Improved Resistance Switching Stability in Fe-Doped ZnO Thin Films Through Pulsed Magnetic Field Annealing
- 677 Downloads
Five percent of Fe-doped ZnO (ZnO:Fe) thin films were deposited on Pt/TiO2/SiO2/Si substrates by a spin-coating method. The films were annealed without (ZnO:Fe-0T) and with a pulsed magnetic field of 4 T (ZnO:Fe-4TP) to investigate the magnetic annealing effect on the resistance switching (RS) behavior of the Pt/ZnO:Fe/Pt structures. Compared with the ZnO:Fe-0T film, the ZnO:Fe-4TP film showed improved RS performance regarding the stability of the set voltage and the resistance of the high resistance state. Transmission electron microscopy and X-ray photoelectron spectroscopy analyses revealed that the ZnO:Fe-4TP film contains more uniform grains and a higher density of oxygen vacancies, which promote the easier formation of conducting filaments along similar paths and the stability of switching parameters. These results suggest that external magnetic fields can be used to prepare magnetic oxide thin films with improved resistance switching performance for memory device applications.
KeywordsRRAM ZnO Fe doping Magnetic annealing
Resistance random access memories
Transmission electron microscopy
X-ray photoelectron spectroscopy
As a potential next-generation nonvolatile memory, transition metal oxide (TMO)-based resistance random access memory (RRAM) has been studied intensively during the last decade and has attracted increasing interest because of its low power consumption, high operation speed, high endurance, and simple structure [1, 2, 3]. Zinc oxide (ZnO), which is a well-known oxide semiconductor, has also been widely studied because of its resistance switching (RS) behaviors [4, 5, 6, 7, 8]. ZnO-based RRAM devices have been reported to show an ultrafast programming speed of 5 ns, an ultrahigh ON/OFF ratio of 10 , a long retention time of more than 107 s, and high reliability at elevated temperatures [2, 5]. However, several problems need to be elucidated before achieving practical device applications. One of the issues is minimizing the dispersion of memory switching parameters, such as the resistance values of the low- and high-resistance states (LRS and HRS, or ON and OFF) and the switching voltages from the HRS to LRS (set voltages, Vset) and vice versa (reset voltages, Vreset) [6, 9]. The dominant cause of the oscillation of the switching parameters is the intrinsic random nature of the formation of defect-dominated conducting filaments (CFs) during the switching process . Many attempts, such as doping impurity elements [11, 12, 13] and interfacial engineering [14, 15], have been reported to be effective for controlling the location of the CFs and therefore stabilizing the switching parameters.
In this work, we report the effect of annealing in a magnetic field on the RS properties of Fe-doped ZnO thin films. A magnetic field can be used not only to study the physical properties but also to synthesize magnetic materials or modify their properties . The application of an external magnetic field during material synthesis can affect the structural and magnetic properties of the prepared materials [17, 18, 19, 20, 21, 22]. Annealing transition metal (TM)-doped ZnO nanoparticles with a high-pulsed magnetic field has been reported to improve the magnetic properties and increase oxygen defects [18, 19, 23, 24], which motivated us to study the magnetic annealing effect on the RS behaviors. In this work, we annealed Fe-doped ZnO (ZnO:Fe) thin films under a pulsed magnetic field of 4 T and determined that the magnetic field annealing process has a dramatic stabilizing effect on the switching parameters of Pt/ZnO:Fe/Pt structures.
Five percent of Fe-doped ZnO thin films were prepared on Pt(111)/TiO2/SiO2/Si substrates using a spin-coating method. Zinc acetate [Zn(CH3COO)2 · 2H2O] and iron nitrate [Fe(NO3)3 · 9H2O] were used as the precursors, and 2-methoxyethanol (HOCH2CH2OCH3) and monoethanolamine (H2NC2H4OH, MEA) were used as the solvent and stabilizer, respectively. The precursor chemicals were first dissolved completely in 2-methoxyethanol, then mixed together with the addition of MEA. The obtained mixture solution with a total metal ion concentration of 0.5 M was stirred at 60 °C for 0.5 h, then aged for 24 h before deposition on the substrate. The spin-coating process was performed at 3000 rpm for 30 s, followed by heating at 100 °C for 10 min to evaporate the solvent and pre-annealing at 400 °C for 10 min to exclude organic residuals. The spin-coating process was repeated several times to obtain a thickness of ~100 nm. The deposited films were annealed at 650 °C for 1 h in air, either with or without a 4-T pulsed magnetic field, labeled as ZnO:Fe-0T and ZnO:Fe-4TP, respectively.
The crystalline orientation and microstructure of the thin films were characterized by X-ray diffraction (XRD) with Cu Kα radiation and transmission electron microscopy (TEM). The chemical states were characterized by X-ray photoelectron spectroscopy (XPS), performed with a monochromatic Al Kα X-ray source (hv = 1486.6 eV) at an energy of 15 kV/150 W. The spot size was 400 μm (Theta Probe AR-XPS System, Thermo Fisher Scientific). Top Pt electrodes with dimensions of 90 × 90 μm  were deposited using e-beam evaporation to fabricate the RRAM devices, and the current-voltage (I-V) characteristics of the RRAM devices were measured using a semiconductor device parameter analyzer (Agilent B1500A).
Results and Discussion
The cumulative distributions of the HRS and LRS of the Pt/ZnO:Fe/Pt devices read at 0.1 V are shown in Fig. 2b. A significant enhancement of stability for the HRS was observed in the ZnO:Fe-4TP film. For the ZnO:Fe-0T film, the resistance varied from 2 to 200 kΩ for the HRS, and the LRS was approximately 40 Ω. In contrast, for the ZnO:Fe-4TP film, the resistance of the HRS remained near 50 kΩ, and the LRS was approximately 25 Ω. The inset shows the endurance properties of both films from 50 switching cycles; it is clear that the resistance of the HRS in the ZnO:Fe-4TP film is more stable than that of the ZnO:Fe-0T film. Because the HRS resistance directly affects the value of Vset during the following set process, the stabilized HRS resistance values were consistent with the stabilization in Vset shown in Fig. 2a . Additionally, the lower LRS resistance in Pt/ZnO:Fe-4TP/Pt implied that stronger CFs existed in this device.
The O 1s spectra from the two films showed quite different profiles, as shown in Fig. 4. The deconvolution results contain three peaks located around 533, 532, and 530 eV, which can be attributed to surface adsorbed O, oxygen deficiency, and lattice oxygen , respectively. Obviously, there is more oxygen deficiency in ZnO:Fe-4TP films. More interestingly, the Fe 2p spectra revealed that Fe3+ ions are dominant in both films (peak located at 711 eV). Additionally, metallic Fe (peak located at 707 eV) is also observed in the ZnO:Fe-4TP film. Although the valence state of Fe should be divalent Fe2+ if it is substituted into a defect-free ZnO crystal lattice, the appearance of Fe3+ implies the existence of Zn vacancies in our spin-coated ZnO thin films. It has been reported that in Fe-doped ZnO nanocrystals, Fe3+ appeared when Zn vacancies were present near the substitutional sites to neutralize the charge imbalance . A similar phenomenon has also been reported for the observation of Cr3+ ions in Cr-Mn-doped ZnO under magnetic annealing .
In summary, Fe-doped ZnO thin films were synthesized by the spin-coating method, and the films were annealed with and without a 4-T pulsed magnetic field. The Pt/ZnO:Fe/Pt structures were prepared to investigate the effect of magnetic annealing on the RS behaviors of ZnO:Fe thin films. Unipolar resistance switching was observed in all samples. Detailed analysis of the switching behaviors revealed that the ZnO:Fe-4TP thin film showed better performance regarding a quite stable set voltage and resistance in the HRS. SEM and TEM indicated the grain size became smaller and more uniform in theZnO:Fe-4TP film and the grain boundary is more clear and regulated. Based on the XPS characterization, the improved switching characteristics of the ZnO:Fe-4TP film were attributed to the increased amount of oxygen vacancies, which provided easier and more stable formation of conducting filaments. Our results suggest that by applying a 4-T pulsed magnetic field during the preparation of Fe-doped ZnO films, the resistance switching performance of the set voltage can be improved greatly.
We would like to thank the Busan Center of the Korea Basic Science Institute (KBSI) for the XPS measurements.
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (2014R1A1A3049826 and 2016K2A9A2A06004723).
HX, CW, and CL designed the whole experiment. HX carried out the sample preparation, XRD, and TEM measurements. CW and HX carried out the I-V measurement and analysis. RJ conducted and analyzed the XPS measurement. ZX participated in the discussion. CL and YL supervised the whole study. All authors read and approved the final manuscript.
The authors declare that they have no competing interests.
- 6.Tseng YT, Tasi TM, Chang TC, Shih CC, Chang KC, Zhang R, Chen KH, Chen JH, Li YC, Lin CY, Hung YC, Syn YE, Zheng JC, Sze SM (2015) Complementary resistive switching behavior induced by varying forming current compliance in resistance random access memory. Appl Phy Lett 106:213505CrossRefGoogle Scholar
- 20.AD S -A, Molodov DA, Garmestani H (2006) Magnetically induced texture development in zinc alloy sheet. Scripta Mater 46:857Google Scholar
- 24.Wang SW, Zhong M, Liu C, Li Y, Zhu MY, Jin HM, Hu YM (2013) Effects of temperature on the microstructure and magnetic property of Cr- doped ZnO DMS prepared by hydrothermal route assisted by pulsed magnetic fields. J Chem 7:2013Google Scholar
- 27.Li YB, Li Y, Zhu MY, Yang T, Huang J, Jin HM, Hu YM (2010) Structure and magnetic properties of Cr-doped ZnO nanoparticles prepared under high magnetic field. Solid State Commun 150:15Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.