In this paper
AbstractScanning Josephson tunneling on monolayer Fe(Te,Se)/STOHalf lattice periodic modulation of the Josephson tunneling conductanceImaging the Josephson tunneling conductance map in 1-UC Fe(Te,Se)/STODiscussionReferences and NotesAcknowledgmentsSupplementary MaterialsMaterials and MethodsSupplementary TextQuasiparticle interference (QPI) experimentsSymmetry analyses on SJTM resultsDiscussions on irreducible pairing functions based on the non-symmorphic space group of the crystal structure with two-Fe UCModel calculations on the SJTM pattern of the nodeless -wave pairing stateModel calculations on the SJTM pattern of the assumed extended pairing stateJosephson interference and phase-sensitivity in SJTM experimentsSJTM results on Pb/Si(111) and characterization of the Pb tipSJTM results on monolayer FeSe/STOSJTM results on bulk Fe(Te,Se)SJTM results on Bi₂Sr₂CaCu₂O₈₊ₓComparison with previous -axis planar Josephson junction experimentsSuppressed Josephson tunneling signal around the defectAbsence of half lattice periodic modulations in other tunneling processesDiscussion on higher harmonic origin of the modulationPossible origins of the finite values on “zero lines”ReferencesWorksPhysics and Astronomy
Direct real-space imaging of sign-reversal superconducting state by scanning Josephson tunneling
Abstract
Phase-sensitive experiments based on Josephson tunneling have been considered as a pivotal method to determine the pairing symmetry in high-temperature (high-Tc) cuprate superconductors. However, in iron-based superconductors, traditional phase-sensitive experiments face setbacks because of the complexity of pairing symmetry in multiband systems. Here, using scanning Josephson tunneling microscopy (SJTM), we report the half-unit-cell periodic Josephson tunneling modulations in monolayer high-Tc Fe(Te,Se) film epitaxially grown on SrTiO3(001). The Josephson tunneling almost vanishes along the lines connecting next-nearest-neighbor Fe atoms under an s-wave superconducting tip. Combined with single-particle tunneling spectra showing the full superconducting gap, our experimental results are consistent with the sign-reversal nodeless d-wave pairing symmetry. Our findings provide sub-unit-cell scale superconducting phase information for understanding the mechanism of high-Tc superconductivity. We anticipate our work as a starting point for atomically resolved Josephson interference experiments using SJTM in various unconventional superconductors.
Main Text:
The determination of the superconducting pairing symmetry is a crucial step in identifying the underlying mechanism of unconventional high-temperature (high-) superconductivity. For high- cuprate superconductors, a series of phase-sensitive experiments [1, 2, 3, 4] provided significant evidence in favor of the -wave symmetry. These experiments are based on Josephson tunneling and flux quantization, which are directly sensitive to the phase of the superconducting order parameter. Iron-based superconductors (FeSCs), the second family of high- superconductors, possess multiple Fermi surfaces which allow for more pairing symmetry candidates [5, 6]. Weakly or moderately doped FeSCs with both electron and hole Fermi pockets are proposed to have pairing gap function where the superconducting order parameter has opposite signs between electron and hole Fermi pockets [7, 8, 9]. On the other hand, the pairing symmetry remains less understood for heavily doped FeSCs with only one type of Fermi pockets, such as the monolayer FeSe and Fe(Te,Se) films grown on SrTiO(001) (STO) substrates [10, 11, 12, 13, 14, 15, 16, 17, 18, 19], which have only electron Fermi pockets [20] and exhibit the highest superconducting transition temperatures among FeSCs [21, 22]. However, phase-sensitive experiments on FeSCs are relatively rare and only focus on moderately doped FeSCs [23, 24, 25] due to the difficulties in fabricating high-quality, homogeneous Josephson junctions at macroscopic or mesoscopic scales. Moreover, the complexity of pairing symmetry in multiband systems presents a significant challenge in interpreting traditional Josephson tunneling experiments [5, 6], which lack both spatial and momentum resolution. Therefore, unambiguously identifying the pairing symmetry of unconventional high- superconductors requires a more accessible phase-sensitive experimental approach with high resolution in space or momentum.
Scanning Josephson tunneling microscopy (SJTM) is a cutting-edge technique in scanning tunneling microscopy (STM) experiments [26, 27, 28], in which a superconducting tip is used to form a spatially tunable microscopic Josephson junction with the superconducting sample. The microscopic Josephson junction in SJTM avoids the inhomogeneity problem present in macroscopic junctions and simultaneously enables subatomic spatial resolution through the manipulation of the tip position. Despite these advantages, the SJTM operates in the dynamical Coulomb blockade regime where the gauge-invariant phase-difference in the Josephson tunneling between the tip and the sample is strongly fluctuating [26, 29, 30, 31]. However, if there are multiple Josephson tunneling channels, their interference depends on the relative phase-difference between the different channels, which can survive in this regime and provide rudimentary phase-sensitivity for the SJTM [32]. Thus, SJTM experiments have the potential to provide Josephson interference information in real space to probe the pairing symmetry of unconventional superconductors.
In this study, by using SJTM, we report the half lattice unit cell periodic modulations of the zero-bias Josephson tunneling conductance between an -wave superconducting tip and the monolayer Fe(Te,Se)/STO film, which is a two-dimensional high- FeSC. The Josephson tunneling signal almost vanishes along the lines connecting the next-nearest-neighbor Fe atoms, which are the diagonal mirror lines of the lattice. Together with the fully opened superconducting gap revealed by single-particle tunneling spectra, our experimental results provide the first microscopic real-space interference evidence for the sign-reversal of the superconducting order parameter that is consistent with the nodeless -wave pairing symmetry. Such SJTM experiments are further performed on several superconductors with different pairing symmetries, including the close cousin monolayer FeSe/STO film, bulk Fe-based superconductor Fe(Te,Se), conventional superconductor Pb film on Si(111), and high- cuprate superconductor BiSrCaCuO. Similar sub-unit-cell Josephson tunneling modulations with vanishing signals along the diagonal mirror lines are observed in monolayer FeSe/STO and BiSrCaCuO with -wave pairing symmetry, whereas absent in -wave superconductors (Pb/Si(111) and bulk Fe(Te,Se)). These findings together demonstrate that our SJTM technique is robust and reliable. Our experimental method can be further extended to a broad class of novel superconducting systems, such as other heavily doped iron-based superconductors and cuprates, heavy-fermion superconductors, and moiré superconductors, offering the potential to provide new insights into the pairing mechanism of unconventional superconductors.
Scanning Josephson tunneling on monolayer Fe(Te,Se)/STO
The high-quality one-unit-cell (1-UC) thick Fe(Te,Se)/STO film is prepared by molecular beam epitaxy (MBE) [33]. Figure 1A displays the crystal structure of the 1-UC Fe(Te,Se) film comprising one Fe layer sandwiched between two Te/Se layers. The Fe atoms are arranged on a square lattice with a lattice constant of and tetrahedrally coordinated with Te/Se atoms above and below the Fe layer. The top chalcogen layer has a square lattice structure with a lattice constant of , which is visible in the STM topography. As shown in the right panel of Figure 1A, we denote the one-Fe unit cell (grey square) axes as and , and the two-Fe unit cell (blue square) axes as and . The corresponding Brillouin zones (BZ) are schematically shown in Figure 1B. Since the primitive unit cell of the 1-UC Fe(Te,Se)/STO film is the two-Fe unit cell, the folded two-Fe BZ in Figure 1B is used throughout this work unless otherwise specified. Due to the heavy electron doping, there are only electron Fermi surfaces around the points in the two-Fe BZ, which is confirmed by our quasiparticle interference experiments (Figure S1). The symmetry elements of the crystal lattice and the Fermi surface structure are marked in Figure 1, A and B, including the diagonal mirror plane (green dashed line), the vertical mirror plane (yellow dashed line), and the four-fold rotation axis (purple dot). Note that the coordinate origin for these symmetry operations is fixed at the projection of the Te/Se atom onto the Fe plane, and thus these operations are point-group symmetry operations of the non-symmorphic P4/nmm crystal structure. Figure 1C shows a typical STM topography of the 1-UC Fe(Te,Se)/STO with the topmost Te/Se lattice clearly resolved. The sites of different atoms are marked for several unit cells. The Te concentration of the film is estimated through the thickness of the 2nd layer (0.58 nm, Figure S2) [34]. The single-particle tunneling spectra measured with a metallic PtIr tip show the fully gapped superconductivity with the gap energy larger than 10 meV at 4.3 K, and the superconducting gap closing temperature of around 58 K (Figure S2), which are consistent with previous results on the interface-enhanced superconductivity [13, 34, 35, 36].

Figure 1. Josephson tunneling spectra on the 1-UC Fe(Te,Se)/STO film. (A) The crystal structure of the 1-UC Fe(Te,Se)/STO film. (B) Schematic of the Fermi surface of 1-UC Fe(Te,Se)/STO. Grey and blue squares represent one-Fe BZ and two-Fe (folded) BZ, respectively. The original and folded Fermi surfaces are plotted by solid and dashed black lines, respectively. In (A) and (B), the diagonal () and vertical () mirror planes and the C axis are marked by green dashed line, yellow dashed line, and purple dot, respectively. The coordinate origin for the symmetry operations is fixed at the projection of the Te/Se atom onto the Fe plane. (C) An atomically resolved topography of the 1-UC Fe(Te,Se)/STO film (6.8 × 3.6 nm, = −40 mV, = 4 nA). The sites of Fe and Te/Se atoms within several unit cells are marked by colored balls. (D) A schematic of the SJTM setup. (E) Current-voltage (-) curves measured with a superconducting tip for different tip-sample distances, in which the local maximum current is marked as . (F) Differential conductance spectra simultaneously measured with (E). Measurement condition of (E) and (F): = 95 mK, = 5 mV, = 50 V. (G) Typical Josephson tunneling spectra measured on the different sites ( = 150 mK, = 5 mV, = 4 nA, = 50 V). The Josephson tunneling almost vanishes on all atomic sites

Figure S1. Quasiparticle interference results on monolayer Fe(Te,Se)/STO film. (A) Schematic of the Fermi surface of monolayer Fe(Te,Se)/STO in the 2-Fe Brillouin zone. The QPI scattering wavevectors () connecting electron pockets are marked by red arrows. The QPI scattering wavevector (dashed green arrow) connects the electron pocket and the absent hole pocket (dashed circle around point). (B-E) Quasiparticle interference patterns measured at 20 mV, 10 mV, -10 mV, -20 mV. QPI scattering wavevectors are observed at that correspond to those shown in (A). The scattering is absent. (F) Linecuts of QPI patterns along the direction in (B) from -30 mV to 30 mV, clearly revealing an electron-type dispersion. The absence of scattering and the electron-type dispersion of demonstrate that the Fermi surface only has electron pockets. The QPI results are measured with a metallic PtIr tip. Measurement conditions: K, mV, pA, mV.

Figure S2. More information about the 1-UC Fe(Te,Se)/STO film. (A) A large-scale topographic image of the Fe(Te,Se)/STO film with terrace of STO substrate and partially grown 2nd-UC Fe(Te,Se) ( nm, V, pA). The right 1-UC area is grown on a higher terrace and is around 0.4 nm (the height of one STO unit cell) higher than the left 1-UC area. (B) The altitude line profile taken along the orange arrow in (A). The thickness of the 2nd-UC FeTeSe is around 0.58 nm, corresponding to . (C) The temperature dependence of the normalized tunneling spectra, which are obtained by dividing the cubic fitting to the spectra for mV (PtIr tip, mV, pA, mV). The inset image shows the zero-bias conductance (ZBC) extracted from the normalized tunneling spectra, yielding an extrapolated gap closing of 58 K when the normalized zero-bias conductance is 1.
To directly probe the Cooper-pair condensate in the 1-UC Fe(Te,Se)/STO, we perform the SJTM experiments with the superconducting Pb-coated PtIr tip at ultralow temperatures (below 150 mK) as shown in Figure 1D. As the superconducting tip approaches the sample surface, two hallmarks of fluctuation dominated Cooper-pair tunneling spectroscopy [26], the local maximum current () and the zero-bias conductance peak due to Josephson tunneling (), emerge in the and curves, respectively (Figure 1, E and F). The is proportional to the zero-bias conductance () (Figure S3), and both and are proportional to the square of the Josephson critical current [26, 29, 30, 31, 37]. As reported before, and are powerful tools to characterize the local superconductivity information [27, 28, 37, 38]. To gain more spatial information of the superconductivity, we compare the Josephson tunneling spectra obtained on different sites as shown in Figure 1G. Strikingly, the Josephson tunneling almost vanishes on all atomic sites but exhibits a large signal at the midpoints of the nearest-neighbor Fe-Fe bonds (also the midpoints between the top and bottom Te/Se sites). We also measure the tunneling spectra in a larger bias range (within mV) at these four kinds of positions (Figure S4). At different positions, the single-particle tunneling spectra show comparable fully opened superconducting gaps at large bias (Figure S4A), while the Josephson tunneling spectra around zero bias exhibit large variations (Figure S4B). It is therefore safe to rule out the possible influence from changes in the normal state tunneling conductance. Such a huge difference occurs when the tip position changes by a distance (around 0.14 nm) that is much smaller than the superconducting coherence length (around 1.7 nm) [39], suggesting a different physical origin other than the inhomogeneous superfluid reported in bulk FeTeSe [37].

Figure S3. Proportional relationship between the and the zero-bias conductance , which are proportional to the Josephson critical current squared.

Figure S4. Tunneling spectra at different positions. (A) Typical tunneling spectra showing the full gap measured at four different kinds of positions ( mK, mV, nA, V). (B) Zoom-in spectra around zero bias of (A). At different positions, the single-particle tunneling conductance spectra (A) are comparable while the Josephson tunneling conductance spectra exhibit large variations (B).
but has large signal at midpoints of the nearest-neighbor Fe sites, which are schematically shown by red dots in the inset.
Half lattice periodic modulation of the Josephson tunneling conductance
We now focus on the unusual spatial variation of the Josephson tunneling signal. As shown in Figure 1G inset, the midpoints of the Fe-Fe bond form a square lattice with axes along and directions and the lattice constant . Thus, high-spatial-resolution Josephson tunneling spectra (Figure 2B) are measured along the direction, passing through the midpoints between nearest-neighbor Fe atoms (orange arrow in Figure 2A). As shown in Figure 2, B and C, the Josephson tunneling spectra show clear periodic modulations with a period of . Figure 2, D and E show the modulation of extracted from Figure 2B and the corresponding Fourier transform. The takes its maximum at midpoints between the nearest-neighbor Fe atoms and almost vanishes at midpoints of nearest-neighbor Fe-Te/Se atoms, exhibiting periodic modulation. We also measure the Josephson tunneling spectra along direction, passing through the nearest-neighbor Fe atoms and their midpoints as shown in Figure S5. The shows periodic modulations, peaks at midpoints and almost vanishes on Fe atomic sites.

Figure 2. Josephson tunneling signal modulation with the period of in the 1-UC Fe(Te,Se)/STO film. (A) A topographic image of the 1-UC Fe(Te,Se)/STO film ( nm, mV, nA). (B) Josephson tunneling spectra measured along the orange arrow in (A) ( mK, mV, nA, µV). The curves are vertically shifted for clarity. The black arrow indicates the Josephson tunneling conductance at zero bias . (C) The color map of (B), which clearly exhibits the spatially modulated Josephson tunneling signals. (D)

Figure S5. Josephson tunneling signal modulation along nearest-neighbor Fe-Fe lines in the 1-UC Fe(Te,Se)/STO film. (A) A topographic image of the 1-UC Fe(Te,Se)/STO film (, , ). (B) Josephson tunneling spectra measured along the orange arrow in (A) (, , , ). The black arrow indicates the Josephson tunneling conductance at zero bias . (C) The color map of (B), which clearly exhibits the spatially modulated Josephson tunneling signals. (D) The extracted along the arrow in (A). The curve shows spatial modulation with the period of the Fe lattice constant and the values at modulation minima are almost zero. The distances in (C) and (D) are defined relative to the beginning of the arrow in (A).
The extracted along the arrow in (A). The curve shows spatial modulations with the period of half the Te/Se lattice constant and the values are almost zero at the modulation minima. The distances in (C) and (D) are defined relative to the beginning of the arrow in (A). (E) The magnitude of the Fourier transform of (D), showing a strong peak at .
Imaging the Josephson tunneling conductance map in 1-UC Fe(Te,Se)/STO
To gain further insights into the spatial modulations of Josephson tunneling, we carry out spectroscopic imaging SJTM measurements on the 1-UC Fe(Te,Se)/STO film. In the region shown in Figure 3A, we measure the zero-bias conductance of Josephson tunneling at every pixel to obtain the spatial map (Figure 3B). The Fourier transform of the topography (Figure 3C) shows distinct Fourier peaks at , which are the reciprocal lattice vectors of the topmost Te/Se lattice (two-Fe unit cell). Intriguingly, as shown in Figure 3D, the dominant modulation wavevectors of are located at , distinct from the modulations at and in the atomic topography (Figure 3C). The different wavevectors between the modulations of and topography demonstrate that the setup effect in the map does not influence the observed modulation in Josephson tunneling. To further confirm this result and exclude possible influence from the setup effect, we perform the SJTM measurements with different setup conditions and obtain similar modulation in the map (fig. S6).

Figure 3. Half lattice periodic modulation of the Josephson tunneling signal with zero lines in the 1-UC Fe(Te,Se)/STO film. (A) A topographic image of the 1-UC Fe(Te,Se)/STO film ( nm, mV, nA). (B) The map measured in the same area as in (A), which shows half lattice periodic modulations ( mK, mV, nA, μV, 816 pixels/nm). The topography (A) and map (B) are obtained simultaneously, showing pixel-to-pixel correspondence between these two datasets. The sites of Fe and Te/Se atoms within several unit cells are marked by colored balls. The green lines show the diagonal mirror plane of the crystal lattice, where is almost zero. (C and D) The magnitude of the Fourier transform of (A) and (B), respectively. The modulation wavevector peaks at the Bragg points of the topmost Te/Se lattice ( and ) are denoted by orange circles in (C). The modulation wavevector peaks at and are denoted by blue circles in (D).
Since the map is obtained simultaneously with the topography , these two maps have pixel-to-pixel correspondence and we can precisely determine the atomic positions in the map. In Figure 3, A and B, the atomic positions within several unit cells are marked by colored balls according to the topography . The map shows that the Josephson tunneling signal almost vanishes not only above all atomic sites, but also along the lines connecting the next-nearest-neighbor Fe atoms (light yellow lines in Figure 3B), which coincide and lie in the diagonal mirror plane of the crystal lattice (green lines in Figure 3, A and B). Such lines of zeroes in the Josephson tunneling signal are also observed under different setup conditions (fig. S6) and using a different microscopic superconducting tip (fig. S7), which demonstrate that they are intrinsic properties of the high- superconducting state in the 1-UC Fe(Te,Se)/STO. The observation of the novel half lattice periodic modulations of the Josephson tunneling with the “zero lines,” which has never been reported in previous SJTM experiments [27, 28, 37, 38, 40], indicates the microscopic phase information of the superconductivity in the 1-UC Fe(Te,Se)/STO.
Discussion
The Josephson tunneling signals and are considered to be approximate estimations of the local superfluid density in previous SJTM studies. However, in these experiments, the superconducting sample and tunneling tip have the same pairing symmetry [27, 28, 38, 40]. For example, the modulations of Josephson tunneling signals observed as evidence for pair density waves in cuprates [28] and 2H-NbSe [38] superconductors are measured with a -wave superconductor BiSrCaCuO tip and an -wave Nb tip, respectively. In this configuration, the reported Josephson tunneling modulations coexist with a large uniform background and have periods of several unit cells, which is fundamentally different from our observations of the half-unit-cell periodic modulations with “zero lines” in the 1-UC Fe(Te,Se)/STO. In contrast, in our measurements, the superfluid density is unlikely to have such large modulations (close to 100%) with a period ( nm) much shorter than the superconducting coherence length (around 1.7 nm) [39].
Since the measured is proportional to , the zero lines in correspond to the vanishing Josephson critical current on such lines, which is possible when the tip and the sample have different pairing symmetries. For the Josephson tunneling between an -wave tip and a superconductor with different pairing symmetry, the Josephson critical current is a coherent superposition resulting from the quantum interference among multiple Josephson tunneling channels [2, 41, 42, 43]. The relative phases between Josephson tunneling channels are determined by the relative phase structure of the pairing function. To the lowest order in the tunneling amplitude, the Josephson critical current can be expressed as [44]:
where is the Fermi distribution function, and are anomalous Green’s functions of the sample and the tip, respectively. Since the tip is an -wave superconductor, the symmetry property of is determined by that of . In the weak coupling limit, where is the normal state Green’s function and is the irreducible representation of the superconducting gap function. Since is invariant, , and thus , transforms according to the pairing matrix of the sample under symmetry operations. Particularly, if the pairing function of the sample is odd under certain symmetry operation , the Josephson critical current transforms as , enabling the symmetry-protected zero Josephson critical current at the high symmetry positions .
In our experiments on the 1-UC Fe(Te,Se)/STO, the zero lines of the Josephson tunneling conductance are observed when the SJTM tip is located in the diagonal mirror plane of the lattice (Figure 3B). This strongly indicates that under the diagonal mirror reflections (), the pairing function is odd, and the Josephson critical current changes sign, i.e., , enforcing the symmetry-protected zero lines of Josephson tunneling along the diagonal mirror lines of the lattice () connecting the next-nearest-neighbor Fe atoms (Figure 4A). Moreover, shown in Figure 3B is maximized at the midpoints of the nearest-neighbor Fe-Fe bond, which are invariant under vertical mirror reflection , suggesting the pairing function is even under . Together with the full superconducting gap revealed by single-particle tunneling spectra (Figure S2), these symmetry properties are fully consistent with the nodeless -wave pairing symmetry (Figure 4B) theoretically proposed for both 1-UC Fe(Te,Se)/STO films and heavily electron-doped bulk FeSCs [18, 19, 45, 46].
Figure 4. Josephson tunneling interference and nodeless -wave pairing symmetry. (A) Schematics for the interference of Josephson tunneling when the tip is located in the diagonal mirror plane. The superconducting phase of the Pb tip is . The superconducting phases on the two Fermi pockets (denoted by red and blue ellipses) of the 1-UC Fe(Te,Se)/STO film are and , respectively. The diagonal mirror plane is denoted by the green plane in real space and the green dashed line in momentum space. The red/blue arrow in real space panel (top) represents the Josephson tunneling associated with the red/blue Fermi pocket in momentum space. (B-D) Schematics for the nodeless -wave, plain -wave, and extended pairing states in the two-Fe BZ. Different colors denote the two Fermi surface pockets with opposite signs of the pairing gap functions. The symmetry elements in k-space, including the mirror planes , and the C4 axis, are marked by green dashed line, yellow dashed line and purple dot, respectively. (E) The calculated Josephson critical current between an -wave superconducting tip and the monolayer Fe(Te,Se)/STO with nodeless -wave pairing symmetry [33]. The map is normalized with respect to its magnitude over the midpoints of the nearest-neighbor Fe atoms. The green and yellow lines show two mirror planes and
Previous theoretical studies have provided the complete classification of superconducting pairing irreducible representations and their symmetry properties at the point in the two-Fe BZ for the non-symmorphic space group of monolayer FeSe [47, 48]. The nodeless -wave pairing state is identified exactly with the pairing sector [33], which is odd under and even under with respect to the mirror planes and defined in Figure 1A. In the -wave pairing state (Figure 4B), the two Fermi pockets hosting different pairing phases cross at four points on the zone boundary, raising the possibility of emerging nodes. However, theoretical studies have shown that the -wave pairing state remains nodeless if the interband spin-orbit coupling (SOC) is smaller than the superconducting gap energy [18, 48, 49, 50]. Experimentally, no mixing of the two Fermi pockets was observed in 1-UC FeSe/STO film with a 5 meV instrument resolution [51], suggesting the interband SOC in this class of superconducting films is smaller than the gap energy (around 10 meV). Furthermore, our SJTM results indicate that the plain -wave (Figure 4C) or extended -wave (Figure 4D) pairing states are highly unlikely since their pairing functions ( type) are even under symmetry operations and thus do not support symmetry-protected zero Josephson tunneling lines in real space (see Supplementary Materials [33] for detailed symmetry analysis and calculations).
The crucial Josephson tunneling interference is schematically illustrated in Figure 4A. When the tip is located in the diagonal mirror plane of the 1-UC Fe(Te,Se)/STO, the microscopic junction is invariant under , which ensures that the Josephson tunneling from the Cooper pairs from the red pocket with pairing phase interfere destructively with those from the blue pocket with pairing phase , resulting in the zero lines in the total Josephson tunneling along the diagonal mirror lines connecting the next-nearest-neighbor Fe atoms. To help visualize the SJTM patterns, we calculate the spatial modulation patterns of the SJTM using approximate Wannier functions in the limit of negligible interband SOC and interband pairing [33]. Figure 4E shows the calculated patterns, exhibiting the lattice periodicity and the sub-unit-cell -wave symmetry. The corresponding map is shown in Figure 4F, which is proportional to the Josephson tunneling conductance . Along the diagonal mirror lines of the lattice connecting the next-nearest-neighbor Fe atoms, the is zero, which is in very good agreement with our experimentally measured map shown in Figure 3B and further supports our observation of a nodeless -wave pairing symmetry in the 1-UC Fe(Te,Se)/STO.
of the crystal lattice. (F) Normalized map calculated from (E), which is consistent with the experimental results of the Josephson tunneling conductance map shown in Figure 3B.
To further support our findings and to test the robustness and reliability of the Josephson tunneling interference technique, we also perform SJTM experiments on various superconductors for comparison, including the monolayer FeSe/STO, the bulk Fe-based superconductor Fe(Te,Se), the conventional superconductor Pb film grown on Si(111), and the high- cuprate . The Josephson tunneling conductance maps as well as the corresponding topography are summarized in Figure 5, and more detailed results are shown in figs. S8 to S11. In the closely related monolayer FeSe/STO, we observe Josephson tunneling modulations with zero lines (Figure 5, A and E), which are indeed similar to the results obtained in the monolayer Fe(Te,Se)/STO, suggesting that the nodeless -wave pairing is not induced by Te substitution, but rather a universal pairing symmetry for the heavily electron-doped monolayer Fe-based superconductors. In the bulk Fe(Te,Se), which exhibits both electron and hole Fermi pockets and is widely recognized as an -wave superconductor with an pairing gap function, the Josephson tunneling conductance is relatively uniform and shows spatial inhomogeneity over length scales much larger than the atomic spacing (Figure 5, B and F). This is consistent with the previous report [37], but totally different than our observations of sub-unit-cell modulations in the monolayer Fe(Te,Se) or FeSe on STO substrate. In the Pb film grown on Si(111), which is a conventional -wave superconductor, the Josephson tunneling signal is uniform (Figure 5, C and G). Strikingly, the SJTM experiments on the BiO surface of the cuprate high- superconductor reveal similar Josephson tunneling modulations to those in the monolayer Fe(Te,Se) or FeSe on STO. As shown in Figure 5, D and H, the Josephson tunneling signal is maximized above the O sites between the nearest-neighbor Cu sites in the plane (the visualized atoms are Bi atoms, which reside directly above the Cu atoms in the plane underneath), but almost vanishes above the lines connecting the next-nearest-neighbor Cu sites, which are the diagonal mirror lines of the lattice (see fig. S11 for more details). This agrees with the -wave pairing symmetry of cuprate superconductors. Together, these SJTM results on superconductors of different pairing symmetries strongly support our discovery of the nodeless -wave pairing in heavily electron-doped monolayer FeSeCs, a pairing symmetry that is different from the bulk Fe(Te,Se) superconductors. Furthermore, they demonstrate that the scanning Josephson tunneling interference technique is powerful enough to distinguish between -wave pairing and -wave pairing at the atomic scale, and thus a useful technique for directly probing the pairing symmetries of unconventional superconductors. In particular, the SJTM technique does not require the fabrication of complicated structures and is well suited for ultrathin superconductors, for which the conventional phase-sensitive experiments, especially those relying on electrical contacts on the side surfaces, are challenging to perform.

Figure 5. SJTM results on various superconductors. (A-D) Atomically resolved topographic images of monolayer FeSe/STO, bulk Fe(Te,Se), Pb/Si(111), and BiSrCaCuO, respectively. (E-H) The Josephson tunneling conductance maps measured in the corresponding areas shown in (A-D). The atoms are marked by colored balls in (A,D,E,H). The maps show sub-unit-cell modulations in monolayer FeSe/STO and BiSrCaCuO that are absent in bulk Fe(Te,Se) and Pb/Si(111). Measurement conditions: (A and E) nm, mK, mV, nA, μV; (B and F) nm, mK, mV, nA, μV; (C and G) nm, mK, mV, nA, μV; (D and H) nm, mK, mV, nA, μV.
To conclude, we report the discovery of half lattice periodic modulations of the Josephson tunneling conductance between an -wave superconducting tip and the monolayer Fe(Te,Se)/STO film. The Josephson tunneling modulation is maximized at the midpoints of the nearest-neighbor Fe atoms and almost vanishes along the next-nearest-neighbor Fe-Fe lines. Combined with single-particle tunneling spectra showing the fully opened superconducting gap, our experimental results are most compatible with a nodeless -wave pairing symmetry for the monolayer Fe(Te,Se)/STO. These unusual modulations could be the first visualization of the -wave superconductivity at the atomic scale through Josephson interference experiments. Similar sub-unit-cell modulations with zero lines are further observed in monolayer FeSe/STO and cuprate BiSrCaCuO, but are absent in bulk Fe(Te,Se) and Pb/Si(111), further supporting our observation and pairing symmetry explanation. Our findings provide precise microscopic evidence for the destructive quantum interference from antiphase superconducting condensates directly in real space within the lattice unit cell and offer new insights into the unconventional superconductors. Our work may also open a new path to study the pairing symmetry of unconventional superconductivity through Josephson interference at the sub-unit-cell scale using the SJTM technique.
References and Notes
Order Parameter Symmetry in a High- Superconductor. Phys. Rev. Lett. 79, 3050–3053 (1997).
M. Mößle, R. Kleiner, -axis Josephson tunneling between BiSrCaCuO and Pb. Phys. Rev. B 59, 4486–4496 (1999).
Acknowledgments
Funding: This work was financially supported by
National Natural Science Foundation of China No. 12488201 (J.W.)
National Natural Science Foundation of China No. 12404215 (Y.L.)
Quantum Science and Technology-National Science and Technology Major Project 2021ZD0302403 (J.W.)
Guangdong Provincial Quantum Science Strategic Initiative GDZX2401001 (Y.L.)
Guangdong Provincial Quantum Science Strategic Initiative GDZX2401009 (Y.L.)
U.S. Department of Energy, Basic Energy Sciences DE-FG02-99ER45747 (Z.W.)
Author contributions: J.W. conceived and supervised the research. T.W., W.R. and Y.L. grew the sample and carried out the experiments. T.W., Y.L., Z.L. and J.W. analyzed the experimental data. T.W., Y.L., Z.W. and J.W. wrote the manuscript.
Competing interests: The authors declare no competing interests.
Data and materials availability: All data are available in the main text or the supplementary materials.
Supplementary Materials
Materials and Methods
Supplementary Text
Figs. S1 to S18
References (52–61)
Materials and Methods
Our experiments were performed on a commercial USM-1600 ultra-high vacuum MBE-STM combined system. The Fe(Te,Se) film was grown as the following recipe: The Nb-doped SrTiO(001) (wt 0.7%) substrate was annealed at 1000°C in the MBE chamber for 1 hour and then cooled down to 330°C for film growth. High purity Fe(99.995%), Te(99.9999%) and Se(99.999%) were co-deposited on the substrate from Knudsen cells for 9 minutes. Then the as-grown film was annealed at 400°C for 17 hours. The thickness of the 1-UC FeTeSe film is around 0.58 nm (Figure S2), corresponding to the composition [34]. Because the monolayer Fe(Te,Se)/STO has a relatively large surface corrugation, which is probably due to the roughness of the substrate, the brighter atoms cannot simply be assigned to Te and the darker to Se [52]. Thus, the Te/Se ratio cannot be determined by counting Te/Se directly from the STM topography [53].
After annealing, the films were transported to the in situ low temperature STM kept at 4.3 K. The SJTM measurements were performed with a superconducting Pb-coated PtIr tip at ultralow temperatures (below 150 mK). The STM/S measurements at 4.3 K and above (shown in Figure S2) were performed with a polycrystalline PtIr tip. The STS spectra were acquired by standard lock-in technique at 983 Hz.
Supplementary Text
Quasiparticle interference (QPI) experiments
The QPI measurements are performed with a PtIr tip at K. As shown in Figure S1, the ring structure in Figure S1B-E corresponds to the intra-pocket scattering and the energy dispersion of (Figure S1F) is consistent with the band dispersion of an electron band, demonstrating the existence of electron pockets at the Fermi level. On the other hand, the scattering (green dashed arrow in Figure S1A,B), which connects the electron pocket and a would-be hole pocket, is absent in the QPI patterns. This suggests the absence of the hole pocket around the point at the Fermi level.
Symmetry analyses on SJTM results
Based on the microscopic theory for the Josephson tunneling in continuum space [44], we provide symmetry analyses on the SJTM results. The experimentally measured Josephson tunneling conductance is proportional to , where is the Josephson critical current of the microscopic junction formed by the sample and the superconducting tip. To the lowest order in the tunneling amplitude, can be expressed as:
where is the Fermi distribution function, and are anomalous Green’s functions of the sample and the tip, respectively. Since the tip is an -wave superconductor, the symmetry property of is determined by that of and thus the pairing symmetry of the superconducting sample. We can discuss how transforms under the symmetry operations in the three prominent unconventional superconductors with different superconducting pairing gap functions.
First, for the cuprate superconductors (Figure S12A), there is only one relevant band with -wave pairing symmetry. The pairing function is odd under , and even under . Therefore, we have , , and . Given these symmetry constraints, at any position of the tunneling tip above the next-nearest-neighbor Cu-Cu line, which is in the diagonal mirror plane, we have and . This requires by symmetry, which is consistent with our observations on bulk cuprate BiSrCaCuO shown in Figure S11 and the calculations in Ref. [44].

Figure S11. SJTM results on . (A) An atomically resolved topography of the on the BiO surface ( nm). (B) The map measured in the same area as in (A), which shows similar modulations as in the monolayer Fe(Te,Se)/STO. (C) Another topographic image of the ( nm). (D) The tunneling spectra measured along the orange arrow in (C). (E) The color map of (D). (F) The extracted Josephson tunneling signal along the arrow in (C). The Josephson tunneling signal is obtained by subtracting the average tunneling conductance measured at from the zero-bias conductance, in order to exclude the contributions from other tunneling processes. The Josephson tunneling signal is maximized at the O sites between the nearest-neighbor Cu atoms in the CuO plane, and almost vanishes at the midpoints of the nearest-neighbor O sites, which are on the next-nearest-neighbor Cu-Cu lines. This is consistent with the pattern expected from the -wave pairing symmetry of . Measurement conditions: mK, mV, nA, .

Figure S12. Symmetry analysis of different pairing gap functions in k-space. (A) Schematic for the nodal -wave pairing in cuprate superconductors. The symmetry elements, including the mirror planes , and the C axis are marked by green dashed line, yellow dashed line and purple dot, respectively. Red and blue colors denote opposite signs of the -wave gap function. (B and C) Schematic for the nodeless -wave pairing in the one-Fe BZ (B) and two-Fe BZ (C) of monolayer Fe(Te,Se)/STO. Red and blue colors denote two Fermi pockets derived from the and orbitals with opposite signs of the gap function. These two Fermi pockets are related by the diagonal mirror reflection or the four-fold rotation . (D and E) Schematic for the pairing in bulk Fe(Te,Se) in the one-Fe BZ (D) and two-Fe BZ (E). The superconducting gap function has opposite signs on the hole pocket around (red) and the electron pockets (blue). Note the -wave pairing symmetry defined by the diagonal mirror reflection or the four-fold rotation . The hole and electron pockets are not related by any symmetry operation. (F) Schematic for the extended pairing scenario in monolayer Fe(Te,Se)/STO. The superconducting order parameter has -wave symmetry with the gap function having opposite signs on the pocket (red) and the pocket (blue). The and pockets are not related by any symmetry operation.
Next, we consider the current study of the monolayer Fe(Te,Se)/STO. In this case, there are two bands derived from the Fe orbitals crossing the Fermi level, as shown in Figure S12B in the one-Fe Brillouin zone (BZ) and in Figure S12C in the two-Fe BZ by different colors. These two bands are dominated by the and orbitals, respectively [5]. Independent of whether we discuss in the one-Fe or the two-Fe BZ, we can consistently define the “diagonal” as the Fe-Fe diagonal ( direction). Both the diagonal mirror reflection and the four-fold rotation exchange these two orbitals/bands, while the does not. This allows the -odd orbital-based nodeless -wave pairing order parameter symmetry, where the gap functions on the two Fermi pockets have opposite signs under rotation. This is the pairing symmetry observed by our SJTM experiments. These symmetry properties ensure that the Josephson critical current transforms according to:
As a result, at any position of the tunneling tip above the next-nearest-neighbor Fe-Fe lines that are in the diagonal mirror plane, the symmetry requires and and hence . This is precisely consistent with our observations in the monolayer Fe(Te,Se)/STO. We note that the superconducting gap is only defined around the two Fermi pockets but not everywhere in the BZ. The -wave symmetry comes from the orbital degrees of freedom under mirror reflection or four-fold rotation. Furthermore, we note that the absence of gap nodes in the orbital-based nodeless -wave scenario makes the zero lines for the Josephson tunneling signal more prominent compared to the case of -space nodal -wave pairing in the cuprates that tends to be influenced by other tunneling processes due to the residual single-particle states near the gap nodes.
Finally, we discuss the pairing scenario as shown in Figure S12D-F. For the usual pairing (Figure S12D in the one-Fe BZ and Figure S12E in the two-Fe BZ) commonly considered for bulk Fe-based superconductors, the superconducting gap function has opposite signs on the hole pocket around (red) and the electron pockets near the BZ boundary (blue). For the extended pairing (Figure S12F), an alternative scenario for the monolayer Fe(Te,Se)/STO, the pairing gap function has opposite signs on the pocket (red) and the pocket (blue) [54]. In both cases of the pairing scenario, the pairing order parameter symmetry is -wave, and the Fermi pockets with opposite signs of the gap function are not related by any symmetry operation. Under the symmetry operations , we have: , and . Therefore, there is no symmetry-protected zero Josephson tunneling positions in real space. As a result, the is expected to be relatively uniform in real space, which is consistent with our experimental observations in bulk Fe(Te,Se) with pairing shown in Figure S9, but completely different and cannot describe our SJTM observations in monolayer Fe(Te,Se)/STO (Figure 3, Figure S6 and Figure S7).

Figure S6. Repeatable SJTM results with different setup conditions. (A) A topographic image of the 1-UC Fe(Te,Se)/STO. (B) The map measured in the same area as in (A). The topography (A) and map (B) are obtained simultaneously, showing pixel-to-pixel correspondence between these two datasets. Measurement condition: mK, mV, nA, µV. The sites of Fe and Te/Se atoms within several unit cells are marked by colored balls. The insets of (A) and (B) are the magnitude of the Fourier transform of (A) and (B), respectively. The modulation wavevector peaks at the Bragg points of the topmost Te/Se lattice ( and ) are denoted by orange circles in the inset of (A). The modulation wavevector peaks at and are denoted by blue circles in the inset of (B). (C and D), Zoom-in maps of the red box regions in (A) and (B) respectively, containing pixels. The zoom-in maps show clear pixel-to-pixel correspondence between the topography and the map.

Figure S7. Repeatable SJTM results with a different microscopic tip. (A) A topographic image of the 1-UC Fe(Te,Se)/STO. (B) The map measured in the same area as in (A). The topography (A) and map (B) are obtained simultaneously, showing pixel-to-pixel correspondence between these two datasets. Measurement condition: mK, mV, nA, µV. The sites of Fe and Te/Se atoms within several unit cells are marked by colored balls. (C and D) The magnitude of the Fourier transform of (A) and (B), respectively. The modulation wavevector peaks at the Bragg points of the topmost Te/Se lattice ( and ) are denoted by orange circles in (C). The modulation wavevector peaks at and are denoted by blue circles in (D).

Figure S9. SJTM results on bulk Fe(Te,Se). (A) An atomically resolved topography of the bulk Fe(Te,Se) (). (B) The map measured in the same area as in (A), which is relatively uniform. (C) The tunneling spectra measured along the orange arrow in (A). (D) Zoom-in spectra around zero bias of (C). (E) The color map of (D). (F) The extracted along the arrow in (A). The Josephson tunneling conductance is relatively uniform and shows inhomogeneity over length scales much larger than the atomic spacing, which are consistent with the previous report [37] and totally different with our observations in the monolayer Fe(Te,Se) or FeSe on STO. Measurement conditions: , , , .
Next, we turn to discuss the corresponding real space pattern under different pairing symmetries and compare to our SJTM observations in monolayer Fe(Te,Se)/STO. Because of the translation symmetry, can be expressed as:
where are integers, are reciprocal lattice vectors along and directions (). For the nodeless -wave pairing symmetry (Figure S12, B and C), requires , , . And requires thus . Therefore, can be expressed up to its lowest-harmonic terms as:
where are the reciprocal lattice vectors of the Fe lattice along and directions (). Since the finite size of the tip will effectively smooth the observed real space map and largely suppress the higher-harmonic terms, we only consider the non-zero lowest-harmonic terms and plot the and maps along with the corresponding lattice positions in Figure S13, A and B. The map (Figure S13B) is in very good agreement with our experimentally measured map shown in Figure 3B and reproduced in Figure S13C.
Figure S13. Josephson critical current map between an s-wave superconducting tip and the monolayer Fe(Te,Se)/STO with nodeless d-wave pairing symmetry based on symmetry analyses. (A) map normalized with respect to its magnitude over the midpoints of nearest-neighbor Fe atoms by symmetry analyses. (B) Normalized map by symmetry analyses, which is consistent with the experimental results of map shown in Figure 3B. (C) The Josephson tunneling conductance map shown in Figure 3B ( mK, mV, nA, µV).
Following the symmetry analysis above, we discuss the map for pairing. In this case, can be expressed up to its second-lowest-harmonic terms as:
The spatial pattern of will most likely show a relatively uniform distribution because the tunneling channels from and bands are not expected to cancel out since there is no symmetry enforcing the cancelation, resulting in a uniform component . This is clearly inconsistent with our experimental SJTM data in monolayer Fe(Te,Se)/STO (Figure 3, Figure S6 and Figure S7).
Discussions on irreducible pairing functions based on the non-symmorphic space group of the crystal structure with two-Fe UC
For the identification of the properties of the pairing function or order parameter, the non-symmorphic symmetry and the 2-Fe unit cell have crucial implications, leading to two-component pairing functions. The complete classification of superconducting pairing irreducible representations and their symmetry properties have been developed in a symmetry-adapted low-energy -point little-group basis in the 2-Fe unit cell for monolayer FeSe [48]. In our experiments, the single-particle tunneling spectra of the monolayer films reveal a full superconducting gap (Figure S2 and Figure S4), which constrains the possible pairing states. Furthermore, because the spin-triplet pairing states do not contribute to the leading-order Josephson tunneling with the spin-singlet -wave superconducting tip we use, the spin-triplet pairing, including the nodeless helical -wave state, can be ignored. Of all the symmetry-derived irreducible pairing states near the point [48], only the spin-singlet (-wave) and the (-wave) have nodeless gap on the Fermi surfaces and can be consistent with our experiments. Using the notations of Ref. [47, 48], the two electronically relevant -point representations are and , forming the basis , where denotes the spin, and denote the two pockets around the point. The pairing matrices for the two states are
where is the Pauli matrix in spin space, and are independent order parameters. They are both intraband pairing states, which is consistent with the absence of mixing of the Fermi surfaces in previous ARPES experiments on monolayer FeSe/SrTiO and a negligible interband spin-orbit coupling (SOC) [48, 51]. We next determine explicitly by symmetry analysis which one of these describes our Josephson tunneling data.
In our SJTM measurements, the Josephson tunneling conductance vanishes along the diagonal mirror (Se-Fe-Se line, in Figure 1A), suggesting the superconducting order parameter is odd under the diagonal mirror reflection . The is maximized at the midpoints between nearest-neighbor Fe atoms, which are in the vertical mirror plane ( in Figure 1A), suggesting the superconducting order parameter is even under the vertical mirror reflection . The four-fold rotation is a combination of the two mirror reflections: , under which the superconducting order parameter is odd. In the above basis at point, the symmetry operations are [48]:
where and are the Pauli matrices in spin space. Under symmetry operations, the irreducible pairing matrices transform as:
The pairing matrix is even under both and , and thus also even under , while the symmetry of the pairing matrix is odd under and even under , and thus odd under . These symmetry properties exclude the pairing states and leave only the pairing state, which is exactly the nodeless -wave pairing state discussed in our work, as the pairing order parameter consistent with our experiments.
Model calculations on the SJTM pattern of the nodeless -wave pairing state
In order to help visualize the SJTM pattern, we perform direct model calculations based on Ref. [44]. For simplicity, we work in the limit of negligible interband SOC and interband pairing as revealed by previous studies [48, 51], under which the 2-Fe UC description is equivalent to the 1-Fe UC description. By extending the theoretical model in Ref. [44] to the multiband case, we can express as:
where labels the two relevant bands crossing the Fermi level. is the Fourier transform of the Wannier function of the orbital/band at momentum . The factor is given by
where is the Fermi distribution function, is the Bogoliubov quasiparticle energy, and is the anomalous Green’s function of the superconducting tip:
where is the tip density of states at the Fermi level and is the -wave gap of the tip. and are calculated according to the five-orbital (3 orbitals of Fe) tight-binding model provided in Ref. [55] for monolayer FeSe/STO. The normal state band dispersions are denoted as with the orbital weights as , where , , , , labels the five 3 orbitals. The Fourier transform of the Wannier functions can be calculated as:
where the Wannier functions can be approximated by the 3 orbitals of a hydrogen-like atom with an effective nuclear charge (the effective nuclear charge for Fe-3 orbitals, [56]). The position is in the plane 0.5 nm above the Fe lattice plane. The two pockets at the Fermi level are dominated by the and orbitals (Figure S12B), respectively, [55]. We take for the band (), and for the band (). The factor can be calculated using the Bogoliubov quasiparticle energy . We take meV for the nodeless -wave gap amplitude in the sample and meV for the -wave tip gap in the calculation. The calculated Josephson critical current is shown in Figure S14A, which is normalized with respect to its magnitude above the midpoints between the nearest-neighbor Fe sites. Figure S14B shows the corresponding map, which is proportional to the Josephson tunneling conductance . The model calculation results are very consistent with our experimentally measured map shown in Figure 3B, supporting our conclusion for the nodeless -wave pairing symmetry in monolayer Fe(Te,Se)/STO.

Figure S14. Josephson critical current map between an s-wave superconducting tip and the monolayer Fe(Te,Se)/STO with nodeless d-wave pairing symmetry based on tight-binding model calculations. (A) map normalized with respect to its magnitude over the midpoints of nearest-neighbor Fe atoms. (B) Normalized map, which is consistent with the experimental results shown in Figure 3B. (C) The Josephson tunneling conductance map shown in Figure 3B ( mK, mV, nA, µV). (D and E) The Josephson critical currents (D) and (E) due to the two Fermi pockets with and , respectively. The current is also normalized by the magnitude of over the midpoints of the nearest-neighbor Fe atoms. The coherent superposition of these two channels (D and E) results in the sub-unit-cell phase structure of the total Josephson critical current shown in (A).
To show the contributions to the Josephson tunneling from different pockets separately, we also calculated and due to the Fermi pockets with pairing function and , respectively. As shown in Figure S14D-E, and are non-zero everywhere in real space and show different spatial distributions. This is consistent with the expectation that these two pockets are delocalized in real space. The coherent superposition of these two channels results in the sub-unit-cell phase structure of the total Josephson critical current (Figure S14A).
Model calculations on the SJTM pattern of the assumed extended pairing state
For the pairing scenario, as discussed above, the Fermi pockets with opposite signs of the gap function are not related by any symmetry operation. Thus, usually symmetry-protected zero Josephson tunneling lines in real space are not expected, which is consistent with our observation of the relatively uniform in bulk Fe(Te,Se) commonly believed to exhibit the gap function. To further examine the extended pairing scenario (Figure S12F), we performed model calculations of the Josephson critical current and ( is proportional to ) by tuning the relative gap amplitudes on the outer pocket and the inner pocket, which are set to have opposite signs. As shown in Figure S15, changing the relative gap amplitudes can produce zero at some positions, but none of the patterns reproduce the zero lines along the diagonal mirror plane with half-unit-cell period observed in our experimental maps. These results further support that the pairing state is highly unlikely to produce the zero Josephson tunneling lines shown in Figure 3B.

Figure S15. Model calculations of Josephson critical current maps between an -wave superconducting tip and the monolayer Fe(Te,Se)/STO with assumed extended pairing symmetry. (A-D) maps normalized with respect to the maximum of , with different relative gap amplitude of the inner () and outer () pockets. (E-H) Normalized maps corresponding to maps in (A-D). can be zero or close to zero at some positions by tuning the relative gap amplitudes, but these positions do not form zero lines along the diagonal mirror with half-unit-cell period as observed in our SJTM results.
Josephson interference and phase-sensitivity in SJTM experiments
In SJTM experiments, the observed Josephson tunneling conductance is proportional to , which only reflects the magnitude of . Therefore, the “phase-sensitivity” in SJTM experiments does not imply the ability to determine the sign of the Josephson critical current . Rather, it indicates that the experimental outcomes (the real space patterns of ) are determined by the phase structure of the superconducting order parameter.
Next, we discuss how the phase-sensitivity can survive considering that the SJTM is in the dynamical Coulomb blockade regime, where the gauge-invariant phase difference across the Josephson junction is strongly fluctuating and not well defined. It seems that due to the strongly fluctuating phase , the Josephson tunneling in SJTM should lose its phase-sensitivity. However, in previous experimental studies, the phase-sensitivity is mainly shown in interference experiments, which only requires a stable relative phase between different Josephson tunneling channels. In this case, a stable phase for a specific tunneling channel is not necessary. For example, in the phase-sensitive experiment in cuprates based on a superconducting quantum interference device (SQUID) [3], the measured quantity is the dynamical resistance () of the interferometer slightly above the Josephson critical current. In this case, there is actually a finite bias voltage on the Josephson junctions and the phase of the two Josephson junctions rotate synchronously. But the relative phase between the two Josephson junctions in SQUID remains stable and the interference is protected. Thus, the key to the phase-sensitive experiments is a stable relative phase between the Josephson tunneling channels, rather than a stable phase of a certain Josephson tunneling channel. In our SJTM experiments, although the gauge-invariant phase difference between the tip and the sample is strongly fluctuating, the relative phase between Josephson tunneling channels from the two Fermi pockets is stable, which is protected by the d-wave pairing symmetry. The symmetry-protected interference of Josephson tunneling channels provides the partial phase-sensitivity explained below.
The “phase-sensitivity” could be a general concept indicating the experimental outcomes are determined by the phase structure of the superconducting order parameter. For example, the quasiparticle interference experiments have been regarded as phase-sensitive in a broad sense because the scattering processes are sensitive to the sign of the order parameter [57, 58]. Next, we discuss to what extent the SJTM results shown in our results are phase-sensitive by comparing our work with the seminal phase-sensitive experiment in cuprates based on SQUID [3]. In both works, multiple coherent Josephson tunneling channels and their interference are used to show the sign-change of the superconducting order parameter. The measured physical quantities ( in our work and dynamic resistance in the SQUID work) only reflect the magnitude of the Josephson critical current and are not sensitive to its sign. The interference of coherent channels is adjusted by certain tunable degree of freedom (spatial position in our work and magnetic flux in the SQUID work). In the SQUID work, the two coherent channels are two Josephson junctions in a SQUID loop. The Josephson critical current exhibits a local minimum at zero magnetic field, which demonstrates the sign-change of the superconducting order parameter in two perpendicular directions related by four-fold rotation. In our work, the coherent channels originate from different Fermi pockets, related by four-fold rotation or diagonal mirror reflection, enabling orbital dependent pairing order parameters of opposite signs. The Josephson tunneling vanishes along the diagonal mirror lines (the next-nearest-neighbor Fe-Fe lines), suggesting the d-wave phase structure of the superconducting order parameter. Thus, in both works, the destructive interference of coherent channels is the key evidence for the d-wave symmetry of the order parameter. The most important difference between these two works is that, in our work, changing the tip position can effectively tune the weights of coherent channels, while the interference is always destructive. In contrast, in the SQUID work, the magnetic flux effectively changes the relative phase between these two channels and can drive the interference from destructive to constructive. Thus, the destructive interference of different Josephson tunneling channels in our work provides partial phase-sensitivity. Indeed, our work offers, for the first time, the sub-unit-cell resolution in real space to resolve and detect the d-wave superconductivity.
SJTM results on Pb/Si(111) and characterization of the Pb tip
We carry out SJTM measurements on Pb films grown on Si(111) substrate with a Pb tip prepared using the same recipe. Figure S10A shows a typical topographic image of Pb/Si(111) film with atomic resolution. The tunneling spectra along the orange arrow shown in Figure S10A are obtained using the superconducting Pb tip (Figure S10C-D) at the base temperature mK. The tunneling spectra show clear features at meV (red dashed line), corresponding to the gap energy of the Pb/Si(111) film. The blue dashed line denotes the voltage corresponding to the sum of the tip and sample gaps meV. Thus, the superconducting gap of the Pb tip is meV. Note that the Josephson tunneling conductance does not show periodic variations at different atomic positions along the orange arrow (Figure S10D-F). The nearly uniform Josephson tunneling conductance is further confirmed in the spatial map shown in Figure S10B, which is obtained in the same area as in Figure S10A. The uniform Josephson tunneling conductance without sub-unit-cell modulations is consistent with the s-wave superconductivity of the Pb/Si(111) film.

Figure S10. Characterization of the superconducting Pb tip on Pb/Si(111) film and SJTM results. (A) An atomically resolved topography of the Pb/Si(111) film ( nm). (B) The map measured in the same area as in (A), which is uniform in the field of view. (C) The tunneling spectra measured along the orange arrow in (A). The red dashed line denotes the gap energy of the Pb/Si(111) film meV and the blue dashed line corresponds to the sum of the tip and sample gaps meV. (D) Zoom-in spectra around zero bias of (C). (E) The color map of (D). (F) The extracted along the arrow in (A). The Josephson tunneling conductance is uniform and does not show sub-unit-cell modulations in real space, consistent with the conventional -wave superconductivity. Measurement conditions: mK, mV, nA, µV.
SJTM results on monolayer FeSe/STO
We carry out SJTM experiments on the monolayer FeSe/STO films. As shown in Figure S8, the Josephson tunneling conductance shows similar half-unit-cell periodic modulations as in Fe(Te,Se)/STO. The curve observed in the monolayer FeSe/STO shows spatial modulations with the period of half the Se lattice constant and the values at the modulation minima are extremely close to zero. The vanishing values (Figure S8E) further support our argument about the zero Josephson tunneling signal on the zero lines coinciding with the diagonal mirror lines of the lattice.

Figure S8. SJTM results on monolayer FeSe/STO film. (A) An atomically resolved topography of the monolayer FeSe/STO film (3.0 × 3.0 nm). (B) The map measured in the same area as in (A), which shows half lattice periodic modulations. (C) The tunneling spectra measured along the orange arrow in (A). (D) The color map of (C), which clearly exhibits the spatially modulated Josephson tunneling signals. (E) The extracted along the arrow in (A). The curve shows spatial modulations with the period of half the Se lattice constant and the values are extremely close to zero at the modulation minima. (F) The magnitude of the Fourier transform of (E), showing a strong peak at . Measurement conditions: mK, mV, nA, μV.
SJTM results on bulk Fe(Te,Se)
The bulk Fe(Te,Se) exhibits both electron and hole pockets at the Fermi level and is widely recognized as an -wave superconductor with an gap function (Figure S12, D and E). Figure S9A shows a typical topographic image of bulk Fe(Te,Se) with atomic resolution. The Josephson tunneling conductance map is relatively uniform in real space as shown in Figure S9B. Figure S9C-E displays the tunneling spectra measured along the orange arrow shown in Figure S9A. The Josephson tunneling conductance is relatively uniform and shows inhomogeneity over length scales much larger than the atomic spacing (Figure S9F). These results are consistent with previous SJTM studies of bulk Fe(Te,Se) [37] and completely different from our results for monolayer Fe(Te,Se) or FeSe on STO and the -wave cases as discussed theoretically above.
SJTM results on Bi₂Sr₂CaCu₂O₈₊ₓ
To further support our experimental technique and theoretical analyses, we also perform SJTM experiments on Bi₂Sr₂CaCu₂O₈₊ₓ, which is a high- cuprate superconductor known to have a -wave pairing symmetry. Figure S11A shows the atomic-resolved topographic image of Bi₂Sr₂CaCu₂O₈₊ₓ on the BiO surface. The visualized atoms are Bi, which reside right above the Cu atoms in the CuO₂ plane underneath. The blue and red balls mark the Cu and O atomic sites in the CuO₂ plane, respectively. The Josephson tunneling conductance map (Figure S11B) shows remarkably similar sub-unit-cell modulations as in the monolayer Fe(Te,Se)/STO. The Josephson tunneling signal is maximized above the O sites between the nearest-neighbor Cu atoms in the CuO₂ plane, and almost vanishes above the lines connecting the next-nearest-neighbor Cu atoms, which are the diagonal mirror lines of the lattice. The small but nonzero values on these lines may originate from other tunneling processes, considering that there are residual single-particle states near the nodes in Bi₂Sr₂CaCu₂O₈₊ₓ. As shown in Figure S11D-E, the tunneling spectra measured along the nearest-neighbor O-O line (orange arrow in Figure S11C) show a small but nonzero conductance background along with the modulating Josephson tunneling signal (peak structure at zero bias). The extracted Josephson tunneling signal (Figure S11F) is maximized over the O sites, and vanishes at the midpoints of the nearest O atoms, which are on the next-nearest-neighbor Cu-Cu lines. The experimental results agree with the pattern expected from the -wave pairing symmetry of the cuprate superconductors [44], suggesting that our experimental techniques and observations, theoretical analyses, findings and conclusions are consistent, robust, and reliable.
Comparison with previous -axis planar Josephson junction experiments
Covering a -wave superconductor with an -wave layer to form a planar Josephson junction has been used in the early studies of the pairing symmetry in the cuprates [2]. Small but finite Josephson effect has been observed in various cuprate/Pb junctions [59, 60, 61], which might originate from spatial inhomogeneities such as dopants, surfaces, interfaces, or steps on a surface. Thus, the macroscopic planar junction experiments could not provide very clear evidence for -wave superconductivity for the cuprates. Moreover, for our monolayer Fe(Te,Se) films grown on STO substrate, such a planar tunnel junction experiment is very challenging.
The monolayer Fe(Te,Se) film is only one unit-cell thick (0.58 nm), which is highly sensitive to air exposure, making the fabrication of a stable planar junction extremely difficult.
Suppressed Josephson tunneling signal around the defect
The sign-reversal superconductivity is sensitive to both magnetic and nonmagnetic defects. As shown in Figure S16, the Josephson tunneling signal is suppressed around a defect marked by the red dashed circle, which is consistent with the sign-reversal superconductivity.

Figure S16. Suppression of Josephson tunneling signal near a defect. (A) A topographic image containing a defect circled by red dashed circle. (B) The map measured in the same area as in (A). The defect position is marked by red dashed circle. The topography (A) and map (B) are obtained simultaneously, showing pixel-to-pixel correspondence between these two datasets. Measurement condition: mK, mV, nA, µV. The Josephson tunneling signal is suppressed around the defect.
Absence of half lattice periodic modulations in other tunneling processes
The differential conductance maps at different voltages are shown in Figure S17. Only maps with extremely close to zero (-0.1 mV to 0.1 mV) show half lattice periodic modulations. This demonstrates that the half lattice periodic modulations are unique patterns of the Josephson tunneling signal and are absent in other non-phase-sensitive tunneling processes. Therefore, the observed half-unit-cell Josephson tunneling modulation cannot be caused by variations in the junction conductance. Furthermore, we display the tunneling spectra at four representative positions (Fe atom, top Te/Se atom, bottom Te/Se atom, and the midpoint) in Figure S4. Across these positions, the fully gapped single-particle tunneling conductance spectra are comparable at large bias (Figure S4A), while the Josephson tunneling conductance spectra around zero bias exhibit pronounced spatial variations (Figure S4B). This further excludes the possibility that the observed zero lines are related to changes in the junction conductance at different positions.

Figure S17. Tunneling conductance maps at different bias voltages. (A) The topographic image of the 1-UC Fe(Te,Se)/STO shown in Figure S6A. (B-H), The maps obtained in the same area as in (A). The topography (A) and maps (B-H) are obtained simultaneously, showing pixel-to-pixel correspondence between these datasets. Only maps with extremely close to zero (-0.1 mV to 0.1 mV) show modulations. This demonstrates that the modulations are unique patterns of the Josephson tunneling signal and are absent in other non-phase-sensitive tunneling processes. Measurement condition: = 99 mK, = 14 mV, = 80 nA, = 50 V.
Discussion on higher harmonic origin of the modulation
The Josephson tunneling signal exhibits modulations, which may cause a question that if the modulations in originate from the higher harmonic modulations of the lattice wavevector . This can be easily addressed because the real space map (Figure 3B and Figure S6B) clearly shows that the modulations are dominant. This is inconsistent with the characteristic of higher harmonic modulations which should be much weaker than the basic frequency modulations. Besides, the higher harmonic scenario cannot explain the “zero lines” observed in our SJTM results.
Possible origins of the finite values on “zero lines”
It is noticed that in Figure 2D, the is slightly larger than zero at the modulation minima, which could have multiple origins. First, there are other tunneling processes besides Josephson tunneling which also contribute to the zero-bias tunneling conductance. The - curves simultaneously measured with the curves in Figure 2B are shown in Figure S18A (black curves), which exhibit Josephson tunneling current coexisting with a linear background (red curve). The tunneling conductance contributed from the Josephson tunneling shows a dip-peak-dip structure in a very small range around zero bias (typically within mV at ultralow temperatures), while other tunneling processes are rather uniform within such a small energy range and contribute to roughly linear background in - curves. The - background shows a positive slope which originates from other tunneling processes and gives rise to a finite conductance μS. After subtracting the conductance from the background, the modulation minima are extremely close to zero as shown in Figure S18B. Moreover, in SJTM measurements, the measurement positions are discrete and inevitably deviate from the precise “zero lines”, leading to a situation where the Josephson tunneling processes cannot cancel out perfectly and exhibit residual non-zero Josephson current at the modulation minima. Such “zero lines” are further supported by the SJTM results in the monolayer FeSe/STO, where the values at modulation minima are extremely close to zero (Figure S8E).

Figure S18. Finite conductance contributed by other tunneling processes. (A) Current-voltage (-) curves (black) simultaneously measured with Figure 2B ( mK, mV, nA, V) at different positions. The red curve shows the background originating from other tunneling processes with a tunneling conductance of S. (B) The - modulation obtained by subtracting from the data in Figure 2D. The modulation minima are extremely close to zero.
References
- [1]D. J. Van Harlingen, Phase-sensitive tests of the symmetry of the pairing state in the high-temperature superconductors—Evidence for dₓ²₋ᵧ² symmetry. Rev. Mod. Phys. 67, 515–535 (1995).DOI
- [2]C. C. Tsuei, J. R. Kirtley, Pairing symmetry in cuprate superconductors. Rev. Mod. Phys. 72, 969–1016 (2000).
- [3]D. A. Wollman, D. J. Van Harlingen, W. C. Lee, D. M. Ginsberg, A. J. Leggett, Experimental determination of the superconducting pairing state in YBCO from the phase coherence of YBCO-Pb dc SQUIDs. Phys. Rev. Lett. 71, 2134–2137 (1993).pubmed.ncbi.nlm.nih.gov/10054591
- [4]C. C. Tsuei, J. R. Kirtley, C. C. Chi, L. S. Yu-Jahnes, A. Gupta, T. Shaw, J. Z. Sun, M. B. Ketchen, Pairing Symmetry and Flux Quantization in a Tricrystal Superconducting Ring of YBa₂Cu₃O₇-δ. Phys. Rev. Lett. 73, 593–596 (1994).
- [5]P. J. Hirschfeld, M. M. Korshunov, I. I. Mazin, Gap symmetry and structure of Fe-based superconductors. Rep. Prog. Phys. 74, 124508 (2011).
- [6]R. M. Fernandes, A. I. Coldea, H. Ding, I. R. Fisher, P. J. Hirschfeld, G. Kotliar, Iron pnictides and chalcogenides: a new paradigm for superconductivity. Nature 601, 35–44 (2022).DOI
- [7]I. I. Mazin, D. J. Singh, M. D. Johannes, M. H. Du, Unconventional Superconductivity with a Sign Reversal in the Order Parameter of LaFeAsO₁-xFₓ. Phys. Rev. Lett. 101, 057003 (2008).pubmed.ncbi.nlm.nih.gov/18764420
- [8]K. Kuroki, S. Onari, R. Arita, H. Usui, Y. Tanaka, H. Kontani, H. Aoki, Unconventional Pairing Originating from the Disconnected Fermi Surfaces of Superconducting LaFeAsO₁-xFₓ. Phys. Rev. Lett. 101, 087004 (2008).pubmed.ncbi.nlm.nih.gov/18764652
- [9]A. V. Chubukov, D. V. Efremov, I. Eremin, Magnetism, superconductivity, and pairing symmetry in iron-based superconductors. Phys. Rev. B 78, 134512 (2008).DOI
- [10]Z. Ge, C. Yan, H. Zhang, D. Agterberg, M. Weinert, L. Li, Evidence for d-Wave Superconductivity in Single Layer FeSe/SrTiO₃ Probed by Quasiparticle Scattering Off Step Edges. Nano Lett. 19, 2497–2502 (2019).DOI
- [11]C. Liu, Z. Wang, Y. Gao, X. Liu, Y. Liu, Q.-H. Wang, J. Wang, Spectroscopic Imaging of Quasiparticle Bound States Induced by Strong Nonmagnetic Scatterings in One-Unit-Cell FeSe/SrTiO₃. Phys. Rev. Lett. 123, 036801 (2019).DOI
- [12]C. Liu, Z. Wang, S. Ye, C. Chen, Y. Liu, Q. Wang, Q.-H. Wang, J. Wang, Detection of Bosonic Mode as a Signature of Magnetic Excitation in One-Unit-Cell FeSe on SrTiO₃. Nano Lett. 19, 3464–3472 (2019).DOI
- [13]C. Chen, C. Liu, Y. Liu, J. Wang, Bosonic Mode and Impurity-Scattering in Monolayer Fe(Te,Se) High-Temperature Superconductors. Nano Lett. 20, 2056–2061 (2020).DOI
- [14]Q. Fan, W. H. Zhang, X. Liu, Y. J. Yan, M. Q. Ren, R. Peng, H. C. Xu, B. P. Xie, J. P. Hu, T. Zhang, D. L. Feng, Plain s-wave superconductivity in single-layer FeSe on SrTiO₃ probed by scanning tunnelling microscopy. Nat. Phys. 11, 946–952 (2015).DOI
- [15]F. Yang, F. Wang, D.-H. Lee, Fermiology, orbital order, orbital fluctuations, and Cooper pairing in iron-based superconductors. Phys. Rev. B 88, 100504 (2013).DOI
- [16]J. Kang, R. M. Fernandes, Superconductivity in FeSe Thin Films Driven by the Interplay between Nematic Fluctuations and Spin-Orbit Coupling. Phys. Rev. Lett. 117, 217003 (2016).DOI
- [17]Y. Gao, Y. Yu, T. Zhou, H. Huang, Q.-H. Wang, Hidden sign-changing s-wave superconductivity in monolayer FeSe. Phys. Rev. B 94, 144512 (2016).DOI
- [18]D. F. Agterberg, T. Shishidou, J. O’Halloran, P. M. R. Brydon, M. Weinert, Resilient Nodeless d-Wave Superconductivity in Monolayer FeSe. Phys. Rev. Lett. 119, 267001 (2017).DOI
- [19]Z.-X. Li, F. Wang, H. Yao, D.-H. Lee, What makes the Tc of monolayer FeSe on SrTiO3 so high: a sign-problem-free quantum Monte Carlo study. Sci. Bull. 61, 925–930 (2016).DOI
- [20]J. J. Lee, F. T. Schmitt, R. G. Moore, S. Johnston, Y.-T. Cui, W. Li, M. Yi, Z. K. Liu, M. Hashimoto, Y. Zhang, D. H. Lu, T. P. Devereaux, D.-H. Lee, Z.-X. Shen, Interfacial mode coupling as the origin of the enhancement of Tc in FeSe films on SrTiO3. Nature 515, 245–248 (2014).DOI
- [21]Q.-Y. Wang, Z. Li, W.-H. Zhang, Z.-C. Zhang, J.-S. Zhang, W. Li, H. Ding, Y.-B. Ou, P. Deng, K. Chang, J. Wen, C.-L. Song, K. He, J.-F. Jia, S.-H. Ji, Y.-Y. Wang, L.-L. Wang, X. Chen, X.-C. Ma, Q.-K. Xue, Interface-Induced High-Temperature Superconductivity in Single Unit-Cell FeSe Films on SrTiO3. Chin. Phys. Lett. 29, 037402 (2012).DOI
- [22]W.-H. Zhang, Y. Sun, J.-S. Zhang, F.-S. Li, M.-H. Guo, Y.-F. Zhao, H.-M. Zhang, J.-P. Peng, Y. Xing, H.-C. Wang, T. Fujita, A. Hirata, Z. Li, H. Ding, C.-J. Tang, M. Wang, Q.-Y. Wang, K. He, S.-H. Ji, X. Chen, J.-F. Wang, Z.-C. Xia, L. Li, Y.-Y. Wang, J. Wang, L.-L. Wang, M.-W. Chen, Q.-K. Xue, X.-C. Ma, Direct Observation of High-Temperature Superconductivity in One-Unit-Cell FeSe Films. Chin. Phys. Lett. 31, 017401 (2014).
- [23]X. Zhang, Y. S. Oh, Y. Liu, L. Yan, K. H. Kim, R. L. Greene, I. Takeuchi, Observation of the Josephson Effect in Pb/Ba₁−xKxFe₂As₂ Single Crystal Junctions. Phys. Rev. Lett. 102, 147002 (2009).pubmed.ncbi.nlm.nih.gov/19392473
- [24]C.-T. Chen, C. C. Tsuei, M. B. Ketchen, Z.-A. Ren, Z. X. Zhao, Integer and half-integer flux-quantum transitions in a niobium–iron pnictide loop. Nat. Phys. 6, 260–264 (2010).DOI
- [25]A. A. Kalenyuk, A. Pagliero, E. A. Borodianskyi, A. A. Kordyuk, V. M. Krasnov, Phase-Sensitive Evidence for the Sign-Reversal s± Symmetry of the Order Parameter in an Iron-Pnictide Superconductor Using Nb/Ba₁−xNaₓFe₂As₂ Josephson Junctions. Phys. Rev. Lett. 120, 067001 (2018).DOI
- [26]O. Naaman, W. Teizer, R. C. Dynes, Fluctuation Dominated Josephson Tunneling with a Scanning Tunneling Microscope. Phys. Rev. Lett. 87, 097004 (2001).
- [27]M. T. Randeria, B. E. Feldman, I. K. Drozdov, A. Yazdani, Scanning Josephson spectroscopy on the atomic scale. Phys. Rev. B 93, 161115 (2016).DOI
- [28]M. H. Hamidian, S. D. Edkins, S. H. Joo, A. Kostin, H. Eisaki, S. Uchida, M. J. Lawler, E.-A. Kim, A. P. Mackenzie, K. Fujita, J. Lee, J. C. S. Davis, Detection of a Cooper-pair density wave in Bi₂Sr₂CaCu₂O₈+x. Nature 532, 343–347 (2016).DOI
- [29]G.-L. Ingold, H. Grabert, U. Eberhardt, Cooper-pair current through ultrasmall Josephson junctions. Phys. Rev. B 50, 395–402 (1994).pubmed.ncbi.nlm.nih.gov/9974556
- [30]Y. M. Ivanchenko, A. Zil’berman, The Josephson effect in small tunnel contacts. Sov Phys JETP 28, 1272–1276 (1969).
- [31]B. Jäck, M. Eltschka, M. Assig, M. Etzkorn, C. R. Ast, K. Kern, Critical Josephson current in the dynamical Coulomb blockade regime. Phys. Rev. B 93, 020504 (2016).DOI
- [32]S. Karan, H. Huang, C. Padurariu, B. Kubala, A. Theiler, A. M. Black-Schaffer, G. Morrás, A. L. Yeyati, J. C. Cuevas, J. Ankerhold, K. Kern, C. R. Ast, Superconducting quantum interference at the atomic scale. Nat. Phys. 18, 893–898 (2022).DOI
- [33]See Supplementary Materials.
- [34]F. Li, H. Ding, C. Tang, J. Peng, Q. Zhang, W. Zhang, G. Zhou, D. Zhang, C.-L. Song, K. He, S. Ji, X. Chen, L. Gu, L. Wang, X.-C. Ma, Q.-K. Xue, Interface-enhanced high-temperature superconductivity in single-unit-cell FeTe1-xSex films on SrTiO3. Phys. Rev. B 91, 220503 (2015).DOI
- [35]Y. Liu, T. Wei, G. He, Y. Zhang, Z. Wang, J. Wang, Pair density wave state in a monolayer high-Tc iron-based superconductor. Nature 618, 934–939 (2023).
- [36]C. Chen, K. Jiang, Y. Zhang, C. Liu, Y. Liu, Z. Wang, J. Wang, Atomic line defects and zero-energy end states in monolayer Fe(Te,Se) high-temperature superconductors. Nat. Phys. 16, 536–540 (2020).
- [37]D. Cho, K. M. Bastiaans, D. Chatzopoulos, G. D. Gu, M. P. Allan, A strongly inhomogeneous superfluid in an iron-based superconductor. Nature 571, 541–545 (2019).DOI
- [38]X. Liu, Y. X. Chong, R. Sharma, J. C. S. Davis, Discovery of a Cooper-pair density wave state in a transition-metal dichalcogenide. Science 372, 1447–1452 (2021).DOI
- [39]T. Wei, Y. Liu, W. Ren, Z. Liang, Z. Wang, J. Wang, Observation of Superconducting Pair Density Modulation within Lattice Unit Cell. Chin. Phys. Lett. 42, 027404 (2025).DOI
- [40]H. Deng, H. Qin, G. Liu, T. Yang, R. Fu, Z. Zhang, X. Wu, Z. Wang, Y. Shi, J. Liu, H. Liu, X.-Y. Yan, W. Song, X. Xu, Y. Zhao, M. Yi, G. Xu, H. Hohmann, S. C. Holbæk, M. Dürrnagel, S. Zhou, G. Chang, Y. Yao, Q. Wang, Z. Guguchia, T. Neupert, R. Thomale, M. H. Fischer, J.-X. Yin, Chiral kagome superconductivity modulations with residual Fermi arcs. Nature 632, 775–781 (2024).DOI
- [41]T. K. Ng, N. Nagaosa, Broken time-reversal symmetry in Josephson junction involving two-band superconductors. Europhys. Lett. 87, 17003 (2009).DOI
- [42]Y. Ota, M. Machida, T. Koyama, H. Matsumoto, Theory of Heterotic Superconductor-Insulator-Superconductor Josephson Junctions between Single- and Multiple-Gap Superconductors. Phys. Rev. Lett. 102, 237003 (2009).pubmed.ncbi.nlm.nih.gov/19658963
- [43]Y. Ota, N. Nakai, H. Nakamura, M. Machida, D. Inotani, Y. Ohashi, T. Koyama, H. Matsumoto, Ambegaokar-Baratoff relations for Josephson critical current in heterojunctions with multigap superconductors. Phys. Rev. B 81, 214511 (2010).DOI
- [44]P. Choubey, P. J. Hirschfeld, Theory of Josephson scanning microscopy with s-wave tip on unconventional superconducting surface: Application to Bi2Sr2CaCu2O8+δ. Phys. Rev. B 110, 184519 (2024).DOI
- [45]T. A. Maier, S. Graser, P. J. Hirschfeld, D. J. Scalapino, d-wave pairing from spin fluctuations in the KxFe2-ySe2 superconductors. Phys. Rev. B 83, 100515 (2011).
- [46]F. Wang, F. Yang, M. Gao, Z.-Y. Lu, T. Xiang, D.-H. Lee, The electron pairing of KₓFe₂₋ᵧSe₂. Europhys. Lett. 93, 57003 (2011).
- [47]V. Cvetkovic, O. Vafek, Space group symmetry, spin-orbit coupling, and the low-energy effective Hamiltonian for iron-based superconductors. Phys. Rev. B 88, 134510 (2013).
- [48]P. M. Eugenio, O. Vafek, Classification of symmetry derived pairing at the M point in FeSe. Phys. Rev. B 98, 014503 (2018).
- [49]T. Nakayama, T. Shishidou, D. F. Agterberg, Nodal topology in d-wave superconducting monolayer FeSe. Phys. Rev. B 98, 214503 (2018).DOI
- [50]M. Khodas, A. V. Chubukov, Interpocket Pairing and Gap Symmetry in Fe-Based Superconductors with Only Electron Pockets. Phys. Rev. Lett. 108, 247003 (2012).
- [51]Y. Zhang, J. J. Lee, R. G. Moore, W. Li, M. Yi, M. Hashimoto, D. H. Lu, T. P. Devereaux, D.-H. Lee, Z.-X. Shen, Superconducting Gap Anisotropy in Monolayer FeSe Thin Film. Phys. Rev. Lett. 117, 117001 (2016).
- [52]X.-L. Peng, Y. Li, X.-X. Wu, H.-B. Deng, X. Shi, W.-H. Fan, M. Li, Y.-B. Huang, T. Qian, P. Richard, J.-P. Hu, S.-H. Pan, H.-Q. Mao, Y.-J. Sun, H. Ding, Observation of topological transition in high-Tc superconducting monolayer FeTe₁₋ₓSeₓ films on SrTiO₃(001). Phys. Rev. B 100, 155134 (2019).
- [53]J.-X. Yin, Z. Wu, J.-H. Wang, Z.-Y. Ye, J. Gong, X.-Y. Hou, L. Shan, A. Li, X.-J. Liang, X.-X. Wu, J. Li, C.-S. Ting, Z.-Q. Wang, J.-P. Hu, P.-H. Hor, H. Ding, S. H. Pan, Observation of a robust zero-energy bound state in iron-based superconductor Fe(Te,Se). Nat. Phys. 11, 543–546 (2015).
- [54]Z. P. Yin, K. Haule, G. Kotliar, Spin dynamics and orbital-antiphase pairing symmetry in iron-based superconductors. Nat. Phys. 10, 845–850 (2014).
- [55]D. Huang, T. A. Webb, S. Fang, C.-L. Song, C.-Z. Chang, J. S. Moodera, E. Kaxiras, J. E. Hoffman, Bounds on nanoscale nematicity in single-layer FeSe/SrTiO₃. Phys. Rev. B 93, 125129 (2016).
- [56]E. Clementi, D. L. Raimondi, Atomic Screening Constants from SCF Functions. J. Chem. Phys. 38, 2686–2689 (1963).DOI
- [57]T. Hanaguri, Y. Kohsaka, M. Ono, M. Maltseva, P. Coleman, I. Yamada, M. Azuma, M. Takano, K. Ohishi, H. Takagi, Coherence Factors in a High-Tc Cuprate Probed by Quasi-Particle Scattering Off Vortices. Science 323, 923–926 (2009).
- [58]T. Hanaguri, S. Nitaka, K. Kuroki, H. Takagi, Unconventional s-Wave Superconductivity in Fe(Se,Te). Science 328, 474–476 (2010).
- [59]A. G. Sun, D. A. Gajewski, M. B. Maple, R. C. Dynes, Observation of Josephson pair tunneling between a high-Tc cuprate (YBa₂Cu₃O₇₋δ) and a conventional superconductor (Pb). Phys. Rev. Lett. 72, 2267–2270 (1994).
- [60]K. A. Kouz netsov, A. G. Sun, B. Chen, A. S. Katz, S. R. Bahcall, J. Clarke, R. C. Dynes, D. A. Gajewski, S. H. Han, M. B. Maple, J. Giapintzakis, J.-T. Kim, D. M. Ginsberg, c-axis Josephson Tunneling between YBa₂Cu₃O₇₋δ and Pb: Direct Evidence for Mixed Order Parameter Symmetry in a High-Tc Superconductor. Phys. Rev. Lett. 79, 3050–3053 (1997).
- [61]M. Mößle, R. Kleiner, c-axis Josephson tunneling between Bi₂Sr₂CaCu₂O₈+x and Pb. Phys. Rev. B 59, 4486–4496 (1999).

