditional perspective, charge density waves and superconductivity compete for the same electronic states at the Fermi level: if a CDW opens an energy gap over a patch of the Fermi surface, it leaves fewer states available to form Cooper pairs, thereby depressing the superconducting transition temperature.
The data from $UTe_2$ demonstrates that rather than acting as adversaries, the modulated charge order and the pair density wave are mutually intertwined manifestations of the same underlying quantum interaction.
[Parent Quantum State]
Strongly Correlated Heavy-Fermion Fluid
|
+-----------------------+-----------------------+
| |
v v
[Uniform Superconductivity] [Pair Density Wave (PDW)]
Delta_0 (k = 0 Cooper Pairs) Delta_p (k = p_i Cooper Pairs)
| |
| |
+-----------------------+-----------------------+
|
v
[Intertwined Descendants]
pi and qi Charge Density WavesIn the traditional BCS model of superconductivity, pairing is an all-or-nothing phenomenon: electrons are either uncorrelated normal quasiparticles or condensed Cooper pairs contributing to macroscopic zero-resistance flow. The $UTe_2$ data establishes that electron pairing can occur in organized sub-phases that precede macroscopic coherence.
This distinction is central to understanding unconventional superconductors, including high-temperature copper oxides (cuprates) and iron pnictides. In cuprates, physicists have debated the nature of the "pseudogap" phase—a state existing above $T_c$ (up to $150\text{–}200\text{ K}$) characterized by a partial depletion of low-energy electronic states. One school of thought attributes the pseudogap to pre-formed Cooper pairs that lack phase coherence; another attributes it to distinct competing orders, such as charge or spin density waves.
The direct confirmation of a vestigial pair density wave persisting at $4.2\text{ K}$ in $UTe_2$ proves that a parent pairing state can directly generate charge density waves while remaining hidden from macroscopic transport measurements. The $UTe_2$ system provides a clean, mathematically tractable platform where this hierarchy can be quantitatively tested, because the magnetic field selectively suppresses the daughter charge orders at fields matching the triplet superconducting critical thresholds.
Implications for Topological Quantum Computing Architectures
Beyond fundamental theory, spin-triplet superconductors with pair density wave order are primary candidates for building fault-tolerant topological quantum computers.
In a standard quantum computer, qubits constructed from superconducting loops (such as transmons) or semiconductor quantum dots are susceptible to environmental noise. Stray magnetic fields, thermal phonons, and charge fluctuations cause quantum states to decohere within tens to hundreds of microseconds, necessitating massive error-correction overheads where thousands of physical qubits are required to maintain a single logical qubit.
Topological quantum computing avoids local decoherence by storing quantum information non-locally in non-Abelian anyons, specifically Majorana zero modes (MZMs). Majorana modes are zero-energy quasiparticles that are their own antiparticles:
$$\gamma_i = \gamma_i^\dagger$$
Because these modes are bound to the topological defects (such as vortex cores or surface edges) of spin-triplet superconductors, quantum information encoded in a pair of spatially separated Majoranas ($\gamma_1, \gamma_2$) cannot be destroyed by local noise sources. Braid operations on these modes perform unitary quantum logic gates whose outcomes depend solely on the topological path of the exchange, not on local timing or pulse shape.
Uranium ditelluride is considered the leading three-dimensional candidate for realizing intrinsic topological superconductivity. However, designing devices has been constrained by the requirement to operate deep inside dilution refrigerators at temperatures below $100\text{ millikelvin}$.
The discovery that Cooper pairs organize into a robust pair density wave at $4.2\text{ Kelvin}$ provides practical guidance for future quantum engineering:
[Thermal and Operating Limits]
Parameter Dilution Refrigerator Regime Vestigial PDW Regime
-------------------------------------------------------------------------------------
Operating Temperature 10 to 50 mK 4.2 K (Liquid He)
Cooling Power Available ~ 10 to 20 microwatts ~ 1.0 to 1.5 Watts
Cryogenic Architecture 3-Stage He-3/He-4 Dilution Closed-Cycle Pulse Tube
Thermal Budget Headroom Extremely Constrained 100,000x Higher
Coherence Protection Base Uniform Gap Only Intertwined Triplet Matrix- Thermal Load Budgets: Modern dilution refrigerators provide cooling powers on the order of $10\text{ to }20\ \mu\text{W}$ at their base temperatures ($10\text{–}20\text{ mK}$). This limits the amount of coaxial wiring, cryogenic amplifiers, and microwave control lines that can interface with the processor. In contrast, at $4.2\text{ K}$, modern closed-cycle cryocoolers deliver between $1.0\text{ and }1.5\text{ Watts}$ of cooling power—a five-order-of-magnitude increase in thermal capacity. Demonstrating that Cooper pairing correlations are thermodynamically stable at $4.2\text{ K}$ shows that the underlying quantum pairing mechanism possesses significantly higher intrinsic energy scales than previously assumed.
- Topological Protection in Modulated States: The presence of an intertwined pair density wave introduces a spatial modulation into the superconducting order parameter. Theorists have predicted that the domain walls and phase discommensurations of a PDW can harbor isolated Majorana zero modes even in the absence of external magnetic fields. The ability to control these modulations via vector magnetic fields opens the door to creating reconfigurable pathways for braiding topological states on the crystal surface.
As experimental methods advance, understanding how these lingering Cooper pairs interact with localized edge states will be critical for assessing their utility in topological devices.
The Next Quantitative Frontiers
While the vector STM measurements in Illinois have confirmed the existence of this vestigial phase, several experimental milestones remain unresolved. The primary question is whether the lingering pair density wave at $4.2\text{ K}$ is an intrinsic bulk phenomenon or a surface-enhanced state.
Scanning tunneling microscopy is an inherently surface-sensitive probe. The tunneling current extracted by the STM tip decays exponentially with tip-sample separation $z$:
$$I(z) \propto e^{-2\kappa z}, \quad \text{where } \kappa = \frac{\sqrt{2m\Phi}}{\hbar}$$
with $\Phi$ representing the local work function. Consequently, more than 95 percent of the tunneling current originates from the outermost atomic layer of the $(01\bar{1})$ cleavage plane.
Because the crystal termination breaks inversion symmetry perpendicular to the surface, the electronic structure and electron-phonon coupling on the surface differ quantitatively from the three-dimensional bulk. It remains possible that the surface environment stabilizes the pair density wave to higher temperatures ($4.2\text{ K}$) than the interior of the crystal can support.
[Experimental Verification Roadmap]
Phase 1 (Completed):
Surface-Sensitive Vector STM ---> Resolved qi peaks at 4.2 K matching Hc2 anisotropy.
Phase 2 (In Progress):
Resonant Elastic X-ray Scattering (REXS) ---> Probing charge modulations in bulk at 4.2 K.
Phase 3 (Upcoming):
Pulsed-Field Vector Ultrasound ---> Measuring elastic moduli across the 40 T - 65 T
re-entrant superconducting pocket.To determine whether the Cheshire Cat grin extends throughout the bulk crystal, several complementary experimental initiatives are currently underway across multiple international facilities:
- Resonant Elastic X-ray Scattering (REXS): Experiments scheduled at synchrotron radiation sources, including the Advanced Photon Source (APS) and Diamond Light Source, aim to probe the $\mathbf{q}_i$ wavevectors using hard and soft X-rays. Because X-rays penetrate hundreds of nanometers into the sample, REXS will directly measure whether the charge modulation persists at $4.2\text{ K}$ in the bulk interior or remains confined to the top atomic monolayer.
- High-Field Vector Magnetometry up to 35 Tesla: Researchers are deploying torque magnetometry and high-frequency contactless conductivity measurements in steady-state hybrid magnets to track the exact evolution of the PDW amplitude as fields approach the bulk $H_{c2}^b$ limit of $35\text{ Tesla}$.
- Thermal Transport Spectroscopy: Measuring the thermal conductivity $\kappa(T)$ and the Righi-Leduc effect (thermal Hall effect) above $T_c$ will indicate whether the pre-formed Cooper pairs can carry heat or contribute to entropy transport in the absence of electrical charge currents.
Tracking these parameters will reveal how Cooper pairs form, interact, and survive when macroscopic phase coherence is stripped away. The detection of the Cheshire Cat grin in uranium ditelluride provides concrete mathematical and physical proof that the architecture of superconductivity begins well before zero resistance appears—and lingers long after it has vanished.
Reference:
- https://xenospectrum.com/en/ute2-pair-density-wave-above-transition/
- https://mrl.illinois.edu/news/cooper-pairs-above-superconducting-critical-temperature-pair-density-waves
- https://arxiv.org/pdf/2402.06850
- https://pubmed.ncbi.nlm.nih.gov/42455673/
- https://www.pnas.org/doi/10.1073/pnas.2602117123
- https://magazine.mindplex.ai/post/physicists-find-cheshire-cat-grin-of-superconductivity-in-uranium-ditelluride
- https://www.sciencedaily.com/releases/2026/10/261005071517.htm
- https://xenospectrum.com/en/ute2-pair-density-wave-above-transition/
- https://arxiv.org/pdf/1905.05181
- https://www.pnas.org/doi/10.1073/pnas.2602117123
- https://arxiv.org/abs/2603.08688
- https://mrl.illinois.edu/news/cooper-pairs-above-superconducting-critical-temperature-pair-density-waves
- https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=926908
- https://www.lt30.es/wp-content/uploads/2025/08/Book-of-abstract-LT-30-1.pdf
- https://www.facebook.com/LifeboatFoundation/posts/physicists-with-the-university-of-illinois-urbana-champaigns-grainger-college-of/1574579224707654/
- https://jzis.org/paper/arxiv%3A2207.09491
