Infrared Nano-Imaging of Electronic Phase across the Metal--Insulator Transition of NdNiO$_3$ Films

Fanwei Liu(刘凡微)$^{1\dag}$, Sisi Huang(黄思思)$^{2\dag}$, Sidan Chen(陈思丹)$^{1}$, Xinzhong Chen(陈欣中)$^{3}$,\\ Mengkun Liu(刘梦昆)$^{3}$, Kuijuan Jin(金奎娟)$^{2\hyperlink{s}{*}}$, and Xi Chen(陈曦)$^{1\hyperlink{s}{*}}$

doi:10.1088/0256-307X/39/7/076801

⇦Chinese Physics Letters, 2022, Vol. 39, No. 7, Article code 076801⇨ Infrared Nano-Imaging of Electronic Phase across the Metal–Insulator Transition of NdNiO$_3$ Films Fanwei Liu (刘凡微)1†, Sisi Huang (黄思思)2†, Sidan Chen (陈思丹)1, Xinzhong Chen (陈欣中)3, Mengkun Liu (刘梦昆)3, Kuijuan Jin (金奎娟)2*, and Xi Chen (陈曦)1* Affiliations 1State Key Laboratory of Low-Dimensional Quantum Physics and Department of Physics, Tsinghua University, Beijing 100084, China 2Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China 3Department of Physics and Astronomy, Stony Brook University, Stony Brook, NY 11794, USA Received 14 April 2022; accepted; published online 17 June 2022 ^†These authors contributed equally to this work.
^*Corresponding authors. Email: kjjin@iphy.ac.cn; xc@mail.tsinghua.edu.cn Citation Text: Liu F W, Huang S S, Chen S D et al. 2022 Chin. Phys. Lett. 39 076801 Abstract NdNiO$_3$ is a typical correlated material with temperature-driven metal–insulator transition. Resolving the local electronic phase is crucial in understanding the driving mechanism behind the phase transition. Here we present a nano-infrared study of the metal–insulator transition in NdNiO$_3$ films by a cryogenic scanning near-field optical microscope. The NdNiO$_3$ films undergo a continuous transition without phase coexistence. The nano-infrared signal shows significant temperature dependence and a hysteresis loop. Stripe-like modulation of the optical conductivity is formed in the films and can be attributed to the epitaxial strain. These results provide valuable evidence to understand the coupled electronic and structural transformations in NdNiO$_3$ films at the nano-scale.

DOI:10.1088/0256-307X/39/7/076801 © 2022 Chinese Physics Society Article Text Perovskite rare-earth nickelates ($R$NiO$_3$) are correlated electron systems with charge-transfer type metal–insulator transition (MIT) driven by temperature.[1–3] The mechanism of MIT in $R$NiO$_3$ involves electronic, magnetic and lattice degrees of freedom[4–6] and is still under debate. The detailed behavior depends on the species of the rare-earth $R$ ion. With decreasing temperature, the transition from a paramagnetic metal to an insulator occurs at $T_{\rm MIT}$.[7–10] A charge order also appears in the insulating phase. At the Néel temperature $T_{\rm N}\le T_{\rm MIT}$, the spin structure becomes antiferromagnetic. As a well-studied member of the $R$NiO$_3$ family, NdNiO$_3$ [Fig. 1(a)] undergoes the metal-to-insulator and paramagnetic-to-antiferromagnetic transitions simultaneously,[11,12] i.e., $T_{\rm N}=T_{\rm MIT}$. The metal–insulator transition is also accompanied by a structural change from orthorhombic ($Pbnm$) to monoclinic ($P2_1/n$) symmetry.[12,13] During the structure transition, the NiO$_6$ octahedra distorts and rotates. Compared with bulk NdNiO$_3$, epitaxial strain and finite film thickness result in more complications during phase transition in films.[14–16] So far, most data about NdNiO$_3$ films was acquired by macroscopic measurements. Nevertheless more and more evidence indicates that the material exhibits strong spatial inhomogeneity at the nano-scale and microscopy is essential to elucidate the MIT mechanism. For example, recent works[17–19] demonstrate the nano-scale phase coexistence when the NdNiO$_3$ films go through the first-order metal–insulator phase transitions. The phase transition is broadened by disorder such as strain field and therefore occurs over a finite range of temperatures where the fraction of low-temperature phase grows from zero to one as the temperature is lowered. Here we report another scenario that features a continuous transition without phase coexistence when NdNiO$_3$ films go through the metal–insulator transition. The nano-infrared (nano-IR) study is performed on a scattering-type scanning near-field optical microscope (s-SNOM). Such infrared near-field microscopy probes local optical conductivity with nanometer resolution[20,21] and has been successfully applied to the investigation of correlated oxides.[22–24] The NdNiO$_3$ films were deposited on the (100)-oriented cubic (La$_{0.3}$Sr$_{0.7}$)(Al$_{0.65}$Ta$_{0.35}$)O$_{3}$ (LSAT, $a = 0.3868$ nm) substrates by pulsed laser deposition (PLD). The pseudo-cubic perovskite lattice parameter of bulk NdNiO$_3$ is estimated to be 0.3807 nm.[2] The in-plane lattice mismatch is $+1.6\%$. Compared with NdNiO$_3$ films grown on (110)-oriented NdGaO$_3$ substrates, the NdNiO$_3$ films on the LSAT substrates are under a larger tensile strain.[16] During the film deposition, the LSAT substrate was kept at 650 ℃ under an oxygen atmosphere of 20 Pa. Then the sample was annealed at room temperature under an oxygen pressure of $\sim $0.01 Pa. The film surface is reasonably flat [see the topography image in Fig. 1(b)] and the RMS roughness is less than 0.5 nm. Figure 1(c) shows the resistance versus temperature ($R$–$T$) curve of the 15-nm-thick NdNiO$_3$ film measured by the conventional four-probe method. The first-order transition between orthorhombic and monoclinic symmetry is smeared likely because of the strain field from the substrate. The warming and cooling cycles form a hysteresis loop. The transition occurs over a finite range of temperature from $130$ K to $200$ K. The resistance at low temperature exponentially depends on temperature, indicating that the film is in the insulating phase. The $R$–$T$ curve suggests that the first-order metal–insulator transition has been broadened by disorder[25] and the thermal cycles should undergo various meta-stable states along the hysteresis loop. However, the transport measurement does not cover the issue of phase coexistence during the transition, which will be addressed by s-SNOM in this study.

Fig. 1. (a) Crystal structure of bulk NdNiO$_3$. (b) AFM topography[26] of NdNiO$_3$ film with 15 nm thickness at 85 K. (c) Resistance versus temperature from the transport measurement in 15 nm NdNiO$_3$ film. Arrows indicate the direction of transitions on warming (red) and cooling (black).

We employ a cryogenic s-SNOM to investigate the evolution of local electronic properties of NdNiO$_3$ films with temperature. The s-SNOM simultaneously maps the surface morphology and local optical response at the 20 nm scale. The CO$_2$ laser (p-polarized, $\lambda = 10.26$ µm) is focused on the atomic force microscope (AFM) tip. The mid-IR excitation at 10.26 µm probes dominantly the free-carrier Drude response of NdNiO$_3$ in accordance with far-field spectroscopy.[27,28] The AFM tip is tapping at frequency $\varOmega$ with an amplitude of $\sim $50 nm. The local optical response is contained in the scattered signal. The higher harmonics of $\varOmega$ in the signal, i.e., $S_n$ $(n=2,3,\dots)$, are demodulated by a lock-in amplifier (Zurich Instruments, HF2LI) to suppress the far-field background. The self-homodyne detection is used in the nano-IR imaging. Considering its nearly perfect reflectivity in the infrared frequencies,[29] 100 nm gold film is deposited at the sample edge as the reference material to quantify the local optical response. Figure 2(a) presents the thermal cycle of the infrared nano-imaging for the same region. The spatial average $\bar{S}$ of the near-field signal normalized by gold [$S = S_2({\rm NdNiO_3})/S_2({\rm Au})$] shows significant temperature dependence and a hysteresis loop [Fig. 2(b)]. At the ends of the loop, $\bar{S}=0.15$ and $0.34$, respectively, corresponding to the insulating and conducting phases. The near-field signal is almost homogeneous when the phase transition is completed. During the transition, the inhomogeneity of $S$ is illustrated by the histogram in Fig. 2(c). For each temperature, the counts of pixel as a function of $S$ shows a single peak centered at $\bar{S}$. The peak width is maximized at 150 K and approaches to zero at the end points of the hysteresis in Fig. 1(c). No double-peak structure signifying phase coexistence is observed in the histogram. The inhomogeneity in nano-IR signal distribution only manifests the local fluctuation of conductivity. The local conductivity fluctuation is more pronounced in the relative nano-IR signal derived from Fig. 2(a). For each temperature of the warming cycle, the image of $\Delta S$ ($\Delta S = S-\bar{S}$) is shown in Fig. 3(a) (more data in Fig. S1 and S2 of the Supplementary Material). The full range of the color scale is set to $\pm15\%$ of $\bar{S}$. The images of $S_3$ are also shown in Fig. S3 of the Supplementary Material. $S_2$ is similar to $S_3$ in the relative signal but with larger absolute magnitude. The stripe-like inhomogeneity along the (011) direction begins to emerge at the onset of phase transition on the $R$–$T$ curve and eventually disappears at the other end point of the hysteresis. The characteristic width of the stripes is about 2 µm. We propose that the stripe-like modulation is most likely given rise by the balance between the strain energy due to inhomogeneity in the structure during the process of phase transition and that exerted by the substrate. The substrate tends to hinder the phase transition. The competition favors a less uniform distribution of structural transition and slows down the distortion and rotation of octahedra. The frustrated structural change gradually opens or closes the energy gap in the electronic band, leading to a second-order-like metal–insulator phase transition. During the thermal cycle, the stripe-like modulation remains the same but the intensity evolves with temperature [see Fig. 3(b)]. We suggest that the evolution of intensity experiences various meta-stable states pinned by disorder in the process of structural transition, which is responsible for the hysteresis. More theoretical investigation is needed to confirm the above model.

Fig. 2. (a) Images of the normalized nano-IR signal ($S$) acquired from NdNiO$_3$ film with 15 nm thickness during the warming (top) and cooling (bottom) transition. (b) The spatial average $\bar{S}$ of nano-IR signal through the MIT. Arrows indicate the warming (red) and cooling (black) transition. (c) Histogram representation of the normalized nano-IR signal ($S$) on warming from 110 K to 200 K.

Fig. 3. (a) Images of relative nano-IR signal ($\Delta S = S-\bar{S}$) during the warming transition from NdNiO$_3$ film with 15 nm thickness. All images were acquired in the same field of view. (b) Line profiles (red dashed lines in (a) from 150 K to 180 K) showing the intensity of relative nano-IR signal ($\Delta S$ is normalized by $\bar{S}$).

To demonstrate that the above observation generally presents in the NdNiO$_3$ thin films, we perform the experiments with different film thicknesses. Figure 4(a) shows the $R$–$T$ curve for a 7 nm NdNiO$_3$ film on LSAT prepared under the same growth condition as that of the 15 nm film. The two kinks around 160 K on the $R$–$T$ curve may be related to an additional phase transition and are still not well understood. The infrared nano-imaging [Fig. 4(b)] and the relative intensity map (Fig. 4(c) and Fig. S4) of the 7 nm film show similar modulation as those of the 15 nm film. The spatial average $\bar{S}$ of the near-field signal normalized by gold ($S = S_2(\rm{NdNiO_3}$)$/S_2(\rm{Au})$) exhibits a hysteresis loop [Fig. 4(d)]. The temperatures for the end points of the near-field signal loop are close to those of the $R$–$T$ loop. The loop from 155 K to 190 K is wider than that from 130 K to 155 K possibly as a result of the kinks at 160 K. Once again no indication of phase coexistence is revealed in the histogram for various temperatures [Fig. 4(e)].

Fig. 4. (a) Temperature dependence of resistance acquired from NdNiO$_3$ films with 7 nm thickness. Arrows indicate the warming (red) and cooling (black) transition. (b) Images of normalized nano-IR signal ($S$) on warming (red arrows) and cooling (black arrows). (c) Images of relative nano-IR signal ($\Delta S$) on warming. (d) The spatial average $\bar{S}$ of nano-IR signal through the MIT. (e) Histogram representation of the normalized nano-IR signal ($S$) during the warming transition.

In summary, we have demonstrated the phase transition in the correlated material NdNiO$_3$ at cryogenic temperature by s-SNOM. The NdNiO$_3$ films on the LSAT substrate undergo a continuous transition without phase coexistence across the MIT. Such observation is different from the previous studies and could be due to the different substrates. The emerging stripe-like modulation of optical conductivity during the phase transition is attributed to the hindered structural change. Our study provides new insights into the interplay between electronic and structural degrees of freedom in NdNiO$_3$ films. References Systematic study of insulator-metal transitions in perovskites RNiO3 (R=Pr,Nd,Sm,Eu) due to closing of charge-transfer gapStructural, magnetic and electronic properties of perovskites (R = rare earth)Progress in perovskite nickelate researchCharge Disproportionation in

R {NiO}_{3}

Perovskites: Simultaneous Metal-Insulator and Structural Transition in

{YNiO}_{3}

Charge Disproportionation without Charge Transfer in the Rare-Earth-Element Nickelates as a Possible Mechanism for the Metal-Insulator TransitionGround-state oxygen holes and the metal–insulator transition in the negative charge-transfer rare-earth nickelatesLandau Theory of Charge and Spin Ordering in the NickelatesPhysics of Ultrathin Films and Heterostructures of Rare-Earth NickelatesImpact of antiferromagnetism on the optical properties of rare-earth nickelatesStructurally triggered metal-insulator transition in rare-earth nickelatesPhotoinduced melting of magnetic order in the correlated electron insulator NdNiO

_{3}

Distortion mode anomalies in bulk

{PrNiO}_{3}

: Illustrating the potential of symmetry-adapted distortion mode analysis for the study of phase transitionsUltrastructural Characterization of the Lower Motor System in a Mouse Model of Krabbe DiseaseMetal-insulator transitions in

{NdNiO}_{3}

thin filmsStrain controlled metal-insulator transition in epitaxial NdNiO₃ thin filmsHeterointerface engineered electronic and magnetic phases of NdNiO3 thin filmsStriped nanoscale phase separation at the metal–insulator transition of heteroepitaxial nickelatesDirect Mapping of Phase Separation across the Metal–Insulator Transition of NdNiO₃Coexisting first- and second-order electronic phase transitions in a correlated oxideMott Transition in VO₂ Revealed by Infrared Spectroscopy and Nano-ImagingNano-optical imaging and spectroscopy of order, phases, and domains in complex solidsAnisotropic Electronic State via Spontaneous Phase Separation in Strained Vanadium Dioxide FilmsNanotextured phase coexistence in the correlated insulator V2O3Multi-messenger nanoprobes of hidden magnetism in a strained manganiteHeterogeneous nucleation and metal-insulator transition in epitaxial films of NdNiO3Gwyddion: an open-source software for SPM data analysisOptical spectroscopy of the metal-insulator transition in

{NdNiO}_{3}

Mott Physics near the Insulator-To-Metal Transition in

{NdNiO}_{3}

Model for quantitative tip-enhanced spectroscopy and the extraction of nanoscale-resolved optical constants

[1]	Torrance J B, Lacorre P, Nazzal A I, Ansaldo E J, and Niedermayer C 1992 Phys. Rev. B 45 8209
[2]	Medarde M L 1997 J. Phys.: Condens. Matter 9 1679
[3]	Catalan G 2008 Phase Transit. 81 729
[4]	Alonso J A, García-Muñoz J L, Fernández-Díaz M T, Aranda M A G, Martínez-Lope M J, and Casais M T 1999 Phys. Rev. Lett. 82 3871
[5]	Johnston S, Mukherjee A, Elfimov I, Berciu M, and Sawatzky G A 2014 Phys. Rev. Lett. 112 106404
[6]	Bisogni V, Catalano S, Green R J, Gibert M, Scherwitzl R, Huang Y, Strocov V N, Zubko P, Balandeh S, Triscone J M, Sawatzky G, and Schmitt T 2016 Nat. Commun. 7 13017
[7]	Lee S, Chen R, and Balents L 2011 Phys. Rev. Lett. 106 016405
[8]	Middey S, Chakhalian J, Mahadevan P, Freeland J W, Millis A J, and Sarma D D 2016 Annu. Rev. Mater. Res. 46 305
[9]	Ruppen J, Teyssier J, Ardizzone I, Peil O E, Catalano S, Gibert M, Triscone J M, Georges A, and Van Der Marel D 2017 Phys. Rev. B 96 045120
[10]	Mercy A, Bieder J, Íñiguez J, and Ghosez P 2017 Nat. Commun. 8 1677
[11]	Caviglia A D, Först M, Scherwitzl R, Khanna V, Bromberger H, Mankowsky R, Singla R, Chuang Y D, Lee W S, Krupin O, Schlotter W F, Turner J J, Dakovski G L, Minitti M P, Robinson J, Scagnoli V, Wilkins S B, Cavill S A, Gibert M, Gariglio S, Z, Triscone J M, Hill J P, Dhesi S S, and Cavalleri A 2013 Phys. Rev. B 88 220401
[12]	Gawryluk D J, Klein Y M, Shang T, Sheptyakov D, Keller L, Casati N, Ph L, Fernández-Díaz M T, Rodríguez-Carvajal J, and Medarde M 2019 Phys. Rev. B 100 205137
[13]	Zhang J Y, Kim H, Mikheev E, Hauser A J, and Stemmer S 2016 Sci. Rep. 6 1
[14]	Catalan G, Bowman R M, and Gregg J M 2000 Phys. Rev. B 62 7892
[15]	Xiang P H, Zhong N, Duan C G, Tang X D, Hu Z G, Yang P X, Zhu Z Q, and Chu J H 2013 J. Appl. Phys. 114 243713
[16]	Liu J, Kargarian M, Kareev M, Gray B, Ryan P J, Cruz A, Nadeem T, Chuang Y D, Guo J H, Rondinelli J M, Freeland J W, Fiete G A, and Chakhalian J 2013 Nat. Commun. 4 2714
[17]	Mattoni G, Zubko P, Maccherozzi F, van der Torren A J, Boltje D B, Hadjimichael M, Manca N, Catalano S, Gibert M, Liu Y, Aarts J, Triscone J M, Dhesi S S, and Caviglia A D 2016 Nat. Commun. 7 13141
[18]	Preziosi D, Lopez-Mir L, Li X, Cornelissen T, Lee J H, Trier F, Bouzehouane K, Valencia S, Gloter A, Barthélémy A, and Bibes M 2018 Nano Lett. 18 2226
[19]	Post K W, McLeod A S, Hepting M, Bluschke M, Wang Y, Cristiani G, Logvenov G, Charnukha A, Ni G X, Padma R, Minola M, Pasupathy A, Boris A V, Benckiser E, Dahmen K A, Carlson E W, Keimer B, and Basov D N 2018 Nat. Phys. 14 1056
[20]	Qazilbash M M, Brehm M, Chae B G, Ho P C, Andreev G O, Kim B J, Yun S J, Balatsky A V, Maple M B, Keilmann F, Hyun-Tak K, and Basov D N 2007 Science 318 1750
[21]	Atkin J M, Berweger S, Jones A C, and Raschke M B 2012 Adv. Phys. 61 745
[22]	Liu M K, Wagner M, Abreu E, Kittiwatanakul S, McLeod A, Fei Z, Goldflam M, Dai S, Fogler M M, Lu J, Wolf S A, Averitt R D, and Basov D N 2013 Phys. Rev. Lett. 111 096602
[23]	McLeod A S, Van Heumen E, Ramirez J G, Wang S, Saerbeck T, Guenon S, Goldflam M, Anderegg L, Kelly P, Mueller A, Liu M K, Schuller I K, and Basov D N 2017 Nat. Phys. 13 80
[24]	Mcleod A S, Zhang J, Gu M Q, Jin F, Zhang G, Post K W, Zhao X G, Millis A J, Wu W B, Rondinelli J M, Averitt R D, and Basov D N 2020 Nat. Mater. 19 397
[25]	Kumar D, Rajeev K P, Kushwaha A K, and Budhani R C 2010 J. Appl. Phys. 108 063503
[26]	Nec̆as D and Klapetek P 2012 Cent. Eur. J. Phys. 10 181
[27]	Katsufuji T, Okimoto Y, Arima T, Tokura Y, and Torrance J B 1995 Phys. Rev. B 51 4830
[28]	Stewart M K, Liu J, Kareev M, Chakhalian J, and Basov D N 2011 Phys. Rev. Lett. 107 176401
[29]	McLeod A S, Kelly P, Goldflam M D, Gainsforth Z, Westphal A J, Dominguez G, Thiemens M H, Fogler M M, and Basov D N 2014 Phys. Rev. B 90 085136