Synthesis and Optical Properties of Co-Doped ZnO Submicrometer

May 27, 2009 - Submicrometer ZnO tubes have been synthesized by a polymer based template approach using sol−gel deposition. Zinc acetate, a precurso...
0 downloads 13 Views 2MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Synthesis and Optical Properties of Co-Doped ZnO Submicrometer Tubes from Electrospun Fiber Templates Frederick Ochanda,† Kevin Cho,† Dickson Andala,† Thomas C. Keane,‡ Ari Atkinson,† and Wayne E. Jones, Jr.*,† †

Department of Chemistry, State University of New York at Binghamton, Vestal Parkway East, Binghamton, New York 13902, and ‡Department of Chemistry and Physics, Russell Sage College, Troy, New York 12180 Received August 22, 2008. Revised Manuscript Received March 9, 2009

Submicrometer ZnO tubes have been synthesized by a polymer based template approach using sol-gel deposition. Zinc acetate, a precursor to ZnO, was deposited on catalytically active electrospun polycarbonate fibers ∼ 250 ( 100 nm in diameter. Thermal degradation of the core fibers resulted in the oxidation of zinc acetate to produce ZnO nanotubes with diameters of ∼500 ( 100 nm and an average wall thickness of ∼100 ( 50 nm. Scanning electron microscopy (SEM), Energy dispersive spectroscopy, themogravimetric analysis, Fourier transform infrared spectroscopy, and UV-visible spectroscopy were used to characterize the composition, structure, and morphology of the tubes. Powder X-ray diffraction results showed that a wurtzite crystalline phase was obtained. The UV-visible absorption spectrum was red-shifted by 25 nm due to narrowing of the ZnO band gap (∼3.22 eV) as a result of Co doping. Similarly, green band emission was not observed in the emission spectrum, while emission lifetime was determined to be 620 ps from photoluminescence studies.

Introduction Zinc oxide based nano/microstructures have attracted much attention as a result of novel electrical, optoelectronic, magnetooptic, and photoelectrochemical properties.1-5 The electronic and optical properties of ZnO nanostructures are largely dependent on their composition/doping, crystal structure, dimension, and morphology. In particular, one-dimensional ZnO structures with controlled morphology offer great potential in efficient assembly and performance of nanoscale devices due to quantum confinement of charge carriers in small dimensions.6-9 Zinc oxide is a wide band gap (3.37 eV) semiconductor with a high exciton binding energy (60 meV) and hence exhibits efficient excitonic emission at room temperature and low excitation energy.10,11 It crystallizes into a wurtzite structure and displays piezoelectric properties when its c-axis is oriented perpendicular to a substrate. As a result, it is found in many electroacoustic applications such as sound sensors, SONAR emitters and detectors, pressure transducers, and catalysis. ZnO is also used as the clear top electrode in solar cells, since it is transparent to visible light.12 *To whom correspondence should be addressed. E-mail: wjones@ binghamton.edu. Fax: (607) 777-4478. Telephone: (607) 777-2421. (1) Schmidt-Mende, L; MacManus-Driscoll, L. J. Mater. Today 2007, 10, 40. (2) Gudiksen, M. S.; Lasuson, L. J.; Wang, J.; Smith, D. C.; Leiber, C. M. Nature (London) 2002, 415, 617. (3) Keis, K.; Vayssieres, L.; Rensmo, H.; Lindquist, S.-E.; Hagfeldt, A. J. Electrochem. Soc. 2001, 148(2), A149. (4) Grabowska, J.; Nanda, K. K.; McGlynn, E.; Mosnier, J. P.; Henry, M. O.; Beaucamp, A.; Meaney, A. J. Mater. Sci.: Mater. Electron. 2005, 16, 397. (5) Xing, Y. J.; Xi, Z. H.; Xue, Z. Q.; Zhang, X. D.; Song, J. H.; Wang, R. M.; Xu, J.; Song, Y.; Zhang, X. L.; Yu, D. P. Appl. Phys. Lett. 2003, 83, 1689. (6) Huang, M. H.; Feick, H; Weber, E.; Wu, Y.; Tran, N.; Yang, P. Adv. Mater. 2001, 13, 113. (7) Kong, X. Y.; Wang, Z. L. Appl. Phys. Lett. 2004, 84, 975. (8) Hughes, W. L.; Wang, Z. L. J. Am. Chem. Soc. 2004, 126, 6703. (9) Wang, Z. L.; Kong, X. Y.; Zuo, J. M. Phys. Rev. Lett. 2003, 91, 185502. (10) Huang, M. H.; Mao, S.; Feick, H.; Yan, H.; Wu, Y.; Kind, H.; Weber, E.; Russo, R.; Yang, P. Science 2001, 292, 1987. (11) Mulligan, R. F.; Iliadis, A. A.; Kofinas, P. J. Appl. Polym. Sci. 2003, 89, 1058. (12) Hu, J.; Gordon, R. G. Sol. Cells 1991, 30, 437.

Langmuir 2009, 25(13), 7547–7552

Transition-metal-doped ZnO has been investigated as a promising dilute magnetic semiconductor for implementing spintronic device concepts. Doping is a widely used method to improve the electrical, magnetic, and optical properties of semiconductors, which are crucial for their practical applications.13 Various group III metals such as Al and Ga,14 transition metals such as Mn15 and Fe,16 and rare earth elements such as Eu17 and Er18 have been doped into ZnO phosphors and thin films for different applications. Research on the doping of ZnO has concentrated on thin films or powder forms; however, reports on doped one-dimensional nanotubes ZnO remain rare. Many methods for fabrication of ZnO nanostructured materials have been developed, ranging from lithographic techniques, molecular beam epitaxy to chemical techniques such as solidvapor decomposition.6,9,19,20 The template based approach provides a versatile and low-cost method to prepare nanometerlength scale templates in high volume and with control of dimension and surface morphology. Building on our previously successful polymer based template approach, one-dimensional Co-doped ZnO submicrometer tubes were fabricated using electrospun fiber templates and sol-gel technology, followed by high temperature thermal treatment to remove the template fibers.21-23 Electrospinning is a nonmechanical, electrostatic method that produces solid fibers in the nanometer to micrometer (13) Yi, G. C.; Wang, C.; Park, W. Semicond. Sci. Technol. 2005, 20, 522. (14) Yamamoto, T.; Yoshida, H. K. Phys. B 2001, 302/303, 155. (15) Han, J.; Senos, A. M. R.; Mantas, P. Q. Mater. Chem. Phys. 2002, 75, 117. (16) Han, S.-J.; Song, J. W.; Yang, C.-H.; Park, S. H.; Park, J.-H.; Jeong, Y. H.; Rhie, K. W. Appl. Phys. Lett. 2002, 81, 4212. (17) Park, Y. K.; Han, J. I.; Kwak, M. G.; Yang, H.; Ju, S. H.; Cho, W. S. J. Lumin. 1998, 78, 87. (18) Zhao, X.; Komuro, S.; Isshiki, H.; Aoyagi, Y.; Sugano, T. J. Lumin. 2000, 87-89, 1254. (19) Ozin, G. A. Adv. Mater. 1992, 4, 612. (20) Kong, X. Y.; Ding, Y.; Wang, Z. L. J. Phys. Chem. B 2004, 108, 570. (21) Ochanda, F.; Jones, W. E.Jr. Langmuir 2005, 21, 10791. (22) Reneker, D. H.; Yarin, A. L.; Koombhongse, S. J. Appl. Phys. 2001, 89, 3018. (23) Hou, H.; Jun, Z.; Reeuning, A.; Schaper, A.; Wendorf, J. H.; Greiner, A. Macromolecules 2002, 35, 2429.

Published on Web 05/27/2009

DOI: 10.1021/la802753k

7547

Article

Ochanda et al.

range from polymer solutions or melts.22,24 It is envisaged that tuning the band gap and reducing point defects in ZnO nanostructures by Co doping will be important in their application in nanoscale electronics, biochemical sensing, and optoelectronics. Additionally, the large surface area to volume ratio makes them suitable for application as catalysts and photoelectrochemical solar cells.

Scheme 1. Schematic Diagram of Tubes by Fiber Templates (TUFT) Process Formation of a Tube

Experimental Section Materials. Polycarbonate (PC) pellets (Mw = 64 000), methylene chloride, N,N-dimethylformamide (DMF), palladium chloride, cobalt(II) nitrate (all from Aldrich), hydrochloric acid (J.T. Baker), and zinc acetate (ZnAc) (Merck GR grade) were used as received from the manufacturer. Template Fiber Sample Preparation. A polymer solution, 180 mg/mL, was prepared by dissolving polymer pellets (Mw = 64 000) in a CH2Cl2/DMF (0.65/0.35) solvent mixture. Electrospinning was done with an applied voltage of 20 kV and a distance of 20 cm between the collector screen and the spinneret.21 Electrospun fibers were peeled from the aluminum foil collector screen after soaking in 1 M HCl solution and then rinsed in deionized water. ZnO polycarbonate coaxial fibers were synthesized through wet chemical methods by hydrolysis of zinc acetate dihydrate, a ZnO precursor, followed by condensation in sol-gel deposition. The nanofiber surface was first sensitized and then activated by immersion for 20 min in 3.0 mM PdCl2 aqueous solution containing 0.01 M HCl. A 0.1 M zinc acetate, Zn(CH3CO2)2 3 2H2O, solution was made by dissolving 0.439 g into 20 mL of DMF. A 0.058 g (0.01 M) solution of cobalt(II) nitrate was added to this, and the resulting mixture was stirred for 24 h to form a homogeneous and stable solution. The activated fibers were rinsed and then immersed in the zinc acetate solution, and the reaction proceeded with stirring for 24 h at room temperature. The resulting coaxial fibers were rinsed in deionized water and dried before heating to 650 °C at a ramp rate of 10 °C/min and annealed at that temperature for 3 h. Instrumentation. The morphology of template and submicrometer ZnO tubes was examined by scanning electron microscopy (SEM) (model Hitachi S-570LB microscope) using an accelerating voltage of 10-15 kV. For nonconducting samples, a thin film of Au/Pd was coated on the surface. Transmission electron microscopy (TEM) was done using a Jeol TEM 2010 instrument. Electronic absorption spectra were recorded on a Hewlett-Packard 8452A UV-visible spectrophotometer. IR spectra were recorded on a Digilab FTS-40 PRO as a KBr pellet or fiber mat. The tensile stress and strain test of spun fibers and coaxial fibers was measured using an Instron 5543 instrument. The samples were cut into strips of five specimens with a gauge length 2 in. by 0.5 in. and with 0.002 in. thickness. The X-ray diffraction spectra were measured on fine ZnO tubes using a Scintag X-ray diffractometer with an X-ray wavelength of 1.5418 A˚ (Cu KR) radiation source. The samples for powder X-ray diffraction (PXRD) were compacted to at least 1 mm in thickness to prevent penetration of the X-ray beam. Emission lifetime measurements were obtained on a Photochemical Research Associate (PRA) system 3000 time-correlated pulsed single photon counting apparatus. The sample was excited with light from a PRA model 510 nitrogen flash lamp that had been transmitted through a SA Inc. H-10 monochromator, and emission was detected perpendicular to the incident beam via a second H-10 monochromator with a water cooled red sensitive Hamamatsu R955 photomultiplier tube. The resulting photon counts were recorded on a Tracor Northern 7200 microprocessor based multichannel analyzer.

Fiber Template Method and Formation. The template approach employed is a general method for the preparation of a broad range of materials. Typically, the method entails synthesis of the desired material within the pores or the walls of a preformed porous membrane or another solid. In this work, electrospun polycarbonate fibers were used as templates. The polymer was selected such that, following deposition of a metal oxide or other wall material, thermal treatment can be used to remove the original template polymer, resulting in the formation of a tube, as shown schematically in Scheme 1. The fiber (core) templates in Scheme 1 were prepared by electrospinning of preformed polymers from room temperature solutions in volatile organic solvents.22 The PC polymer was selected as the template polymer based on its favorable solubility in organic solvents and relatively low decomposition temperature. The average diameter of the fibers prepared was found to be 250 ( 100 nm. The morphology of the polymer fibers, as determined by SEM analysis of dry fiber mats (Figure 1) was smooth and uniform without beads. Varying the solvent type, concentration, and polymer molecular weights can be used to optimize and control the fiber diameter and morphology.22,24 Sol-Gel Deposition. The colloidal ZnO particles were softly deposited via dip coating leading to submicrometer coaxial fibers with better mechanical properties. Sol-gel technique offers the advantage of low cost, low processing temperatures, potential for high quality deposits over large and irregular substrates, and high process throughput over chemical vapor deposition technique.25 The procedure employed in this work was simpler than established methods, since the sols consisted of only three components: a zinc acetate precursor, a dopant salt, and the solvent. Zinc acetate is more soluble in highly polar

(24) Dong, H.; Macdiarmid, A. G.; Jones, W. E.Jr. J. Polym. Sci., Part B: Polym. Phys. 2004, 42, 3934.

(25) Xu, L.; Zhou, K.; Xu, H.; Zhang, H.; Huagig, L.; Liao, J.; Xu, A.; Gu, N.; Shen, H.; Liu, J. Appl. Surf. Sci. 2001, 183, 58.

7548 DOI: 10.1021/la802753k

Figure 1. SEM image of electrospun polycarbonate template fibers.

Results and Discussion

Langmuir 2009, 25(13), 7547–7552

Ochanda et al.

Article

Figure 2. SEM image of ZnAc-PC coaxial fibers.

Figure 4. SEM image ZnO nanotubes at high magnification.

Figure 3. SEM image of ZnO nanotubes at low magnification.

solvents such as DMF than in alcohols, so it was not necessary to heat the starting mixture. The relatively low vapor pressure of DMF prevented premature and uneven drying, which can cause cracking and disrupt uniaxial crystallization of ZnO. In addition, no additive was required to improve sol stability and homogeneity. Heating the mixture (which is necessary when dissolving in alcohols) increases the possibility of inadvertently altering the oxidation state of the dopants or forming dopant oxide secondary phases prior to ZnO deposition. To circumvent this problem, all sols were prepared at room temperature. Heating was performed only after the amorphous doped zinc acetate coating was formed. This could only be achieved when using a solvent that stabilized the sol and did not require heating to dissolve the precursor, the reason why this method was developed. Sol-gel deposition of zinc acetate was found to increase the mechanical (tensile) strength of electrospun polycarbonate fibers as determined from tensile stress and strain analysis. The modulus for the polycarbonate template was found to be ∼3.82 mPa, whereas the modulus for the ZnAc-PC coaxial fibers was found to be ∼18.90 mPa. This observed increase in modulus was attributed to hardening of ZnO gel. Morphology Characterization. The SEM image of ZnO precursor fibers shown in Figure 2 indicates a smooth morphology of the ZnAc-PC coaxial fibers with a larger diameter compared to polycarbonate template fibers (Figure 1). The SEM images of ZnO nanotubes are shown in Figures 3 and 4. The broken ends of the tubes in Figure 4 confirm the tubular morphology of these materials and with uniform size distribution of the nanotubes (Figure 3). The ZnO nanotubes were Langmuir 2009, 25(13), 7547–7552

Figure 5. TEM image of ZnO nanotubes.

Figure 6. EDS spectrum of Co-doped ZnO nanotubes.

found to have an average diameter of 450 nm with a wall thickness of ∼100 nm. TEM images also suggest that the tubes were hollow and not solid nanowires (Figure 5). The TEM results were complicated by the rapid degradation of the sample under the electron beam which limited the quality of the resulting images. Energy dispersive spectroscopy (EDS) profiling over the tube surfaces confirmed the uniformity of elemental composition of the Co-doped ZnO tubes with Zn and O peaks consistent with formation of zinc oxide after heat treatment (Figure 6). The presence of the palladium peak was attributed to the surface activation of the nanofibers in the palladium chloride bath. The cobalt peak was attributed to the addition of cobalt nitrate during the sol-gel process as a dopant to help reduce oxygen vacancy defects. DOI: 10.1021/la802753k

7549

Article

Ochanda et al.

Figure 7. FTIR spectrum of PC fibers before thesol-gel process.

Figure 10. Thermogravimetric analysis of PC-Zn acetate fibers.

Figure 8. FTIR spectrum of ZnAc-PC coaxial fibers before pyrolysis.

Figure 11. PRXD curve for wurtzite ZnO after calcination of ZnAc-PC at 650 °C.

Figure 9. FTIR spectrum of ZnO nanotubes after calcination.

FTIR spectra of the PC fibers before sol-gel deposition and ZnAc-PC coaxial fibers before calcination are shown in Figures 7 and 8, respectively. The absorption peaks of the template fibers were maintained in the coaxial fibers containing zinc acetate. The absorption peak at 1600 cm-1 from the FTIR spectrum of coaxial fibers in Figure 8 was due to chelating the zinc compound, since acetate is a well-known chelating ligand. This property was further confirmed by the broadening of peaks around 1300-1700 cm-1 in Figure 8 as compared to Figure 7, indicating interaction of acetate with PC functional groups. From the FTIR spectrum of ZnO nanotubes after calcination (Figure 9), the peak at 872 cm-1 corresponded to the optical mode due to electronic transitions in ZnO, while doublets at 500 and 436 cm-1 were assigned to νZn-O consistent with the formation on inorganic ZnO nanotubes.26 The strong bands observed between 500 and 3000 cm-1 in Figures 7 and 8 can be assigned to the bending and stretching frequencies of PC and acetate. After calcination, all these strong PC bands disappeared, as shown in Figure 9, and no sign of adsorbed water or hydroxyl, carbonate, or hydrocarbon impurities was observed. It was evident that all of the organic components were completely removed from PC-zinc acetate composite fibers after calcination at 650 °C. (26) Wahab, R.; Ansari, S. G.; Kim, Y-S.; Seo, H-K.; Shin, H-S. Appl. Surf. Sci. 2007, 253 (18), 7622.

7550 DOI: 10.1021/la802753k

Thermogravimetric analysis (TGA) of the PC/zinc acetate fibers was conducted to determine the decomposition temperature of the precursor and the calcination temperature for ZnO fibers and tubes as shown in Figure 10. The initial weight loss around 100 and 300 °C was attributed to the loss of solvent, the decomposition of zinc acetate, and the pyrolysis of PC by dehydration of the polymer side chain. From the TGA curve, most of the organic component belonged to PC and the CH3COOH group of zinc acetate which was removed at a temperature below 450 °C. There was no further weight loss after 450 °C, indicating the formation of inorganic ZnO. This was further confirmed by PXRD shown in Figure 11, corresponding to wurtzite crystalline phase with nine reflection peaks at 2θ = 31.42°(100), 34.08°(002), 35.90°(101), 47.20°(102), 56.24°(110), 62.52°(103), 66.20°(200), 67.60°(112), and 68.76°(201).27 The peak positions exhibited good agreement with the standard diffraction spectrum JCPDS card 36-1451, suggesting the formation of single phase ZnO solid solutions. The absence of the XRD peak shift on doping of ZnO with Co is reasonable given that the ionic radii of tetrahedrally coordinated Co2+ and Zn2+ are quite similar at 0.072 and 0.074 nm, respectively.27 Optical Properties of Zinc Oxide Nanotubes. The UVvisible absorption maximum for ZnO nanotubes occurred at 380 nm compared to that for macrocrystalline (bulk) ZnO which normally occurs around 362 nm. The red shift in the nanotube sample reflected a decrease in the band gap of the semiconductor, indicating the formation of nanoscale ZnO material with similar optical properties to those reported in literature for ZnO nanostructures, nanowires, (27) Hirano, T.; Kozuka, H. J. Mater. Sci. 2003, 38, 4203.

Langmuir 2009, 25(13), 7547–7552

Ochanda et al.

Article

Figure 12. (a) UV-visible spectrum of ZnO nanotubes and (b) plot of hν2/cm2 versus hν for determination of band gap of nanotubes.

and nanobelts.28-41 Figure 12a shows the UV-vis absorbance spectrum of zinc oxide nanotubes, and it was used to determine the band gap for these nanotubes (Figure 12b). A band gap of ∼3.22 eV was obtained which is slightly smaller than the wellknown band gap of 3.37 eV for macrocrystalline ZnO.10,11 Schwartz et al. predicted the energy for ligand metal to charge transfer (LMCT) transitions (estimated as the energy at which the charge transfer intensity equals that of the ligandfield maximum) in cobalt-doped ZnO to be ∼26 300 cm-1 (379.77 nm).42 This is in reasonable agreement with our experimental transition energy of ∼26 316 cm-1 (380 nm) and hence supports assignment of the observed band as a LMCT transition, although it does not preclude the presence of additional transitions in the same energy region such as the band gap transition that occurs at ∼25 975 cm-1.42 Nanocrystalline ZnO exhibits a characteristic luminescence spectrum with two maxima around 380 and 500 nm. The absorption at 380 nm has been attributed to the free exciton emission band, while the visible emission, 500 nm, was believed to stem from point defect levels associated with oxygen vacancies or zinc interstitial sites.5,43 The qualitative fluorescence of our fabricated ZnO nanotubes showed a broad and strong ultraviolet emission with a maximum at around 405 nm following excitation at 300 nm (Figure 13). The peak maximum can be attributed to band gap emission; the broadness of the emission with its long tail to lower energy indicates a broad nanotube size distribution, polycrystallinity of the nanotubes, and high level of impurities in the ZnO nanotubes. When compared with the (28) Yang, C. L.; Wang, J. N.; Ge, W. K.; Guo, L.; Yang, S. H.; Shen, D. Z. J. Appl. Phys. 2001, 90, 4489. (29) Koch, U.; Fojtik, A.; Weller, H.; Henglein, A. Chem. Phys. Lett. 1985, 122, 507. (30) Haase, M.; Weller, H.; Henglein, A. J. Phys. Chem. 1988, 92, 482. (31) Wong, E. M.; Hoertz, P. G.; Liang, C. J.; Shi, B. M.; Meyer, G. J.; Searson, P. C. Langmuir 2001, 17, 8362. (32) Keis, K.; Roos, A. Opt. Mater. 2002, 20, 35. (33) Borgohain, K.; Mahamuni, S. Semicond. Sci. Technol. 1998, 13, 1154. (34) Spanhel, L.; Andersen, A. J. Am. Chem. Soc. 1991, 113, 2826. (35) Meulenkamp, E. A. J. Phys. Chem. B 1998, 102, 5566. (36) Sakohara, S.; Ishida, M.; Anderson, M. A. J. Phys. Chem. B 1998, 102, 10169. (37) Rabani, J.; Behar, D. J. Phys. Chem. 1989, 93, 2559. (38) Kamat, P. V.; Patrick, B. J. Phys. Chem. 1992, 96, 6829. (39) Dijken van, A.; Meulenkamp, E. A.; Vanmaekelbergh, D.; Meijerink, A. J. Phys. Chem. B 2000, 104, 1715. (40) Kuo, T.-J.; Lin, C.-N.; Kuo, C.-L.; Huang, M. H. Chem. Mater. 2007, 19, 5143. (41) Sun, H.; Zhang, Q.-F.; Wu, J. L. Nanotechnology 2006, 17, 2271. (42) Schwartz, D. A.; Norberg, N. S.; Nguyen, P. Q.; Parker, J. M. Garnelin, D. R. J. Am. Chem. Soc. 2003, 125, 13205. (43) Bahnemann, D. W.; Kormann, C.; Hoffmann, M. R. J. Phys. Chem. 1987, 91, 3789.

Langmuir 2009, 25(13), 7547–7552

Figure 13. Emission spectrum of ZnO nanotubes.

UV emission peak (380 nm) from commercial ZnO powders detected by the same instrument, the peak shifted about 25 nm toward longer wavelength and broadened significantly.43 The broadening of the emission peak can be explained by the formation of band tailing in the band gap, which is often induced by the introduction of impurities, cobalt, into the semiconductor as in our experiment.44 The red shift of the UV emission peak was attributed to the narrowing of the band gap, Eg, which mainly appears in the photoluminescence spectra of ZnO powders heavily doped by donors, such as In.44,45 The UV emission can be attributed to the near band edge emission of the wide band gap zinc oxide.45 In the present case, the properties of the nanotubes are affected by the abundant surface states, although the size of the nanotubes, several tens of nanometers in diameter, was not expected to generate a strong carrier confinement effect which may widen the band gap. The UV emission from the band edge excitonic recombination should be easy to observe in the one-dimensional nanostructures due to the high surface state density.5 The green band emission, which is attributed to the presence of a singly ionized oxygen vacancy, was not observed in the ZnO nanotubes. This is indicative of virtually no oxygen vacancy in the ZnO nanotubes due to doping by cobalt. This observation can also be attributed to annealing of the nanotubes at 650 °C for 3 h in air after the sol-gel process which results in a decrease in the concentration of the oxygen vacancies. Emission lifetime measurements indicated that the nanotubes possessed a short-lived state in the visible region of the spectrum. Generally, the reported values had χ2 ∼ 1.5 and Durbin Watson (DW) = 1.54, and all values represent the (44) Jie, J. S.; Wang, G. Z.; Han, X. H.; Yu, Q. X.; Liao, Y.; Li, G. P.; Hou, J. G. Chem. Phys. Lett. 2004, 387, 466. (45) Kim, K. J.; Park, Y. R. Appl. Phys. Lett. 2001, 78, 475.

DOI: 10.1021/la802753k

7551

Article

Ochanda et al.

The lifetime measurements show that the radiative recombination of the excitons was slower than ZnO thin films with a lifetime of 200 ps.48 The luminescence lifetime was mainly determined by the concentration of defects, which trap the electrons and possibly holes and eventually cause their nonradiative recombination. Although the exact origin of the luminescence decay remains unclear at this stage, the measured lifetime for these nanotubes was consistent with high crystal quality achieved with the nanotube growth process since the lifetime was more similar to crystalline powders and needles than the thin films.46,47

Figure 14. Room temperature photoluminescence spectra of (a) Co-doped ZnO nanotubes and (b) glycogen in water to illustrate instrumental response at excitation wavelength of 300 nm.

average of at least four readings. The lifetime, τ, of the ZnO was determined to be 620 ps. The decay time constant probably represents the free exciton recombination at room temperature, which is similar to the reported lifetime (585 ps)46 of free excitons in single-crystal ZnO and that of ZnO nanoneedles (560 ps),47 and the luminescence decay data for the ZnO nanotubes is shown in Figure 14. :: :: :: (46) Teke, A.; Ozgur, U.; Dogan, S.; Gu, X.; Markoc, H.; Nemeth, B.; Nause, J.; Everitt, H. O. Phys. Rev. B 2004, 70, 195207. (47) Cao, B.; Cai, W.; Duan, G.; Li, Y.; Zhao, G.; Yu, D. Nanotechnology 2005, 16, 2567.

7552 DOI: 10.1021/la802753k

Conclusions ZnO nanotubes were fabricated from electrospun PC as template fibers and zinc acetate using a sol-gel process. The thermal degradation of the core PC template fibers yields polycrystalline wurtzite ZnO nanotubes. Emission measurements suggested that ZnO nanotubes formed were low in point defects, and exhibited a red shift in the UV emission peak which was attributed to the narrowing of the band gap, Eg, due to Co doping. Tuning the electric band structure of ZnO nanostructures by doping will be important in future applications in electronic, optical, sensor, and optoelectronic devices. Future work will aim at controlling size distribution, shape, crystal structure, and defect distribution and understanding their growth mechanisms. Acknowledgment. We thank Henry Eichelberger of the Biological Science Department for assistance with SEM collection, Debbie Dittrich for EDS collection, Dr. Jie Xiao for PXRD, and Dr. Justin Martin for useful discussion. (48) Cao, H. Phys. Rev. Lett. 2000, 84, 5584.

Langmuir 2009, 25(13), 7547–7552