Electronic Materials Letters, Vol. 9, No. 2 (2013), pp. 207-211 DOI: 10.1007/s13391-012-2106-y
Effect of DC Bias on Dielectric Properties of Nanocrystalline CuAlO2 T. Prakash,1,* S. Ramasamy,2 and B. S. Murty3 1
Department of Medical Bionanotechnology, Chettinad Hospital and Research Institute Kelambakkam, Tamil Nadu 603103, India 2 Crystal Growth Center, Anna University, Chennai 600025, India 3 Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600036, India (received date: 21 June 2012 / accepted date: 23 July 2012 / published date: March 2013) Grain boundary effect on the room temperature dielectric behavior in mechanically alloyed nanocrystalline CuAlO2 has been investigated using impedance spectroscopy under the applied DC bias voltages 0 V to 4.8 V in a periodic interval of 0.2 V. Analysis of impedance data confirms the existence of double Schottky potential barrier heights (Φb) between two adjacent grains (left and right side) with grain boundary and its influences in dielectric relaxation time (τ), dielectric constant (ε') and dielectric loss (tan δ) factor. Also, clear evidence on the suppression of Φb was demonstrated in the higher applied bias voltages with the parameter τ. At equilibrium state, τ is 0.63 ms and it was reduced to 0.13 ms after the 3.2 V applied DC bias. These observed DC bias voltage effects are obeying ‘brick layer model’ and also elucidates Φb is playing a crucial role in controlling dielectric properties of nanomaterials. Keywords: double schottky barrier, dielectric properties, transparent semiconductor
1. INTRODUCTION The physics of grain boundaries in semiconductors is of interest both from a fundamental as well as technical point of view because they play a crucial role in controlling properties. The grain boundary effect on the nanocrystalline materials is found more because the volume fraction of atoms lying at the grain boundaries of the nanocrystalline materials is more as compared with conventional coarse-grained polycrystalline materials. Delafossite CuAlO2 is one of the most promising material for transparent electronic devices because of its optical transparency, p-type semiconducting nature and electrical conductivity.[1] Gao et al.[2] have observed the enhancement of electrical conductivity of this material from 1 S/cm to 2.4 S/cm when the crystallite size is reduced to nano-scale. Dielectric permittivity is one of the most important physical properties of semiconductors, since it provides information on the behavior of localized electric charge carriers. Generally, the method of preparation, processing temperature, compositions, and grain dimension are the key parameters that affect the dielectric permittivity of any material. In our recent papers, impedance spectroscopic investigation on temperature dependent grain and grain boundary contributed electrical conductivities of nanocrystalline CuAlO2 (45 nm)[3] and the existence of Debye type dielectric *Corresponding author:
[email protected] ©KIM and Springer
relaxation behavior in this material[4] were discussed. In the present work, the existence of grain boundary double Schottky potential barrier height (Φb) and its influence on the dielectric permittivity of nanocrystalline CuAlO2 were studied under applied DC bias voltages.
2. EXPERIMENTAL PROCEDURE The nanocrystalline CuAlO2 was synthesized by mechanically alloying the oxide precursor powders Cu2O and αAl2O3 in the molar ratio of 1 : 1 for 20 h in toluene medium with tungsten carbide balls and vials in Fritsch pulveristte-5 planetary ball mill. The milling was carried out at 300 rpm with a ‘balls to sample weight ratio’ of 10 : 1. The as-milled sample was annealed at 1100°C using a platinum crucible in air for 20 h to form delafossite phase. The selected area electron diffraction and microstructure of the sample was recorded by using Philips CM12 transmission electron microscope (TEM). In order to perform TEM measurement, the sample was prepared by placing a drop of nanocrystals dispersion in ethanol on a carbon coated copper grid and kept at room temperature to evaporate the ethanol. To study the dielectric properties of this material using impedance spectroscopic measurements, powder was pressed into cylindrical pellets of 8 mm diameter and 1 mm thickness with the pressure of 4 ton using a hydraulic press. Poly vinyl alcohol (PVA) was used as a binder to reduce the brittleness of the pellet. It was burnt out by sintering the pellet under at 150°C
208
T. Prakash et al.
for 15 min. The impedance measurement was performed at room temperature in the frequency range of 1 Hz to 1 MHz with the applied AC potential of 50 mV using Solartron 1260 impedance/gain phase analyser under DC bias voltages from 0 V to 4.8 V with a step of 0.2 V. Prior to the measurements, the sample is spring loaded between two circular silver electrodes having similar dimensions and shielded test leads were used for electrical connections from the analyser to the electrode to avoid any parasitic impedance due to connecting the cables. The impedance data were acquired using a dedicated computer with automated data collection Z-plot software and analysed using Z-view software version 2.2.
3. RESULTS AND DISCUSSION The single phase formation was confirmed using x-ray diffraction (XRD) and selected area electron diffraction (SAD) techniques. The average crystallite size of 45 nm was estimated from peak broadening of the most intense XRD peak after eliminating instrumental and strain broadening.[4] The SAD patterns of nanocrystalline CuAlO2 is shown in Fig. 1. The analysis of SAD pattern confirms the presence of delafossite phase CuAlO2 without any secondary or impurity phases. From the transmission electron microscopic image of the sample shown in Fig. 2, the grain size distribution was estimated and the average crystallite size was found to be the same that one obtained from the powder XRD pattern analysis. The grain size distribution plot is shown as an inset in Fig. 2. In general, accumulation of defects at the grain boundaries of nanomaterials can act as traps or recombination centers. The trapped charge carriers are creating a potential barrier with the adjacent grains identical to Schottky barrier. But it is often considered as double (or back-to-back) Schottky barrier because the interface is between the two adjacent grains. At the equilibrium state, the two space charge layers of a
Fig. 1. The SAD pattern of nanocrystalline CuAlO2. The pattern analysis confirms the presence of delafossite phase without any secondary or impurity phase.
Fig. 2. TEM image of nanocrystalline CuAlO2 (Inset: Grain size distribution plot.).
Fig. 3. Pictorial representation of the existence of double Schottky barrier structure of grain boundary with adjacent grains (L - left, R Right), here CB is the conduction band of grain, GB is the grain boundary and Φb is the potential barrier height.
grain boundary are symmetrical but while applying bias, one of the layers is depressed and the other one is extended. Such a situation should cause strong variations in the capacitance[5] and resistance of the depletion layers. If the bias is increased further then within the space charge region the population of interface states are considerably changed, on taking the band bending into account and this leads to the suppression of grain boundary potential barrier height.[6] Schematically the band bending in polycrystalline semiconductors as per ‘Grain Boundary Double Schottky Barrier Model’ is illustrated in Fig. 3. The effect of frequency on dielectric constant (ε') and dielectric loss (tan δ) of nanocrystalline CuAlO2 at different applied bias voltages are shown in Figs. 4 and 5 respectively. The value of ε' and tan δ decreases with frequency, which is a normal behavior of semiconductors since this provides information on the behavior of localized electric charge carriers, which give rise
Electron. Mater. Lett. Vol. 9, No. 2 (2013)
T. Prakash et al.
209
Fig. 4. Frequency dependence of dielectric constant (ε') of nanocrystalline CuAlO2 for various DC bias voltages.
Fig. 6. Grain and grain boundary contributed dielectric constant (ε') of nanocrystalline CuAlO2 as a function of DC voltages.
Fig. 5. Frequency dependence of dielectric loss (tan δ) of nanocrystalline CuAlO2 for various applied DC bias voltages.
Fig. 7. Grain and grain boundary contributed dielectric loss (tan δ) of nanocrystalline CuAlO2 as a function of DC voltages.
to the better understanding of the dielectric polarization mechanism. The present observed behavior clearly reveals
reduction of ε' and raise of tan δ are acceptable. The observed results are also showing evidence for considerable changes influenced by bias within the space charge region and this leads to the suppression of grain boundary potential barrier height. So, materials with Schottky barrier of the grain boundary can be affected by the grain size and width of grain boundaries, because for sufficiently large DC applied bias voltages there is a significant voltage drop across each grain boundary. Since dielectric constant is tunable with applying bias voltage, the relaxation time (τ) also shows its inference according to the relation[7]
that electronic and ionic polarisations are main contributors at high frequencies but at low frequencies the dipolar and interfacial polarisations are the only contributors. On each increment of DC bias voltage slight variation was observed in the low frequency region where the grain boundary contribution of dielectric constant is resolved. Such an influence of bias on dielectric constant (ε') and dielectric loss (tan δ) at 105 and 104 Hz frequencies are illustrated Figs. 6 and 7 respectively. In both of the plots, the data at 105 Hz frequency is independent of applied bias voltages. But in the case of 104 Hz frequency there is a reduction in the value of the ε' and an increase in the value of tan δ after 3.2 V. Since ε' and tan δ are inversely proportional to each other such
ε′ε τ = ---------O σ where εo is the vacuum permittivity (8.854 e−14 F/cm) and σ is
Electron. Mater. Lett. Vol. 9, No. 2 (2013)
210
T. Prakash et al.
the conductivity. The complex permittivity [ε*(ω) = ε'(ω) − iε"(ω)] and complex modulus [M*(ω) = M'(ω) + iM"(ω)] formalisms are interrelated as M*(ω) = 1/ε*(ω), where M'(ω) and M"(ω) are the real and imaginary parts of the electric modulus respectively, ω (= 2πf) is the relaxation frequency and f is the applied frequency.[8] The variation of real M'(ω) and imaginary M"(ω) parts of modulus M*(ω) spectra as a function of frequency for different applied bias voltages is shown in Fig. 8. The advantage of representing relaxation in modulus formalism is that the electrode polarization effects are suppressed in this representation. The variation in M"(ω) as a function of frequency over a range of bias is characterized by clearly resolved peaks appearing at a unique frequency have a tendency to shift towards the higher frequency side with the rise in bias. The low frequency side of the M"(ω) peak represents the range of frequencies in which charge carriers can move over a long distance, i.e., charge carriers can perform successful hopping from one site to the neighboring sites. The high frequency side of the M"(ω) peak represents the range of frequencies in which the charge carriers are spatially confined to their potential wells and thus could make a localized motion within the well. The region where the peak
Fig. 9. The variations of dielectric relaxation time of nanocrystalline CuAlO2 as a function of DC voltages.
occurs is an indication of the transition from long range to short range mobility with increase in frequency.[3] One can estimate the relaxation time from the M"(ω) plot peak frequency for each bias voltage. The bias dependence on the relaxation time is plotted in Fig. 9. At equilibrium state the τ is 0.63 ms and it was reduced to 0.13 ms after 3.2 V applied bias. Such a drastic reduction of relaxation time was observed after suppressing the grain boundary Schottky potential barrier height (Φb).
4. CONCLUSIONS This paper demonstrates the influence of DC bias voltage on the grain boundaries of nanocrystalline delafossite CuAlO2 by studying the dielectric properties using impedance spectroscopy. Analysis of impedance data confirms the existence of back-to-back Schottky potential barrier heights (Φb) each between an adjacent grain with grain boundary at the equilibrium condition later its suppression in the higher applied bias voltages. At equilibrium state, τ is 0.63 ms and it was reduced to 0.13 ms after the application of 3.2 V DC bias, which reveals the suppression of Φb. These observed DC bias voltage effects are obeying ‘brick layer model’ and also elucidate that Φb is playing a crucial role in controlling dielectric properties of nanomaterials.
ACKNOWLEDGEMENTS
Fig. 8. The real and imaginary parts of the modulus spectra of nanocrystalline CuAlO2 as a function of DC voltages (M'(ω) is represented by filled and M"(ω) by unfilled circles).
The authors would like to thank Mrs D. Kanchanamala, Senior Technical Assistant, Department of Metallurgical and Materials Engineering, Indian Institute of Technology Madras, Chennai 600 036, for her assistance in the TEM measurement.
Electron. Mater. Lett. Vol. 9, No. 2 (2013)
T. Prakash et al.
REFERENCES 1. H. Kawazoe, M. Yasukawa, H. Hyodo, M. Kurita, H. Yanagi, and H. Hosono, Nature 389, 939 (1997). 2. S. Gao, Y. Zhao, P. Gau, N. Chen, and Y. Xie, Nanotechnology 14, 538 (2003). 3. T. Prakash, K. Padma Prasad, S. Ramasamy, and B. S. Murty, J. Nanosci. Nanotech. 8, 4273 (2008). 4. T. Prakash, K. Padma Prasad, R. Kavitha, S. Ramasamy,
211
and B. S. Murty, J. Appl. Phys. 102, 104104 (2007). 5. W. Li and R. Schwartz, Phys. Rev. B 75, 012104 (2007). 6. I.-D. Kim, A. Rothschild, and H. L. Tuller, Appl. Phys. Lett. 88, 072902 (2006). 7. P. Balaya, J. Jamik, J. Fleig, and J. Maier, Appl. Phys. Lett. 88, 062109 (2006). 8. J. R. Macdonald (Ed.) Impedance Spectroscopy: Emphasizing Solid Materials and Systems, p. 8, Wiley, New York (1987).
Electron. Mater. Lett. Vol. 9, No. 2 (2013)
