Diode Effect May Assist Finding Proper Superconductivity Mechanism in Copper Oxides Armen Gulian, Serafim Teknowijoyo, Vahan Nikoghosyan Abstract We present measurements demonstrating that copper-oxide high-temperature superconductors can exhibit broken time-reversal symmetry in the absence of external magnetic fields. Using T ​ l 2 ​ B ​ a 2 ​ C ​ a ​ C ​ u 2 ​ O 8 Tl_{2}Ba_{2}CaCu_{2}O_{8} microbridges, we observe a pronounced superconducting diode effect at 100 K under strictly zero-field conditions. This nonreciprocal response remains unchanged in magnetic fields up to ± 100 ​ O ​ e \pm 100Oe . Our results are consistent with recent reports of zero-field diode behavior in B ​ i 2 ​ S ​ r 2 ​ C ​ a ​ C ​ u 2 ​ O 8 + δ Bi_{2}Sr_{2}CaCu_{2}O_{8+{\delta}} and together indicate that time-reversal symmetry breaking may be an intrinsic property of the cuprate superconducting state. These findings significantly constrain theoretical models of high-temperature superconductivity that rely on time-reversal-symmetric mechanisms. I Introduction High-temperature superconductivity (HTSC) in copper oxides has been known for nearly forty years, yet the mechanism behind this phenomenon remains unresolved. Since the time of Sir Neville Mott, it has often been said that there are as many theories as there are theorists. Despite decades of progress, the number of proposed or modified theoretical models has not decreased significantly, although experimental results have helped identify several leading candidates. However, their validation is critically dependent on targeted experimental tests. A particularly informative development has emerged recently from research on the superconducting diode effect. This effect, reminiscent of the semiconductor diode, manifests itself as the suppression of one unipolar branch of an alternating current flowing through a superconducting system, as illustrated in Fig. 1. Figure 1: Superconducting diode effect in presence of external magnetic field: the voltage down-spikes in response to applied sinusoidal current (shown in arb. units, circles). These data were obtained in our laboratory with Nb_{3}Sn bridge at 2K ; the magnetic field is applied perpendicular to the film surface, more details can be found in ref1 . First clearly demonstrated in Nb/V/Ta superlattices ref2 through the observation of nonreciprocal critical currents, the effect has since been reported in a wide range of systems, including artificially structured materials (such as those with conformal holes ref3 ), Josephson junctions, and microbridges (see the recent review ref4 ). A characteristic feature of systems exhibiting this effect is the simultaneous breaking of time-reversal symmetry (TRS) and inversion symmetry (IS). This combination is readily identified within the Ginzburg-Landau framework ref5 . Typically, TRS breaking is associated with the presence of a magnetic field, while IS may be absent due to either the crystal structure or environmental factors. In most experimental settings the magnetic field is externally applied, although exceptions exist, such as chiral order parameters in triplet superconductors ref6 or magnetic inclusions. Although initial studies focused on conventional superconductors, diode-effect research has now extended to the cuprate HTSCs ref7 . A particularly notable result is that the diode behavior in Bi_{2}Sr_{2}CaCu_{2}O_{8+{\delta}} flakes appears in the absence of any externally applied magnetic field, with the effect observed up to 72K for samples with T_{c}^{onset}=93.6K , and T_{c}^{zero}=75.5K . The appearance of a diode effect under strictly zero-field conditions requires careful interpretation. The authors of Ref. ref7 attributed this to TRS breaking caused by internal electronic effects - specifically, loop-current states associated with the pseudogap phase in Varma’s model ref8 . In the present paper we extend the investigation of Ref. ref7 to lithographically patterned microbridges fabricated from the related cuprate superconductor Tl_{2}Ba_{2}CaCu_{2}O_{8} , with measurements performed at 100K . Section 2 describes the physical properties of the films and bridges. We show that this simple microbridge structure exhibits a diode effect at 100K under strictly zero magnetic field. Furthermore, measurements carried out under fields of \pm 100Oe show no deviation from the zero-field behavior. Since our results are consistent with those reported in Ref. ref7 and more recently in Ref. ref9 , they are likely to be further corroborated, and significantly constrain theoretical models of HTSC in cuprates that do not involve TRS breaking. Section 3 discusses the classes of theories compatible or incompatible with these observations. Section 4 summarizes the implications of our findings and outlines possible directions for future work. Section 5 describes the methods employed. II Experimental details The films of Tl_{2}Ba_{2}CaCu_{2}O_{8} were deposited as described in ref10 on LaAlO_{3} substrates using Ba-Ca-Cu-O precursors. Bridges were defined by wet photolithography using iron chloride as an etchant. The resistive and magnetic properties of the films were measured with a PPMS DynaCool (Quantum Design). Bridge transport measurements used a Keithley 6221 current source and a Keithley 2182A nanovoltmeter for low-voltage measurements, and a Keithley DMM 2000 for higher-voltage acquisitions. All instruments were connected externally to the DynaCool cryostat. In DC mode, the I+ branch was swept from zero to its maximum value; the polarity of the current leads was then reversed to obtain the corresponding I- branch. III Results Our principal finding - the appearance of a diode effect at 100K - is presented in Fig. 2. Figure 2: Appearance of the diode effect in Tl_{2}Ba_{2}CaCu_{2}O_{8} superconducting bridge ( T=100K and H=0 ) at the increase of the bias current. The amplitude of sinusoidal AC current is 30mA in panel (a) (shown by circles, arb. units) and 35mA in panel (b); the frequency is 0.2Hz in both cases. This temperature was chosen for two reasons: 1) to stay closer to T_{c} for keeping the critical current relatively low – at higher currents the transition to normal state could have irreversibly damaged the precious bridge; 2) at this temperature, as follows from Fig. 3, we have full resistive transition in zero fields. The microbridge was fabricated from a Tl_{2}Ba_{2}CaCu_{2}O_{8} film with T_{c}\approx 101K (Fig. 3(a)). The resistive transition of the bridge is shown in Fig. 3(b). Figure 3: Superconducting transition in Tl_{2}Ba_{2}CaCu_{2}O_{8} film recorded by AC susceptibility (panel (a); measurement frequency 400Hz, AC amplitude 5Oe ), and in lithographically processed bridge via resistivity (panel (b); measurement current 1\mu A ). The film thickness was approximately 1{\mu}m , and the bridge dimensions were 20{\mu}m in width and 2mm in length. A defining aspect of these measurements is that they were performed under strictly zero external magnetic field, H=0 . Measurements under H=+100Oe and H=-100Oe showed no observable difference from the zero-field case (Fig. 4). Figure 4: Nonreciprocal superconducting transition in Tl_{2}Ba_{2}CaCu_{2}O_{8} bridge is almost similar for H=0 and H=\pm 100Oe (panel (a)). At the same time, as follows from panel (b), 100Oe exceeds the value of H_{c1} of this material. These results therefore indicate that copper-oxide superconductors can exhibit TRS breaking through internal mechanisms intrinsic to the superconducting state. IV Discussion The central experimental observation of this work is the emergence of a pronounced superconducting diode effect in Tl_{2}Ba_{2}CaCu_{2}O_{8} microbridges at T=100K under strictly zero applied magnetic field (Figs. 2 and 4). The nonreciprocal response remains unchanged within experimental resolution under applied fields of \pm 100Oe , exceeding the nominal H_{c1} of the material. Similar zero-field diode behavior has recently been reported in Bi_{2}Sr_{2}CaCu_{2}O_{8+{\delta}} flakes ref7 ; ref9 . The present results demonstrate that this phenomenon is not limited to a single bismuth-cuprate family but also appears in a structurally related thallium-based compound. Given the chemical and electronic similarities between Bi- and Tl- based cuprates, the consistency between these independent observations strengthens the case that the effect reflects an intrinsic property of the superconducting state rather than a material-specific artifact. IV.1 Symmetry considerations Nonreciprocal superconducting transport generally requires the simultaneous breaking of TRS and IS ref5 . In many experimental realizations of the superconducting diode effect, TRS breaking is introduced by an external magnetic field ref1 ; ref2 ; ref3 ; ref4 . In contrast, the present measurements were conducted under strictly zero applied field. If the observed diode response is intrinsic, it implies that TRS breaking may arise internally in the superconducting state. Such a possibility has been discussed in the context of loop-current models of the pseudogap phase ref8 ; ref11 and in scenarios involving chiral or complex order parameters ref12 ; ref13 ; ref14 . At the same time, we note that inversion symmetry in our microbridge geometry may be broken extrinsically by structural asymmetry at the film–substrate interface or by lithographic processing. In that case, intrinsic TRS breaking combined with extrinsic IS breaking would suffice to generate nonreciprocal response without invoking inversion symmetry breaking in the bulk crystal. IV.2 Constraints on theoretical descriptions Rather than conclusively ruling out specific microscopic mechanisms, the present results provide a symmetry-based constraint. Superconducting states that strictly preserve TRS in equilibrium would require additional mechanisms to explain zero-field nonreciprocity. Preformed-pair and phase-fluctuation approaches ref15 ; ref16 ; ref17 , RVB-based models ref18 ; ref19 ; ref20 , and many density-wave descriptions of the pseudogap ref21 ; ref22 ; ref23 ; ref24 ; ref25 ; ref26 do not, in their simplest formulations, require spontaneous TRS breaking in the superconducting state. If the diode effect observed here is intrinsic and reproducible across cuprate families, these frameworks may require extension to incorporate TRS-breaking components or nonequilibrium effects. Conversely, models allowing spontaneous TRS breaking, including loop-current states ref8 ; ref11 , or chiral superconducting order parameters ref12 ; ref13 ; ref14 , are naturally compatible with the symmetry requirements of the diode effect. However, further experimental discrimination between these possibilities remains necessary. IV.3 Nonequilibrium aspects An additional observation is that the diode effect appears above a finite current threshold (Fig. 5). Figure 5: Demonstration of possible nonequilibrium effects in the resistive state of the bridge. The top datasets show absolute values of voltage measured at linear current sweep to 16mA and 27mA with the relatively slow rate 1mA/s (the lines are guide for eyes). These curves (and many others measured) coincide at intersecting current values. As soon as the current exceeds 25 mA, irreversibility appears. At the sinusoidal bias current, the level of voltage is significantly lower (the bottom data points) because of much higher scanning speed (indicated on labels), and the irreversibility comes in at noticeably higher current amplitudes. This raises the possibility that nonequilibrium superconducting dynamics may enhance or reveal symmetry-breaking tendencies not prominent in the equilibrium state. Most microscopic theories of cuprate superconductivity address equilibrium properties. If finite current density plays a role in stabilizing or amplifying TRS-breaking components, the observed diode behavior may reflect a current-driven phenomenon rather than a purely equilibrium order parameter symmetry. Further systematic studies, including current and temperature dependence, thickness scaling, and complementary probes of TRS breaking, will be necessary to clarify this distinction. The present measurements, together with recent reports in Bi-based cuprates ref7 ; ref9 , suggest that zero-field superconducting diode behavior may be a more general feature of high-temperature superconductors than previously recognized. While the data do not uniquely identify a microscopic mechanism, they impose a robust symmetry constraint: any comprehensive theory of cuprate superconductivity must accommodate the possibility of time-reversal symmetry breaking, whether intrinsic to the equilibrium state or emerging under finite current. Continued experimental investigation across different cuprate families and measurement geometries will be essential for determining the origin and universality of this phenomenon. V Summary The above considerations lead to the following conclusion: a viable theoretical model of HTSC in cuprates may need to incorporate TRS breaking. This significantly narrows the set of compatible theoretical proposals, although substantial uncertainty remains. Further constraints may arise from examining additional signatures predicted by specific theories – for example, structural modifications detectable through X-ray or neutron reflections. It is also possible that nonequilibrium electronic effects play a role in TRS breaking, as can be deduced from Fig. 5 . As follows from this figure, the diode effect emerges at sufficiently high current densities. We should also refer to Fig. 2, which shows that the same bridge with the diode effect at the applied current amplitude 35mA displays bipolar symmetry and markedly lower resistance at the current amplitude smaller by 30\% . Continued studies of the diode effect in HTSCs may therefore yield further insight into the underlying mechanism of superconductivity. Acknowledgements. This research was supported in part by the ONR Grant No. N00014-24-1-2595. We are grateful to the Physics Art Frontiers for the provided technical assistance. *