Suppression of antiferromagnetic order in the electron-doped cuprate $T^{\prime }\text{−}{\mathrm{La}}_{2-x}{\mathrm{Ce}}_x\mathrm{Cu}{\mathrm{O}}_{4\pm \delta }$

We performed systematic angle-resolved photoemission spectroscopy measurements in situ on $T^{\prime }\text{−}{\mathrm{La}}_{2-x}{\mathrm{Ce}}_x\mathrm{Cu}{\mathrm{O}}_{4\pm \delta }$ (LCCO) thin films over the extended doping range prepared by the refined ozone/vacuum annealing method. Electron doping level ($n$), estimated from the measured Fermi surface (FS) volume, varied from 0.05 to 0.23, fully encompassing the superconducting dome. We observed an absence of the insulating behavior around $n\sim 0.05$ and the shift of FS reconstruction to $n\sim 0.11$ in LCCO from $n\sim 0.15$ in other electron-doped cuprates, suggesting that the antiferromagnetic order is strongly suppressed in this material. The possible explanation may lie in the enhanced next-nearest-neighbor hopping in LCCO because of the largest ${\mathrm{La}}^{3+}$ ionic radius among all the lanthanide elements.

Figure 1

(a) Phase diagram of $T^{\prime }\text{−}{\mathrm{La}}_{2-x}{\mathrm{Ce}}_x\mathrm{Cu}{\mathrm{O}}_{4\pm \delta }$ (LCCO).The superconducting dome is referenced from transport measurements, and the purple antiferromagnetic region is from angular magnetoresistance data, respectively [8, 15]. The gray Fermi surface (FS) reconstruction (FSR) line is based on Hall measurements, indicating the phase transition of the FSR [16]. The blue marks represent the doping levels estimated from the FS volume of films by ozone/vacuum annealing, with the Ce concentrations $x$ corresponding to 0.1 (dots) and 0.19 (squares). Error bars are estimated from the uncertainty of determining the nodal $k_{\mathrm{F}}$. (b) Reflection high-energy electron diffraction patterns record the process of refreshing the film surface and tuning the doping level, and the patterns are from a pristine LCCO film exposed to air and after ozone/vacuum annealing. (c) Temperature dependence of resistivity for as-grown LCCO films on $\mathrm{SrTi}{\mathrm{O}}_3$ (STO) substrate with $x=0.1,0.19$. (d) X-ray diffraction pattern of a LCCO film with $x=0.1$. The peaks of Nb-doped STO substrate are marked in black color and the (00l) peaks of $T$′-LCCO are marked in red.

Figure 2

(a) The representative band dispersions measured at 10 K, which are referred to as the node, the hotspot, near the hotspot, and the antinode, respectively, as indicated in (b). The black arrows in (a) indicate the positions of the kinks. (b) Fourfold symmetrized FS measured at 30 K by integrating over $E_{\mathrm{F}}\pm 15\mathrm{meV}$; the locus of the highest intensity in the FS map is fitted by the tight-binding model, which displays as the red dashed curve. The estimated doping level is $n\sim 0.06$. The solid red lines are the antiferromagnetic Brillouin-zone boundaries. The absent spectral weight at the upper half of the FS is due to the matrix element effect. (c) The half width at half maximum of the momentum distribution curves (MDCs) for the four cuts.

Figure 3

(a1)–(a7) The evolution of FS for the films with $n\sim 0.05-0.23$. (b) The schematic FSs from the highly underdoped to heavily overdoped films. (c) and (d) Nodal and antinodal band dispersions as indicated in (a7) were acquired at $n\sim 0.23$. The red lines are the tight-binding fitting results with $t=0.3$. (e) Momentum distribution curves (MDCs) at the Fermi level ($E_{\mathrm{F}}$) for the band in (c) and (d).

Figure 4

(a)–(e) Doping evolution of band dispersions at the hotspot. Spectra at $n\sim 0.06$ and 0.084 were acquired at 10 K; others were measured at 30 K. The white dashed lines represent the $E_{\mathrm{F}}$. The energy distribution curves (EDCs) at $k_{\mathrm{F}}$ are summarized in (f). The black arrow indicates the leading-edge shift is ∼25 meV for $n\sim 0.05$. (g) Comparison of $-t^{\prime }/t$ for different systems LCCO, NCCO, SCCO, and ECCO at their optimal doping. Solid and dashed lines reflect the assumption $t^{\prime \prime }=0$ and $-t^{\prime \prime }/t^{\prime }=0.5$, respectively.

Figure 5

The FS for $n\sim 0.10$, 0.11, and 0.12. The FS reconstruction is clear for $n\sim 0.10$ but disappears for $n\sim 0.11$ and 0.12. The samples at $n\sim 0.10$ and 0.12 are annealed from a $x=0.15$ film.