Advertisement

Bipolar resistive switching with coexistence of mem-elements in the spray deposited CoFe2O4 thin film

  • T. D. Dongale
  • A. A. Bagade
  • S. V. Mohite
  • A. D. Rananavare
  • M. K. Orlowski
  • R. K. Kamat
  • K. Y. Rajpure
Article

Abstract

In the present investigation, we have experimentally demonstrated the bipolar resistive switching with the coexistence of three fundamental memelements in the Ag/CoFe2O4/FTO thin film metal-insulator-metal (MIM) device. The device shows the analog resistive switching behavior and charge transport follows the Ohmic and space charge limited conduction (SCLC) mechanisms. The device transforms from asymmetric to symmetric resistive switching when the SCLC conduction mechanism change to the Ohmic conduction mechanism at higher voltage sweep rates. It was observed that the I–V crossing location of MIM device shifted towards the higher voltage range with increasing voltage sweep rates for both bias regions due to the nanobattery effect. The significant tunneling gap between immature conductive filament(s) and percolation channels was responsible for the coexistence of memelements and nanobattery effect in the Ag/CoFe2O4/FTO thin film MIM device.

1 Introduction

In 1971, Prof. L. Chua predicted the memristor as a fourth fundamental circuit element with rigorous mathematical proof [1] and generalized the memristor devices to a broader class of memristive systems which is a specialized set of nonlinear dynamical devices in 1976 [2]. Memristor can be generalized into the class of memristive system whose functionality depends upon internal state variable(s) and can be defined as current controlled or voltage controlled memristive system, as given in Eqs. (1 and 2) [2, 3]

  1. a.

    n th order current controlled memristive system:

     
$$\left. \begin{gathered} {V_M}(t){\text{ }}=R(x,{\text{ }}{I_M},{\text{ }}t){I_M}(t) \hfill \\ \dot {x}=f(x,{\text{ }}{I_M},{\text{ }}t) \hfill \\ \end{gathered} \right\}~$$
(1)
  1. b.

    n th order voltage controlled memristive system:

     
$$\left. \begin{gathered} {I_M}(t){\text{ }}=G(x,{\text{ }}{V_M},{\text{ }}t){V_M}(t) \hfill \\ \dot {x}=f(x,{\text{ }}{V_M},{\text{ }}t) \hfill \\ \end{gathered} \right\}$$
(2)
Where x is a state variable, V M (t) and I M (t) denote the voltage across the device and current through the device, R is a scalar quantity termed as memristance and G is a memductance [3].

In the original paper, Prof. Chua defined ‘w’ as a state variable to define the physical properties of memristor in time-varying conditions [1] and HP team defined drifting of oxygen vacancies as a Chua’s state variable ‘w’ for the realization of first TiO2 memristor [4]. The HP memristor was a special case of current controlled memristor [4]. Both the equations 1 and 2 suggests that any physical device whose resistance and conductance depends upon internal state variable(s) termed as memristive device [3, 5]. Recently, Di Ventra et al. have extended the memristive system to memcapacitive and meminductive systems based on state-dependent relationships between charge-voltage and current-flux respectively [6, 7]. In the recent years, many researchers around the globe were striving hard to realize memcapacitor [8] and meminductor [9] memory devices. However, very few reports were available that show the concurrent memristance, meminductance, and memcapacitance (collectively called as ‘memelements’) memory effects in single device [10, 11].

The memelements possess the passivity, nonlinearity, and intrinsic memory properties with simple two terminal structure. These properties are very important for the emerging low-power electronic circuits, memory, logic, and neuromorphic computing applications [12, 13, 14, 15, 16, 17, 18]. In the recent year, metal oxide based memristors are popular among the researchers due to its small footprint, lower power consumption, better endurance and low cost [18, 19, 20]. Recently, spinel ferrites have gained interest for the resistive random access memory (RRAM) applications [21]. These are the class of magnetic oxide materials and their magnetic, as well as electrical properties, can be utilized to develop high-performance RRAM device. Recently, Mustaqima et al. [22] have developed spin-coated CoFe2O4 thin films RRAM structure. Their device shows unipolar resistive switching effect with improved stable set voltage. Xiahou et al. [23] have studied the effect of Cu doping on the resistive switching behaviors of sol–gel synthesized CoFe2O4 thin films. Their results suggested that the appropriate Cu doping can lead to improved resistive switching voltages, endurance, and data retention characteristics.

However, in many applications, bipolar resistive switching is required to utilize the power of memristive device for memory, logic, and computing applications. Furthermore, very few reports show the coexistence of three memelements in a single device. In view of this, the present report deals with the structural, morphological, and electrical studies of spray deposited CoFe2O4 bipolar resistive switching device. The rest of paper is as organized as follows: after a brief introduction in the first section, the second section deals with the development of Silver (Ag)/CoFe2O4/Fluorine doped tin oxide (FTO) thin film metal-insulator-metal (MIM) device using the cost effective spray pyrolysis technique. The structural, morphological, and magnetic characterizations are reported in the third section. The bipolar resistive switching, analog memory, conduction mechanism, and coexistence of memelements have been put forth in the fourth section. Finally, in the last section, we draw the main conclusions.

2 Experimental

The FTO substrate was used to deposit the CoFe2O4 thin film by employing the spray pyrolysis technique. The sheet resistance of FTO substrate was ~ 15 Ω/cm2. The spraying solution was prepared by using a ferric nitrate and cobalt nitrate in double distilled water having the ratio 2:1. The deposition temperature 400 °C, spray rate 8 ml/min, the quantity of spraying solution 100 ml and the substrate to nozzle distance 32.5 cm were kept constant. Phase identification and structural properties were studied by Bruker D2-Phaser X-ray powder diffractometer with CuKα radiation (λ = 1.5406 Å). For the X-ray diffraction (XRD) study, CoFe2O4 deposited on a separate quartz substrate. The diffraction pattern was analyzed with the help of the FullProf program by employing Rietveld refinement technique using the Fd3m space group. The quality of fitting experimental data was assessed by calculating the parameters such as the ‘goodness of fit’ χ2 and R factors. The scanning electron microscopy (SEM) (JEOL-JSM 6360 A) was used to investigate the surface morphology of CoFe2O4 thin film. Vibrating sample magnetometer (VSM) measurement was performed by the magnetic field applied parallel to film surface.

Figure 1a represents the structure of Ag/CoFe2O4/FTO thin film MIM device. It consist of top Ag layer, middle CoFe2O4 active layer and bottom FTO layer. The Ag paste was used to create the top contact and spray pyrolysis method was employed to develop the FTO and CoFe2O4 active layers. The electrical characterizations of Ag/CoFe2O4/FTO MIM device were recorded using programmable electrochemical workstation (Autolab N-Series). For all measurements, the top Ag electrode was biased while bottom FTO electrode was grounded. The typical pinched current–voltage (I–V) characteristic of MIM device was recorded by applying the bias voltage ± 4 V across the device with sweep rate increased from 100 to 500 mV/s. The device was first set to high resistance state (HRS) then a 0 to + 4 V and back to 0–4 V triangular waveform pulse was applied to the device. The typical nature of triangular waveform is shown in Fig. 1b. The compliance current was set to 1 mA for the soft breakdown.

Fig. 1

a Basic structure of Ag/CoFe2O4/FTO MIM device. b Triangular waveform input stimulus for the developed device

3 Characterizations of CoFe2O4 thin film memristive device

Figure 2 shows the XRD spectra along with Rietveld refined data for the CoFe2O4 thin film. In XRD pattern the Bragg peak positions are shown in a vertical line. The open circles indicate observed intensity; solid line represents Rietveld refined calculated intensity and the bottom line shows the difference between the observed and refined calculated intensities. The major reflection planes (111), (220), (311), (222), (400), (422), (511) and (440) in the refined XRD pattern gives the indication of the formation of spinel cubic crystal structure. It is observed that the sample is single-phase without any undesirable phase and the diffraction peaks can be indexed with a spinel crystal structure [24]. The oxygen positions have been taken as free parameters and atomic fractional positions have been taken as fixed. The lattice constant, temperature parameters, occupancies, scale factor, and shape parameters have been taken as free parameters. The superiority of the refinement was quantified by the goodness of fit (χ2). The Pseudo-Voigt function was employed for background correction. The atomic position coordinates and occupancies of different atoms of CoFe2O4 thin films are shown in Table 1. From this table, it is seen that the Fe and Co atoms are distributed in both A-site and B-site. The Rietveld refined factors such as χ2, Rwp, Rexp, RB, RF, D, a, and ρ are summarized in Table 2. The average crystallite sizes (D) and lattice constant (a) was presented in Table 2. The value of crystallite size and lattice constant of CoFe2O4 thin film was found to be 46 nm and 8.3766 Å, respectively. These values are slightly greater than previously reported values [25].

Fig. 2

Rietveld refined X-ray diffraction pattern of CoFe2O4 thin film

Table 1

Position coordinates and occupancies of different atoms of CoFe2O4 thin film

Spinel lattice sites

Atom

Position coordinates

Atom occupancy

x

y

z

 

Tetrahedral (A-site)

Fe

0.125

0.125

0.125

0.850

Co

0.125

0.125

0.125

0.140

Octahedral (B-site)

Fe

0.500

0.500

0.500

1.150

Co

0.500

0.500

0.500

0.860

Table 2

Rietveld refinement factors, lattice constant, unit cell volume, density and oxygen position parameters of CoFe2O4 thin film

Reitvield refinement factors

CoFe2O4 thin film

χ2

1.02

RB (%)

3.85

RF (%)

5.19

Rwp

24.2

Rexp

23.5

D (nm)

46

a (Å)

8.3866

V (a3)

589.87

ρ (g/cm3)

5.846

Oxygen position (x = y = z)

0.2455

The SEM micrograph of the CoFe2O4 thin film is shown in Fig. 3. The micrographs show the compact nature of the thin film. The crack-free and densely packed structures are observed which suggests the uniformity of the thin film. The cross sectional view of the Ag/CoFe2O4/FTO MIM device is shown in the inset of Fig. 3. The cross sectional analysis suggested that the thickness of the CoFe2O4 active layer is nearly equal to 5.60 µm, whereas the thickness of top Ag and bottom FTO layer is found to be ~ 3.08 and ~ 1.05 µm respectively. The active area between top Ag contact and CoFe2O4 active layer is nearly equal to 25 mm2. The magnetic property of the CoFe2O4 thin film was characterized by VSM at room temperature and the corresponding hysteresis loop is shown in Fig. 4. A coercivity value of 44 Oe is shown in the left-upper inset of Fig. 4. The curve shows that the film is well magnetically saturated in external magnetic field. The values of saturation magnetization (Ms), remanence magnetization (Mr), coercive field (Hc) and squareness ratio (Mr/Ms) are listed in Table 3. The magnetic study shows that the material possesses soft magnetic phase with high saturation magnetization and very low coercivity. The soft ferrites can be readily saturated in a low magnetic field; therefore they show low coercive [26]. The low coercivity due to the decrease in magnetic anisotropy since an applied field at a given temperature should be able to overcome the energy barrier and change the orientation of magnetization [27].

Fig. 3

SEM image of CoFe2O4 thin film (inset shows the cross sectional view of Ag/CoFe2O4/FTO MIM device)

Fig. 4

Hysteresis loop of CoFe2O4 thin film (inset represents the enlarge view)

Table 3

Magnetic parameters of CoFe2O4 thin film

Saturation magnetization (Ms), emu/cc

Remanance magnetization (Mr), emu/cc

Coercivity (Hc), Oe

Squareness ratio (Mr/Ms)

163

6.14

48

0.037

4 Results and discussions

In general, the memristive device has two different resistance state viz. High resistance state (HRS) and low resistance state (LRS). If the resistance state of the device changed from HRS to LRS then the device undergoes in an ON state and corresponding switching voltage is termed as VSET. On the contrary, a transition occurs from LRS to HRS, result in an OFF state and corresponding switching voltage is termed as VRESET. Based on these resistive switching properties, memristive devices can be classified as bipolar or unipolar resistive switching devices [5]. The bipolar resistive switching devices require different VSET and VRESET with different polarity, however, the same polarity is required for the unipolar resistive switching devices. In the present investigations, the developed device shows bipolar resistive switching behavior. Furthermore, the asymmetric resistive switching transforms to symmetric resistive switching with an increase in the sweep rates.

Figure 5 represents a typical bipolar resistive switching of Ag/CoFe2O4/FTO MIM device with sweep cycles from 100 to 500 mV/s sweep rate. It was observed that the programmed and erased device required only ± 4 V resistive switching voltages (VSET and VRESET) predetermined by the maximum range of the applied voltage waveform. Furthermore, the developed device shows asymmetric analog memory property in which transition from HRS to LRS and vice versa occurs smoothly. Such kind of inherent analog memory is required for the development of high-performance electronic synapse for neuromorphic computing applications [28, 29]. The synaptic weights of biological synapses are increasing or decreasing depending upon the rate of discharge of chemical signals in the synaptic cleft [17, 18, 28] and such kind of synaptic weights behavior were observed in developed devices as shown in Figs. 5 and 6, hence the current i(t) can be considered as a synaptic weight for the neuromorphic application. The maximum magnitude of current i(t) was observed at 100 mV/s sweep rate and it decreases as the sweep rate increases. This result may be explained by considering the flux φ [1, 2] applied to the device: \(\phi =\int_{0}^{{4/rr}} {rrt} dt=8/rr\) in units of [Vs] where rr is the ramp rate in [V/s] and 4 V is the maximum voltage during the sweep. In order to induce a certain conductive phase in the device a finite flux is needed. As seen from the above equation this required flux φ decreases with increasing ramp rate. Therefore, a device programmed at a lower ramp rate is more conductive than the same device programmed at a higher ramp rate with the same maximum voltage. This is consistent with space charge limited conduction as explained below.

Fig. 5

Bipolar resistive switching property of Ag/CoFe2O4/FTO MIM device with 100–500 mV/s sweep rates. The developed device shows analog memory property in which transition from low resistance state to high resistance state and vice versa occurs smoothly. The arrow indicates the direction of resistive switching

Fig. 6

Maximum current at VSET and VRESET of Ag/CoFe2O4/FTO MIM device at positive and negative biases with different sweep rates

In order to understand the conduction mechanism of the developed device, we have plotted the I–V characteristics on the double-logarithmic scale as shown in Fig. 7a and b. It was observed that the values of the slopes are not greater than 0.85 at lower voltages for both bias regions. This confirms that the switching mechanism was approaching Ohmic behavior but is unable to form pure conductive filament(s). In other words, the incomplete conductive filament(s) may form during the device operation. Furthermore, the current was suddenly increased at higher voltage region which leads to the higher slope. This kind of data was well fitted with the space charge limited conduction (SCLC) mechanism [30]. The results were in good agreement with the existing results reported in the literature [22, 31, 32]. Furthermore, the asymmetric bipolar resistive switching was changed to symmetric resistive switching as the sweep rate increased from 100 to 500 mV/s. The change in the nature of pinched hysteresis loop may be due to the different switching mechanisms occur during different sweep rates. In order to examine the switching mechanism(s), we have calculated the slopes at different sweep rates. The details of fitting were shown in Fig. 8a and b. The results suggested that the slopes decrease as the sweep rate increases which results in a decrease of the absolute magnitude of the current [i(t)]. In a nutshell, the SCLC mechanism tends towards Ohmic conduction mechanism as the sweep rate increases. From the results, it was confirmed that the Ohmic conduction mechanism dominant at higher sweep rates which leads to symmetric resistive switching in I–V plane.

Fig. 7

ab Double-logarithmic plots of Ag/CoFe2O4/FTO MIM device at positive and negative biases with sweep rate equals to 100 mV/s

Fig. 8

Slopes of Ag/CoFe2O4/FTO MIM device at a positive and b negative bias with different sweep rates and voltage range

It would be interesting to know that the I–V crossing points were shifted to the first and third quadrant as shown in Fig. 7a and b. The recent results suggested that the shifting of I–V crossing point to the first and third quadrant leads to coexistence of memristance, memcapacitance, and meminductance memory effects [10, 11]. The coexistence of memristance, memcapacitance, and meminductance memory effects was illustrated in the inset of Fig. 7a and b. The results of I–V crossing locations are summarized in Table 4. It was observed that the I–V crossing location of Ag/CoFe2O4/FTO MIM device were shifted towards the higher voltage range as the sweep rate increases from 100 to 500 mV/s for both bias regions. These non-ideal I–V crossings suggest that the device shows non-zero-crossing property which has been identified as a nanobattery effect [33]. This non-ideal behavior is due to the inhomogeneous charge carrier distribution within the active layer of the MIN device [34]. For the present case, lower value of slopes during LRS and inhomogeneous distribution of oxygen vacancies leads to the formation of the incomplete conductive filament(s).

Table 4

I–V crossing locations of Ag/CoFe2O4/FTO MIM device

Sweep rate (mV/s)

Crossing point voltage (V) at positive bias region

Crossing point

Voltage (IVI) at

Negative bias region

100

0.13

1.78

200

0.37

1.47

300

0.51

1.56

400

0.71

1.64

500

0.73

1.79

The coexistence of triple state mem-effects can be explained using formation and breaking of the conductive filament(s) that occurs within the active layer of MIM device due to the valency change mechanism (VCM). In VCM, oxygen vacancies called as anions drifted in the active layer and formed conductive filament(s). In general, as the positive voltage applied to the top electrode (Ag), positively charged oxygen vacancies were attracted towards negatively charged bottom electrode (FTO). This results in a formation of the conductive filament as shown in the Fig. 9a. In this case, the MIM device will be switched from HRS to LRS as the conductivity of the filament increased due to charge change of the cations (Fe2+/3+, Co2+/3+) [32, 35]. On the other hand, when a negative voltage was applied to the top electrode, the oxygen vacancies were detached from the conductive filament. The detached oxygen vacancies were concentrated near to top electrode and form many percolation channels below the top Ag electrode. Such kind of mechanism resulted in the HRS i.e. device switchbacks to OFF state, as shown in Fig. 9b. The analog nature of I–V switching suggests the possibility of the interfacial type of resistive switching (RS) may also coexist with the filamentary RS at lower sweep rate. The interfacial type of RS could observe at lower carrier concentration due to lower sweep rate, which results in non-symmetric I–V characteristics, as shown in Fig. 5. However, the Schottky-like barrier reduces at the higher sweep rate due to increase in the carrier concentrations [36]. On the other hand, the switching voltages VSET and VRESET = 4V, determined by the interval of the voltage waveform were also higher than sweep rate for the present device hence there is a possibility that filamentary RS mechanism was dominating on interfacial type in RS mechanism. Recently, Münstermann et al. [37] and Biju et al. [38] have experimentally demonstrated that the counter-clockwise switching represents the homogeneous or interfacial mechanism, whereas clockwise switching represents the filamentary RS mechanism. In the present case, the direction of resistive switching was found to be clockwise hence switching mechanism was filamentary RS in nature.

Fig. 9

Schematic illustration of the filamentary resistive switching mechanism of the Ag/CoFe2O4/FTO MIM device: a ON or LRS state and b OFF or HRS state

The parasitic memcapacitance and meminductance may arise from the incomplete formation of the conductive filament(s). In LRS, the incomplete conductive filament(s) may introduce the memcapacitance effect due to the significant tunneling gap between conductive filament(s) and the bottom electrode. On the other hand, stochastic nature of the conductive filament(s) may produce many percolation channels over the larger area in the HRS. These percolation channels facilitate to form the winding conductive filaments which result in the significant meminductance effect in the MIM device [10]. According to Di Ventra et al. theoretical suggestions, the memristive dynamics are always accompanied by memcapacitive and meminductive behavior at the nanoscale and it cannot be avoided [6, 7, 8, 9, 10, 11]. In the nutshell, the results reported herein will be important for the development of self-resonating nanoelectronic devices for the development of nanoscale adaptive circuits and neuromorphic architecture.

5 Conclusion

The present manuscript deals with the development Ag/CoFe2O4/FTO thin film MIM device using the cost-effective spray pyrolysis technique. The bipolar with analog resistive switching was observed in the MIM device and such kind of behavior is important for the development of the electronic synapse for the neuromorphic application. The conduction mechanism suggests that the Ohmic charge transportation mechanism was dominant at lower voltage range for both bias regions whereas, SCLC conduction mechanism was observed at the higher voltage range. Moreover, the SCLC conduction mechanism approaches Ohmic conduction mechanism during the higher sweep rates which results in the device transform from asymmetric RS to symmetric RS behavior. The coexistence of memristance, memcapacitance, and meminductance memory effects was observed in the MIM device. Furthermore, I–V crossing location shifted towards the higher voltage range with an increase in the sweep rates, suggested that the nanobattery effect present in the device. Such kind of parasitic effects may occur due to tunneling gap between conductive filament(s) and the bottom electrode, incomplete formation of the conductive filament(s) and percolation channels. The results reported herein will be important for the development of nanoscale adaptive circuits and synaptic devices.

Notes

Acknowledgements

The authors extend their appreciation to the Staff and Students of Physics Instrumentation Facility Centre (PIFC), Shivaji University, Kolhapur for valuable discussion and characterizations.

References

  1. 1.
    L.O. Chua, IEEE Trans. Circuit Theory 18, 507 (1971)CrossRefGoogle Scholar
  2. 2.
    L.O. Chua, S.M. Kang, Proc. IEEE 64, 209 (1976)CrossRefGoogle Scholar
  3. 3.
    Y.V. Pershin, M. Di Ventra, Adv. Phys. 60, 145 (2011)CrossRefGoogle Scholar
  4. 4.
    D.B. Strukov, G.S. Snider, D.R. Stewart, R.S. Williams, Nature 453, 80 (2008)CrossRefGoogle Scholar
  5. 5.
    L.O. Chua, Appl. Phys. A 102, 765 (2011)CrossRefGoogle Scholar
  6. 6.
    M. Di Ventra, Y.V. Pershin, L.O. Chua, Proc. IEEE 97, 1717 (2009)CrossRefGoogle Scholar
  7. 7.
    M. Di Ventra, Y.V. Pershin, Nanotechnology 24, 255201 (2013)CrossRefGoogle Scholar
  8. 8.
    R.E. Meade, G.S. Sandhu, U.S. Patent No. 8,867,261. (U.S. Patent and Trademark Office, Washington, DC, 2014)Google Scholar
  9. 9.
    J. Han, C. Song, S. Gao, Y. Wang, C. Chen, F. Pan, ACS Nano 8, 10043 (2014)CrossRefGoogle Scholar
  10. 10.
    L. Qingjiang, A. Khiat, I. Salaoru, C. Papavassiliou, X. Hui, T. Prodromakis, Sci. Rep. 4, 4522 (2014)CrossRefGoogle Scholar
  11. 11.
    S. Sarma, B.M. Mothudi, M.S. Dhlamini, J. Mater. Sci. Mater. Electron. 27, 4551 (2016)CrossRefGoogle Scholar
  12. 12.
    T.D. Dongale, K.V. Khot, S.S. Mali, P.S. Patil, P.K. Gaikwad, R.K. Kamat, P.N. Bhosale, Mater. Sci. Semicond. Process 40, 523 (2015)CrossRefGoogle Scholar
  13. 13.
    T.D. Dongale, S.V. Mohite, A.A. Bagade, P.K. Gaikwad, P.S. Patil, R.K. Kamat, K.Y. Rajpure, Electron. Mater. Lett. 11, 944 (2015)CrossRefGoogle Scholar
  14. 14.
    T.D. Dongale, S.S. Shinde, R.K. Kamat, K.Y. Rajpure, J. Alloys Compd. 593, 267 (2014)CrossRefGoogle Scholar
  15. 15.
    T.D. Dongale, K.V. Khot, S.V. Mohite, S.S. Khandagale, S.S. Shinde, V.L. Patil, S.A. Vanalkar, A.V. Moholkar, K.Y. Rajpure, P.N. Bhosale, P.S. Patil, P.K. Gaikwad, R.K. Kamat, J. Nano-Electron. Phys. 8, 04030 (2016)CrossRefGoogle Scholar
  16. 16.
    T.D. Dongale, P.J. Patil, N.K. Desai, P.P. Chougule, S.M. Kumbhar, P.P. Waifalkar, P.B. Patil, R.S. Vhatkar, M.V. Takale, P.K. Gaikwad, R.K. Kamat, Nano Converg. 3, 1 (2016)CrossRefGoogle Scholar
  17. 17.
    T.D. Dongale, N.D. Desai, K.V. Khot, N.B. Mullani, P.S. Pawar, R.S. Tikke, V.B. Patil, P.P. Waifalkar, P.B. Patil, R.K. Kamat, P.S. Patil, P.N. Bhosale, J. Solid State Electr. 21, 2753 (2017)CrossRefGoogle Scholar
  18. 18.
    P.S. Pawar, R.S. Tikke, V.B. Patil, N.B. Mullani, P.P. Waifalkar, K.V. Khot, A.M. Teli, A.D. Sheikh, T.D. Dongale, Mater. Sci. Semicond. Process 71, 102 (2017)CrossRefGoogle Scholar
  19. 19.
    Q. Xia, M.D. Pickett, J.J. Yang, X. Li, W. Wu, G. Medeiros-Ribeiro, R.S. Williams, Adv. Funct. Mater. 21, 2660 (2011)CrossRefGoogle Scholar
  20. 20.
    J.P. Strachan, A. C.Torrezan, G. Medeiros-Ribeiro, R.S. Williams, Nanotechnology 22, 505402 (2011)CrossRefGoogle Scholar
  21. 21.
    W. Hu, N. Qin, G. Wu, Y. Lin, S. Li, D. Bao, J. Am. Chem. Soc. 134, 14658 (2012)CrossRefGoogle Scholar
  22. 22.
    M. Mustaqima, P. Yoo, W. Huang, B. Lee, C. Liu, Nanoscale Res. Lett. 10, 1 (2015)CrossRefGoogle Scholar
  23. 23.
    Z. Xiahou, D.H. Kim, H. Xu, Y. Li, B. Lee, C. Liu, J. Mater. Sci. Mater. Electron. 27, 2255 (2016)CrossRefGoogle Scholar
  24. 24.
    V.S. Sawant, A.A. Bagade, K.Y. Rajpure, Phys. B 474, 47 (2015)CrossRefGoogle Scholar
  25. 25.
    A.A. Bagade, K.Y. Rajpure, J. Alloys Compd. 657, 414 (2016)CrossRefGoogle Scholar
  26. 26.
    F. Liu, Y. Hou, S. Gao, Chem. Soc. Rev. 43, 8098 (2014)CrossRefGoogle Scholar
  27. 27.
    Q. Song, Z.J. Zhang, J. Am. Chem. Soc. 126, 6164 (2004)CrossRefGoogle Scholar
  28. 28.
    S.H. Jo, T. Chang, I. Ebong, B.B. Bhadviya, P. Mazumder, W. Lu, Nano Lett. 10, 1297 (2010)CrossRefGoogle Scholar
  29. 29.
    Y. Kaneko, Y. Nishitani, M. Ueda, IEEE Trans. Electron Dev. 61, 2827 (2014)CrossRefGoogle Scholar
  30. 30.
    T.D. Dongale, K.P. Patil, P.K. Gaikwad, R.K. Kamat, Mater. Sci. Semicond. Process 38, 228 (2015)CrossRefGoogle Scholar
  31. 31.
    W. Hu, X. Chen, G. Wu, Y. Lin, N. Qin, D. Bao, Appl. Phys. Lett. 101, 063501 (2012)CrossRefGoogle Scholar
  32. 32.
    C. Jiang, L. Wu, W. Wei, C. Dong, J. Yao, Nanoscale Res. Lett. 9, 584 (2014)CrossRefGoogle Scholar
  33. 33.
    I. Valov, E. Linn, S. Tappertzhofen, S. Schmelzer, J. Van den Hurk, F. Lentz, R. Waser, Nat. Commun. 4, 1771 (2013)CrossRefGoogle Scholar
  34. 34.
    S. Tappertzhofen, E. Linn, U. Bottger, R. Waser, I. Valov, IEEE Electron Device Lett. 35, 208 (2014)CrossRefGoogle Scholar
  35. 35.
    C. Jin, D. Zheng, P. Li, W. Mi, H. Bai, Appl. Surf. Sci. 263, 678 (2012)CrossRefGoogle Scholar
  36. 36.
    K. Shibuya, R. Dittmann, S. Mi, R. Waser, Adv. Mater. 22, 411 (2010)CrossRefGoogle Scholar
  37. 37.
    R. Muenstermann, T. Menke, R. Dittmann, R. Waser, Adv. Mater. 22, 4819 (2010)CrossRefGoogle Scholar
  38. 38.
    K.P. Biju, X. Liu, S. Kim, S. Jung, J. Park, H. Hwang, Phys. Status Solidi Rapid Res. Lett. 5, 89 (2011)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2017

Authors and Affiliations

  • T. D. Dongale
    • 1
  • A. A. Bagade
    • 2
  • S. V. Mohite
    • 2
  • A. D. Rananavare
    • 1
  • M. K. Orlowski
    • 3
  • R. K. Kamat
    • 4
  • K. Y. Rajpure
    • 2
  1. 1.Computational Electronics and Nanoscience Research Laboratory, School of Nanoscience and BiotechnologyShivaji UniversityKolhapurIndia
  2. 2.Electrochemical Materials Laboratory, Department of PhysicsShivaji UniversityKolhapurIndia
  3. 3.Bradley Department of Electrical and Computer EngineeringVirginia Tech.BlacksburgUSA
  4. 4.Department of ElectronicsShivaji UniversityKolhapurIndia

Personalised recommendations