betaFeedback

WorksPhysics and Astronomy

Electronic nematic order in the normal state of strontium ruthenate

Abstract

Despite significant achievements in characterizing the properties of Sr2_2RuO4_4 over the last three decades, the precise nature of its electronic ground state is still unresolved. In this work, we provide a missing piece of the puzzle by uncovering evidence of electronic nematic order in the normal state of Sr2_2RuO4_4, revealed by ultrafast time-resolved optical dichroism measurements of uniaxially strained thin films. This nematic order, whose domains are aligned by the strain, spontaneously breaks the four-fold rotational symmetry of the crystal. The temperature dependence of the dichroism resembles an Ising-like order parameter, and optical pumping induces a coherent oscillation of its amplitude mode. A microscopic model of intra-unit-cell nematic order is presented, highlighting the importance of Coulomb repulsion between neighboring oxygen pp-orbitals. The existence of electronic nematic order in the normal state of Sr2_2RuO4_4 may have consequences for the form and mechanism of superconductivity in this material.

The study of Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4 entered a new era after the recent observation of a reduction in the Knight shift at the onset of superconductivity [1], which overturned the long-favored chiral pp-wave triplet pairing scenario. Researchers are presently faced with an array of sometimes contradictory results, and it is clear that a fundamental piece of the Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4 puzzle is still missing [2]. A complete understanding of the superconducting phase in any material depends upon a detailed knowledge of the normal state electronic structure out of which it emerges. It has long been believed that the normal state of Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4 is a conventional quasi-two-dimensional Fermi liquid with moderate correlations [2, 3]. Evidence of any deviation from conventional Fermi liquid behavior would have significant consequences for both theoretical proposals and the interpretation of experiments. In this work, we present such evidence. Motivated by recent transverse resistivity experiments indicating possible electronic nematicity [4], we use optical dichroism measurements of high-quality epitaxially strained thin films to uncover evidence of electronic nematic order in the normal state of Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4.

Nematic order spontaneously breaks rotational symmetry while preserving translational symmetry. Nematicity in unconventional superconductors is far from rare, occurring in both the iron-based superconductors and the cuprates [5, 6, 7]. A nematic transition in these systems, however, is usually accompanied by a structural or magnetic transition. Such a transition is absent in Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4. In a tetragonal crystal, nematicity results in a lowering of symmetry to an orthorhombic unit cell where the aa and bb directions become nonequivalent. The spontaneous nature of the symmetry breaking typically leads to the formation of opposing domains related by 90∘90^\circ rotations, whose average over long length and time scales retains the parent tetragonal symmetry. Thus, directly detecting electronic nematic symmetry breaking with macroscopic probes generally requires stabilizing the formation of one type of domain over the other, akin to structural detwinning. Uniaxial strain is an ideal external conjugate field by which one can achieve this effect. In the iron-based superconductors, for example, uniaxial strain has been used successfully to study nematic order with optical, transport, and photoemission experiments [8, 9, 10, 11, 12]. Here, we take advantage of uniaxial epitaxial strain to enable macroscopic transient optical reflectivity measurements of electronic nematic order in Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4.

We study high-quality thin films of (001)-oriented Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4 grown by molecular-beam epitaxy, as described in Ref. 13. The films are coherently strained by growing them on (110) NdGaO3\mathrm{NdGaO}_3 single-crystal substrates [14], guaranteeing a clean, uniform strain field over the entire sample area. Electrical transport measurements confirm the films are superconducting, with a critical temperature of Tc≈1.4T_c \approx1.4 K (see Supplemental Materials [15]). NdGaO3\mathrm{NdGaO}_3 induces a small uniaxial (B1gB_{1g}) strain of a/b−1≈0.3%a/b - 1 \approx0.3\% at low temperatures, where aa and bb are the in-plane lattice constants of the coherently strained Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4 film. Specifically, aa and bb correspond to the [11ˉ0][1\bar{1}0] and [001][001] spacings, respectively, of the (110) NdGaO3\mathrm{NdGaO}_3 substrate in the PbnmPbnm setting [14], as illustrated in Figure 1(a). At room temperature, the uniaxial strain is reduced to ∼0.2%\sim0.2\%. Strain has been used extensively in recent years to study the Lifshitz transition in Sr2RuO4\mathrm{Sr}_2\mathrm{RuO}_4 and to enhance the superconducting critical temperature [16, 17, 18, 19, 20, 21, 22]. We emphasize, however, that the magnitude of uniaxial strain that we employ is not sufficient to induce a Lifshitz transition (estimated in Ref. 22 to be εxx−εyy≈0.7%\varepsilon_{xx}-\varepsilon_{yy} \approx0.7\%), and moreover the epitaxial strain is biaxially compressive (−0.1%-0.1\% in the aa direction and −0.4%-0.4\% in the bb direction at low temperatures) relative to bulk Sr2RuO4\mathrm{Sr}_{2}\mathrm{RuO}_{4} [23, 24], which is known to move the Fermi level away from the van Hove singularity rather than towards it [17]. Thus, the effects that we measure are due to a relatively weak uniaxial perturbation of the electronic structure and are not expected to be driven by density of states enhancements or changes in band topology.

Figure 1. Static and transient optical measurements of Sr2RuO4\mathrm{Sr}_{2}\mathrm{RuO}_{4}. (a) Crystallographic directions in the film and substrate. For Sr2RuO4\mathrm{Sr}_{2}\mathrm{RuO}_{4} on (110) NdGaO3_3, the aa-axis corresponds to [11ˉ0][1\bar{1}0] NdGaO3_3 and the bb-axis corresponds to [001][001] NdGaO3_3. (b) Static reflectivity anisotropy Ra/RbR_a/R_b versus temperature for films grown on NdGaO3_3 (SRO/NGO) and LSAT (SRO/LSAT). A fit to an Ising model of nematic order is shown for the NdGaO3_3 sample. Also shown is the temperature-dependent lattice constant ratio a/ba/b for films strained to NdGaO3_3 from Ref. 14. (c) Illustration of the pump-probe transient reflectivity experimental setup. The pump and probe beams remain fixed in a cross-polarized configuration while the sample is rotated about its surface normal. θ=0∘\theta= 0^\circ when the probe is polarized along the aa-axis (as shown) and θ=90∘\theta= 90^\circ when polarized along the bb-axis. (d) Raw transient reflectivity curves for several values of θ\theta between 0∘0^\circ and 90∘90^\circ at 12 K. A clear anisotropy is observed in the picosecond-scale transient optical response, with a peak at 0∘0^\circ and a flat step at 90∘90^\circ. (e) Polar plot of static reflectivity versus θ\theta, showing approximate isotropy. (f) Polar plot of ΔR/R\Delta R/R integrated from 0 to 2 ps [gray shaded region in (c)] versus θ\theta, showing pronounced anisotropy.

To determine whether signatures of electronic nematic order are evident in the optical response of Sr2RuO4\mathrm{Sr}_{2}\mathrm{RuO}_{4} at our probe wavelength of 800 nm, we first examine the static reflectivity anisotropy, which we define as the ratio of optical reflectivity with electric field polarized along

the aa-axis to that along the bb-axis of the film. Figure 1(b) shows the measured reflectivity anisotropy as a function of temperature, where a striking dichroism is apparent. We find a decrease in reflectivity along the (long) aa-axis relative to that along the (short) bb-axis of the film. Figure 1(b) shows the measured reflectivity anisotropy as a function of temperature, where a striking dichroism is apparent. We find a decrease in reflectivity along the (long) aa-axis relative to that along the (short) bb-axis, which becomes more pronounced at lower temperatures. At the lowest temperatures, we measure a maximum reflectivity anisotropy of 1.7%, more than five times larger than the uniaxial strain imposed by the substrate (0.3%). Similar behavior has been observed in detwinned FeSe, where the optical reflectivity anisotropy was taken as a proxy for the nematic order parameter [12]. To eliminate the possibility that the dichroism is merely due to structural orthorhombicity, we include in the figure the temperaturedependent lattice constant ratio a/ba/b for films strained to NdGaO3_3. Under the hypothesis that the reflectivity anisotropy is entirely due to simple a≠ba \ne b structural effects, we would expect a linear proportionality with the substrate orthorhombicity as it changes with temperature (see Supplemental Materials [15]). This is not what we observe. Instead, the dichroism grows by an order of magnitude upon cooling from room temperature to low temperature, while the substrate orthorhombicity only increases by 30%. As a control, we also measure the reflectivity anisotropy of a Sr2_2RuO4_4 thin film grown on (001) (LaAlO3_3)0.29_{0.29}(SrTa1/2_{1/2}Al1/2_{1/2}O3_3)0.71_{0.71} (LSAT), which is tetragonal and induces no uniaxial strain (a=ba=b) [25]. In this case, we detect no appreciable dichroism. Our static optical anisotropy measurements suggest that the electronic structure of Sr2_2RuO4_4 possesses either spontaneous nematic order (scenario I) or a large nematic susceptibility (scenario II). In scenario I, an electronic instability results in the formation of microscopic nematic domains fluctuating in space and time, with one orientation favored by the uniaxial strain over the other. This imbalance induces a net macroscopic anisotropy in the optical response of the electrons. In scenario II, the weak uniaxial strain drives a strong electronic nematic response via a large nematic susceptibility. Without the uniaxial strain, however, the electronic structure would retain tetragonal symmetry even at low temperatures. Both scenarios are consistent with an absence of dichroism in films grown on LSAT.

Now that signatures of electronic nematic order in the optical dichroism have been established, we next turn to measuring the ultrafast response of the order. As we will show, this reveals much larger signatures of nematicity than the static reflectivity and enables the extraction of dynamical information about the order. We use a standard pump-probe transient reflectivity technique to measure changes in the dichroism induced by an ultrafast (∼50\sim50 fs) pump pulse of wavelength 760 nm and fluence ∼25\sim25 μ\muJ/cm2^2. To eliminate any possible anisotropy due to the polarization of the pump, we keep the pump and probe pulses fixed in a cross-polarized configuration and rotate the sample about its surface normal, as illustrated in Figure 1(c). Like the static reflectivity, we uncover a pronounced anisotropy of the transient optical response. Figure 1(d) shows raw transients of the relative change in reflectivity ΔR/R\Delta R/R versus pump-probe time delay at 12 K for several angles between the aa-axis (0°) and the bb-axis (90°). A clear breaking of tetragonal symmetry is apparent: for optical polarization parallel to the aa-axis, we observe a sharp peak in ΔR/R\Delta R/R that decays over a ∼1\sim1 ps timescale, whereas for polarization parallel to the bb-axis, the peak vanishes and only a flat step remains. This nematic (C4→C2C_4 \to C_2) rotational symmetry breaking can be seen more clearly in a polar plot of the area under the transient reflectivity curve between 0 and 2 ps, as shown in Figure 1(f). Here, no symmetrization has been performed; independent reflectivity transients were collected over a full 360° angular range. While the static reflectivity shows a modest 1.7% anisotropy, the transient reflectivity exhibits a giant anisotropy exceeding 100%. Ultrafast optical anisotropies of this magnitude are routinely observed in the nematic phase of iron-based superconductors [26, 27, 28, 29, 30]. Sub-picosecond timescales are dominated by low-energy electronic degrees of freedom, affirming that the dichroism we observe is a result of electronic nematic order rather than structural orthorhombicity.

A saturation at low temperatures is evident in the static dichroism shown in Figure 1(b). This behavior resembles an Ising-like order parameter. To quantitatively analyze the data, we therefore develop a heuristic mean-field Ising model of the nematic order in Sr2_2RuO4_4. This model can describe both scenarios discussed above. Within the model, we define an order parameter ϕ\phi that takes the values +1+1 or −1-1, corresponding to a nematic director oriented along the aa- or bb-axis, respectively. The Hamiltonian is given by

H=−12∑i,jU(ri−rj)ϕiϕj−F∑iϕi,\mathcal{H}=-\frac{1}{2}\sum_{i,j}U(\mathbf{r}_i-\mathbf{r}_j)\phi_i\phi_j-F\sum_i\phi_i,

where ii and jj label unit cells, U(r)U(\mathbf{r}) is the interaction energy, related to the stiffness of the nematic order and the size of domains, and FF is the uniaxial strain field, which favors +1+1 domains over −1-1 domains. Within mean-field theory, the average order parameter ⟨ϕ⟩\langle\phi\rangle satisfies the self-consistency equation

⟨ϕ⟩=tanh⁡(U⟨ϕ⟩+FkBT),\langle\phi\rangle=\tanh\left(\frac{U\langle\phi\rangle+F}{k_{\mathrm{B}}T}\right),

where U=∑iU(ri)/2=kBTcU=\sum_i U(\mathbf{r}_i)/2=k_{\mathrm{B}}T_c defines the critical temperature of the nematic phase transition in the absence of the external strain field FF. Henceforth, we drop the angle brackets and call the Ising nematic order parameter ϕ\phi. By assuming the measured optical anisotropy is proportional to the order parameter (Ra/Rb−1∝ϕ)(R_a/R_b-1\propto\phi), we can fit the data to this functional form. Such a least-squares fit, included in Figure 1(b), yields the fitting parameters U=53±8U=53\pm8 K and F=53±7F=53\pm7 K. Reasonably good agreement with the data is achieved despite the relative simplicity of our model. Taking the model at face value, we predict a critical temperature of Tc≈50T_c\approx50 K, where electronic nematic order would spontaneously condense following scenario I described above. In the measured sample, however, this transition occurs gradually over a wide temperature range due to the large magnitude of the symmetry-breaking uniaxial strain field (i.e., F≈UF\approx U). Scenario II, where nematic order does not occur without strain, would require U≤0U\le0, which is not consistent with our data. Based on the success of this simple Ising nematic model, we will continue to use it to extract quantitative information about our data.

Turning now to the temperature dependence of the nematicity, Figure 2(a) shows transient reflectivity measurements at selected temperatures between 12 K and 300 K for polarization along the aa-axis (RaR_a) and bb-axis (RbR_b). To isolate temperature-dependent changes in the anisotropy from other effects, we also show the difference in ΔR/R\Delta R/R between the aa- and bb-axis (Ra−RbR_a - R_b) as well as the sum of the two (Ra+RbR_a + R_b). From this process, we see that the sharp peak in ΔR/R\Delta R/R is entirely due to the nematicity of the material, while the flat step after time-zero represents the isotropic component of the transient optical response. Moreover, while the isotropic response monotonically decreases with increasing temperature, as shown in Figure 2(b), we uncover an unusual non-monotonic temperature dependence of the nematic component of the response, displayed in Figure 2(c). The isotropic transient response is likely caused by an increase in temperature ΔT\Delta T after absorption of the pump. Under the assumption that each pump pulse deposits a fixed amount of energy ΔE\Delta E into the electronic subsystem (with heat capacity C=γTC = \gamma T), one can show that

Figure 2. Temperature dependence of transient reflectivity measurements. (a) ΔR/R\Delta R/R measurements for selected temperatures between 12 K and 300 K for polarization along the aa-axis (upper left) and the bb-axis (upper right), as well as their difference (lower left) and their sum (lower right). (b) Temperature dependence of the average value of Δ(Ra+Rb)/R\Delta(R_a + R_b)/R between 1.5 and 2 ps, representing the isotropic component of the dynamical response of the material. The response monotonically decreases with increasing temperature. The gray line is a fit to the sample temperature change, as described in the text. (c) Temperature dependence of the average value of Δ(Ra−Rb)/R\Delta(R_a - R_b)/R between 0 and 2 ps, representing the nematic component of the dynamical response of the material. An unusual non-monotonic response is observed. The gray line is a guide to the eye. (d) Two-dimensional depiction of the joint time delay and temperature dependence of Δ(Ra−Rb)/R\Delta(R_a - R_b)/R. (e) Corresponding depiction generated by fitting to the Ising model of nematic order discussed in the text.

ΔTmax⁡(T)=T2+(2ΔE/γ)−T.\Delta T_{\max}(T) = \sqrt{T^2 + (2\Delta E/\gamma)} - T.

Using the value of γ\gamma from Ref. 31 and ΔE\Delta E calculated from our pump fluence, we find (2ΔE/γ)≈(80 K)2(2\Delta E/\gamma) \approx(80\ \mathrm{K})^2 (see Supplemental Materials [15]). A function of the form g(T)=α+βΔTmax⁡(T)g(T) = \alpha+ \beta\Delta T_{\max}(T) with only two free parameters, α\alpha and β\beta, fits the isotropic component extremely well, as shown by the gray line in Figure 2(b). We therefore use this fact to scale the data at each temperature to match g(T)g(T) and correct for laser fluctuations during the duration of the experiment, where, for example, the two data points near ∼35\sim35 K are otherwise anomalously low. This normalization procedure is used for the remainder of our data analysis, including Figure 2(c).

Figure 1(b) shows that the onset of nematic order reduces the optical reflectivity along the aa-axis. After the pump pulse, this reflectivity increases. It follows that the effect of the pump is to suppress the nematic order, likely through a transient increase in temperature. To model this, we assume that the dichroism is proportional to the nematic order parameter ϕ\phi. The measured change in dichroism is therefore proportional to the change in ϕ\phi:

Δ(Ra−Rb)/R∝(1+δ)ϕ(T+ΔT)−ϕ(T),\Delta(R_a - R_b)/R \propto(1+\delta)\phi(T+\Delta T)-\phi(T),

where ϕ(T)\phi(T) is the equilibrium value of the order parameter at the temperature TT of the measurement, ΔT\Delta T is the increase in electronic temperature after absorption of the pump energy, and δ<0\delta< 0 represents a possible non-thermal suppression of the order parameter amplitude by the pump pulse. The previously developed Ising model provides a concrete expression for ϕ(T)\phi(T). Due to equilibration, the temperature increase ΔT\Delta T is not necessarily instantaneous. Instead, we allow the temperature of the electrons participating in the nematic order to rise to some time-dependent fraction 0≤f(t)≤10 \le f(t) \le1 of ΔTmax⁡\Delta T_{\max} calculated from the pump fluence as discussed above [ΔT(t)=f(t)ΔTmax⁡][\Delta T(t) = f(t)\Delta T_{\max}]. In doing so, we assume a temperature-independent electron equilibration rate. With this model, we fit the transient optical data at all time delays and all temperatures simultaneously. The fit includes three global parameters (UU, FF, and a fixed scaling factor) and two time-dependent parameters [δ(t)\delta(t) and f(t)f(t)]. The results of the fit, shown in Figure 3(a), are excellent (R2=0.994R^2=0.994).

Figure 3. Ising model fit of optical anisotropy. (a) Measured Δ(Ra−Rb)/R\Delta(R_a-R_b)/R versus temperature for selected time delays after the pump pulse. Curves are offset by 10−410^{-4} for clarity. Gray lines show the global fit to the Ising model of electronic nematic order, as described in the text. (b) Confidence regions for the Ising parameters UU and FF computed through an FF-test. U>0U>0 with greater than 99% confidence, implying the existence of spontaneous nematic order. The inset shows the order parameter ϕ(T)\phi(T) together with the critical temperature TcT_c for spontaneous nematic order when F=0F=0. (c) Electronic temperature increase versus time delay as a fraction of the calculated maximum increase. Also shown is a fit to the solution of the heat equation df/dt=k(f0−f)df/dt=k(f_0-f) (purple curve) with background electronic temperature f0f_0 (gray curve). (d) Relative suppression of the nematic order versus time delay, showing a coherent oscillation of the order parameter amplitude. Also shown is a fit to a damped harmonic oscillator (purple curve) with a pump-induced displacement of the equilibrium amplitude (gray curve).

A number of salient features of the data are uncovered by the fitting analysis: (1) The unusual non-monotonic temperature dependence of the dichroism [Figure 2(c)] is a result of the flattening of the nematic order parameter at low temperatures (dϕ/dT→0d\phi/dT \to0 as T→0T \to0), which reduces the effect of pump heating. (2) Uncertainties in the fit parameters UU and FF are correlated, which is a result of their coupled influence on the functional form of ϕ(T)\phi(T). The FF-test-derived confidence regions displayed in Figure 3(c), however, show that U>0U>0 with high confidence. Within the applicability of our Ising model, this implies that the nematic order we observe is spontaneous (scenario I).

taneuous (scenario I). (3) While we extract a value of FF that is similar to that of the static reflectivity measurement (46 K versus 53 K), the value of UU in the transient optical measurements is substantially smaller (23 K versus 53 K). We conjecture that repeatedly exciting the sample with high-frequency pump and probe pulses does not allow the sample to completely return to equilibrium, causing a net suppression of the order parameter. This subsequently results in a renormalization of the average interaction energy UU. (4) Figure 3(c) shows that the nematic electron temperature rises to ∼\sim50% of the maximum calculated increase ΔTmax⁡\Delta T_{\max} over a timescale of ∼\sim0.5 ps. This suggests that the energy of the pump is first absorbed by background electrons during the lifetime of the pulse (\sim$50 fs) and then is quickly transferred to the nematic electrons via thermal equilibration mediated by electron-electron interactions. At a much slower rate dictated by electron-phonon interactions, thermal equilibration with the lattice commences, causing a gradual cooling. To make this picture more quantitative, we fit f(t)f(t) to the solution of the heat equation df/dt=k(f0−f)df/dt=k(f_0-f), where f0(t)=e−t/τf_0(t)=e^{-t/\tau} represents the background electronic temperature, with lattice equilibration timescale τ\tau. We find excellent agreement with the data for τ=0.82\tau=0.82 ps. (5) The nonthermal suppression of the order parameter amplitude after the pump pulse [Figure 3(d)] reveals a coherent oscillation. This behavior can be understood within the context of a simple damped harmonic oscillator model: d2δ/dt2+2γdδ/dt+ω02(δ−δ0)=0d^2\delta/dt^2+2\gamma d\delta/dt+\omega_0^2(\delta-\delta_0)=0, where γ\gamma is the damping rate, ω0\omega_0 is the natural frequency of oscillation, and δ0\delta_0 is the pump-induced displacement of the equilibrium order parameter amplitude. We find that a solution to this equation fits the data well if δ0(t)=αe−t/τ+β\delta_0(t)=\alpha e^{-t/\tau}+\beta, representing a sharp impulsive suppression at t=0t=0 followed by a rapid recovery (τ=0.09\tau=0.09 ps) to a displaced equilibrium amplitude β=−5.0%\beta=-5.0\% relative to the initial value, as shown by the gray curve in Figure 3(d). We determine a frequency of ω0/2π=0.89\omega_0/2\pi=0.89 THz for the amplitude mode of the nematic order.

The foregoing optical experiments suggest that Sr2_2RuO4_4 supports a high-temperature Ising-like nematic order parameter with an order-disorder phase transition below ∼\sim50 K. To put forth a possible microscopic origin for this order, we turn to the Emery model, first used to describe hole doping of CuO2_2 planes in cuprate superconductors away from half filling [32]. Here, we use the model to study intra-unit-cell nematic order in RuO2_2 planes with fillings adjacent to the van Hove singularity of the γ\gamma band [33]. Our mean-field analysis closely follows Ref. 34 (see Supplementary Materials [15]). Figure 4(a) shows the hopping and interaction parameters defined within the model, and Figure 4(b) shows the resulting band structure. Of central importance are the Coulomb repulsion between neighboring oxygen pp-orbitals (VppV_{pp}), which favors a nematic charge order η=nxp−nyp\eta=n_x^p-n_y^p with different electron densities on the pxp_x and pyp_y orbitals, and the van Hove singularity of the γ\gamma band at the XX point, which leads to a diverging density of states that supports electronic instabilities by reducing the energy cost to redistribute orbital density. Figure 4(c) shows the mean-field ground state phase diagram as a function of VppV_{pp} and filling. A nematic phase is stabilized for sufficiently large values of VppV_{pp} that depend on the proximity to the van Hove filling. Bulk Sr2_2RuO4_4 has n−nvH≈−0.2n - n_{\mathrm{vH}} \approx-0.2 [33], which requires an interaction strength Vpp>1.7V_{pp} > 1.7 eV to stabilize nematic order. The full temperature–filling phase diagram is displayed in Figure 4(d), where it is seen that such order, if it exists, condenses well above room temperature. The model shows that Sr2_2RuO4_4 is exceptionally close to a nematic instability, and offers a plausible microscopic picture of an Ising-like (η=±η0\eta= \pm\eta_0) nematic charge order. We conjecture that at high temperatures this nematicity is spatially disordered at the nanoscale, but shows an order-disorder phase transition to a globally nematic state at lower temperatures.

Figure 4. Microscopic model of nematic order in Sr2_2RuO4_4. (a) Unit cell of the RuO2_2 plane showing the three-orbital basis (Ru dxyd_{xy} and O pxp_x and pyp_y) of the Emery model. Hopping integrals (solid lines) and interaction parameters (dashed lines) are defined following Ref. 34. (b) Resulting tight-binding band structure, with parameters taken from Ref. 35. (c) Interaction–filling phase diagram showing the critical value of VppV_{pp} necessary to produce a nematic ground state (η≠0\eta\ne0). (d) Temperature–filling phase diagram for different values of VppV_{pp}. Panels (c) and (d) were calculated for Up=4.4U_p = 4.4 eV [35].

In conclusion, we have presented static and ultrafast time-resolved optical dichroism measurements of epitaxially strained thin films of Sr2_2RuO4_4 that strongly support the existence of electronic nematic order. By fitting our data to a simple Ising model, we conjecture that this order emerges spontaneously at low temperatures through an order-disorder transition, rather than through a large nematic susceptibility driven by strain. These results corroborate the angle-resolved transverse resistivity measurements that first uncovered electronic nematicity in Sr2_2RuO4_4 [4]. Electrical transport and optical reflectivity are disparate probes operating at different energy scales. In both cases, however, clear signatures of two-fold electronic anisotropy are evident. Furthermore, while transport can be sensitive to non-local effects such as percolation and boundary scattering, especially in a material with ostensible microscopic nematic domains, optical reflectivity is a local, bulk-sensitive probe. This difference may explain the minor discrepancies between experiments, such as the absence of nematic order in tetragonal Sr2_2RuO4_4/LSAT and the alignment of the nematic director with respect to the crystallographic directions. Our findings also offer a new perspective on the Fermi liquid crossover at TFL∼40T_{\mathrm{FL}} \sim40 K observed by Hall transport [36, 37, 38], optical spectroscopy [39], and nuclear magnetic resonance [40, 41]. We speculate that the crossover is directly related to nematicity, which appears to develop at the same temperature. Indeed, such a Fermi liquid crossover has been theoretically shown to emerge in the vicinity of an Ising nematic quantum critical point [42, 43, 44]. Nematic order also offers an explanation for the checkerboard charge order observed at the surface of Sr2_2RuO4_4 by scanning tunneling microscopy [45]. Rather than surface nematic order emerging as a secondary effect, we find it likely that the checkerboard charge order is the secondary response, emerging from a combination of bulk nematicity and surface octahedral rotations. An important open question is to what extent electronic nematic order in the normal state of Sr2_2RuO4_4 influences the likelihood of different forms of superconductivity at lower temperatures. Future experimental and theoretical work is necessary to address this principal question.

Acknowledgments

We thank Dr. Ludi Miao for performing electrical resistivity measurements. This work was primarily supported by the U.S. Department of Energy, Office of Basic Energy Sciences under Award No. DE-SC0019414. Thin film synthesis was funded in part by the Gordon and Betty Moore Foundation’s EPiQS Initiative through Grant Nos. GBMF3850 and GBMF9073 to Cornell University. Substrate preparation was, in part, facilitated by the Cornell NanoScale Facility, a member of the National Nanotechnology Coordinated Infrastructure (NNCI), which is supported by the National Science Foundation (Grant No. NNCI-2025233). Additional support was provided by AFOSR FA9550-21-1-0168.

References

References

  1. [1]A. Pustogow, Y. Luo, A. Chronister, Y.-S. Su, D. A. Sokolov, F. Jerzembeck, A. P. Mackenzie, C. W. Hicks, N. Kikugawa, S. Raghu, E. D. Bauer, S. E. Brown. Constraints on the superconducting order parameter in Sr₂RuO₄ from oxygen-17 nuclear magnetic resonance. Nature 574, 72 (2019).DOI
  2. [2]A. P. Mackenzie, T. Scaffidi, C. W. Hicks, Y. Maeno. Even odder after twenty-three years: the superconducting order parameter puzzle of Sr₂RuO₄. npj Quantum Mater. 2, 40 (2017).arxiv.org/abs/1706.01942
  3. [3]A. P. Mackenzie, Y. Maeno. The superconductivity of Sr₂RuO₄ and the physics of spin-triplet pairing. Rev. Mod. Phys. 75, 657 (2003).DOI
  4. [4]J. Wu, H. P. Nair, A. T. Bollinger, X. He, I. Robinson, N. J. Schreiber, K. M. Shen, D. G. Schlom, I. Božović. Electronic nematicity in Sr₂RuO₄. Proc. Natl. Acad. Sci. 117, 10654 (2020).DOI
  5. [5]T.-M. Chuang, M. P. Allan, J. Lee, Y. Xie, N. Ni, S. L. Bud’ko, G. S. Boebinger, P. C. Canfield, J. C. Davis. Nematic Electronic Structure in the “Parent” State of the Iron-Based Superconductor Ca(Fe₁−ₓCoₓ)₂As₂. Science 327, 181 (2010).
  6. [6]R. Daou, J. Chang, D. LeBoeuf, O. Cyr-Choinière, F. Laliberté, N. Doiron-Leyraud, B. J. Ramshaw, R. Liang, D. A. Bonn, W. N. Hardy, L. Taillefer. Broken rotational symmetry in the pseudogap phase of a high-Tc superconductor. Nature 463, 519 (2010).
  7. [7]A. E. Böhmer, J.-H. Chu, S. Lederer, M. Yi. Nematicity and nematic fluctuations in iron-based superconductors. Nat. Phys. 18, 1412 (2022).
  8. [8]A. Dusza, A. Lucarelli, F. Pfuner, J.-H. Chu, I. R. Fisher, L. Degiorgi. Anisotropic charge dynamics in detwinned Ba(Fe₁−ₓCoₓ)₂As₁−ₓ. EPL 93, 37002 (2011).
  9. [9]M. A. Tanatar, E. C. Blomberg, A. Kreyssig, M. G. Kim, N. Ni, A. Thaler, S. L. Bud’ko, P. C. Canfield, A. I. Goldman, I. I. Mazin, R. Prozorov. Uniaxial-strain mechanical detwinning of CaFe₂As₂ and BaFe₂As₂ crystals: Optical and transport study. Phys. Rev. B 81, 184508 (2010).DOI
  10. [10]J.-H. Chu, J. G. Analytis, K. De Greve, P. L. McMahon, Z. Islam, Y. Yamamoto, I. R. Fisher. In-Plane Resistivity Anisotropy in an Underdoped Iron Arsenide Superconductor. Science 329, 824 (2010).DOI
  11. [11]M. Yi, D. Lu, J.-H. Chu, J. G. Analytis, A. P. Sorini, A. F. Kemper, B. Moritz, S.-K. Mo, R. G. Moore, M. Hashimoto, W.-S. Lee, Z. Hussain, T. D. Devereaux, I. R. Fisher, Z.-X. Shen. Symmetry-breaking orbital anisotropy observed for detwinned Ba(Fe₁−ₓCoₓ)₂As₂ above the spin density wave transition. Proc. Natl. Acad. Sci. 108, 6878 (2011).DOI
  12. [12]M. Chinotti, A. Pal, L. Degiorgi, A. E. Böhmer, P. C. Canfield. Optical anisotropy in the electronic nematic phase of FeSe. Phys. Rev. B 96, 121112(R) (2017).
  13. [14]D. Savytskii, L. Vasylenchko, A. Senyshyn, A. Matkovskii, C. Bähtz, M. L. Sanjuán, U. Bismayer, M. Berkowski. Low-temperature structural and Raman studies on rare-earth gallates. Phys. Rev. B 68, 024101 (2003).
  14. [15]See Supplemental Material at [URL will be inserted by publisher] for experimental methods, nematicity across several relaxation timescales, measurements of film on an LSAT substrate, calculation of pump-induced temperature change, and microscopic model of nematic charge order.
  15. [16]C. W. Hicks, D. O. Brodsky, E. A. Yelland, A. S. Gibbs, J. A. N. Bruin, M. E. Barber, S. D. Edkins, K. Nishimura, S. Yonezawa, Y. Maeno, A. P. Mackenzie. Strong Increase of T_c of Sr₂RuO₄ Under Both Tensile and Compressive Strain. Science 344, 283 (2014).DOI
  16. [17]B. Burganov, C. Adamo, A. Mulder, M. Uchida, P. D. C. King, J. W. Harter, D. E. Shai, A. S. Gibbs, A. P. Mackenzie, R. Uecker, M. Bruetzam, M. R. Beasley, C. J. Fennie, D. G. Schlom, K. M. Shen. Strain Control of Fermilogy and Many-Body Interactions in Two-Dimensional Ruthenates. Phys. Rev. Lett. 116, 197003 (2016).DOI
  17. [18]A. Steppke, L. Zhao, M. E. Barber, T. Scaffidi, F. Jerzembek, H. Rosner, A. S. Gibbs, Y. Maeno, S. H. Simon, A. P. Mackenzie, C. W. Hicks. Strong peak in T_c of Sr₂RuO₄ under uniaxial pressure. Science 355, 148 (2017).
  18. [19]V. Sunko, E. A. Morales, I. Marković, M. E. Barber, D. Milosavljević, F. Mazzola, D. A. Sokolov, N. Kikugawa, C. Cacho, P. Dudin, H. Rosner, C. W. Hicks, P. D. C. King, A. P. Mackenzie. Direct observation of a uniaxial stress-driven Lifshitz transition in Sr₂RuO₄. npj Quantum Mater. 4, 46 (2019).DOI
  19. [20]V. Grinenko, S. Ghosh, R. Sarkar, J.-C. Orain, A. Nikitin, M. Elender, D. Das, Z. Guguchia, F. Brückner, M. E. Barber, J. Park, N. Kikugawa, D. A. Sokolov, J. S. Bobowski, T. Miyoshi, Y. Maeno, A. P. Mackenzie, H. Luetkens, C. W. Hicks, H.-H. Klauss. Split superconducting and time-reversal symmetry-breaking transitions in Sr₂RuO₄ under stress. Nat. Phys. 17, 748 (2021).DOI
  20. [21]Y.-S. Li, N. Kikugawa, D. A. Sokolov, F. Jerzembeck, A. S. Gibbs, Y. Maeno, C. W. Hicks, J. Schmalian, M. Nicklas, A. P. Mackenzie. High-sensitivity heat-capacity measurements on Sr₂RuO₄ under uniaxial pressure. Proc. Natl. Acad. Sci. 118, e2020492118 (2021).DOI
  21. [22]M. E. Barber, F. Lechermann, S. V. Streltsov, S. L. Skornyakov, S. Ghosh, B. J. Ramshaw, N. Kikugawa, D. A. Sokolov, A. P. Mackenzie, C. W. Hicks, I. I. Mazin. Role of correlations in determining the Van Hove strain in Sr₂RuO₄. Phys. Rev. B 100, 245139 (2019).
  22. [23]T. Vogt, D. J. Buttrey. Low-temperature structural behavior of Sr₂RuO₄. Phys. Rev. B 52, R9843 (1995).pubmed.ncbi.nlm.nih.gov/9980128
  23. [24]O. Chmaissem, J. D. Jorgensen, H. Shaked, S. Ikeda, Y. Maeno. Thermal expansion and compressibility of Sr₂RuO₄. Phys. Rev. B 57, 5067 (1998).DOI
  24. [25]M. Steins, J. Doerschel, P. Reiche. Crystal structure of aluminium lanthanum strontium tantalum oxide, (La₀.₂₇₂Sr₀.₇₂₈)(Al₀.₆₄₈Ta₀.₃₅₂)O₃. Zeitschrift für Kristallographie - New Crystal Structures 212, 77 (1997).
  25. [26]L. Stojchevska, T. Mertelj, J.-H. Chu, I. R. Fisher, D. Mihailovic. Doping dependence of femtosecond quasiparticle relaxation dynamics in Ba(Fe,Co)₂As₂ single crystals: Evidence for normal-state nematic fluctuations. Phys. Rev. B 86, 024519 (2012).DOI
  26. [27]C.-W. Luo, P. C. Cheng, S.-H. Wang, J.-C. Chiang, J.-Y. Lin, K.-H. Wu, J.-Y. Juang, D. A. Chareev, O. S. Volkova, A. N. Vasiliev. Unveiling the hidden nematicity and spin subsystem in FeSe. npj Quantum Matter 2, 32 (2017).
  27. [28]S. Liu, C. Zhang, Q. Deng, H. Wen, J. Li, E. E. M. Chia, X. Wang, M. Xiao. Transient electronic anisotropy in overdoped NaFe1−xCoxAs superconductors. Phys. Rev. B 97, 020505(R) (2018).DOI
  28. [29]E. Thewalt, I. M. Hayes, J. P. Hinton, A. Little, S. Patankar, L. Wu, T. Helm, C. V. Stan, B. Tamura, J. G. Analytis, J. Orenstein. Imaging Anomalous Nematic Order and Strain in Optimally Doped BaFe2(As,P)2. Phys. Rev. Lett. 121, 027001 (2018).DOI
  29. [30]M.-C. Lee, I. Kwak, Y. Lee, B. Lee, B. C. Park, T. Wolf, T. W. Noh, K. Kim. Nematic response revealed by coherent phonon oscillations in BaFe2As2. Phys. Rev. B 105, 024501 (2022).DOI
  30. [32]V. J. Emery. Theory of high-Tc superconductivity in oxides. Phys. Rev. Lett. 58, 2794 (1987).pubmed.ncbi.nlm.nih.gov/10034851
  31. [33]K. M. Shen, N. Kikugawa, C. Bergemann, L. Balicas, F. Baumberger, W. Meevasana, N. J. C. Ingle, Y. Maeno, Z.-X. Shen, A. P. Mackenzie. Evolution of the Fermi Surface and Quasiparticle Renormalization through a van Hove Singularity in Sr2−yLayRuO4. Phys. Rev. Lett. 99, 187001 (2007).
  32. [35]K. Rościszewski, A. M. Oleś. Charge-transfer model for the electronic structure of layered ruthenates. Phys. Rev. B 91, 155137 (2015).DOI
  33. [36]N. Shirakawa, K. Murata, Y. Nishihara, S. Nishizaki, Y. Maeno, T. Fujita, J. G. Bednorz, F. Lichtenberg, N. Hamada. Novel Hall-Coefficient Behavior in Superconducting Sr2RuO4. J. Phys. Soc. Jpn. 64, 1072 (1995).DOI
  34. [37]A. P. Mackenzie, N. E. Hussey, A. J. Diver, S. R. Julian, Y. Maeno, S. Nishizaki, T. Fujita. Hall effect in the two-dimensional metal Sr2RuO4. Phys. Rev. B 54, 7425 (1996).pubmed.ncbi.nlm.nih.gov/9984367
  35. [38]M. Zingl, J. Mravlje, M. Aichhorn, O. Parcollet, A. Georges. Hall coefficient signals orbital differentiation in the Hund’s metal Sr2RuO4. npj Quantum Mater. 4, 35 (2019).
  36. [39]D. Stricker, J. Mravlje, C. Berthod, R. Fittipaldi, A. Vecchione, A. Georges, D. van der Marel. Optical Response of Sr2RuO4 Reveals Universal Fermi-Liquid Scaling and Quasiparticles Beyond Landau Theory. Phys. Rev. Lett. 113, 087404 (2014).
  37. [40]T. Imai, A. W. Hunt, K. R. Thurber, F. C. Chou. 17O NMR Evidence for Orbital Dependent Ferromagnetic Correlations in Sr2RuO4. Phys. Rev. Lett. 81, 3006 (1998).
  38. [41]A. Chronister, M. Zingl, A. Pustogow, Y. Luo, D. A. Sokolov, F. Jerzembeck, N. Kikugawa, C. W. Hicks, J. Mravlje, E. D. Bauer, J. D. Thompson, A. P. Mackenzie, A. Georges, S. E. Brown. Tuning the Fermi liquid crossover in Sr2RuO4 with uniaxial stress. npj Quantum Mater. 7, 113 (2022).arxiv.org/abs/2111.05570
  39. [42]X. Wang, E. Berg. Scattering mechanisms and electrical transport near an Ising nematic quantum critical point. Phys. Rev. B 99, 235136 (2019).DOI
  40. [43]V. S. de Carvalho, R. M. Fernandes. Resistivity near a nematic quantum critical point: Impact of acoustic phonons. Phys. Rev. B 100, 115103 (2019).DOI
  41. [44]L. E. Vieira, V. S. de Carvalho, H. Freire. DC resistivity near a nematic quantum critical point: Effects of weak disorder and acoustic phonons. Ann. Phys. (NY) 419, 168230 (2020).DOI
  42. [45]C. A. Marques, L. C. Rhodes, R. Fittipaldi, V. Granata, C. M. Yim, R. Buzio, A. Gerbi, A. Vecchione, A. W. Rost, P. Wahl. Magnetic-Field Tunable Intertwined Checkerboard Charge Order and Nematicity in the Surface Layer of Sr2RuO4. Adv. Mater. 33, 2100593 (2021).DOI

Paper details

vlnc:W_u37eq2r8
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents