Observation of two distinct pairs fluctuation lifetimes and supercurrents in the pseudogap regime of cuprate junctions

Pairs fluctuation supercurrents and inverse lifetimes in the pseudogap regime are reported. These were measured on epitaxial c-axis junctions of the cuprates, with a ${\text{PrBa}}_2{\text{Cu}}_3{\text{O}}_{7-\delta }$ barrier sandwiched in between two ${\text{YBa}}_2{\text{Cu}}_3{\text{O}}_{7-\delta }$ or doped ${\text{YBa}}_2{\text{Cu}}_3{\text{O}}_y$ electrodes, with or without magnetic fields parallel to the a-b planes. All junctions had a $T_c\left(\text{high}\right)\approx 85–90$ K and a $T_c\left(\text{low}\right)\approx 50–55$ K electrodes, allowing us to study pairs fluctuation supercurrents and inverse lifetimes in between these two temperatures. In junctions with a pseudogap electrode under zero field, an excess current due to pair fluctuations was observed which persisted at temperatures above $T_c\left(\text{low}\right)$, in the pseudogap regime, and up to about $T_c\left(\text{high}\right)$. No such excess current was observed in junctions without an electrode with a pseudogap. The measured conductance spectra at temperatures above $T_c\left(\text{low}\right)$ were fitted using a modified fluctuations model by Scalapino [Phys. Rev. Lett. 24, 1052 (1970)] of a junction with a serial resistance. We found that in the pseudogap regime, the conductance vs voltage consists of a narrow peak sitting on top of a very broad peak. This yielded two distinct pairs fluctuation lifetimes in the pseudogap electrode which differ by an order of magnitude up to about $T_c\left(\text{high}\right)$. Under in-plane fields, these two lifetime values remain separated in two distinct groups, which varied with increasing field moderately. We also found that detection of Amperian pairing [Phys. Rev. X 4, 031017 (2014)] in our cuprate junctions is not feasible, due to Josephson vortices penetration into the superconducting electrodes which drove the necessary field above the depairing field.

Figure 1

(a) A schematic drawing of a $c$-axis junction cross section. The trilayer base electrode (of the CJ-2 wafer here) is comprised of 100-nm-thick ${\text{YBa}}_2{\text{Co}}_{0.3}{\text{Cu}}_{2.7}{\text{O}}_y$ on 25 nm ${\text{PrBa}}_2{\text{Cu}}_3{\text{O}}_{7-\delta }$ on 200 nm ${\text{YBa}}_2{\text{Cu}}_3{\text{O}}_{7-\delta }$, and the gold cover electrode is 400 nm thick. (b) Top view of the whole wafer. The black squares and the gold coating (yellow) are for the $4\times 10$ contacts while the ten junctions are located in the central part of the wafer. (c) Atomic force microscope image of a single junction.

Figure 2

Transport results of junctions on the CJ-2 wafer. (a) Top right panel shows zero field cooled resistance versus temperature of five junctions, where the right inset shows a zoom in on the proximity region of YBCO-PrBCO just below 90 K, and the left inset a zoom in on the YBCoCO transition at 55 K. (b) Conductance spectra of the J2-3 junctions on this wafer under zero field and different temperatures, while (c) shows the corresponding supercurrent $I_c=V_c\times \mathrm{\Delta }(dI/dV)$ versus temperature.

Figure 3

Transport results of junctions J2-3 on the CJ-2 wafer at 76.5 K under different in-plane $H_{\left|\right|}$ magnetic fields. (a) The conductance spectra while (b) shows the corresponding supercurrents versus field.

Figure 4

(a) Conductance spectra at different temperatures of the J10 junction on the CJ-5 wafer. The $T_c$ of the overdoped electrode is estimated to be 54 K from the junction resistance drop versus temperature. (b) The conductance spectra at 74 K under different parallel magnetic fields. In (c) the conductance values of (b) at a constant 50 mA bias current (at about 5 mV) are plotted versus $H_{\left|\right|}$, with a parabolic fit. The dominant $H^2$ term in this fit indicates the existence of flux flow conductivity originating in supercurrent in the YBCO lead to the junction.

Figure 5

Conductance spectra at zero field and different temperatures of junction J10 on the CJ-6 wafer are shown in (a). (b) A zoom-in on spectra of (a) at high temperatures, where the top-hat peak at 45.6 K narrows down with increasing $T$ and eventually vanishes at 52.8 K. (c) The excess current $I_c=V_c\times \mathrm{\Delta }(dI/dV)$ extracted from (a) and (b) as described in Fig. 2. In (d), the low-$V$ spectrum of (b) at 49.9 K is fitted using Eqs. (5) and (6) for a single fluctuations peak and a superposition of two such peaks. See Fig. S2 of the Supplemental Material for details [29].

Figure 6

Conductance spectra of junctions J6-8 of CJ-2 under zero field at different temperatures showing the development of the broad background peak with increasing temperature. The inset shows that under a low-$V$ scan, the shape of the near zero bias peak is changed from top hat at 76.4 K into a standard fluctuations peak at 80.2 K [19]. Both fits at 69.5 K in (a) and at 80.2 K in the inset are fits to a double peak with the details given in the text and Supplemental Material [29].

Figure 7

(a) Conductance spectra at 79 K of the J6-8 junctions on the CJ-2 wafer under 0 and 8 T in-plane magnetic fields versus the voltage drop on the junctions $V_J$ after solving Eq. (9) (first iteration), together with the corresponding fits to a superposition of two peaks as in Eq. (6). (b)–(d) Summarize the results of the fit parameters versus field (a constant $R_0=0.16\mathrm{\Omega }$ was used). Raw conductance spectrum at a similar temperature of 79.3 K and 0 T can be seen in Fig. 6. Clearly there is a strong compression of the $V$ scale of the raw data compared to the $V_J$ scale here, by a factor of about 6–7.