Department of Physics, Faculty of Science, King Abdul Aziz University, P.O. Box 80203, Jeddah, Saudi Arabia 2

Transcription

1 RESEARCH ARTICLE Copyright 2011 American Scientific Publishers All rights reserved Printed in the United States of America Advanced Science Letters Vol. 4, 1 5, 2011 Electrical Properties of p-si/n-zno Nanowires Heterojunction Devices S. Al-Heniti 1, R. I. Badran 1 2, A. A. Al-Ghamedi 1, and F. A. Al-Agel 1 1 Department of Physics, Faculty of Science, King Abdul Aziz University, P.O. Box 80203, Jeddah, Saudi Arabia 2 Department of Physics, The Hashemite University, P. O. Box 1459, Zarqa, Jordan This paper explores the temperature dependent heterojunction behavior of n-type zinc oxide (ZnO) nanowires/p- Si diodes. The device behavior at different temperatures in forward as well as reverse biased conditions are studied and reported. From the detailed electrical properties, it is confirmed that the fabricated p n diode showed a good stability over the temperature range of 130 C. The turn-on and breakdown voltage of the device slightly decreases with an increase of temperature whereas the saturation current of the device increases. The effective potential barrier height is found increasing with the increase in temperature. The quality factor is found with and without the consideration of barrier height inhomogenity. The mean potential barrier is also determined. Moreover, a value of activation energy of 53 mev which is close to the exciton binding energy of ZnO, is estimated. Keywords: ZnO, Nanowires, Photoluminescence, Heterojunction Diodes. 1. INTRODUCTION Among various semiconductor metal oxide nanostructures, the nanostructures of II VI semiconductor ZnO possess special place due to its own merits and properties. It is an important n-type semiconductor and have various specific properties such as direct and wide band gap (3.37 ev), large saturation velocity, high breakdown voltage, high-mechanical and thermal stabilities, large exciton binding energy (60 mev) at room temperature. The strong exciton binding energy of ZnO compared to thermal energy at room-temperature (26 mev) ensures an efficient exciton emission at room-temperature under low-excitation energy. 1 3 ZnO exhibits biocompatibility hence recently used for the fabrication of various kinds of biosensors. 4 6 Due to the lack of center of symmetry in ZnO combined with large electromechanical coupling, the ZnO exhibits the piezoelectric and pyroelectric properties, hence used efficiently for the fabrication of piezoelectric sensors and actuators. 7 In addition to this, the ZnO is also efficiently used in various applications such as catalysts, solar cells, field emission devices, sensors and actuators, hydrogen storage, solar cells, electronic and photovoltaic devices, transparent conductive films and so on Because of various excellent properties and so many high-technological applications, variety of ZnO nanostructures have been synthesized by using several fabrication techniques and reported in Author to whom correspondence should be addressed. the literature. Some of such techniques like thermal evaporation process, solution process, simple gas reactions, template directed wet and dry methods, vapor-phase transport process, metal organic chemical vapor deposition (MOCVD), hydrothermal process, arc discharge, solution-based synthesis, sol gel process and so forth might have a special attention The ZnO possesses variety of nanostructures, hence it is believed that the ZnO is probably richest family of nanostructures Few nanostructures of ZnO are nanorods, nanowires, nanocombs, nanonails, nanotubes, flower-shaped and hierarchical nanostructures, nanohelixes, nanosprings and nanorings, doughnuts-like nanostructures and so on Among various nanostructures, the one-dimensional (1D) nanostructure systems such as nanowires, nanotubes, and nanobelts have stimulated much attention due to their unique and specific properties and their applications to nanoscale devices. In this paper, the electrical properties of as-grown aligned hexagonal-shaped ZnO nanowires on silicon substrate are studied with more details. Moreover, the as-grown n-zno nanowires on p-silicon substrate were used to fabricate p n heterojunction diode. 23 The temperature-dependant (from C 130 C) behavior of as-fabricated n-zno/p-si heterojunction diode were also examined in this manuscript. Here, high values of quality are determined at different temperatures. Lateral fluctuations that exist in the Schottky contact are modeled by Gaussian distribution of barrier height V b with standard deviation s around the mean value. The corrected mean barrier height is obtained, Adv. Sci. Lett. Vol. 4, No. 1, /2011/4/001/005 doi: /asl

2 RESEARCH ARTICLE Adv. Sci. Lett. 4, 1 5, 2011 when the barrier height inhomogenity is considered. In addition, corrected quality factors are also obtained at different temperatures. 2. EXPERIMENTAL DETAILS The fabrication process of n-zno/p-si heterojunction diodes is reported elsewhere. 23 The I V characteristics of the fabricated n-zno/p-si heterojunction diode ( cm 2 ) were studied by using KEITHLEY 6517A under vacuum using Neytech Qex furnace at 2 torr at various temperatures (i.e., C 130 C) RESULTS AND DISCUSSION The detailed structural and optical properties of as-synthesized ZnO nanowires has already been investigated and published elsewhere. 23 The electrical properties of the as-grown hexagonal-shaped ZnO nanowires, a heterojunction device based on n-zno nanowires/p-si substrate, via the current voltage characteristics, were studied. The current voltage (I V ) characteristics of the fabricated heterojunction diode were studied at several temperatures (from 130 C) in the forward and reverse bias conditions. 23 The general I V characteristics of the fabricated n-zno/p-si heterojunction diode with the variation of temperature (from 130 C) have already been published in Ref. [23]. This manuscript demonstrates the further in-depth studies on the of the fabricated n-zno/p-si heterojunction diode at various temperatures. The voltage dependence of the junction current may be written as: I = I o T e ev /mkt (1) where, e, k, and I o T ) are electronic charge, Boltzman constant and temperature-dependent saturation current. Here, n is the quality factor that measures the conformity of the diode to pure thermionic emission. The temperature-dependent saturation current can be related to the effective barrier height V eff by the relation: I o T = AA T 2 e ev eff /kt (2) Here, A and A are the contact area and Richardson constant, respectively. In a thermionic emission model, a more realistic relation than that shown in Eq. (1) may be adopted and written as: 24 I = I o T e ev /mkt 1 e ev /kt (3) A non-linear and rectifying behavior is obtained in the I V characteristics of n-zno/p-si heterojunction. At C, n-zno/p-si heterojunction attains moderate turn-on voltage of 0.5 V and delivers reasonable least current in the order of 10 4 A. At 130 C a little increase in current occurs at same turn-on voltage. The fabricated n-zno/p-si heterojunction has a turn-on voltage of 1.0 V at higher current. It also exhibits far broader breakdown voltage ( 3 V) with good leakage current of A at C. On the other hand, at 130 C, the diode has high leakage current of A at a voltage of 6 V. Due the increase in temperature, the turn-on voltage, breakdown voltage and leakage current have exhibited similar behavior in reverse bias. Such behavior may be attributed to the tunneling effect of ZnO nanowires at the junction. The obtained characteristics of the junction exhibit a diode-like behavior (which indicates a reasonable Ohmic behavior of Ag contact with ZnO wires) with a turn-on voltage of 1.0 V at all temperatures. This latter value lies between a larger value obtained in Refs. [27, 28] and a lower value found by others This turn-on voltage allows very small value of current 0.1 na. Such turn-on voltage indicates the formation of a Schottky junction between the top of metal contact and the ZnO wires. This is because such value is close to the expected turn-on voltage (about 1.5 V) for n-zno/p-si heterojunction. Moreover the low value of current may be attributed to nonuniform growth geometry of nanowires. Both the gradual change of turn-on voltage and the small change in the slope of forward bias I V plots via the change in temperature are an indication for attaining a rectifying behavior of the device. For example at 3.0 V the junction exhibits a rectifying behavior with rectifying ratio of 1.0. The plots of ln I versus ln V, in Figure 1, reveal two regions which have different slopes at all temperatures and may be attributed to two transport mechanisms. This is in agreement with that found by others. 26 The average slope in region I (with V<3 5 V) is larger than that in region II (with V>3 5 V) at all temperatures. The mild fluctuations around the slope in region II may be an indication for the stability in the device. Region I exhibits a linear dependency of current on the applied voltage. This may indicate that a field emission or tunneling is the dominated mechanism. Moreover, region II shows an exponential relation between the current and the applied voltage. This is mainly due to recombination-tunneling mechanism. Figure 2(a) shows the plots of ln I versus V for both forward and reverse bias of heterojunction diode. However, the plots of ln I versus V in forward bias (Fig. 2(b)) shows again two different regions. Both saturation current I o and diode quality factor n can be obtained from the intercept and slope of region I shown in Figure 2(b), respectively. The obtained saturation current and quality factor are drawn against temperature and shown in Figure 3. The saturation current, in region I, exhibits a trend of increase when the temperature increases except at T = 130 C. This trend is found similar to that revealed by others. 29 The quality factor evaluated, at region I, also exhibits a trend similar to that of others 29 when the temperature increases. However, n

3 Adv. Sci. Lett. 4, 1 5, 2011 RESEARCH ARTICLE (a) Temperature (K) ln (I) T = C Voltage (V) ln (I o ) E a = 53 mev R s (nω) ln (I) (b) Region I V<3.5 v T = C Voltage (V) Region II V>3.5 v Fig. 2. (a) Plots of ln I versus V for forward and reverse bias of n-zno/p- Si heterojunction diode at all temperatures. (b) Plots of lni versus V for forward bias of n-zno/p-si heterojunction diode at all temperatures showing two regions only. exhibits a fall off with the increase in temperature in this region. Here, n is considerably high at low temperatures. The temperature sensitivity of n is reduced to a much lower value when the temperature becomes high. The values of n, in region II, are very I o (µa) Temperature (K) Quality factor Fig. 4. LnI o (solid squares) versus 1000/T and series resistance (open circles) versus temperature. much larger than the latter values. Such a trend at higher voltages is in agreement with that shown by others. 27 The high values of n at high voltages may be an indication for the existence of high concentration of defect states in ZnO nanowires. Figure 4 shows the variation of ln I o versus 1000/T. The activation energy of E a = 53 mev is extracted from this plot. This latter value is close to the exciton binding energy of ZnO (of about 60 mev). However the global behavior, here, is similar to that obtained in Ref. [29]. The series resistance which is obtained from the slope of high voltage region of I V plots increases with the increase in temperature exhibits behavior similar to that found by others. Employing a better model than the one used earlier in this paper is necessary to explain the inhomogenity with some possible Schottky barrier patches which may exist in the Schottky diode. The resultant current, at low temperatures, through the Schottky junction may be determined by the current component flowing through localized low-barrier height regions. As a result of that, the quality factor derived from the I V characteristic deviates far from unity. Moreover, the increase in temperature may cause the gradual increase in the current component flowing through the homogenous region of the Schottky contacts with higher barrier height. This is the reason behind the reduction of quality factor at high temperatures. Figure 5 shows the variance of effective barrier height with quality factor at each temperature. The linear relationship may be attributed to lateral inhomogenities of the barrier height. A homogenous barrier height of approximately 0.6 ev can be obtained from the extrapolation of the straight line to n = 1. The inhomogenity in the barrier height for n-zno/p-si may be resulted from the considerable mismatch in the lattice constant and thermal expansion coefficient between n-zno and p-si at the interface. Such mismatch may introduce different defects like dislocation, stacking faults, etc which are likely the origin of localized inhomogenity of the barrier height. 32 Schottky contacts are commonly known to exhibit lateral fluctuations. It is worth investigating such behavior in our device. The effective barrier height V eff was correlated to the mean barrier height V b as follows: 33 Fig. 3. Saturation current I o (solid squares) and quality factor (open circles) against temperature. V eff = V b e 2 s 2kT (4) 3

4 RESEARCH ARTICLE Adv. Sci. Lett. 4, 1 5, Effective barrier height (ev) (n 1 1) Quality factor Fig. 5. Effective barrier height versus quality factor due to the consideration of barrier height inhomogeneity. Here, the lateral fluctuations may be modeled by Gaussian distribution of barrier heights with standard deviation s around the mean value V b. A plot of effective barrier height versus inverse temperature is shown in Figure 6. The linear fit gives values of 0.68 ev and 0.13 ev to mean barrier height and standard deviation, respectively. This means that the measure of barrier height homogeneity is approximately around 13% of the mean barrier height. The obtained value of mean barrier height is consistent with the energy difference of the work function between Si and ZnO; the Fermi level below the vacuum level is 4.97 ev for p-si and 4. ev for ZnO where the difference between them is 0.72 ev. Moreover, the model also tackles the barrier height inhomogenity via the relation between the quality factor and temperature: n 1 1 = i e 2 (5) 2kT Here 1 and 2 are temperature-independent parameters. Figure 7 shows the plot of n 1 1 versus the inverse of temperature, which is fitted to a straight line. The values of Fig. 7. Plot of n 1 1 versus the inverse of temperature due to barrier height inhomogeneity. 1 = and 2 = V are obtained. Thus, the plot in Figure 8 confirms that the quality factor denotes the voltage deformation of the Gaussian distribution of the barrier height. ln (I o /T 2 ) (a) (b) Effective barrier height (ev) ln(i o /T 2 ) (eσ s ) 2 /2(kT) Fig Effective barrier height versus the inverse of temperature. Fig. 8. (a) The plot of ln I o /T 2 versus 1000/T ; (b) The corrected Richardson plot of ln I o /T 2 e s / 2kT 2 versus 1000/T. 4

5 Adv. Sci. Lett. 4, 1 5, 2011 RESEARCH ARTICLE The plot of ln I o /T 2 versus inverse temperature, in Figure 8(a), shows scattered points, especially at low temperatures. The deviation of Richardson plot from linearity due to the barrier inhomogenity may require the use of a modified formula. Using Eqs. (2), (3) and (4), an expression that accounts for the modified Richardson constant can be expressed by: ( ) ( ) 2 Io e s ln = ln AA ev b (6) T 2 2kT kt Figure 8(b) shows the modified Richardson plot using Eq. (6). When the obtained points are fitted to a straight line, the mean barrier height V b = 0 57 ev is obtained from the slope. It is interesting to note that the modified Richardson plot has a good linearity over the whole temperature range corresponding to single activation energy around V b. Moreover, the mean barrier potential is found close to those of 0.6 ev and 0.68 ev found earlier in our analysis and is consistent with the value of 0.72 ev of the energy difference of the work function between Si and ZnO. 4. CONCLUSIONS Hexagonal-shaped aligned n-zno nanowires were grown on p-silicon substrate by thermal evaporation process using metallic zinc powder and oxygen gas as source materials for zinc and oxygen, respectively. The as-grown n-zno nanowires grown on p-silicon substrate were used to fabricate p n heterojunction diode. The fabricated n-zno/p-si heterojunction diode attains moderate turn-on voltage of 1.0 V. Temperature-dependant (from C 130 C), I V characteristics for the fabricated device was also examined and presented in this paper. From the obtained results, it can be concluded that the simply grown ZnO nanostructures can be used to fabricate efficient heterojunction diodes. The p n diode showed a good stability over the temperature range of 130 C. The turn-on and breakdown voltage of the device slightly decreased and the saturation current increased with the increase in temperature. A measure of barrier height homogeneity is found approximately around 13% of the mean barrier height of 0.68 ev. High values of quality factor may be attributed to be an indication for the existence of high concentration of defect states in ZnO nanowires or to the lateral inhomogenity of barrier height. Acknowledgments: The authors are highly thankful for the financial support of King Abdulaziz University Research Centre through grant number 430/ References and Notes 1. H. S. Nalwa (ed.), Encyclopedia of Nanoscience and Nanotechnology (2004), Vol S. J. Pearton, B. S. Kang, B. P. Gila, D. P. Norton, O. Kryliouk, F. Ren, Y.-W. Heo, C. Y. Chang, G.-C. Chi, W.-M. Wang, and L.-C. Chen, J. Nanosci. Nanotechnol. 8, 99 (2008). 3. (a) Z. L. Wang, J. Nanosci. Nanotechnol. 8, 27 (2008); (b) T. N. Soitah, Y. Chunhui, and S. Liang, Sci. Adv. Mater. 2, 534 (2010). 4. (a) Y. Song, M. Zheng, L. Ma, and W. Shen, J. Nanosci. Nanotechnol. 10, 426 (2010); (b) A. Khan, S. N. Khan, and W. M. Jadwisienczak, Sci. Adv. Mater. 2, 572 (2010). 5. A. Umar, M. M. Rahman, A. Al-Hajry, and Y. B. Hahn, Electrochem. Commun. 11, 278 (2009). 6. (a) E. Rayón and C. Ferrer, J. Nanosci. Nanotechnol. 10, 1371 (2010); (b) W. Wu, S. Bai, N. Cui, F. Ma, Z. Wei, Y. Qin, and E. Xie, Sci. Adv. Mater. 2, 402 (2010) 7. C.-Y. Lu, S.-P. Chang, S.-J. Chang, C.-L. Hsu, Y.-Z. Chiou, and I.-C. Chen, J. Nanosci. Nanotechnol. 10, 1135 (2010). 8. (a) S. Kim, S. Paek, and S. Lee, J. Nanosci. Nanotechnol. 10, 60 (2010). 9. Y. Song, M. Zheng, L. Ma, and W. Shen, J. Nanosci. Nanotechnol. 10, 426 (2010). 10. H. Zhang, N. Du, B. Chen, D. Li, and D. Yang, Sci. Adv. Mater. 1, 13 (2009). 11. Z. L. Wang, Mater. Today 10, 20 (2007). 12. V. Dallacasa, F. Dallacasa, P. Di Sia, E. Scavetta, and D. Tonelli, J. Nanosci. Nanotechnol. 10, 1043 (2010). 13. C.-X. Wang, G.-W. Yang, H.-W. Liu, Y.-H. Han, J.-F. Luo, C.-X. Gao. and G.-T. Zou, Appl. Phys. Lett. 84, 2427 (2004). 14. A. Dejneka, A. Churpita, Z. Hubika, V. Trepakov, Z. Potek, L. Jastrabík, G. Suchaneck, and G. Gerlach, J. Nanosci. Nanotechnol. 9, 4094 (2009). 15. A. Umar, B. K. Kim, J. J. Kim, and Y. B. Hahn, Nanotechnology 18, (2007). 16. K. Vanheusden, C. H. Seager, W. L. Warren, D. R. Tallant, and J. A. Voigt, J. App. Phys. 79, 7983 (1996). 17. X. Y. Kong, Y. Ding, R. S. Yang, and Z. L. Wang. Science 303, 1348 (2004). 18. W. I. Park, G. C. Yi, M. Kim, and S. J. Pennycook, Adv. Mater. 15, 526 (2003). 19. K. Huo, J. Fu, H. Ni, Y. Hu, G. Qian, P. K. Chu, and Z. Hu J. Nanosci. Nanotechnol. 9, 3848 (2009). 20. Y. Song, M. Zheng, L. Ma, and W. Shen, J. Nanosci. Nanotechnol. 10, 426 (2010). 21. A. Umar and Y. B. Hahn, Appl. Phys. Lett. 88, (2006). 22. P. X. Gao and Z. L. Wang, J. Phys. Chem. B 108, 7534 (2004). 23. S. Al-Heniti, J. Nanosci. Nanotechnol. 10, 6606 (2010). 24. H. C. Card and E. H. Rhoderick, J. Phys. D: Appl. Phys. 4, 1589 (1971).. S. M. Sze, Physics of Semiconductor Devices, 2nd edn., John Wiley & Sons, New York (1981). 26. R. Ghosh and D. Basak, Appl. Phys. Lett. 90, (2007). 27. N. K. Reddy, Q. Ahsanulhaq, J. H. Kim, and Y. B. Hahn, Appl. Phys. Let. 92, (2008). 28. A. Osinsky, J. W. Dong, M. Z. Kauser, B. Hertog, A. M. Dabiran, P. P. Chow, S. J. Pearton, O. Lopatiuk, and L. Chernyak, Appl. Phys. Lett. 85, 4272 (2004). 29. N. K. Reddy, Q. Ahsanulhaq, and Y. B. Hahn, Appl. Phys. Let. 93, (2008). 30. S. Majumdar, S. Chattopdhyay, and P. Banerji, Appl, Sur. Sci. 5, 6141 (2009). 31. N. K. Reddy, Q. Ahsanulhaq, J. H. Kim, and Y. B. Hahn, Exploring the Frontiers of Physics 81, (2008). 32. S. Roy, C. Jacob, and S. Basu, Solid State Sci. 6, 377 (2004). 33. S. Altindal, I. Dokme, M. M. Bulbul, N. Yalcin, and T. Serin, Microelectr. Eng. 83, 499 (2006). Received: 10 May Accepted: 17 July