Next Article in Journal
Multi-Graph Convolutional Network for Fine-Grained and Personalized POI Recommendation
Previous Article in Journal
Eight-Element Antenna Array with Improved Radiation Performances for 5G Hand-Portable Devices
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Electronics
Volume 11
Issue 18
10.3390/electronics11182963
electronics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Bipolar Resistive Switching in Hafnium Oxide-Based Nanostructures with and without Nickel Nanoparticles

by
Markus Otsus
*,
Joonas Merisalu
,
Aivar Tarre
,
Anna-Liisa Peikolainen
,
Jekaterina Kozlova
,
Kaupo Kukli
and
Aile Tamm
*
Institute of Physics, University of Tartu, W. Ostwaldi 1, 50411 Tartu, Estonia
*
Authors to whom correspondence should be addressed.
Electronics 2022, 11(18), 2963; https://doi.org/10.3390/electronics11182963
Submission received: 25 August 2022 / Revised: 13 September 2022 / Accepted: 16 September 2022 / Published: 19 September 2022
(This article belongs to the Section Circuit and Signal Processing)
Browse Figures
Versions Notes

Abstract

:
As research into additives and intentionally introduced impurities in dielectric thin film for enhancing the resistive switching based random access memories (RRAM) continues to gain momentum, the aim of the study was to evaluate the effects of chemically presynthesised Ni nanoparticles (NPs) embedded in a dielectric layer to the overall structure and resistive switching properties. HfO2-based thin films embedded with Ni NPs were produced by atomic layer deposition (ALD) from tetrakis(ethylmethylamino)hafnium (TEMAH) and the O2 plasma ALD process onto a TiN/Si substrate. The Ni NPs were separately synthesised through a continuous flow chemistry process and dispersed on the dielectric layer between the two stages of preparing the HfO2 layer. The nanodevices’ morphology and composition were analysed with physical characterisation methods and were found to be uniformly dispersed across the sample, within an amorphous HfO2 layer deposited around them. When comparing the resistive switching properties of otherwise identical samples with and without Ni NPs, the ILRS/IHRS ratio rose from around a 4 to 9 at 0.2 V reading voltage, the switching voltage dropped from ~2 V to ~1.5 V, and a distinct increase in the endurance characteristics could be seen with the addition of the nanoparticles.

1. Introduction

Research investigating the resistive switching (RS) phenomena in solid thin films has gained a lot of interest over the last decade, offering the potential for fabricating a new generation of nonvolatile memory devices named resistive random access memories (RRAMs) [1]. Among various thin metal oxide films studied as RS media for memristor cells, HfO2 films have been mentioned as a media where both unipolar and bipolar RS could be observed depending on the cell design [2,3,4,5]. Additionally, hafnium dioxide has a high dielectric constant and exhibits good chemical stability, making it widely studied in RRAM structures [6,7].
Even though many different materials have been extensively studied and evaluated as potential dielectrics for RRAMs, the effects of different additives to or within the dielectric layer have been receiving more and more attention in the last few years to further enhance their properties. It has been proposed that locally controlled impurities within the dielectric layer aid in the formation of conductive filaments (CF). Recent studies have shown the positive effects of doping of the dielectric [8,9,10,11] or adding nanodots [6], particles [12], crystals [13], even nanoislands [14] and graphene [15] to the dielectric layer, lowering the SET/RESET voltages of RRAMs, increasing endurance, and helping guide the formation of conductive filaments. Recently, Wang et al. [12] have even proposed the mechanism of a hybrid conductive filament, combining metal and oxygen vacancies-based models, brought about by the addition of a Ag and Cu nanoparticle layer. Introducing controlled impurities has been shown to help modify not only the characteristics relevant to resistive switching but also the magnetic properties of materials, possibly making way for new applications [16]. For example, we recently showed that layering HfO2 with Fe2O3 allows us to control the magnetic and resistive switching properties of a structure [17]. Nanocomposites of HfO2 and foreign metal particles have also been studied as RS media [6]. However, the addition of such impurities can also lead to the appearance of different types of RS that can coexist in these structures [18], which may destabilise the RS and complicate programming-reading operations.
There are a number of different ways for depositing HfO2 when using atomic layer deposition (ALD), but one of the most common ones is Hf[N(CH3)(C2H5)]4 (TEMAH). It, alongside other alkylamide-based precursors, is mainly used due to its low deposition temperature and increased nucleation density of HfO2. Moreover, for example in comparison to another widely used precursor, HfCl4, there is no risk of chlorine residue [15,19]. ALD allows one to deposit conformal nanolayers and effectively coat nanostructures, with a thickness in ideal cases in the Å range, while not being dependent on the size and complexity of the surfaces being coated. ALD is a chemical vapor phase technique, relying on self-limiting reactions controlled by exact purges of reactants and inert gases [20,21]. All of these factors make it an ideal method for building various complex nanodevices.
There have been a number of works looking at different ways of combining HfO2 and Ni, among them works going over pure Ni [22] and combined metal electrodes [23], as well as embedded Ni nanocrystals [24,25]. Z. Tan et al. showed a clear increase in the memory window of HfO2 based devices for nonvolatile memory applications after embedding Ni nanocrystals into the dielectric layer. All in all, however, Ni and HfO2 pairings, be it with Ni as a separate electrode or an additive, remain relatively little researched [26], even though it might provide interesting additional opportunities, for example in relation to Ni magnetic properties. To the best of our knowledge, there have been no previous studies on using presynthesised Ni nanoparticles to be embedded within an ALD HfO2 dielectric layer. This approach might allow for the use of novel materials, a wider range of synthesis temperatures for both the particles and the nanostructures, as well as reduce the risk of residue from the synthesis of these controlled impurities. Additionally, HfO2 + Ni NPs as a system has not been extensively studied, even though Ni has been shown to enhance resistive switching properties [9].
In the present work, we aim to study how the electrical characteristics of RS and formation mechanisms of CFs in hafnium oxide thin films with and without presynthesised nickel nanoparticles (Figure 1a) differ. Ni nanoparticles (NP) were embedded into a metal insulator metal-like (MIM) nanostructure where the main insulator layer was made using atomic layer deposition (ALD) of hafnium dioxide. The aim of the study was to investigate the physical structure and resistive switching characteristics in otherwise identical samples with and without Ni nanoparticles and show how the addition of Ni NPs affects the nanodevices.

2. Materials and Methods

The HfO2 thin films were deposited from Hf[N(CH3)(C2H5)]4 (TEMAH) and O2 plasma in a commercial Picosun R-200 Advanced ALD system (Picosun OY, Helsinki, Finland). An initial HfO2 film was deposited at 220 °C with the first 50 cycles carried out with pulse times 0.3-4-15-4 s (TEMAH-N2 (purge)–O2 (plasma)–N2, respectively). The Ni nanoparticles were dispersed onto the sample by the dip-coating method. For that, the semiprepared wafer was submerged into a solution containing Ni NPs for 90 s, after which a top layer of HfO2 was deposited on top of the nanoparticles. For the top layer, 8 cycles of only TEMAH were deposited directly onto the particles, with cycle times 0.3-8-0-0 s, followed by 70 full cycles of TEMAH with an O2 plasma similar to the first 50 cycles. All films were deposited on crystalline TiN-covered substrates cut out of Si(100) wafers with a resistivity of 0.014–0.020 Ω·cm. The Si wafers were boron-doped to concentrations ranging from 5 × 1018 to 1 × 1019 cm–3. The TiN layer of 10 nm was pregrown by pulsed chemical vapor deposition using a batch TiCl4/NH3 process at temperatures of 450–500 °C in an ASM A412 Large Batch 300 mm reactor (Advanced Semiconductor Materials, Almere, The Netherlands) at Fraunhofer IPMS-CNT (Dresden, Germany). Samples without Ni NPs were prepared alongside samples with Ni NPs, effectively having an identical nanolaminate structure with the exception of having been submerged in the solution with dispersed Ni NPs (Figure 1a).
Ni nanoparticles were separately synthesised using a continuous flow synthesis reactor FlowSyn Multi-X™ (Uniqsis, Royston, UK) from nickel(II) 2,4-pentanedionate (Ni(acac)2, 95%, Alfa Aesar, Haverhill, MA, USA) in continuous mode, using benzyl alcohol as a reducing and particle-stabilising agent. An amount of 30 mM Ni(acac)2 solution in N,N-dimethylformamide (DMF, reagent grade, Honeywell, Charlotte, NC, USA) was mixed with benzyl alcohol in a volume ratio of 1:1 in flow at 30 °C. The reaction was conducted at 170 °C, with system pressure being 3.5–4 bar and a reaction time of 22 min. The dispersion of Ni particles was collected into a glass vial. The powder was then dried under vacuum, and then redispersed in 2-propanol:benzyl alcohol mixture (1000:1).
The surface morphology was characterised by using a high-resolution scanning electron microscope (SEM) Helios NanoLab 600 (FEI) in a high-resolution mode (10 kV, 86 pA). Compositional analysis was performed using an energy-dispersive X-ray (EDX) spectrometer INCA Energy 350 (Oxford Instruments, Abingdon, UK), attached to the same microscope. Scanning transmission electron microscopy (STEM) and elemental analysis of the nanostructure in cross-sectional orientation were performed in a Cs-corrected Titan Themis 200 microscope (FEI, Hillsboro, OR, USA) equipped with a Super-X EDX system (FEI/Bruker) at 200 kV. EDX maps were acquired using Esprit software version 1.9 (Bruker, Billerica, MA, USA). Thin cross-sectional samples for STEM observations were prepared using the in situ lift-out technique using the Helios Nanolab 600 scanning electron microscope/focused ion beam (FIB) system (FEI, Hillsboro, OR, USA). The same SEM-FIB system was used to carry out initial cross-sectional analysis of the samples.
A diffractometer SmartLab (Rigaku, Tokyo, Japan) with a rotating Cu anode (λ = 0.15406 nm) working at 8.1 kW was used for X-ray diffraction (XRD) measurements for evaluating the structure of the nanostructures. A small glancing angle of 0.35 degrees was used to reduce the background intensity of the TiN/Si substrate and maximise that of HfOx thin film.
For the electrical measurements, samples were supplied with titanium top electrodes, prepared by electron-beam evaporator (EBE), and using maskless photolithography (μMLA Heidelberg Instruments, Zurich, Switzerland). The top electrodes used in the measurements had a diameter of 50 µm and an area of 0.002 mm2 (Figure 1b). Electrical measurements were carried out in a light-proof and electrically shielded box on MPS150 probe station (Cascade Microtech, Beaverton, OR, USA), using a 2636 A source-meter (Keithley Instruments, Cleveland, OH, USA). Endurance measurements, carried out in addition to the characterisation measurements, were done in four sequential source (V) and measure (I) steps with 0.6 s intervals consisting of SET at 1.5 V, READ at −0.2 V, READ at 0.2 V, and RESET at 1.5 V, respectively, making the time interval between SET and RESET 1.8 s.

3. Results and Discussions

3.1. Initial Characterisation of the Samples

The preliminary size and structure of the particles were found to be relatively uniform, with no large clustering and a polycrystalline structure of Ni. Figure 2a shows a STEM bright field (BF) image of the prepared Ni NPs after synthesis, before being dispersed onto the aimed nanostructure surface. Particles were centrifuged out of the initial solution in order to separate, dry, and disperse the Ni NPs onto a TEM grid. Therefore, the row of particles seen in Figure 2a is not representative of the particle placement on the final device. As can be seen, the as-synthesised Ni NPs had spherical morphology and were polycrystalline. The average NP diameter estimated from STEM images was 81 nm (σ = ±3 nm). Figure 2b shows the overall distribution of the Ni nanoparticles (seen as white dots) after they had been dispersed onto the surface of the initial HfO2 layer before the deposition of a secondary hafnium layer and the Ti top electrode. As can be seen, the distribution was mainly uniform across the substrate, even though some larger clusters of the NPs were seen. According to the SEM measurements, the average size of the nanoparticles dispersed on the surface was 88 nm, being only slightly larger than the size of the as-synthesised NPs. The average distance between particles was 1.7 µm (σ = ±0.1 µm), and the estimated average density of the Ni NPs was 3 NPs per 10 µm2. When comparing the distribution of NPs to those found in previous works [13,24,25], our method of submerging the wafers results in a much sparser distribution, reducing the impact of the particles to the overall surface morphology of the devices. We are also able to say that the NPs are distinctly isolated and separate, so the effect of an individual Ni NP can be expected to be more limited to certain regions of the device and not cover the whole surface. This aspect makes it harder to detect and evaluate their amount and distribution in later stages with various methods, such as XRD, for example, as the concentration of Ni is restricted to discrete locations. However, this also allows us to investigate the effects of separate nanoparticles, not more or less connected webs of particles, or even a distinct layer, which can be assumed to happen in a situation where densely situated metal nanoparticles are prepared through annealing and/or diffuse into the dielectric surrounding it [23,24].
The particles were confirmed to be the presynthesised Ni particles by EDX mapping. As can be seen from Figure 3, there is a local increase in nickel concentration in the areas with nanoparticles.
After the ALD of the second HfO2 layer, Ni NPs became effectively embedded within the structure, and therefore more difficult to locate. Even though larger particles were noticeable from a top-down view, as those seen within the highlighted areas in Figure 4a, the extent to which the Ti top electrode was raised varied significantly, with many particles being very difficult or impossible to see embedded within the structure when examining the electrodes from above. Note, for example, the minimal effect the Ni NP has on the surface of the finished Ti/HfO2-Ni-HfO2/TiN/Si structure in Figure 4b.
Figure 5 depicts XRD patterns of nanostructures with and without Ni nanoparticles measured on a TiN substrate. The measurements indicate an amorphous layer of hafnium oxide in both samples, correlating also with Zhang et al. [27]’s process study of HfO2 deposited from a similar TEMAH + O2 plasma process.
Figure 6a presents the results of STEM imaging carried out on a representative Ti/HfO2/TiN/Si nanostructure. One can see that the constituent HfO2 films with thicknesses of ~20 nm have formed with defined interfaces between the switching medium and mounting electrodes. The hafnium dioxide layer remains amorphous, as also observed by Zhang et al. [27], whilst the bottom TiN electrode (Figure 6b) and top Ti electrode show a polycrystalline structure.
A cross-sectional STEM-EDX analysis of the nanostructure with embedded Ni NPs was also performed. The EDX mapping of the region confirmed that we indeed have Ni nanoparticles and could also observe an increased concentration of hafnium and oxygen on the surface of the nanoparticle, suggesting a formation of HfO2 on top of the Ni NPs. This indicates that the Ni NPs were, in fact, embedded in the dielectric layer and were not merely a part of the top electrode.
All in all, we have shown that we managed to disperse the Ni NPs across the surface with uniformity (Figure 2a) and embed them between a continuous dielectric layer (Figure 4b), thus achieving the goal of creating a basis for the investigation of the effects of discrete NPs on the appearance of conductive filaments within the nanostructure. It is clear that the preparation of the nanoparticles separately as discrete particles can cause issues such as clustering/agglomeration and also tends to lead to larger particle size, compared to preparing the particles by, for example, annealing a deposited layer within the sample [13,28] or creating nanodots using lithography [6]. It does however allow us to produce our nanodevices at lower temperatures compared to annealing and relieves issues related to lithography.

3.2. Electrical Measurements

The electrically evaluated Ti/HfO2-Ni-HfO2/TiN/Si had a MIM-like structure, where commercial highly conductive titanium nitride (TiN) was used as the common bottom electrode and titanium as the top electrode (Figure 1a). Ti/HfO2-Ni-HfO2/TiN/Si nanostructure containing several hundred devices as the top electrodes were produced by maskless photolithography as circular dots covering the sample, each dot defining one device (Figure 1b). Several devices per sample were electrically characterised to evaluate the resistive switching performance. RRAMs rely on the initial formation and switching from a high resistive state (HRS) to a low resistive state (LRS) in a so-called SET procedure at a corresponding voltage named the set voltage (USET). Vice versa, switching the device from LRS to HRS is called a RESET procedure, and the corresponding voltage is denoted as the reset voltage (URESET). This allows information to be stored in two distinctive resistance states.
The studied nanostructures with and without Ni nanoparticles showed clockwise (CW) bipolar resistive switching (RS) properties (Figure 7 and Figure 8), meaning that the transition from HRS to LRS (SET) takes place at a negative voltage polarity and the RESET at a positive polarity. Some I-V measurements implied the presence of seldom observed counterclockwise (CCW) RS or unipolar RS behaviour. However, it was not possible to achieve a controlled unipolar RS. Supporting the polarity inversion of bipolar RS, G. Vinuesa et al. [29] demonstrated that >13 nm thick HfO2 thin films exhibit a CW bipolar RS, and less than 8 nm of the same dielectric layer tend to switch in the CCW direction. Based on the latter observation, it could be tempting to assume that the anomalous behaviour could be caused by bipolar RS polarity changes. However, it cannot be unambiguously ruled out that the unipolar RS and the CW RS or the CW and the CCW RS are competing, because the unipolar RS has also been demonstrated earlier by other researchers [30,31].
The nanostructures containing Ni nanoparticles tended to demonstrate RS more easily and at lower voltages, i.e., at around 1.5 V compared to the 2 V necessary for the samples without NPs (Figure 7). The device with Ni NPs demonstrated nearly 400 clockwise RS cycles, with set and reset voltages adjusted to −1.5 V and 1.5 V, respectively. Even though in comparison with many other devices, the stability of ours is not outstanding, it is not far off from other works already performed on nanodot based memories [32], and allows us to compare its performance to the samples without NPs. When carefully focusing on the endurance characteristic, one can notice how the memory window gradually diminishes and as can be seen in the figures, the switching voltages demonstrate a difference between LRS and HRS over slightly more than one order of magnitude with an average ILRS/IHRS ratio of around 9 at 0.2 V reading voltage.
It must be pointed out that 400 RS cycles is not yet significant from a device engineering point of view. Nevertheless, when compared to the performance of the sample without Ni NPs, the behaviour of the latter device remained inferior, even though the nanostructures were prepared during the same HfO2 deposition process. Figure 8 demonstrates that the nanostructures without nanoparticles have a memory window smaller than one order of magnitude, with an average ILRS/IHRS ratio of around 4 at a 0.2 V reading voltage. With a rapid deterioration of the memory window, only a scarce endurance performance could be measured (Figure 8).
Discussing the underlying RS mechanisms remains somewhat speculative, as studying the CF with direct methods like HRTEM was unsuccessful in our current study, which is not surprising, as it is known to be a challenging objective [33]. The Ni NPs are a source of an electrochemically active metal, supporting RS mechanisms like thermochemical metallisation, dominating in unipolar RS, and electrochemical metallisation mechanism (ECM), where ionic migration of active metal ions causes forming and disruption of metallic filaments by changing the direction of the electric field [31,33,34]. In addition to ECM, oxygen deficiency, i.e., the existence of oxygen vacancies can play a key role in the formation of CFs. In the current study, the top electrode of Ti can have the role of an oxygen scavenger aiding to provide oxygen vacancies in the main host media of hafnia. Y. Zhang et al. [35] have demonstrated that, in structures with symmetric inert electrodes, for example, Pt/HfO2/Pt and also in asymmetric Ti/HfO2/Pt structures, the valence change mechanism (VCM), in addition to Joule heating, is of paramount importance in the formation mechanism of CFs. VCM is based on the migration of oxygen vacancies, with the temperature-aided formation of CFs, consisting of conductive phases with a changed valence state in host media, relying on redox electrochemistry [16,35]. From another perspective, S. Wang et al. [12] offer a third possibility of hybrid CFs consisting of two intrinsic regions—one formed by the migration of oxygen vacancies (VCM) and another region formed by metallic ions (ECM). Explaining the rather low endurance performance for the devices with Ni NPs, J. Munoz-Goriz et al. [36] demonstrated the degradation of unipolar RS in Ni/HfO2/n+-Si devices owing to the diffusion of Ni into the host media, gradually increasing SET currents due to the strengthening CF and cascading the degradation.
The study demonstrated that, even though the samples without NPs already showed clockwise bipolar resistive switching, adding the Ni nanoparticles to the HfO2 RS medium lowered the switching voltages and enhanced the overall RS performance.

4. Conclusions

In this study, the effect of Ni NPs embedded in a 20 nm thick dielectric HfO2 film was investigated. The particles were found to be uniformly distributed across the surface and embedded in a discrete continuous layer of amorphous HfO2 thin film. A clear distinction between resistive switching properties of nanostructures with and without the addition of Ni NPs to the hafnium dioxide dielectric layer was made. The study demonstrated that all samples could show both clockwise and counterclockwise bipolar resistive switching. Adding Ni nanoparticles to the HfO2 RS medium lowered switching voltages from ~2 V to ~1.5 V, with an ILRS/IHRS ratio of 9, and enhanced the overall RS performance. The MIM-like stack structured as Ti/HfO2-Ni-HfO2/TiN/Si was able to carry out approximately 400 bipolar switching cycles before degradation. At the same time, even though the nanostructure without Ni NPs also exhibited a distinct memory window with an ILRS/IHRS ratio of 4, it remained markedly smaller and only unstable endurance measurements were realised. The present study provided valuable information and possible inspiration for further investigations, as further studies should evidently comprise investigations of the effect of the average size and density of metallic nanodots as well as the influence of the thickness and processing temperature of the host medium layer.

Author Contributions

Conceptualisation, A.T. (Aile Tamm) and M.O.; formal analysis, A.T. (Aivar Tarre) and J.M.; investigation, M.O., J.M., J.K., and A.-L.P.; resources, A.T. (Aile Tamm) and K.K.; writing—original draft preparation, M.O., J.M., J.K., A.T. (Aile Tamm), A.-L.P., K.K., and A.T. (Aivar Tarre); funding acquisition, K.K. and A.T. (Aile Tamm). All authors have read and agreed to the published version of the manuscript.

Funding

The work was partially supported by the European Regional Development Fund Projects “Emerging Orders in Quantum and Nanomaterials” (TK134) and Estonian Research Agency (PRG753 and PRG4).

Data Availability Statement

Data is contained within the article.

Acknowledgments

Tauno Kahro is thanked for his assistance in using photolithography and Aarne Kasikov is thanked for his assistance in using EBE. The authors are indebted to the European Regional Development Fund Project “Centre of Nanomaterials Technologies and Research” (NAMUR+, Project No. 2014–2020.4.01.16-0123).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Lanza, M.; Wong, H.-S.P.; Pop, E.; Ielmini, D.; Strukov, D.; Regan, B.C.; Larcher, L.; Villena, M.A.; Yang, J.J.; Goux, L.; et al. Recommended Methods to Study Resistive Switching Devices. Adv. Electron. Mater. 2019, 5, 1800143. [Google Scholar] [CrossRef]
  2. Pedretti, G.; Ielmini, D. In-Memory Computing with Resistive Memory Circuits: Status and Outlook. Electronics 2021, 10, 1063. [Google Scholar] [CrossRef]
  3. Lim, E.W.; Ismail, R. Conduction Mechanism of Valence Change Resistive Switching Memory: A Survey. Electronics 2015, 4, 586–613. [Google Scholar] [CrossRef]
  4. Chen, A. A review of emerging non-volatile memory (NVM) technologies and applications. Solid-State Electron. 2016, 125, 25–38. [Google Scholar] [CrossRef]
  5. Zahoor, F.; Zulkifli, T.Z.A.; Khanday, F.A. Resistive Random Access Memory (RRAM): An Overview of Materials, Switching Mechanism, Performance, Multilevel Cell (mlc) Storage, Modeling, and Applications. Nanoscale Res. Lett. 2020, 15, 1–26. [Google Scholar] [CrossRef]
  6. Li, Y.; Tang, J.; Gao, B.; Sun, W.; Hua, Q.; Zhang, W.; Li, X.; Zhang, W.; Qian, H.; Wu, H. High-Uniformity Threshold Switching HfO 2 -Based Selectors with Patterned Ag Nanodots. Adv. Sci. 2020, 7, 2002251. [Google Scholar] [CrossRef]
  7. Tomer, S.; Vandana; Panigrahi, J.; Srivastava, R.; Rauthan, C. Importance of precursor delivery mechanism for Tetra-kis-ethylmethylaminohafnium/water atomic layer deposition process. Thin Solid Film. 2019, 692, 137629. [Google Scholar] [CrossRef]
  8. Qin, Y.; Wang, Z.; Ling, Y.; Cai, Y.; Huang, R. A TaOx-Based RRAM with Improved Uniformity and Excellent Analog Characteristics by Local Dopant Engineering. Electronics 2021, 10, 2451. [Google Scholar] [CrossRef]
  9. Yan, L.; Ruan, L.; Luo, F.; Tong, J.; Sun, C.; Zheng, Y.; Han, X.; Zhang, Y.; Zhang, X. Enhanced resistive switching behavior of CH3NH3PbI3 based resistive random access memory by nickel doping. Vacuum 2022, 198, 110862. [Google Scholar] [CrossRef]
  10. Guo, T.; Huang, H.; Huang, X.; Wang, Y.; Duan, L.; Xu, Z. Modulation of resistive switching and magnetism of HfOx film by Co doping. J. Alloy. Compd. 2022, 921, 166248. [Google Scholar] [CrossRef]
  11. Han, G.; Chen, Y.; Liu, H.; Wang, D.; Qiao, R. Impacts of LaOx Doping on the Performance of ITO/Al2O3/ITO Transparent RRAM Devices. Electronics 2021, 10, 272. [Google Scholar] [CrossRef]
  12. Wang, S.; Ning, X.; Hao, A.; Chen, R. Metal nanoparticles layer boosted resistive switching property in NiFe2O4-based memory devices. J. Alloy. Compd. 2022, 908, 164569. [Google Scholar] [CrossRef]
  13. Yeh, P.; Chen, L.; Liu, P.; Wang, D.; Chang, T. Metal nanocrystals as charge storage nodes for nonvolatile memory devices. Electrochimica Acta 2007, 52, 2920–2926. [Google Scholar] [CrossRef]
  14. Wang, J.; Li, L.; Huyan, H.; Pan, X.; Nonnenmann, S.S. Highly Uniform Resistive Switching in HfO 2 Films Embedded with Ordered Metal Nanoisland Arrays. Adv. Funct. Mater. 2019, 29, 1808430. [Google Scholar] [CrossRef]
  15. Kahro, T.; Tarre, A.; Käämbre, T.; Piirsoo, H.-M.; Kozlova, J.; Ritslaid, P.; Kasikov, A.; Jõgiaas, T.; Vinuesa, G.; Dueñas, S.; et al. Hafnium Oxide/Graphene/Hafnium Oxide-Stacked Nanostructures as Resistive Switching Media. ACS Appl. Nano Mater. 2021, 4, 5152–5163. [Google Scholar] [CrossRef]
  16. Müller, C.; Deleruyelle, D.; Ginez, O.; Portal, J.-M.; Bocquet, M. Design challenges for prototypical and emerging memory concepts relying on resistance switching. In Proceedings of the 2011 IEEE Custom Integrated Circuits Conference (CICC), San Jose, CA, USA, 19–21 September 2011. [Google Scholar] [CrossRef]
  17. Kalam, K.; Otsus, M.; Kozlova, J.; Tarre, A.; Kasikov, A.; Rammula, R.; Link, J.; Stern, R.; Vinuesa, G.; Lendínez, J.M.; et al. Memory Effects in Nanolaminates of Hafnium and Iron Oxide Films Structured by Atomic Layer Deposition. Nanomaterials 2022, 12, 2593. [Google Scholar] [CrossRef] [PubMed]
  18. Chung, Y.-L.; Cheng, W.-H.; Jeng, J.-S.; Chen, W.-C.; Jhan, S.-A.; Chen, J.-S. Joint contributions of Ag ions and oxygen vacancies to conducting filament evolution of Ag/TaOx/Pt memory device. J. Appl. Phys. 2014, 116, 164502. [Google Scholar] [CrossRef]
  19. Kamiyama, S.; Miura, T.; Nara, Y.; Arikado, T. Atomic Layer Deposition of Hafnium Silicate Gate Dielectric Films Using Hf[N(CH3)(C2H5)]4 and SiH[N(CH3)2]3 Precursors. Electrochem. Solid-State Lett. 2005, 8, G215. [Google Scholar] [CrossRef]
  20. Johnson, R.W.; Hultqvist, A.; Bent, S.F. A brief review of atomic layer deposition: From fundamentals to applications. Mater. Today 2014, 17, 236–246. [Google Scholar] [CrossRef]
  21. Leskelä, M.; Ritala, M. Atomic Layer Deposition Chemistry: Recent Developments and Future Challenges. Angew. Chem. Int. Ed. 2003, 42, 5548–5554. [Google Scholar] [CrossRef]
  22. Gonzalez, M.B.; Acero, M.C.; Beldarrain, O.; Zabala, M.; Campabadal, F. Investigation of the resistive switching behavior in Ni/HfO2-based RRAM devices. In Proceedings of the 2015 10th Spanish Conference on Electron Devices (CDE), Aranjuez, Spain, 11–13 February 2015. [Google Scholar] [CrossRef]
  23. Chen, Y.Y.; Pourtois, G.; Wang, X.P.; Adelmann, C.; Goux, L.; Govoreanu, B.; Pantisano, L.; Kubicek, S.; Altimime, L.; Jurczak, M.; et al. Switching by Ni Filaments in a HfO2 Matrix: A New Pathway to Improved Unipolar Switching RRAM. In Proceedings of the 2011 3rd IEEE International Memory Workshop (IMW), Monterey, CA, USA, 22–25 May 2011. [Google Scholar] [CrossRef]
  24. Tan, Z.; Samanta, S.K.; Yoo, W.J.; Lee, S. Self-assembly of Ni nanocrystals on HfO2 and N-assisted Ni confinement for nonvolatile memory application. Appl. Phys. Lett. 2005, 86, 013107. [Google Scholar] [CrossRef]
  25. Yang, F.M.; Chang, T.C.; Liu, P.-T.; Chen, U.S.; Yeh, P.H.; Yu, Y.C.; Lin, J.Y.; Sze, S.M.; Lou, J.C. Nickel nanocrystals with HfO2 blocking oxide for nonvolatile memory application. Appl. Phys. Lett. 2007, 90, 222104. [Google Scholar] [CrossRef]
  26. Napolean, A.; Sivamangai, N.M.; Rajesh, S.; NaveenKumar, N.; Nithya, N.; Kamalnath, S.; Aswathy, N. Review on role of nanoscale HfO2 switching material in resistive random access memory device. Emergent Mater. 2022, 5, 489–508. [Google Scholar] [CrossRef]
  27. Zhang, X.-Y.; Hsu, C.-H.; Cho, Y.-S.; Zhang, S.; Lien, S.-Y.; Zhu, W.-Z.; Xiong, F.-B. Rapid thermal processing of hafnium dioxide thin films by remote plasma atomic layer deposition as high-k dielectrics. Thin Solid Film. 2018, 660, 797–801. [Google Scholar] [CrossRef]
  28. Liu, C.-Y.; Huang, J.-J.; Lai, C.-H. Resistive switching characteristics of a Pt nanoparticle-embedded SiO2-based memory. Thin Solid Film. 2012, 529, 107–110. [Google Scholar] [CrossRef]
  29. Vinuesa, G.; García, H.; González, M.B.; Kalam, K.; Zabala, M.; Tarre, A.; Kukli, K.; Tamm, A.; Campabadal, F.; Jiménez, J.; et al. Effect of Dielectric Thickness on Resistive Switching Polarity in TiN/Ti/HfO2/Pt Stacks. Electronics 2022, 11, 479. [Google Scholar] [CrossRef]
  30. González-Cordero, G.; González, M.; Zabala, M.; Kalam, K.; Tamm, A.; Jiménez-Molinos, F.; Campabadal, F.; Roldán, J. Study of RTN signals in resistive switching devices based on neural networks. Solid-State Electron. 2021, 183, 108034. [Google Scholar] [CrossRef]
  31. Muñoz-Gorriz, J.; Miranda, E.; Suñé, J.; Acero, M.C.; Gonzalez, M.B.; Campabadal, F. Microelectronic Devices and Technologies. In Proceedings of the 1st International Conference on Microelectronic Devices and Technologies (MicDAT’2018), Barcelona, Spain, 20–22 June 2018. [Google Scholar]
  32. Banerjee, W.; Liu, Q.; Hwang, H. Engineering of defects in resistive random access memory devices. J. Appl. Phys. 2020, 127, 051101. [Google Scholar] [CrossRef]
  33. Petzold, S.; Zintler, A.; Eilhardt, R.; Piros, E.; Kaiser, N.; Sharath, S.U.; Vogel, T.; Major, M.; McKenna, K.P.; Molina-Luna, L.; et al. Forming-Free Grain Boundary Engineered Hafnium Oxide Resistive Random Access Memory Devices. Adv. Electron. Mater. 2019, 5, 1900484. [Google Scholar] [CrossRef]
  34. Kärkkänen, I. Resistive Switching in ZrO2 Based Metal-Oxide-Metal Structures; Forschungszentrum Jülich: Jülich, Germany, 2014. [Google Scholar]
  35. Zhang, Y.; Mao, G.-Q.; Zhao, X.; Li, Y.; Zhang, M.; Wu, Z.; Wu, W.; Sun, H.; Guo, Y.; Wang, L.; et al. Evolution of the conductive filament system in HfO2-based memristors observed by direct atomic-scale imaging. Nat. Commun. 2021, 12, 1–10. [Google Scholar] [CrossRef]
  36. Muñoz-Gorriz, J.; González, M.; Miranda, E.; Suñé, J.; Campabadal, F. Impact of the forming and cycling processes on the electrical and physical degradation characteristics of HfO2-based resistive switching devices. Thin Solid Film. 2020, 706, 138027. [Google Scholar] [CrossRef]
Electronics 11 02963 g001 550
Figure 1. (a) Schematic representation of the nanostructures with and without Ni nanoparticles. (b) Optical microscope top view image of the nanodevices prepared using photolithography.
Figure 1. (a) Schematic representation of the nanostructures with and without Ni nanoparticles. (b) Optical microscope top view image of the nanodevices prepared using photolithography.
Electronics 11 02963 g001
Electronics 11 02963 g002 550
Figure 2. (a) BF STEM image of the as-synthesised Ni NPs measured as a powder. (b) SEM image of Ni NPs dispersed on the substrate before the deposition of the second layer of HfO2.
Figure 2. (a) BF STEM image of the as-synthesised Ni NPs measured as a powder. (b) SEM image of Ni NPs dispersed on the substrate before the deposition of the second layer of HfO2.
Electronics 11 02963 g002
Electronics 11 02963 g003 550
Figure 3. (a) SEM image of Ni nanoparticles on HfO2/TiN substrate and (b) EDX mapping of Ni of the same region indicating the presumed particles to be, in fact, nickel.
Figure 3. (a) SEM image of Ni nanoparticles on HfO2/TiN substrate and (b) EDX mapping of Ni of the same region indicating the presumed particles to be, in fact, nickel.
Electronics 11 02963 g003
Electronics 11 02963 g004 550
Figure 4. (a) Top-down SEM image of Ni NPs (highlighted) embedded in HfO2 without the Ti electrode and (b) cross-sectional SEM image of a Ni NP in the finished Ti/HfO2-Ni-HfO2/TiN/Si nanostructure (Pt on top was deposited to protect the structure from the FIB milling).
Figure 4. (a) Top-down SEM image of Ni NPs (highlighted) embedded in HfO2 without the Ti electrode and (b) cross-sectional SEM image of a Ni NP in the finished Ti/HfO2-Ni-HfO2/TiN/Si nanostructure (Pt on top was deposited to protect the structure from the FIB milling).
Electronics 11 02963 g004
Electronics 11 02963 g005 550
Figure 5. GiXRD patterns from nanostructures with and without Ni nanoparticles and a pure TiN/Si substrate. For the detailed inset color code, see Figure 1a.
Figure 5. GiXRD patterns from nanostructures with and without Ni nanoparticles and a pure TiN/Si substrate. For the detailed inset color code, see Figure 1a.
Electronics 11 02963 g005
Electronics 11 02963 g006 550
Figure 6. (a) Cross-section of the nanostructure as seen by BF STEM. All layers of the nanostructure are clearly discrete and visible. (b) HfO2 remains amorphous, while crystal structures are visible in the TiN layer.
Figure 6. (a) Cross-section of the nanostructure as seen by BF STEM. All layers of the nanostructure are clearly discrete and visible. (b) HfO2 remains amorphous, while crystal structures are visible in the TiN layer.
Electronics 11 02963 g006
Electronics 11 02963 g007 550
Figure 7. Electrical characterisation of the Ti/HfO2-Ni-HfO2/TiN/Si nanostructure with Ni nanoparticles: (a) clockwise I-V curves and (b) endurance measurements at 0.2 V reading voltage. For the detailed inset color code, see Figure 1a.
Figure 7. Electrical characterisation of the Ti/HfO2-Ni-HfO2/TiN/Si nanostructure with Ni nanoparticles: (a) clockwise I-V curves and (b) endurance measurements at 0.2 V reading voltage. For the detailed inset color code, see Figure 1a.
Electronics 11 02963 g007
Electronics 11 02963 g008 550
Figure 8. Electrical characterisation of the Ti/HfO2/TiN/Si nanostructure without Ni nanoparticles: (a) clockwise I-V curves and (b) endurance measurements at 0.2 V reading voltage. For the detailed inset color code, see Figure 1a.
Figure 8. Electrical characterisation of the Ti/HfO2/TiN/Si nanostructure without Ni nanoparticles: (a) clockwise I-V curves and (b) endurance measurements at 0.2 V reading voltage. For the detailed inset color code, see Figure 1a.
Electronics 11 02963 g008
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Otsus, M.; Merisalu, J.; Tarre, A.; Peikolainen, A.-L.; Kozlova, J.; Kukli, K.; Tamm, A. Bipolar Resistive Switching in Hafnium Oxide-Based Nanostructures with and without Nickel Nanoparticles. Electronics 2022, 11, 2963. https://doi.org/10.3390/electronics11182963

AMA Style

Otsus M, Merisalu J, Tarre A, Peikolainen A-L, Kozlova J, Kukli K, Tamm A. Bipolar Resistive Switching in Hafnium Oxide-Based Nanostructures with and without Nickel Nanoparticles. Electronics. 2022; 11(18):2963. https://doi.org/10.3390/electronics11182963

Chicago/Turabian Style

Otsus, Markus, Joonas Merisalu, Aivar Tarre, Anna-Liisa Peikolainen, Jekaterina Kozlova, Kaupo Kukli, and Aile Tamm. 2022. "Bipolar Resistive Switching in Hafnium Oxide-Based Nanostructures with and without Nickel Nanoparticles" Electronics 11, no. 18: 2963. https://doi.org/10.3390/electronics11182963

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop