Volume 10 Number 12 28 March 2022 Pages 6253–6818 Journal of Materials Chemistry A Materials for energy and sustainability rsc.li/materials-a ISSN 2050-7488 PAPER José Santiso, Mónica Burriel et al. Role of pO2 and film microstructure on the memristive properties of La 2 NiO4+δ/LaNiO3−δ bilayers Journal of Materials Chemistry A This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online PAPER View Journal | View Issue Role of pO2 and film microstructure on the memristive properties of La2NiO4+d/LaNiO3d Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. Cite this: J. Mater. Chem. A, 2022, 10, 6523 bilayers† Klaasjan Maas,a Chloé Wulles,ab Jose Manuel Caicedo Roque,b Belén Ballesteros, b Valentin Lafarge,ab José Santiso *b and Mónica Burriel *a LaNiO3/La2NiO4 bilayers deposited at varying pO2 conditions resulted in remarkable differences in film microstructure and cell parameters, directly affecting the electrical behaviour of Pt/LaNiO3/La2NiO4/Pt Received 1st December 2021 Accepted 29th January 2022 devices. The devices deposited at low pO2 showed the largest memristance. We propose this is due to the formation of a p-type Schottky contact between LaNiO3 and La2NiO4, where the extent of its carrier DOI: 10.1039/d1ta10296f depletion width can be modulated by the electric-field induced drift of interstitial oxygen ions acting as rsc.li/materials-a mobile acceptor dopants in La2NiO4. 00 (Oi ), as is the case in La2NiO4.5,6 The additional interstitial Introduction oxygen ions (usually expressed as d) can be stabilized in the Mixed ionic electronic conducting materials (MIEC) based on form of over-stoichiometric La2NiO4+d, partially oxidizing Ni2+ transition metal oxides are key elements in many different into Ni3+ and resulting into a larger concentration of electronic applications, such as ceramic gas separation membranes, gas (hole) carriers. The p-type semiconducting properties and high sensing or as cathode materials for solid oxide fuel cells.1,2 oxygen diffusion of La2NiO4+d make it an interesting material Recently, a new application has come into play for the use as for the fabrication of VCM-based resistive switching devices. MIEC oxides in resistive switching devices, more particularly, in La2NiO4 has a layered Ruddelsden–Popper-type structure those based on the valence change mechanism (VCMs). The consisting of alternate La–Ni–O perovskite and La–O rock salt combined electronic conductivity and oxide ion mobility in layers along the c-axis of the structure.7 The oxygen diffusion these materials are key aspects in the functioning of these has been proven to be highly anisotropic in this layered struc- memristive devices. The dri of oxygen ions induced by the ture.8–11 Indeed, the interstitial oxygen anions located in the application of an external electric eld triggers internal redox rock salt layers diffuse preferentially along the a–b plane of the reactions, locally changing the structure and composition of the structure through an interstitialcy mechanism.8 Similarly, the material and therefore also its electrical properties. Typically, electronic conductivity takes place preferentially along the VCMs are based on the formation of conducting laments perovskite layers, leading to about three orders of magnitude across an insulating oxide such as SrTiO3, TiO2, Ta2O5 or HfO2. higher electrical conductivity along the a–b plane compared to This insulator is sandwiched between two metal electrodes; the the c-axis direction at room temperature.12,13 In this sense the conducting lament is produced by the ion migration of mobile control over the microstructure of the material, and more donors such as oxygen vacancies (V$$ O ) or cation interstitials particularly the preferential orientation of the crystal domains, under high electric elds and can be considered as a so is essential to understand the memristive properties of this dielectric breakdown.3 In addition, interface-type switching has material when used in VCM-type devices. been demonstrated for some MIEC oxides, such as Pr1xCax- For this study LaNiO3, a perovskite-structure oxide material MnO3 (PCMO), where the mobile defects are oxygen vacancies.4 showing metallic conductivity,14 has been selected as template However, very few studies are devoted to other types of MIEC electrode due its chemical affinity and crystallographic match oxides where the main mobile species are oxygen interstitials with La2NiO4. To the best of our knowledge the electrical and memristive behavior of the LaNiO3/La2NiO4 bilayer interface remains yet to be explored. a Univ. Grenoble Alpes, CNRS, Grenoble INP, LMGP, 38000 Grenoble, France. E-mail: This work aims at studying the microstructure and electrical monica.burriel@grenoble-inp.fr properties of LaNiO3/LaNiO4 epitaxial bilayers deposited under different oxygen partial pressure (pO2) during lm growth in b Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and the Barcelona Institute of Science and Technology (BIST), Campus UAB, 08193 Bellaterra, Barcelona, order to tune the oxygen content and therefore the contact Spain. E-mail: jose.santiso@icn2.cat † Electronic supplementary information (ESI) available. See DOI: properties and memristive behaviour of these complex-oxide 10.1039/d1ta10296f heterojunctions. This journal is © The Royal Society of Chemistry 2022 J. Mater. Chem. A, 2022, 10, 6523–6530 | 6523 View Article Online Journal of Materials Chemistry A Paper This study demonstrates that not only the oxygen stoichi- cooled down under the pO2 used for the growth of the top ometry of the lms, but also the overall crystal domain orien- La2NiO4 layer. For a proper comparison single layers of LaNiO3 tations are largely affected by the deposition conditions, leading and La2NiO4 were also grown at the two different pO2 conditions. to substantial differences in the memristive properties of these The XRD measurements (2q/u and f scans) were carried out This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. MIEC-based devices. The results presented here show that using a Malvern-Panalytical X'Pert-PRO MRD system with 4- oxygen dri across the material, as well as oxygen transport angle goniometer, and Cu Ka1 radiation (l ¼ 1.5406 Å, across the interface, are key aspects of the memristive response, 2xGe(220) monochromator). and could have implications both as memristive synapses for Microstructural characterization of the lm was carried out neuromorphic computing in low-energy consumption elec- by high-resolution transmission electron microscopy (HRTEM) tronics, as well as for a wider type of electrochemical devices using a FEI Tecnai G2 F20 microscope with eld emission gun Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. making use of this type of materials, such as solid oxide fuel working at 200 kV. Cross-section samples were prepared by FIB cells and electrolysers. on regions with top Pt electrodes (aer the electrical measurements). Experimental methodology During lm deposition one part of the sample was covered in situ by a metal mask aer the growth of the LaNiO3 preventing the LaNiO3/La2NiO4 epitaxial bilayers were grown by Pulsed laser growth of the subsequent La2NiO4 layer. In this way, aer depo- deposition by using a pulsed UV excimer laser (Lambda Physik sition, there is one part of the sample which has exposed LaNiO3 Compex Pro 201, l ¼ 248 nm) in a PLD-Workstation from layer. For the electrical characterisation of the bilayers an array of Surface-Tec, GmbH, and corresponding ceramic targets 50–200 mm diameter Pt electrodes were deposited by e-beam prepared by solid-state reaction. SrTiO3(001) single crystals (10 evaporation through a hard mask. Some electrodes were located  10 mm2, from Crystec, GmbH) were used as substrates. directly on the exposed part of the LaNiO3 bottom electrode layer, LaNiO3 was chosen as bottom electrode material, providing while the rest were on the La2NiO4 top layer. Electrodes of 200 mm a crystal matching for the subsequent epitaxial growth of in diameter were selected to form the devices, with a distance La2NiO4. The deposition temperatures were 600  C and 700  C between bottom and top electrodes between 1400 mm and 1800 for LaNiO3 and La2NiO4, respectively, as previously determined mm depending on the sample. Electrical measurements were for optimal crystal quality of single layers deposited on SrTiO3 carried out with a Keithley 2604B SourceMeter in a dark box, substrates. The values of the oxygen pressure during deposition using two tips to apply a voltage and simultaneously measure the were 12 mTorr and 70 mTorr. The pO2 range was limited by the resulting current. The voltage was applied to the Pt/LaNiO3 elec- PLD system as it can only operate in high vacuum conditions. trode while the La2NiO4/Pt electrode remained grounded. First, Other conditions for the sequential deposition of LaNiO3 and the initial resistance state (IRS) of the device was measured at low La2NiO4 were the same: target-to-substrate distance of 55 mm, voltage. Then, the device was cycled by triangular voltage sweeps and laser uence of 1.2 J cm2 (spot size 7.3 mm2), and pulse (0 V / +Vmax / Vmax / 0 V), while progressively increasing repetition rate of 2 Hz. Given the different ablation rates of the jVmaxj by 1 V every ve cycles, in order to trigger the memristive two targets and the different pO2 conditions the number of laser behavior, until jVmaxj ¼ 15 V was reached. The sweep rate was pulses was adjusted to obtain an approximate thickness of xed to 10 V s1. The current was measured aer each cycle (at 30 nm for both LaNiO3 and La2NiO4 layers. Three sets of bila- a voltage of 500 mV) to determine the resistance state. No elec- yers were prepared combining the growth of LaNiO3 and troforming step was required to trigger the memristance, and subsequent growth of La2NiO4 at either “High” pO2 (70 mTorr) thus these devices can be considered forming-free. or “Low” pO2 (12 mTorr), resulting in three bilayers with different pressure combinations, namely H/L, L/L and L/H Results and discussion samples. The attempted growth of a H/H bilayer was dis- carded from the study as it showed insufficient crystal quality. Sample preparation and characterisation The deposition protocols (shown in Fig. S1†) were designed to Structural characterization. The XRD measurements per- avoid as much as possible some parasitic effects. For instance, formed on the bilayers indicate a high degree of orientation of the difference in the optimal growth temperature between both LaNiO3 and La2NiO4 layers. Only 00L peaks of both LaNiO3 and La2NiO4 (600  C and 700  C, respectively) made it very compounds were observed in the 2q scans as depicted in Fig. 1a. important to minimize the time interval between subsequent The oscillations in the background of the XRD scans conrm depositions, since LaNiO3 is prone to vary its oxygen stoichiom- the high crystal quality of the layers. From the position of the etry and eventually decompose at high temperatures and low pO2, peaks it is possible to determine the corresponding out-of-plane when heating up from 600  C to 700  C. Therefore, this was an parameters, as shown in Table 1. As far as the LaNiO3 layers are important parameter to control. For the bilayer grown under concerned, the c-parameter is 3.83 Å for the layer deposited at constant low pO2 (L/L), La2NiO4 was deposited as soon as the high pO2 of 70 mTorr (H/L sample), while for the LaNiO3 sample reached 700  C aer a heating ramp of 10  C min1 (10 deposited at low pO2 we obtain c ¼ 3.86 and 3.87 Å, for L/L and min). For the growth of the H/L and L/H bilayers, the pO2 was only L/H samples, respectively. It seems that the c-axis parameter of changed aer having heated up the sample from 600  C to LaNiO3 layer mainly depends on the pO2 during its growth and 700  C, and then waiting 1 min for the pO2 to stabilize before less so on the pO2 present during the subsequent growth of growing the second layer in different pO2. The samples were all La2NiO4 (when grown at a different pO2). 6524 | J. Mater. Chem. A, 2022, 10, 6523–6530 This journal is © The Royal Society of Chemistry 2022 View Article Online Paper Journal of Materials Chemistry A Fig. S2 and S3†) in the sense that the lattice parameters measured for lms grown in low pO2 are expanded (about 1%) with respect to those obtained at high pO2. This can be explained by a larger oxygen vacancy concentration, as expected This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. for LaNiO3d. The crystal quality of the LaNiO3 layer is preserved at both pO2. However, when comparing the cell parameter values of LaNiO3 in the single layer deposited at low pO2 with those ob- tained in the bilayers at the same conditions, a slight reduction of the cell parameter between 0.2–0.4% is observed. This Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. reduction has been observed regardless of the pO2 used during the growth of the subsequent La2NiO4 layer. This means that LaNiO3 deposited at low pO2 oxidizes during the deposition of La2NiO4 (either at low or high pO2), while it maintains a constant oxygen content when deposited at high pO2. This is consistent with the observed oxygen transport previously re- ported in LaNiO3/La2NiO4 bilayers at elevated temperatures,15 although, a possible effect of the impinging energy of the growing species to scavenge oxygen vacancies from the under- lying LaNiO3 cannot be discarded.16,17 Therefore, the LaNiO3 layers deposited at low pO2 (L/L and L/H bilayers) could potentially develop an oxygen vacancy gradient at the interface with La2NiO4. A different situation occurs for the La2NiO4 layers. At low pO2 a clear c-axis orientation is observed for La2NiO4 (for the H/L and L/L bilayers), with an intense 006 reection for both lms. The c-axis parameter for the L/L samples is about 12.64 Å, similar to that obtained for La2NiO4 single layers directly grown on the SrTiO3 substrate (Fig. S3† and Table 1) for the same pO2 conditions. This could be another indication of a certain oxygen ion transport happening between the LaNiO3 and La2NiO4 layers. However, in this case it could not be ruled out that part of the Fig. 1 (a) High resolution X-ray diffraction 2q/u scans in the vicinity of changes in La2NiO4 cell parameter are due to a different strain 002 SrTiO3 substrate reflection of the three bilayer samples combi- when growing on LaNiO3 instead of directly on SrTiO3 nation of the growth at low and high pO2. The peaks labelled as 006 substrates. La2NiO4 and 002 LaNiO3 correspond to c-axis oriented domains. Note When growing La2NiO4 at high pO2, in the L/H bilayer, no that the weak 006 La2NiO4 peak in the L/H sample is due to the dominant a-axis orientation of this layer. (The peak with * corresponds 006 reection is observed. Therefore, this La2NiO4 lm does not to a double-crystal diffraction and is an indication of high crystal present domains with c-axis orientation, contrary to single coherence of the whole bilayer) (b) f-scans of the 101 La2NiO4 La2NiO4 lms which showed c-axis orientation when directly reflections indicating either dominant out-of-plane c-orientation or a- grown on SrTiO3 substrates (Fig. S2†). This type of scan does not orientation of the La2NiO4 film, for the extreme H/L and L/H samples, preclude the existence of other orientations of La2NiO4, i.e. a- respectively. axis orientation, which would overlap with the intense substrate peaks. In order to clarify whether La2NiO4 grows with a different The LaNiO3 c-parameter values are consistent with the orientation F-scans of asymmetric reections, such as 101 results obtained for the single layers, as shown also in Table 1, La2NiO4, were performed at a sample stage tilt angle c ¼ 16.7 , (the corresponding XRD patterns are in the supplementary part: which is the position expected for the a-oriented domains, as Table 1 Values of the c parameter of LaNiO3 and La2NiO4 as well as observed orientation of La2NiO4 domains depending on the bilayer deposition conditions (varying the oxygen partial pressure) LaNiO3 c-parameter Sample pO2 (mTorr) (Å) La2NiO4 c-parameter (Å) La2NiO4 orientation Bilayer (LaNiO3/La2NiO4) 70/10 (H/L) 3.832 12.664 Pure ct 10/10 (L/L) 3.862 12.637 Mixed ct and at 10/70 (L/H) 3.870 — Pure at Single layer 10 (L) 3.877 12.632 Pure ct 70 (H) 3.830 12.686 Pure ct This journal is © The Royal Society of Chemistry 2022 J. Mater. Chem. A, 2022, 10, 6523–6530 | 6525 View Article Online Journal of Materials Chemistry A Paper depicted in Fig. 1b. Indeed, the scan over the L/H sample shows contrast, presumably because in this case the c-axis lays the presence of broad 101 peaks with 4-fold symmetry. This perpendicular to the image. The corresponding FFT is consis- indicates the existence of a-axis orientation with perfect in- tent with pure at-orientation, as already shown by the f-scans plane alignment with the substrate [100]/[010] directions. For in Fig. 1b. The third column of Fig. 2 corresponds to colour- This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. comparison, the corresponding F-scan of the 101 reection for ltered images obtained by selecting the spots in the FFT for the c-axis oriented domains in the H/L sample, in this case at c either LaNiO3 phase (red), ct-La2NiO4 (green) and at (c//)- ¼ 72.1 , was also measured indicating again a four-fold La2NiO4 (blue). The H/L sample shows full ct-La2NiO4 (full symmetry with a large degree of in-plane arrangement. There- green colour), while the L/L sample shows mixed ct- and at- fore, the growth of La2NiO4 at high pO2 promotes the growth of La2NiO4 (green with blue patches). The L/H sample shows a-axis domains, while c-axis orientation is obtained at low pO2. a columnar arrangement with an intimate mixture of blue and Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. This situation was different to the direct growth of La2NiO4 red stripes, with nanometer size domains, corresponding to at- on SrTiO3(001) substrates where a high degree of c-axis orien- La2NiO4. The red stripes correspond in this case to at-La2NiO4 tation is achieved at both pO2 values (Fig. S3†). Therefore, it is domains with c-axis pointing perpendicular to the image, which likely that the exposure of underlying LaNiO3 to the high pO2 are not distinguished from a standard perovskite structure (red before the subsequent growth of La2NiO4 affects the LaNiO3 colour), but obviously do not correspond to LaNiO3 although topmost surface to inhibit the perfect cube-on-cube arrange- they show the same FFT pattern. The right column in Fig. 2 ment of perovskite blocks of c-axis oriented La2NiO4 on LaNiO3. summarizes these observations in the form of a schematic of In agreement with the La–Ni–O thermodynamic phase the samples' microstructures. The regions where La2NiO4 diagram,2 it is possible that some secondary phases, resulting domains with different orientation meet together form grain from LaNiO3 decomposition into La2NiO4 + NiO are also present boundaries. In the pure ct-axis oriented lm (H/L) there are not at the interface, which could alter the subsequent growth of grain boundaries in the La2NiO4 layer, whereas in the mixed at/ La2NiO4. ct-oriented La2NiO4 layer (in the L/L sample) there are vertical Bilayer microstructure. In order to address the microstruc- and horizontal (001)/(100) boundaries between at and ct ture of the bilayers cross-sectional samples were prepared from domains. In the pure at La2NiO4 layer (in the L/H sample) the the same set of bilayers. in-plane rotation by 90 of their corresponding c-axis denes Fig. 2 shows the cross-sections oriented along [100]-zone axis also vertical (001)/(100) domain boundaries. The vertical and at the interface between bottom LaNiO3 and top La2NiO4 layers horizontal domains have been marked with green and red for the three different deposition conditions: H/L (top), L/L colours, respectively, in Fig. 2 (right column). (middle), and L/H (bottom). The le column shows HRTEM To conclude this section, different growth conditions (pO2) images of the bilayers where clear differences in microstructure induce remarkable differences not only in the oxygen concen- are revealed in top and bottom layers. All the bilayers show tration proles across the bilayer, with possible gradients at the highly oriented LaNiO3 with a large density of vertical planar LaNiO3d/La2NiO4+d interface, but also in the bilayer nano- defects, which very likely correspond to Ruddlesden–Popper structure, with the appearance of a particular type of (001)/ type defects, a sort of antiphase boundary defect with (100) grain boundaries within La2NiO4. These differences will a displacement of half primitive perovskite cell along the h111i surely affect the electrical properties of these bilayers and could direction, as previously shown in.18,19 ultimately dictate the memristive behaviour of Pt/LaNiO3/ On the other hand the nano-structural differences of the La2NiO4/Pt devices, as will be discussed in the following section. La2NiO4 lms of the bilayers are more pronounced. The H/L sample shows a sharp interface with highly c-axis oriented La2NiO4. The corresponding FFT of the top La2NiO4 layers are Role of the pO2 on the electrical properties of Pt/LaNiO3/ shown in the second column in Fig. 2. For the H/L sample the La2NiO4/Pt memristive devices pattern is consistent with a pure ct-orientation, and a- or b- axis Fig. 3 presents an overview of the electrical response of the three parallel to the image, aligned with in-plane main [100]/[010] bilayers using circular Pt contacts with a diameter of 200 mm. substrate directions. For the L/L sample, the rst layers at the The voltage was applied to the Pt electrodes directly on top of interface with LaNiO3 start growing with ct-orientation, but LaNiO3 bottom layer (bright part of the image in Fig. 3a), while aer 5–8 nm additional domains with a different orientation the Pt contacts on top of La2NiO4 (dark side of Fig. 3a) were start to nucleate, resulting in a FFT which consists of a mixture grounded, as shown in the schematic of Fig. 3b. The shape of of ct- and at-orientations. The existence of a ct-component each voltage sweep cycle versus time is depicted in Fig. 3c. The was already observed in the XRD pattern of this lm. However, I(V) measurements were reproducible when cycling within the existence of a mixed at-orientation was not detected a xed voltage range. because of the overlap with the substrate peaks. For the L/H The I(V) curves measured for the three samples are shown in bilayer the majority of the La2NiO4 lm seems to be formed Fig. 3d–f, corresponding to every h cycle at the 1 V, 5 V, by narrow columns of at-oriented domains, some of them with 10 V and 15 V sweeps. Interface-type memristive devices the c-axis parallel to the image (c//). This orientation is identied typically show rectifying I–V characteristics in their initial at atomic scale by the arrangement of bright vertical atomic resistance state (IRS). This nonlinear behaviour is usually planes corresponding to the a/b plane of the Ruddlesden– explained by the presence of a Schottky-like barrier at the metal/ Popper structure, while other domains do not show such oxide interface necessary for the emergence of the memristive 6526 | J. Mater. Chem. A, 2022, 10, 6523–6530 This journal is © The Royal Society of Chemistry 2022 View Article Online Paper Journal of Materials Chemistry A This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. Fig. 2 HRTEM images (left column) of cross sections of the different bilayers from top to bottom: H/L, L/L and L/H LaNiO3/La2NiO4 bilayers, as well as corresponding FFT patterns of the La2NiO4 layer (second column). The third column shows colour-filtered images for the selected spots in the FFT corresponding to cubic perovskite (red), ct-Ruddlesden–Popper (RP) (green), at-RP (blue). The column on the right shows a schematic of the film's microstructure and domain orientation (the c-axis is indicated with blue arrows) for the three types of samples, showing the formation of boundaries between at- and ct-domains: perpendicular (marked in red colour) and parallel (green) to the vertical axis. It also 00 shows the high concentration or low concentration of oxygen interstitials (Oi ) in the initial state depending on the deposition conditions. behavior.5 Both the H/L and the L/L samples show nonlinear form of hysteresis which is most pronounced at Vmax ¼ 10 V. It I(V) characteristics in their IRS (I–V sweeps between 1 V). Since should be noticed that the small hysteretic behavior disappears Pt forms an ohmic contact with both LaNiO3 and La2NiO4 (as when Vmax is increased to 15 V for the H/L and L/H samples, as shown in Fig. S4†) the nonlinearity can only be explained by the clearly shown in the resistance versus voltage plots in Fig. S5.† presence of a Schottky-like barrier at the LaNiO3/La2NiO4 The disappearing of the hysteresis seems to be irreversible as complex-oxide interface. The increased nonlinearity measured the subsequent application of a lower voltage does not restore for the L/L sample could signify the presence of a larger the original opening in two resistance states (not shown here). Schottky barrier. On the other hand the L/H sample shows On the other hand the L/L sample progressively shows an almost linear I(V) characteristics between 1 V and should increasingly rectifying behaviour with two very distinct resis- therefore present the smallest LaNiO3/La2NiO4 contact resis- tance states. The Low Resistance State (LRS – higher current tance. As will be later explained, this difference is likely due to branch), shows little rectication within the 2 V sweeping the high electronic conductivity expected for the La2NiO4 top range, i.e. before the devices switches to the High Resistance layer of the L/H sample due to its at orientation, high oxygen State (HRS). Conversely, the HRS shows highly rectifying I(V) overstoichiometry (d) and concomitant high concentration of characteristics with an increased threshold voltage when electronic hole carriers. compared to the LRS. This hysteretic behavior and the distinct Other interesting differences can be noticed between the LRS and HRS states are well maintained even for the largest samples when increasing Vmax gradually from 1 V to 15 V. voltage sweep range (15 V). The evolution of the HRS and LRS Both the HRS and the LRS for all three samples present some values (measured at 500 mV immediately aer each sweep) as This journal is © The Royal Society of Chemistry 2022 J. Mater. Chem. A, 2022, 10, 6523–6530 | 6527 View Article Online Journal of Materials Chemistry A Paper This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. Fig. 3 (a) Optical image of the top surface of the Pt/LaNO3/La2NiO4/Pt sample showing the circular Pt electrodes. (b) Schematic of the measuring configuration from grounded top Pt electrode on the La2NiO4 layer to biased Pt electrode on the LaNiO3 bottom layer. (c) Applied triangular voltage sweeps. (d)–(f) Evolution of the I(V) characteristics of the H/L, L/L and L/H bilayers, respectively, when increasing the maximum voltage from left to right (1 V, 5 V, 10 V and 15 V). The L/L sample shows clear analog-type switching characteristics. (Arrows indicate the direction of the voltage sweep). well as the corresponding HRS/LRS ratios are shown in Fig. S6† conducting character of La2NiO4+d, and its anisotropy, and we for the three bilayer samples. For the H/L and L/H samples both present two possible scenarios or mechanisms which can the HRS and LRS show very low values (<103 U) below Vmax ¼ explain the observed memristance. The layered Ruddlesden– 7 V. For larger voltage sweep ranges, in the H/L sample both Popper structure of La2NiO4+d is expected to induce a high states increase to values up to 104 U, while for the L/H sample anisotropy both in the electronic transport and oxygen diffusion they remain relatively low. However, in both cases the HRS/LRS properties.10–13 At room temperature the resistivity along the c- ratios stay between 1 and 3 for the whole measured range. axis was measured to up to three orders of magnitude larger Interestingly the L/L sample shows a completely different than that within the basal plane,12,13 mainly attributed to behaviour where both the HRS and LRS values continuously differences in mobility due to the variation in interplanar increase all the way from the lowest values of Vmax ¼ 1 V until separation within the crystallographic structure. In addition, as Vmax ¼ 15 V. At Vmax ¼ 7 V, the LRS reaches a stable value of a p-type conductor, the resistivity highly depends on the oxygen around 104 U, while the corresponding HRS continues content, and more particularly to the interstitial oxygen content, increasing above 105 U. Therefore, the corresponding HRS/LRS which is directly related to the concentration of electronic hole ratio shows a pronounced increase from 1–2 below Vmax ¼ 7 V carriers. We recently showed that for epitaxial La2NiO4+d thin up to 20–30 at Vmax ¼ 15 V. lms prepared by Pulsed Injection Metal Organic Chemical Focusing on the LaNiO3/La2NiO4 interface we interpret the Vapor Deposition, we could easily change the oxygen interstitial resistive switching behavior based on the mixed ionic-electronic concentration by annealing in different gas atmospheres (from 6528 | J. Mater. Chem. A, 2022, 10, 6523–6530 This journal is © The Royal Society of Chemistry 2022 View Article Online Paper Journal of Materials Chemistry A H2/Ar to O2), leading to changes in the c-axis out of plane cell Joule heating. Finally, it is possible that the two mechanisms parameter from 12.62 to 12.69 Å.9 The resistivity values of the evoked previously are cohabitating, both playing a role in the different lms showed three orders of magnitude increase in measured change in resistance of the device. resistivity (from 5.3  103 to 5.7 U cm) when decreasing the While the hysteretic behavior is much more pronounced for This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. oxygen content (smaller cell parameter). Thus, in the case of the the L/L bilayer, the comparison of the lm's microstructure L/H sample studied here a higher electronic conductivity is ex- between H/L and the L/L samples, as observed by HRTEM and pected across the top La2NiO4+d lm due to: (i) an increased depicted in the schematic in Fig. 2, does not reveal any relevant oxygen content (compared to the other two samples) leading to difference at the LaNiO3/La2NiO4 interface itself. Both bilayers a larger hole carrier concentration, and (ii) its at-orientation show sharp interfaces between ct-LaNiO3 and ct-La2NiO4 (opposite to the other samples), and thus along the fastest layers across the entire interfacial region. However, the differ- Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. electronic transport path in this anisotropic material. This ences in their respective LaNiO3 and La2NiO4 cell parameters could signicantly reduce the carrier depletion width on the seem to indicate a variation in their corresponding oxygen La2NiO4+d side of the interface Schottky contact, therefore defect concentrations, higher V$$O concentration in the LaNiO3 reducing the overall contact resistance. Consequently, this side for the L/L sample (larger cell parameter, see Table 1) and 00 prevents the building up of a memristive behavior in this bilayer higher Oi concentration in the La2NiO4 layer for the H/L sample (small HRS/LRS ratio at Vmax ¼ 10 V that then disappears at (larger cell parameter, see Table 1). The reduced concentration higher voltages). of V$$ O within LaNiO3 in the H/L sample, is expected to limit the In the following we focus only on the H/L and L/L samples amount of oxygen ions that can be incorporated into LaNiO3 which, do present a Schottky-like characteristics and present (from the La2NiO4 lm) and their dri within the lm towards the rst potential mechanism. Besides the dri of highly mobile the Pt electrode. This could be one of the reasons for the smaller electronic charges, typically forming a depletion zone at the HRS/LRS ratios measured for this sample compared to the L/L interface, in a material such as La2NiO4+d the application of an sample. The second main difference between these two bila- external electrical eld above a certain voltage facilitates the yers is found in the nano-structure of the top part of the La2NiO4 dri of charged oxygen defects (negatively-charged interstitial lm. A mixed orientation of the La2NiO4 domains in the L/L 00 oxygen, Oi , in this case).10 Then, the accumulation of oxygen sample generates a high density of (001)/(100) boundaries interstitials at the interface with LaNiO3 may produce a change within the La2NiO4 layer. It is likely that these internal (001)/ in the Schottky barrier as well as in the material properties in (100) nano-domain boundaries form additional barriers, effec- this location, which result in a change in the contact resistance tively blocking both electronic and ionic dri particularly those and ultimately in the overall resistance of the device. Indeed, in boundaries laying perpendicular to the electric eld (red colour the L/L sample, that shows the largest memristive behavior, in the schematic presented in the right column of Fig. 2), and a strong nonlinearity was observed at low voltage, probably therefore, accumulating traps in the volume of the La2NiO4 caused by this potential barrier. The resistance-change mecha- layer. If the presence of these domain boundaries helps nism in Pt/LaNiO3/La2NiO4/Pt could therefore be explained as increasing the rectifying and hysteretic I(V) characteristics, this follows: initially, the device is in a HRS. When applying a high could bring an additional strategy for engineering lm micro- enough positive voltage (on the LaNiO3 side) the negatively- structures with enhanced and better controlled memristive charged oxygen ions dri towards the LaNiO3/La2NiO4 inter- properties. face, locally changing the carrier density (the mobile interstitial 00 oxygen ions, Oi , act as hole dopants in La2NiO4) and reducing Conclusions the charge depletion width in this location. The device now nds itself in its LRS as the ow of current is now facilitated High crystal quality preferentially oriented LaNiO3/La2NiO4 through a thinner potential barrier (additional conduction bilayers were grown using PLD under varying pO2 conditions. mechanisms can be present, i.e. tunneling). The opposite The results conrmed that the oxygen partial pressure used occurs when applying a high enough negative voltage. Oxygen during the growth of the lms clearly impacts their micro- interstitials in La2NiO4 dri away from the LaNiO3/La2NiO4 structure, severely affecting the domain orientation of the interface reestablishing the barrier width, and therefore the La2NiO4 layer. At the same time electrical measurements carried HRS. out on Pt/LaNiO3/La2NiO4/Pt devices fabricated out of these Although, in the previous explanation, the oxygen interstitial bilayers conrmed important pO2-dependent differences in the accumulation region has been limited to the La2NiO4 part at the resistive switching behavior of these complex-oxide stacks. Two interface with LaNiO3, a possible exchange of oxygen ions switching scenarios have been proposed to explain the mem- between the LaNiO3d and the La2NiO4+d layers, may result in an ristive behavior of the bilayers. The proposed mechanisms extension of the depletion region through an internal redox could be occurring independently or simultaneously, depend- mechanism, also modifying the contact resistance. This corre- ing on the voltage range. While the rst one has been attributed sponds to the second mechanism proposed. This recombina- to the electric eld-induced dri of oxygen interstitial ions in tion at the interface of LaNiO3d and the La2NiO4+d layers has the La2NiO4 layer, accumulating at the LaNiO3/La2NiO4 inter- been observed at 600  C by Moreno et al.,15 and it is likely that face, and modulating the depletion width of the Schottky the exchange of oxygen ions between the two materials can barrier present; the second one would be related to the occur at room temperature, assisted by the electric eld and by exchange of oxygen ions between the top La2NiO4 and the This journal is © The Royal Society of Chemistry 2022 J. Mater. Chem. A, 2022, 10, 6523–6530 | 6529 View Article Online Journal of Materials Chemistry A Paper bottom LaNiO3 lms. This study also reveals the presence of 4 C. Baeumer, T. Heisig, B. Arndt, K. Skaja, F. Borgatti, F. Offi, a high density of boundaries between domains with ct and at- F. Motti, G. Panaccione, R. Waser, S. Menzel and orientation, under particular lm growth conditions, which R. Dittmann, Faraday Discuss., 2019, 213, 215–230. could also play a signicant role in the accumulation of oxygen 5 K. Maas, E. Villepreux, D. Cooper, C. Jiménez, H. Roussel, This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. interstitials in the volume of the La2NiO4 layer, possibly L. Rapenne, X. Mescot, Q. Raay, M. Boudard and improving the memristive properties of these novel bilayer M. Burriel, J. Mater. Chem. C, 2020, 8, 464–472. nickelate-based devices. 6 K. Maas, E. Villepreux, D. Cooper, E. Salas-Colera, J. Rubio-Zuazo, G. R. Castro, O. Renault, C. Jimenez, Author contributions H. Roussel, X. Mescot, Q. Raay, M. Boudard and M. Burriel, Adv. Funct. Mater., 2020, 30, 1909942. Open Access Article. Published on 02 February 2022. Downloaded on 10/31/2022 12:30:42 PM. Conceptualization: KM, MB and JS; thin lm deposition: CW, 7 A. Aguadero, J. A. Alonso, M. J. Martı́nez-Lope, JMCR, VL; TEM analysis: BB, formal electrical characterisation: M. T. Fernández-Dı́az, M. J. Escudero and L. Daza, J. Mater. CW, KM; writing –review & editing: KM, MB, JS. Chem., 2006, 16, 3402–3408. 8 A. Chroneos, D. Partt, J. A. Kilner and R. W. Grimes, J. Conflicts of interest Mater. Chem., 2010, 20, 266–270. 9 J. M. Bassat, P. Odier, A. Villesuzanne, C. Marin and There are no conicts to declare. M. Pouchard, Solid State Ionics, 2004, 167, 341–347. 10 M. Burriel, G. Garcia, J. Santiso, J. A. Kilner, R. J. Chater and Acknowledgements S. J. Skinner, J. Mater. Chem., 2008, 18, 416–422. 11 M. Burriel, H. Téllez, R. J. Chater, R. Castaing, P. Veber, This work has been nanced by the ANR funded project M. Zaghrioui, T. Ishihara, J. A. Kilner and J.-M. Bassat, J. “MICROSWITCH” (ANR-14-ACHN-0012) and by the “FUN-TO- Phys. Chem. C, 2016, 120, 17927–17938. BE” CNRS-CSIC supported International Program for Scientic 12 J. M. Bassat, F. Gervais, P. Odier and J. P. Loup, Mater. Sci. Cooperation (PICS, no. PICS07294). Part of this work was Eng., B, 1989, 3, 507–514. carried out through the access to PLD and HRTEM facilities 13 J. M. Honig and D. J. Buttrey, in Localization and Metal- granted by the NFFA-Europe Infrastructure (proposal ID 886) Insulator Transitions, Springer US, Boston, MA, 1985, pp. under Horizon 2020 EU Funding Program. JS acknowledges the 409–418. grant PID2019-108573GB-C21 funded by MCIN/AEI/10.13039/ 14 T. Yu, Y. F. Chen, Z. G. Liu, X. Y. Chen, L. Sun, N. Ben Ming 501100011033. This work was also supported in part by the and L. J. Shi, Mater. Lett., 1996, 26, 73–76. Spanish Ministry of Economy, Industry and Competitiveness 15 R. Moreno, P. Garcı́a, J. Zapata, J. Roqueta, J. Chaigneau and (projects MAT2016-77100-C2-1-P), and by the Catalan AGAUR J. Santiso, Chem. Mater., 2013, 25, 3640–3647. agency (project: 017-SGR-579). ICN2 is funded by the CERCA 16 J. Zhu, J.-W. Lee, H. Lee, L. Xie, X. Pan, R. A. De Souza, programme/Generalitat de Catalunya and by the Severo Ochoa C.-B. Eom and S. S. Nonnenmann, Sci. Adv., 2019, 5, programme (SEV-2017-0706). eaau8467. 17 E. Mikheev, B. Hoskins, D. Strukov and S. Stemmer, Nat. References Commun., 2014, 5, 3990. 18 E. Detemple, Q. M. Ramasse, W. Sigle, G. Cristiani, 1 I. Riess, Solid State Ionics, 2003, 157, 1–17. H.-U. Habermeier, B. Keimer and P. A. van Aken, J. Appl. 2 J. A. Kilner and M. Burriel, Materials for intermediate- Phys., 2012, 112, 013509. temperature solid-oxide fuel cells, 2014, vol. 44. 19 S. Bagdzevicius, M. Boudard, J. M. Caicedo, X. Mescot, 3 C. Lenser, R. Dittmann and J. P. Strachan, in Resistive R. Rodrı́guez-Lamas, J. Santiso and M. Burriel, Solid State Switching, Wiley-VCH Verlag GmbH & Co. KGaA, Ionics, 2019, 334, 29–35. Weinheim, Germany, 2016, pp. 437–456. 6530 | J. Mater. Chem. A, 2022, 10, 6523–6530 This journal is © The Royal Society of Chemistry 2022