Polyaniline/carbon nanotube composite Schottky contacts.
Schottky diodes fabricated from solvent casting (1-3), electropolymerization (4), (5), Langmuir-Blodgett deposition (6), (7), vacuum deposition (8), and pellets (9) of polyaniline (PANI) have been investigated by several different groups. It is believed that nano-engineered conducting polymer composites containing carbon nanotubes, which have a high aspect ratio and high conductivity, will enhance the thermal and mechanical stability as well as increase the overall conductivity of devices made with such materials (10), (11). It has been suggested that uniformly distributed nanotubes in the polymer act as nanometric heat sinks, preventing the buildup of large thermal effects and thus reducing material and device degradation (12). In particular, composites containing 1 weight percent carbon nanotubes have been found to possess good mechanical properties along with low surface roughness and enhanced DC conductivity (13). In this paper, we report on the fabrication and characterization of a polyaniline/multiwalled carbon nanotube composite, and investigate the electrical characteristics of Schottky contacts to this material.
PANI was chemically synthesized by a previously reported method as follows (14-16). Ammonium persulfate was added to a stirred aniline solution (in 1.0 M HCI). The aniline:oxidant molar ratio was 0.7:0.5. Total reaction time was 6 hours. The resultant emeraldine salt was washed with deionized water, and subsequently deprotonated to emeraldine base (EB) by prolonged stirring in 3-wt% ammonium hydroxide. Purification of the EB was accomplished by several continuous washings with copious amounts of methanol followed by de-ionized water. This purified EB powder was vacuum dried at 50[degrees]C, and the yield obtained was approximately 40%. The molecular weight was determined by gel permeation chromatography using polystyrene standards with n-methyl pyrrolidone (NMP) and LiBr (0.05 M) as the eluent at a column temperature of 85[degrees]C. The weight average molecular weight determined by this secondary weight determination method was found to be 135,000, with a polydispersity index of 3.
Multiwalled carbon nanotubes (MWNT) were synthesized by chemical vapor deposition using a mixture of xylene and ferrocene (17). These nanotubes were used in the fabrication of the composite material. A solution of o-xylene containing ferrocene (with ~ 0.75 at. % Fe/99.25 at. % C ratio) was fed continuously into a two-stage tubular quartz reactor (operated at near atmospheric pressure) using a syringe pump at a rate of 1 ml/h. Carbon deposits formed on the walls of the quartz reactor tube and on bare quartz substrates. A dense mat of aligned MWNTs were revealed in scanning electron microscopy (SEM) images, and electron diffraction patterns obtained on individual MWNTs showed the presence of 002 reflections, confirming a high degree of structural order.
COMPOSITE MATERIAL AND DEVICE FABRICATION
A polyaniline (EB) solution (3.5 wt% EB) was prepared at room temperature by dissolving the EB powder in dimethyl-propylene urea (DMPU). The EB powder was added very slowly to the solvent, using a powder funnel, over several days to avoid gelation. The polyaniline (EB) solution was then filtered through a 50 [micro]m filter (16). One weight percent of MWNT was added to this solution and then stirred for 24 hours. Then, 0.2 ml of the composite solution was poured onto ITO coated glass slides to form thin films. The slides were then placed into an oven under a dynamic vacuum of 10 mm Hg at 50[degrees]C in order to remove the solvent by evaporation. The thickness of the resulting films was determined to be 15 [+ or -] 2 [micro]m. These films were then doped using a vapor deposition technique, whereby they were placed in an enclosed chamber with a 1 M HCl solution for 48 hours at room temperature (18), (19).
Polyaniline/carbon nanotube composite diodes were fabricated in order to study the electrical characteristics of the films, as well as to assess their applicability to organic electronic devices in general. As discussed, these films were cast onto ITO coated glass slides, where the ITO forms the backside ohmic contact of the diode. Aluminum was chosen as the Schottky contact metal because of its relatively low work function [[PHI] = 4.1 eV) as compared to polyaniline ([PHI] = 4.1 - 4.45 eV) (20). Schottky contacts to the front side of the polymer films were formed by vacuum evaporation at a base pressure of 5 X 1[0.sup.-6] Torr, using shadow masks to form circular dots. A schematic representation of the device structure is shown in Fig. 1. Dots approximately 1 [micro]m in thickness were deposited. The shadow mask used for the contacts was designed to produce an array of 4 diodes on each sample, with a dot diameter of approximately 3.75 mm. Electrical characterization of the polymer devices was performed inside an enclosed probe station manufactured by Micromanipulator Co., Inc. Current-voltage (IV) measurements were obtained with an HP 4156B Precision Semiconductor Parameter Analyzer, by applying a ramped voltage and measuring the resulting diode current. The conductivity of the films was measured using a four-point probe technique with a Keithley 2000 Multimeter and 220 Current Source. Atomic Force Microscope images were obtained using a Digital Instruments Nanoscope III Atomic Force Microscope (AFM) in the tapping mode. The RMS surface roughness of the composite was evaluated based on data obtained over the entire film surface. AFM measurements were performed using a standard silicon-nitride tip.
[FIGURE 1 OMITTED]
RESULTS AND DISCUSSION
IV measurements were performed on diodes fabricated on the PANI/MWNT composite. Forward current was measured up to 20 volts, with a current compliance limit of the instrument set to 10 mA. Once the current reaches this compliance limit, higher voltage data is not valid. A typical IV curve measured on these devices at both forward and reverse bias is shown in Fig. 2. Considering the work functions of PANI and Al, contacts between these two materials are expected to be rectifying, and this is observed in the IV curves. Although the magnitude of the reverse current in Fig. 2 is starting to increase slightly at a bias of -5 V, the corresponding forward current at +5 V is over an order of magnitude higher. The IV curve in Fig. 2 clearly demonstrates that the contact is rectifying. The measured forward current for the composite diode reached the compliance limit at just above 18 V. Identically prepared devices containing no MWNTs typically produce current levels of nearly an order of magnitude less than these composite devices at the same voltage, showing that the nanotubes enhance conduction (13). The forward IV data is plotted on a semi-log scale in Fig. 3. One of the most interesting characteristics observed in Fig. 3 is the absence of a region of linearity, which we have observed on all our composite devices. A linear region on such a plot is normally expected for the usual transport mechanisms observed in Schottky diodes, such as thermionic emission and diffusion (21). Conductivity measurements on several PANI/MWNT samples ranged between 4 and 6 S/cm; thus it is unlikely that series resistance of the composite is dominating the IV curves. These results imply that additional transport mechanisms are dominant in these devices.
[FIGURE 2 OMITTED]
Further insight into the transport mechanisms operating in these composite diodes can be obtained by plotting the data on a log-log scale, which reveals two regions of linearity with different slopes. This is indicative of a power-law relationship between voltage and current, and this relationship can be expressed as
I = K[V.sup.m], (1)
where K is a constant and m is the exponent, which is determined from the slope of the curve. The transition region between the two linear regions occurs at approximately 2 Volts. In Fig. 4, the IV data is plotted on a log-log scale from just above 0 V to 2 V, along with the best fit power-law curve. The power-law fit to the measured IV curve is excellent, with a correlation co-efficient of 0.9995. From the slope of the curve, the exponent, m, is determined to be 1.05. Since this exponent is so close to unity, the current essentially follows Ohm's law in this range of forward bias. The IV data for voltages greater than 2 V is shown in Fig. 5 plotted on a log-log scale, along with the best fit power-law curve. Again, the fit is good, with a correlation coefficient of 0.9976. From Fig. 5, the transition region between the two power-law characteristics is seen to occur at approximately 2 V. For this second region, the exponent is determined to be 1.53, or approximately 3/2. It appears, therefore, that in these composite diodes the current follows Ohm's law at low voltages and transitions to a power-law at higher voltages.
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Although the low voltage characteristics appear to follow Ohm's law, the higher voltage power-law characteristics are consistent with space-charge-limited emission. The exponent of 3/2, leading to a current-voltage relationship given by
I = K[V.sup.3/2], (2)
is similar to Child's law for thermionic vacuum diodes. Furthermore, the current-voltage relationship of Eq 2 is also consistent with space-charge-limited emission at high fields in the presence of a distribution of shallow traps (22). The Ohm's law relationship observed at lower voltages in this material is consistent with the fundamental nature of the space-charge-limited emission mechanism, and can be explained by the presence of thermal free carriers from localized defect states. The onset of space-charge-limited current injection cannot dominate until some critical field is reached at which the injected excess free-electron concentration becomes comparable to the thermally generated, and consequently neutralized, carrier concentration (22). This occurs at higher voltages, apparently around 2 V in the composite devices described here. Similar transport phenomena have been observed in dielectrics, amorphous semiconductors, wide bandgap semiconductors, and other organic solids (22-26). One of the key points here is that this transport is affected by a distribution of traps in the material. A reduction in the density of localized defect states, therefore, may lead to devices with more ideal diode characteristics. A more exact current density equation would require knowledge of the dependence of drift velocity on applied voltage. In particular, if above some critical field, [E.sub.c], the drift velocity was proportional to the square-root of the applied field (not an uncommon relationship at moderately high fields), a current density expression could be approximated as (22)
J [equivalent] [epsilon][mu][theta] [square root of (E.sub.c) [[V.sup.3/2]/[L.sup.5/2]], (3)
where [epsilon] is the dielectric constant, [mu] is the low-field mobility, [theta] is a constant trap parameter which is inversely proportional to the trap density, L is the film thickness, and the critical field, [E.sub.c], is a constant. The model of Eq 3 contains the 3/2 power law variation with voltage, and it predicts a reduction in current density with increased trap density. To confirm these apparent transport mechanisms and determine the exact distribution and average concentration of the traps, more detailed measurements at different temperatures and film thicknesses, along with transport modeling, will be required; and this will be the subject of future investigations. Finally, it should be mentioned that devices were tested for both consistency and stability. Conductivity values of 30 + samples were found to be consistent and stable over a period of 700 hours.
[FIGURE 5 OMITTED]
The interfacial roughness of the polymer film plays a critical role in the properties of any organic device, since the metal/organic semiconductor interface strongly affects transport across a diode junction. A topographical image of the film produced with an AFM is shown in Fig. 6a. The RMS surface roughness of these composite films was found to be 4 nm, which is similar to the results we obtained for pure PANI films as reported in an earlier paper (27). PANI/MWNT composites of this composition, therefore, do not appear to significantly worsen surface roughness. The actual texture of the composite can be observed in Fig. 6b, which shows a scanning electron microscopy (SEM) micrograph of a cut edge of the composite film, where MWNTs are protruding from the core of the polymer film. From the image, it appears that the nanotubes blend well with the PANI. Further physical characterization and analysis of these composite materials is currently under way in our lab and will be the subject of a future publication (28).
Schottky diodes fabricated using composites of high molecular weight polyaniline and multiwalled carbon nanotubes produce current levels of significantly higher magnitude than pure polyaniline devices. The absence of a single linear region on semi-log IV curves is observed for these devices, which is inconsistent with thermionic emission. Linear regions with two different slopes observed on a log-log scale could explain this behavior, where at lower voltages the charge transport mechanism is consistent with Ohm's law but at higher voltages the charge transport is consistent with Child's law of space-charge-limited emission. This non-ideal diode behavior is strongly influenced by localized defect states. Further experiments and modeling should confirm the transport mechanisms and provide a detailed characterization of the traps. Continuing work is currently under way to study the thermo-electric power in PANI/MWNT composite materials and devices. Finally, the effects of single-walled carbon nanotubes on the electronic and mechanical properties of PANI composites are also being investigated and will be reported at a later date.
[FIGURE 6 OMITTED]
(1) School of Materials Science & Engineering
(2) Holcombe Department of Electrical & Computer Engineering
(3) Department of Physics & Astronomy
Clemson University, Clemson, SC 29634
* Author for correspondence. E-mail: firstname.lastname@example.org
(1.) C. Li and Y. Wang, Synth. Metals, 39, 91 (1990).
(2.) S. A. Chen, Y. Fang, and H. T. Lee, Synth. Metals, 55-57, 4082 (1993).
(3.) S. A. Chen and Y. Fang, Synth. Metals, 60, 215 (1993).
(4.) S. S. Pandey, S. C. K. Misra. B. D. Malhotra, and S. Chandra, J. Appl. Polym. Sci., 44, 911 (1992).
(5.) Y. Renkuan, Y. Hong, Z. Zheng, and Z. Youdou, Synth. Metals, 41, 731 (1991).
(6.) A. Riu Jr., C. A. Mills, and D. M. Taylor, J. Mater. Chem., 10, 91 (2000).
(7.) M. K. Ram, R. Gowri, and B. D. Malhotra, J. Appl. Polym. Sci., 63, 141 (1997).
(8.) S. C. K. Misra, M. K. Ram, S. S. Pandey, B. D. Malhotra, and S. Chandra, Appl. Phys. Lett., 61, 1219 (1992).
(9.) S. S. Pandey, M. K. Ram, V. R. Srivastava, and B. D. Malhotra, J. Appl. Polym. Sci., 65, 2745 (1997).
(10.) A. Hassanien, M. Gao, M. Tokumoto, and L. Dai, Chem. Phys. Lett., 342, 479 (2001).
(11.) M. Cochet, W. K. Maser, A. M. Benito, M. A. Callejas, M. T. Martinez, J. M. Benoit, J. Schreiber, and O. Chauvet, Chem. Commun., 1450 (2001).
(12.) S. A. Curran, P. M. Ajayan, W. J. Blau, D. L. Carroll, J. N. Coleman, A. B. Dalton, A. P. Davey, A. Drury, B. McCarthy, S. Maier, and A. Strevens, Adv. Mater., 10, 1091 (1998).
(13.) P. C. Ramamurthy, W. R. Harrell, R. V. Gregory, B. Sadanadan, and A. M. Rao, Synth. Metals, 137, 1497 (2003).
(14.) P. M. Beadle, Y. F. Nicolau, E. Banka P. Rannou, and D. Djurado, Synth. Metals, 95, 29 (1998).
(15.) Y. Wei, K. F. Hsueh, and G. W. Jang, Polymer, 16, 3572 (1994).
(16.) M. Angelopoulos, R. Dipietro, W. G. Zheng, A. G. MacDiarmid, and A. J. Epstein, Synth. Metals, 84, 35 (1997).
(17.) R. Andrews, D. Jacques, A. M. Rao, F. Derbyshire, D. Qian, X. Fan, E. C. Dickey, and J. Chen, Chem. Phys. Lett., 303, 467 (1999).
(18.) Z. X. Bao, C. X. Liu, P. K. Kahol, and N. J. Pinto, Synth. Metals, 106. 107 (1999).
(19.) J. P. Long, S. E. Bullock, L. J. Buckley, and J. N. Russell Jr., Synth. Metals, 126, 317 (2002).
(20.) S. Bandyopadhyay, A. Bhattacharyya, and K. Sen, J. Appl. Phys., 85, 3671 (1999).
(21.) S. M. Sze, Physics of Semiconductor Devices, 2nd Ed., pp. 254-63, Wiley, New York (1981).
(22.) M. A. Lampert and P. Mark, Current Injection in Solids, Academic Press, New York (1970).
(23.) J. Zhang and W. R. Harrell, J. Vac. Sci. Tech. B, 21, 872 (2003).
(24.) S. Ashok, A. Lester, and S. J. Fonash, IEEE Electron Dev. Lett., EDL-1, 200 (1980).
(25.) H. Meier, Organic Semiconductors, pp. 299-300, Verlag Chemie, Weinheim, Germany (1974).
(26.) P. R. Somani, R. Marimuthu, and A. B. Mandale, Polymer, 42, 2991 (2001).
(27.) A. Lodha, S. M. Kilbey, P. Ramamurthy, and R. V. Gregory, J. Appl. Poly. Sci., 82, 3602 (2001).
(28.) P. C. Ramamurthy, W. R. Harrell, R. V. Gregory. B. Sadanadan, and A. M. Rao, to be published.
PRAVEEN C. RAMAMURTHY (1), WILLIAM R. HARRELL (2), *, RICHARD V. GREGORY (1), BINDU SADANADAN (3), and APPARAO M. RAO (3)
|Printer friendly Cite/link Email Feedback|
|Author:||Ramamurthy, Praveen C.; Harrell, William R.; Gregory, Richard V.; Sadanadan, Bindu; Rao, Apparao M.|
|Publication:||Polymer Engineering and Science|
|Date:||Jan 1, 2004|
|Previous Article:||Measurement of interfacial tension in PS/LDPE melts saturated with supercritical C[O.sub.2].|
|Next Article:||Effect of erucamide and silica loadings on COF behavior of POP multilayer films in repetitive testing.|