WorksPhysics and Astronomy
Site-selective polar compensation of Mott electrons in a double perovskite heterointerface
Abstract
Double perovskite oxides (DPOs) with two transition metal ions (O) offer a fascinating platform for exploring exotic physics and practical applications. Studying these DPOs as ultrathin epitaxial thin films on single crystalline substrates can add another dimension to engineering electronic, magnetic, and topological phenomena. Understanding the consequence of polarity mismatch between the substrate and the DPO would be the first step towards this broad goal. We investigate this by studying the interface between a prototypical insulating DPO NdNiMnO and a wide-band gap insulator SrTiO. The interface is found to be insulating in nature. By combining several experimental techniques and density functional theory, we establish a site-selective charge compensation process that occurs explicitly at the Mn site of the film, leaving the Ni sites inert. We further demonstrate that such surprising selectivity, which cannot be explained by existing mechanisms of polarity compensation, is directly associated with their electronic correlation energy scales. This study establishes the crucial role of Mott physics in polar compensation process and paves the way for designer doping strategies in complex oxides.
The epitaxial growth has been further confirmed by a cross-sectional STEM experiment on a 20 uc film on STO. High-angle annular dark field (HAADF) image along the zone axis [100] confirmed coherent epitaxial growth of the film, devoid of any defect or dislocation [Fig. 1(c)].
Atomically resolved electron energy loss spectroscopy (EELS) mapping shows that the extent of cationic intermixing across the interface is limited to 1 uc on the film side, also consistent with the film/substrate interface roughness obtained from XRR fitting. Since minuscule intermixing, likely to be inherent to all films due to same growth condition, cannot explain the systematic electronic structure changes with film thickness, shown in latter part of the paper. Another notable observation is that we do not find any NiO precipitation, which was reported to be the polar compensation mechanism for a thick ( nm) LaNiMnO film [33].
The smaller band gap of NNMO (-1.5 eV) [36, 37] compared to STO ( eV) should result in a straddled band alignment, similar to the well-studied LaMnO/SrTiO (LMO/STO) heterostructure [16, 19, 38, 39, 40, 41]. While for LMO/STO, the electronic reconstruction should occur within the LMO film [19, 38], there are also reports of charge compensation within the STO substrates [39, 40]. Therefore, it is crucial to probe the electronic structures of both film and substrate for our heterostructures. For checking the valence state of the TM sites, we have performed element-sensitive XAS experiments for all samples grown on both substrates [42]. All XAS data discussed in the main text were recorded in surface-sensitive total electron yield (TEY) mode with a grazing incidence geometry (Figure 1(d)).

Figure 1. Reciprocal space mapping (RSM) in (a) plane (b) plane for a 20 uc film on STO around Bragg reflection. (c) HAADF-STEM image of the film. (d) Schematic of the XAS measurement setup. XAS data recorded at 300 K at (e) Ti edge for 2, 3 and 5 uc films on STO, compared with standard Ti spectra of STO.
The origin of the metallic interface for LAO/STO heterostructure is often attributed to the polar catastrophe-driven electronic reconstruction, resulting in Ti in STO near the interface [17] or oxygen vacancy in STO leading to electron doping in Ti [43]. Our measurements on NNMO/STO films confirm that the spectral features of Ti edges are very similar to Ti of bulk stoichiometric STO (Figure 1(e)). Further peak analysis of the and edges, known to detect even small amount of Ti ions [40] do not find any signature of electron doping on Ti. (also supported by Ti edge EELS).These results are further corroborated by electrical transport measurements with wire-bonded contacts, which do not find any measurable interfacial conductivity, similar to LaMnO/SrTiO [19], LaCrO/SrTiO heterostructures [24].The presence of any bulk oxygen vacancy within the substrates can also be discarded as all films were annealed in-situ for 30 minutes under 500 Torr ultrapure oxygen atmosphere post deposition.
In the absence of any charge compensation in STO substrate, we proceed to the film side. In Figure 2(a),(Figure 2b) we show Ni edge XAS spectra of several films on NGO and STO along with a Ni standard [44]. Admittedly, Ni remains in +2 state for all films on both substrates, akin to bulk NNMO [36]. EELS analysis further verifies this, asserting that the polar mismatch does not affect Ni ions.
Figure 2. XAS data recorded at 300 K at (a) Ni edge for 5, 10 and 20 uc films on NGO (b) Ni edge for 2, 3, 5, 10 and 20 uc films on STO. Reference for Ni spectrum adapted from Ref. [44]. (c) Mn edge for 5, 10 and 20 uc films on NGO (d) Mn edge for 2, 3, 5, 10 and 20 uc films on STO. References Mn, Mn and Mn adapted from Ref. [45]
In sharp contrast, we find a systematic change in Mn XAS spectra [Fig. 2(d)]. For easy comparison, Mn XAS spectra for the standard of Mn, Mn and Mn have also been shown in Figure 2(d). The spectra for films on NGO (no polar mismatch) [Fig. 2(c)] confirm Mn state, similar to bulk NNMO [36]. However, a small feature of Mn around 639.4 eV is present, becoming much more prominent for the 5 uc film. On the contrary, there is an additional change in Mn XAS for the films on STO(Fig. 2(d)). Apart from the Mn feature also found here with a similar trend with film thickness, an additional feature for Mn states is observed, increasing on lowering the film thickness. These Mn signatures become comparable to Mn for the 5 uc film and even become the dominant contributor for the 3 uc film. Bulk-sensitive total fluorescence yield (TFY) data also verifies the same trend. Notably, surface oxygen vacancies leading to ions are common for manganite thin films [38, 46, 47, 48, 49] and has also been designated to provide electrons for polarity compensation in LMO/STO heterostructures [38]. However, such surface cannot compensate for polar catastrophe here as it is present in films on both STO and NGO. Rather, the compensation is happening through the formation of , being exclusively present only for films on STO. We also note that TEY mode probes nm length scale from the surface and contribution of the deeper layers decreases exponentially from the surface [42]. Thus, the decrease of feature with the increase in NNMO film thickness indicates non-uniformity of Mn valence along the out-of-plane film direction with hosted in a few layers.
The ionic radii of , and in octahedral environment with high spin configuration are 0.83 Å, 0.645 Å, and 0.53 Å, respectively [50]. Thus, the layer-resolved would vary according to the Mn oxidation state if present in different layers. Figure 3(a) shows the variation of the layer resolved , evaluated using HAADF-STEM image, demonstrating that the film has three types of layers: for intermediate layers (4-14 uc from the interface) are similar to the value obtained from RSM measurement (Figure 1(a),(Figure 1b)), few outermost layers have higher value of validating near film surface and most importantly, the value for the first 3 uc of the film is slightly higher, compared to the intermediate layers, implying that 2-3 uc near the interface regions are enriched with . This aspect has been further tested by Mn EELS as the ratio of the area under and edge () strongly depends on the Mn oxidation state [51, 52, 53]. Similar analysis of Mn EELS recorded from different regions of the 20 uc NNMO film on STO clearly demonstrate the interfacial regions are enriched with , whereas surface layers have [Figure 3(b)-(Figure 3c)]. We also performed annular bright field imaging (ABF-STEM), sensitive to the oxygen atoms, revealing reduced oxygen octahedral distortion near the interface, in line with enrichment, compared to the more distorted bulk layers. Overall, our comprehensive experiments demonstrate that NNMO films on STO exhibit site-selective electron doping at interfacial Mn states, accompanied by modulations in octahedral rotations on both sides of the interface.

Figure 3. (a) Layer resolved for 20 uc NNMO on STO, evaluated from HAADF-STEM image. (b) EELS spectra have been averaged region-wise, marked by 1 to 8. (c) Mn plotted region-wise. The dashed lines correspond to Mn references for , and .
While, oxide interfaces with interfacial polar mismatch have been widely studied in the last 20 years, the novelty of our work lies in the selective participation of Mn and inertness of Ni towards compensation. To examine the role of on-site electron correlation (), we performed first-principles DFT calculations for bulk NNMO using SCAN and SCAN+. While in the SCAN framework the lowest unoccupied states () arise from both Ni and Mn states, SCAN+ shows only Mn contribution [Fig. 4(a)]. To elucidate the polar catastrophe’s impact on electronic structure, we performed DFT + calculations for a symmetric 4 uc NNMO/3.5 uc STO/4 uc NNMO slab using both PBE-sol and SCAN meta-GGA functionals [54, 55, 56]. To determine surface oxygen vacancy () effects, we simulated another slab with a surface [Fig. 4(b)]. The surface Mn magnetic moment is also enhanced to which is considerably higher than the bulk Mn states () corroborating the surface ions. In both cases: pristine and surface [Fig. 4(c)], layer-resolved DOS confirm Mn 3-dominated states are just above the Fermi level (), while the Ni states remain higher in energy, precisely matching the experimentally observed

Figure 4. (a) Projected density of states (PDOS) for Ni and Mn states for bulk NNMO using SCAN (left), SCAN+ (right). (b) Structure of 4 uc NNMO/ 3.5 uc STO/ 4 uc NNMO with one oxygen vacancy created on each surface of the NNMO slabs (marked as V), post relaxation used in DOS calculations. Top view of surface with the V has also been shown. (c) Layer resolved projected density of states of Ni and Mn states using SCAN + functional corresponding to the layers marked in (b). Purple arrow showing shift of the Mn states towards Fermi is guide to the eye.
site-selectivity. Moreover, the same analysis reveals that the minima of the conduction band move towards from L (surface) to L (interface) layers due to the internal electric field, estimated to 0.14 V/Å Thus, a further reduction in interfacial conduction band energy in thicker slabs would lead to Mn at the interface.
Based on our results of XAS and STEM experiments, a charge compensation, schematically depicted in Figure 5(a), can be considered to alleviate polar catastrophe via transfer of 1/2 electron per uc to the interfacial Mn forming Mn cations [22] and surface oxygen vacancies act as electron donors. However, this conventional mechanism fails to explain why only the Mn site participates in this mechanism as the potential slope is independent of the nature of the cation. In order to explain such site selectivity, we invoke the physics of strong electron correlation for TMOs. In oxides, has a typical eV and eV [see page 102, Ref. [59]]. The corresponding parameters for are eV and 1 eV, respectively. In Figure 5(b), we show the upper Hubbard band (UHB) and lower Hubbard band (LHB) for both and with a common O band. The Mn UHB is at an energetically lower value than Ni UHB, making it a preferred host to electron addition. The resultant energy level diagram after addition of an electron to the Mn is also depicted in Figure 5(b), highlighting the significant overlap between UHBs of and [ eV and eV for , Ref. [59]].
Figure 5. (a) Schematic of proposed charge compensation for NNMO on STO demonstrating transfer of 1/2 per uc to the interface (green highlighted region). (b) Energy band representation of UHB and LHB of Ni and Mn cations (left) and of Ni and Mn cations (right). (c) O edge XAS of 2, 3, 5, 10 and 20 uc films on STO at 300 K compared with reference of AlO/STO[57] and NiO[58]. Pink shaded line represents Mn -O and Mn -O , orange shaded line represents Ti - O , and blue shaded line represents Ni UHB - O and Ti - O contributions to the spectra, meant as guide to the eye.
The above modification in unoccupied electronic states is verified by O -edge XAS, which identifies relative cation energy positions above Fermi energy [60]. O -edge XAS spectra for NNMO on STO are shown in Figure 5(c) along with reference features related to Ti and Ni states [57, 58]. The feature of Ti - O hybridized states of STO substrates are strongly present in the spectra of 5 uc film along with the films with lower thicknesses, and is weakly visible for the 10 uc film. The feature related to Ni UHB states appears around 532 eV for all NNMO films [58]. The most notable observation is the evolution of feature around 528 eV with film thickness, which is contributed by the Mn -O and Mn -O states [61]. The gradual suppression of this peak with decreasing film thickness establishes the shift of Mn UHB to higher energy, as depicted in Figure 5(b), due to charge compensation on Mn sites near the NNMO/STO interface. This is further supported by our observation that the 528 eV feature remains unaltered with the lowering of NNMO thickness grown on NGO.
To summarize, the present study unveils a surprising site-selective charge compensation in thin films on (001) substrates via electron transfer from surface to the interfacial Mn states resulting in cations at the interface. Despite being uniformly distributed as Mn, Ni sites remain unaffected by the process. By using O edge XAS, we further demonstrate that this site-selectivity is related to the relative energy position of the upper Hubbard band of w.r.t. the . Thus, our results provide direct experimental evidence of strongly correlated electron physics’ crucial role and inevitability in understanding polar compensation for oxide heterostructures. Further, exploring such selective doping will be an exciting avenue to realize exotic quantum phases in complex oxides.
Acknowledgement
The authors acknowledge XRD and wire bonding facilities of the Department of Physics, IISc Bangalore. SM acknowledges funding support from a SERB Core Research grant (Grant No. CRG/2022/001906). SM also acknowledges the seed funding from Quantum Research Park (QuRP), funded by Karnataka Innovation and Technology Society (KITS), K-Tech, Government of Karnataka. SM also acknowledges funding from a DST Nano Mission consortium project [DST/NM/TUE/QM-5/2019]. NB acknowledges funding from the Prime Minister’s Research Fellowship (PMRF), MoE, Government of India. A. Sinha acknowledges JNCASR for providing a research fellowship and Param Yukti computing facility, NSM, JNCASR for providing computational resources. JM acknowledges UGC, India for fellowship. This research used resources of the Advanced Photon Source, a U.S. Department of Energy Office of Science User Facility operated by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. This research used resources of the Advanced Light Source, which is a Department of Energy Office of Science User Facility under Contract No. DE-AC02-05CH11231. ZY acknowledges the National Natural Science Foundation of China (Grant No. 52202235).
References
References
- [16]Y. Anahory, L. Embon, C. J. Li, S. Banerjee, A. Meltzer, H. R. Naren, A. Yakovenko, J. Cuppens, Y. Myasoedov, M. L. Rappaport, M. E. Huber, K. Michaeli, T. Venkatesan, Ariando, and E. Zeldov, Nature Communications 7, 12566 (2016).
- [17]N. Nakagawa, H. Y. Hwang, and D. A. Muller, Nature Materials 5, 204 (2006).
- [19]Z. Chen, Z. Chen, Z. Q. Liu, M. E. Holtz, C. J. Li, X. R. Wang, W. M. Lü, M. Motapothula, L. S. Fan, J. A. Turcaud, L. R. Dedon, C. Frederick, R. J. Xu, R. Gao, A. T. N’Diaye, E. Arenholz, J. A. Mundy, T. Venkatesan, D. A. Muller, L.-W. Wang, J. Liu, and L. W. Martin, Phys. Rev. Lett. 119, 156801 (2017).
- [22]J. A. Mundy, Y. Hikita, T. Hidaka, T. Yajima, T. Higuchi, H. Y. Hwang, D. A. Muller, and L. F. Kourkoutis, Nature communications 5, 3464 (2014).
- [24]S. A. Chambers, L. Qiao, T. C. Droubay, T. C. Kaspar, B. W. Arey, and P. V. Sushko, Phys. Rev. Lett. 107, 206802 (2011).
- [33]S. R. Spurgeon, P. V. Sushko, A. Devaraj, Y. Du, T. Droubay, and S. A. Chambers, Phys. Rev. B 97, 134110 (2018).
- [36]S. Pal, S. Jana, S. Govinda, B. Pal, S. Mukherjee, S. Keshavarz, D. Thonig, Y. Kvashnin, M. Pereiro, R. Mathieu, P. Nordblad, J. W. Freeland, O. Eriksson, O. Karis, and D. D. Sarma, Phys. Rev. B 100, 045122 (2019).
- [37]M. Nasir, S. Kumar, N. Patra, D. Bhattacharya, S. N. Jha, D. R. Basaula, S. Bhatt, M. Khan, S.-W. Liu, S. Biring, et al., ACS Applied Electronic Materials 1, 141 (2019).
- [38]T. C. Kaspar, P. V. Sushko, S. R. Spurgeon, M. E. Bowden, D. J. Keavney, R. B. Comes, S. Saremi, L. Martin, and S. A. Chambers, Advanced Materials Interfaces 6, 1801428 (2019).
- [39]J. Garcia-Barriocanal, F. Y. Bruno, A. Rivera-Calzada, Z. Sefrioui, N. M. Nemes, M. Garcia-Hernández, J. Rubio-Zuazo, G. R. Castro, M. Varela, S. J. Pennycook, et al., Advanced Materials 22, 627 (2010).
- [40]J. Garcia-Barriocanal, J. C. Cezar, F. Y. Bruno, P. Thakur, N. B. Brookes, C. Utfeld, A. Rivera-Calzada, S. R. Giblin, J. W. Taylor, J. A. Duffy, S. B. Dugdale, T. Nakamura, K. Kodama, C. Leon, S. Okamoto, and J. Santamaria, Nature Communications 1, 82 (2010).
- [41]X. R. Wang, C. J. Li, W. M. Lü, T. R. Paudel, D. P. Leusink, M. Hoek, N. Poccia, A. Vailionis, T. Venkatesan, J. M. D. Coey, E. Y. Tsymbal, n. Ariando, and H. Hilgenkamp, Science 349, 716 (2015).
- [42]J. Stöhr and H. C. Siegmann, Solid-State Sciences. Springer, Berlin, Heidelberg 5, 236 (2006).
- [43]G. Herranz, M. Basletic, M. Bibes, C. Carrétero, E. Tafra, E. Jacquet, K. Bouzehouane, C. Deranlot, A. Hamzić, J.-M. Broto, A. Barthélémy, and A. Fert, Phys. Rev. Lett. 98, 216803 (2007).
- [44]R. Morrow, K. Samanta, T. Saha Dasgupta, J. Xiong, J. W. Freeland, D. Haskel, and P. M. Woodward, Chemistry of Materials 28, 3666 (2016).
- [45]S. Pal, S. Govinda, M. Goyal, S. Mukherjee, B. Pal, R. Saha, A. Sundaresan, S. Jana, O. Karis, J. W. Freeland, and D. D. Sarma, Phys. Rev. B 97, 165137 (2018).
- [46]M. J. Calderón, L. Brey, and F. Guinea, Phys. Rev. B 60, 6698 (1999).
- [47]Z. Li, M. Bosman, Z. Yang, P. Ren, L. Wang, L. Cao, X. Yu, C. Ke, M. B. H. Breese, A. Rusydi, W. Zhu, Z. Dong, and Y. L. Foo, Advanced Functional Materials 22, 4312 (2012), https://onlinelibrary.wiley.com/doi/pdf/10.1002/adfm.201200143.
- [48]A. Tebano, C. Aruta, S. Sanna, P. G. Medaglia, G. Balestrino, A. A. Sidorenko, R. De Renzi, G. Ghir-inghelli, L. Braicovich, V. Bisogni, and N. B. Brookes, Phys. Rev. Lett. 100, 137401 (2008).
- [49]D. Pesquera, G. Herranz, A. Barla, E. Pellegrin, F. Bondino, E. Magnano, F. Sánchez, and J. Fontcuberta, Nature Communications 3, 1189 (2012).
- [50]R. D. Shannon, Acta crystallographica section A: crystal physics, diffraction, theoretical and general crystallography 32, 751 (1976).
- [51]R. D. Leapman and L. A. Grunes, Phys. Rev. Lett. 45, 397 (1980).
- [52]M. Varela, M. P. Oxley, W. Luo, J. Tao, M. Watanabe, A. R. Lupini, S. T. Pantelides, and S. J. Pennycook, Phys. Rev. B 79, 085117 (2009).
- [53]K. Bawane, P. Manganaris, Y. Wang, J. Sure, A. Ronne, P. Halstenberg, S. Dai, S. K. Gill, K. Sasaki, Y.-c. K. Chen-Wiegart, et al., Scripta Materialia 197, 113790 (2021).
- [54]J. Sun, A. Ruzsinszky, and J. P. Perdew, Phys. Rev. Lett. 115, 036402 (2015).
- [55]G. Kresse and J. Furthmüller, Phys. Rev. B 54, 11169 (1996).
- [56]G. Kresse and J. Furthmüller, Computational Materials Science 6, 15 (1996).
- [57]S. K. Ojha, S. Hazra, P. Mandal, R. K. Patel, S. Nigam, S. Kumar, and S. Middey, Phys. Rev. Appl. 15, 054008 (2021).
- [58]P. Kuiper, G. Kruizinga, J. Ghijsen, G. A. Sawatzky, and H. Verweij, Phys. Rev. Lett. 62, 221 (1989).
- [59]D. Khomskii, Transition metal compounds (Cambridge University Press, 2014).DOI
- [60]F. Frati, M. O. Hunault, and F. M. De Groot, Chemical reviews 120, 4056 (2020).
- [61]A. Galdi, C. Aruta, P. Orgiani, C. Adamo, V. Bisogni, N. B. Brookes, G. Ghiringhelli, D. G. Schlom, P. Thakur, and L. Maritato, Phys. Rev. B 85, 125129 (2012).

