Epitaxial stabilization of thin films of the frustrated Ge-based spinels
Denis M. Vasiukov, Mikhail Kareev, Fangdi Wen, Liang Wu, Padraic Shafer, Elke Arenholz, Xiaoran Liu, Jak Chakhalian
FFabrication and electronic properties of GeNi O epitaxial thin film Denis M. Vasiukov, a) Mikhail Kareev, Fangdi Wen, Liang Wu, Padraic Shafer, ElkeArenholz,
2, 3
Xiaoran Liu, and Jak Chakhalian Department of Physics and Astronomy, Rutgers University, Piscataway,New Jersey 08854, USA Advanced Light Source, Lawrence Berkeley National Laboratory, Berkeley,CA 94720, USA Cornell High Energy Synchrotron Source, Cornell University, Ithaca, NY 14853,USA (Dated: 19 July 2019)
Frustrated magnets can host numerous exotic many-body quantum and topologicalphenomena. Bulk GeNi O spinel is a frustrated magnet with an unusual two-stagetransition to antiferromagnetic state and spontaneous breaking of cubic symmetry.Here we report on the first growth of (001)-oriented GeNi O thin films on MgAl O substrate. The fabricated films were characterized by X-ray scattering and advancedphotoelectron spectroscopies confirming epitaxial growth of high crystallinity stoi-chiometric films. The GeNi O films with S = 1 pyrochlore sublattice can help toclarify the complex mechanism of the magnetic ordering in the bulk GeNi O andopens the door towards studies of 2D frustrated magnetism in ultra-thin limit. a) [email protected] a r X i v : . [ c ond - m a t . s t r- e l ] J u l hysics of frustrated magnetic systems is a very active and fruitful subfield with a greatpotential for emergent phenomena, new states of matter and exotic excitations (exemplifiedby spin ice, quantum spin liquids, spin-charge separation and so on ). Despite a plethoraof interesting theoretical proposals there is still no definitive evidence of quantum spin liquidand the fundamental challenge is to map theory to material systems that can host the exoticphenomena and states . As an alternative, ultra-thin films with frustrated magnetism maybe a viable option to answer the challenge.In this context, complex oxides with spinel structure are of special interest. Spinel oxideshave the general formula AB O with structure formed by close-packed sublattice of anions inwhich one-eight of the tetrahedral (A-site) and one-half of the octahedral (B-site) intersticesare filled by cations. As shown in Fig. 1a, in the absence of distortions, spinel crystallizes incubic space group F d ¯3 m and B-site cations form a network of corner-shared tetrahedra or thepyrochlore sublattice that has the strongest known frustration . Interestingly, along { } directions the pyrochlore lattice consists of alternating sequence of kagome and triangularplanes which are also strongly frustrated in 2D. This feature opens the door for the so-called“geometrical lattice engineering” approach to fabricate novel artificial systems with theselattices .From the practical viewpoint, it is desirable to find material classes with a pure realizationof the frustrated lattices, e.g. without concomitant structural or electronic variations, whichcan hinder effects of frustrated magnetism or even destroy it. For instance, many spineloxides undergo structural phase transition at low temperatures due to the cooperative Jahn-Teller effect that relieves the frustration and promotes a long-range magnetic ordering (forexamples see [5 and 7]).In this work we focus on GeNi O (GNO) spinel. There are three crucial features thatmake magnetic properties of this spinel interesting from the frustrated magnetism viewpoint:( i ) GNO belongs to a so-called “4–2” family with germanium on the A-site (with generalformula Ge B O ). This class of compounds has magnetically inactive A-site, hence onlythe pyrochlore sublattice is responsible for magnetic properties. ( ii ) Ni has 3 d electronicconfiguration which adopts a non-degenerate electronic term ( A g ) meaning absence of theJahn-Teller effect. ( iii ) The pyrochlore lattice with integer spin statistics ( S = 1) next tothe most interesting quantum spin limit of S = 1 / IG. 1. a) Crystal structure of GeNi O with pyrochlore sublattice of Ni atoms highlightedin gray. b) and c) RHEED patterns of MgAl O substrate at 650 ◦ C and GeNi O film at roomtemperature, respectively. d) X-ray reflectivity data of the same sample show that film thicknessis 11.79(4) nm. e) 2Θ − ω X-ray diffraction scan along 00 l direction. The scan contains allowedreflections of the substrate and fabricated film together with two forbidden reflections (002 and006) of the substrate which arise because of Umweganregung effect . Asterisks mark (004) and(008) peaks of the substrate due to residual K β radiation. transitions. At low temperatures GNO reaches an antiferromagnetic state with propagationvector (
12 12 12 ) corresponding to rhombohedral lattice system . Upon this transition the B-sitesplits into two crystallographically non-equivalent positions which correspond to the kagomeand triangular planes. More detailed recent studies show that this antiferromagnetic stateis developed through two consecutive first-order phase transitions separated by ∼ . T N = 12 . T N = 11 . .A muon-spin relaxation experiment suggested that these two transition correspond to aseparate ordering of two magnetic subsystems . Based on the single-crystal neutron elastic3 IG. 2. a) High resolution XPS scan of Ni 2 p states shows the spin-orbit split doublet (green)together with satellite features (light blue). The background line is shown in gray color. Unsplitteddoublet confirms that film contain Ni only in a single valence state. b) Oxygen K -edge XASspectrum demonstrates absence of the sharp pre-peak feature that corresponds to the d electronconfiguration of Ni without significant admixture of configurations with oxygen p -holes (like d L ). scattering data it was concluded that the first transition ( T N ) corresponds to a magneticordering only within the kagome planes (each plane has ferromagnetic ordering stackedantiferromagnetically). The triangular planes become magnetically ordered at the secondtransition ( T N ). On the other hand, specific-heat data show several puzzling features whichyet remain unexplained . For instance, the coexistence of gapped and gapless spin wavesand the presence of substantial magnetic correlations in the paramagnetic state despite thelow frustration ratio Θ CW /T N ( ∼ . ions relative to the kagome lattice).In light of the above, the thin films of GNO are a promising candidate to probe emergentquantum states by applying epitaxial strain to modify the magnetic ground state of GNO.Additionally, such experiments can help in understanding of peculiar magnetic orderingmechanism of the S = 1 pyrochlore sublattice in the bulk GNO. In this letter we report on thefirst fabrication of high-quality (001)-oriented GNO films on MgAl O (MAO) (001)-orientedsubstrate using pulsed laser deposition. The surface morphology, film thickness and crystal4tructure were measured by in-situ reflection high-energy electron diffraction (RHEED),atomic force microscopy (AFM), X-ray reflectivity (XRR) and X-ray diffraction (XRD).Film stoichiometry and electronic state of nickel were investigated by X-ray photoelectronspectroscopy (XPS) and resonant X-ray absorption spectroscopy (XAS) on Ni L-edge andO K-edge. The combined results confirm the successful epitaxial growth of high quality(001)-oriented GNO thin film.First we describe the details of GNO growth. GNO films were fabricated under 6 mTorr ofan oxygen atmosphere with KrF excimer laser (248 nm wavelength). The samples were grownon (001)-oriented MgAl O substrate with average surface roughness of S a ∼
100 pm froma stoichiometric (GeNi O ) target. The UV-laser was operated at 3 Hz repetition rate with2.7 J/cm energy density per pulse. Films were deposited on the substrate at temperature700 ◦ C (measured by pyrometer). After deposition samples were annealed at the growthcondition and then cooled down (with initial rate 15 ◦ C/min) to room temperature at thesame oxygen pressure.Growth of the films was monitored in-situ by high-pressure RHEED. The representativeRHEED patterns with an indexed zero-order Laue zone of the substrate and fabricatedfilm are plotted in Fig. 1b,c. As expected for two isostructural compounds the patterns aresimilar including identity of higher order Laue zones. The intensity of the specular reflectionshows only an initial reduction with the start of the deposition followed by the recovery. Itremains constant without any evident oscillations during further deposition. Taking intoaccount the developed streak pattern shown in Fig. 1c we conclude that the fabricated GNOfilm grows by a step-flow mode .X-ray reflectivity and diffraction measurements were performed on Empyrean diffractome-ter using Cu K α radiation. Fit of XRR data (Fig. 1d) yields roughness S a = 240(20) pm,which is in a good agreement with the AFM results ( ∼
200 pm). Both XRR and AFMconfirm the development of smooth surface morphology of the GNO film. A fit of the XRRdata shown in Fig. 1d yields 11.79(4) nm film thickness. The 2Θ − ω scan (Fig. 1e) containsreflections of the substrate and two reflections of the GNO film. These peaks are indexedas 004 and 008 reflections confirming (001)-orientation and epitaxial growth of the film.As the bulk lattice parameter of GNO is a = 8 .
22 ˚A [18], growth on the MAO substrate( a = 8 .
08 ˚A) should result in 1.7 % in-plane compressive strain for the fully coherent in-terface. Indeed, from our diffraction data we have determined that the out-of-plane lattice5onstant is 8.376(5) ˚A, which is consistent with compressive strain.The XPS spectra were measured using a Thermo Scientific K-Alpha XPS spectrometerwith monochromated Al K α radiation. Stoichiometry of the GNO film was determinedusing high-resolution core-shell scans around 2 p state for Ge and Ni and 1 s state for O.The analysis of XPS data results in Ge : Ni : O = 1 : 1.98 : 4.24 which is consistent withinexperimental uncertainty with GeNi O composition (the deviation in the oxygen value fromthe ideal stoichiometric ratio is due to the surface exposed to air). Figure 2a shows a core-shell Ni 2 p scan. As seen, the energy position of the Ni 2 p / peak at 856.5 eV and theabsence of any peak splitting imply single valency of Ni in the film.To further investigate the electronic state of Ni and O away from the surface, we carriedout a XAS experiment at beamline 4.0.2 of the Advanced Light Source (Lawrence BerkeleyNational Laboratory). Oxygen K -edge scan shown in Fig. 2b is very similar to the theoreticalline-shape and does not show any sizable pre-edge intensity around 529 eV implying onlya small contribution of configurations with ligand holes (e.g. d L ) to the electronic state ofNi ion. Further, in accord with the XPS data the Ni L -edge lineshape corresponds to Ni in2+ state. The simulated spectrum of Ni shown in Fig. 3 corroborate the divalent state ofNi . An optimization of the lineshape to the experimental data yields the following atomicparameters: crystal field splitting 10 Dq = 0 . V e g and V t g FIG. 3. Ni L -edge XAS spectrum of the GNO film. Blue dots are the measured data whereasorange line show simulated spectrum of Ni . O spinelwith the Ni pyrochlore sublattice. A combination of advanced probes confirmed highstructural and chemical quality of the GNO thin films. The GeNi O films with S = 1pyrochlore sublattice can help to clarify the complex mechanism of the magnetic orderingin the bulk GeNi O and also opens the door towards studies of 2D frustrated magnetismin ultra-thin limit. ACKNOWLEDGMENTS
The work in Rutgers University was supported by the Gordon and Betty Moore Founda-tion’s EPiQS Initiative through Grant No. GBMF4534, and by the Department of Energyunder Grant No. de-sc0012375. This research used resources of the Advanced Light Source,which is a Department of Energy Office of Science User Facility under Contract No. DE-AC0205CH11231.
REFERENCES C. Lacroix, P. Mendels, and F. Mila,
Introduction to Frustrated Magnetism: Materials,Experiments, Theory , Springer Series in Solid-State Sciences (Springer Berlin Heidelberg,2011). C. Castelnovo, R. Moessner, and S. Sondhi, “Spin ice, fractionalization, and topologicalorder,” Annu. Rev. Condens. Matter Phys. , 35–55 (2012). L. Savary and L. Balents, “Quantum spin liquids: a review,” Rep. Prog. Phys. , 016502(2016). Y. Zhou, K. Kanoda, and T.-K. Ng, “Quantum spin liquid states,” Rev. Mod. Phys. ,025003 (2017). H. Takagi and S. Niitaka, “Highly frustrated magnetism in spinels,” in
Introduction toFrustrated Magnetism (Springer, 2011) pp. 155–175. X. Liu, S. Middey, Y. Cao, M. Kareev, and J. Chakhalian, “Geometrical lattice engineeringof complex oxide heterostructures: a designer approach to emergent quantum states,” MRSCommun. , 133–144 (2016). 7 S.-H. Lee, H. Takagi, D. Louca, M. Matsuda, S. Ji, H. Ueda, Y. Ueda, T. Katsufuji, J.-H. Chung, S. Park, et al. , “Frustrated magnetism and cooperative phase transitions inspinels,” J. Phys. Soc. Jpn. , 011004 (2010). E. Bertaut, V. Van Qui, R. Pauthenet, and A. Murasik, “Structure magn´etique et pro-pri´et´es magn´etiques de GeNi O ,” J. Phys. , 516–521 (1964). M. Crawford, R. Harlow, P. Lee, Y. Zhang, J. Hormadaly, R. Flippen, Q. Huang, J. Lynn,R. Stevens, B. Woodfield, J. Boerio-Goates, and R. Fisher, “Structure and properties ofthe integer-spin frustrated antiferromagnet GeNi O ,” Phys. Rev. B , 220408 (2003). R. Stevens, B. F. Woodfield, J. Boerio-Goates, and M. K. Crawford, “Heat capacities,third-law entropies and thermodynamic functions of the geometrically frustrated antifer-romagnetic spinels GeCo O and GeNi O from t = (0 to 400) k,” J. Chem. Thermodyn. , 359–375 (2004). S. Diaz, S. De Brion, G. Chouteau, B. Canals, V. Simonet, and P. Strobel, “Magneticfrustration in the spinel compounds GeCo O and GeNi O ,” Phys. Rev. B , 092404(2006). P. T. Barton, M. C. Kemei, M. W. Gaultois, S. L. Moffitt, L. E. Darago, R. Seshadri,M. R. Suchomel, and B. C. Melot, “Structural distortion below the N´eel temperature inspinel GeCo O ,” Phys. Rev. B , 064105 (2014). M. Renninger, ““Umweganregung”, eine bisher unbeachtete Wechselwirkungserscheinungbei Raumgitterinterferenzen,” Z. Phys. , 141–176 (1937). T. Lancaster, S. Blundell, D. Prabhakaran, P. Baker, W. Hayes, and F. Pratt, “Magnetismin the S = 1 frustrated antiferromagnet GeNi O studied using implanted muons,” Phys.Rev. B , 184436 (2006). M. Matsuda, J.-H. Chung, S. Park, T. Sato, K. Matsuno, H. A. Katori, H. Takagi, K. Kaku-rai, K. Kamazawa, Y. Tsunoda, I. Kagomiya, C. Henley, and S.-H. Lee, “Frustratedminority spins in GeNi O ,” EPL , 37006 (2008). J. Lashley, R. Stevens, M. Crawford, J. Boerio-Goates, B. Woodfield, Y. Qiu, J. Lynn,P. Goddard, and R. Fisher, “Specific heat and magnetic susceptibility of the spinelsGeNi O and GeCo O ,” Phys. Rev. B , 104406 (2008). S. Hasegawa, “Reflection high-energy electron diffraction,” Charact. Mater. , 1925–1938(2002). K. Hirota, T. Inoue, N. Mochida, and A. Ohtsuka, “Study of germanium spinels (part 3),”8. Ceram. Soc. Jpn. , 976–986 (1990). K. Persson, “Materials data on Ni GeO (sg:227) by materials project,” (2016). Simulation was done using Crispy software version 0.7.2 developed by Marius Retegan.In the simulation we used 0.9 reduction factor of default Slater integrals for Ni in thesoftware and the parameters U d, d and U p, d were 7.3 and 8.5 eV, respectively. M. Haverkort, M. Zwierzycki, and O. Andersen, “Multiplet ligand-field theory using Wan-nier orbitals,” Phys. Rev. B85