Zr Doping on One-Dimensional Titania Nanomaterials Synthesized in

Nov 23, 2010 - Rahima A. Lucky, Yaocihuatl Medina-Gonzalez, and Paul A. Charpentier*. Department of Chemical and Biochemical Engineering University of...
0 downloads 0 Views 3MB Size
pubs.acs.org/Langmuir © 2010 American Chemical Society

Zr Doping on One-Dimensional Titania Nanomaterials Synthesized in Supercritical Carbon Dioxide Rahima A. Lucky, Yaocihuatl Medina-Gonzalez, and Paul A. Charpentier* Department of Chemical and Biochemical Engineering University of Western Ontario London, Ontario, Canada N6A 5B9 Received May 20, 2010. Revised Manuscript Received November 6, 2010 The growth mechanism of one-dimensional metal oxide nanotubular structures is of tremendous current interest to tailor materials using “green” synthetic procedures for emerging industries in alternative energy and biomaterials. In this study, ZrO2-modified TiO2 nanorods and tubular structures were successfully synthesized via a surfactant-free sol-gel route using supercritical carbon dioxide (scCO2) as the solvent/drying agent. The effect of metal alkoxide concentration (0.35-1.4 mol/L), acid/metal alkoxide ratio (R = 3-7), and Zr ratio (0-20%) was examined on the morphology and crystallinity of the resulting nanostructures as measured by electron microscopy (SEM and TEM), EDX, XPS, and XRD. The electron microscopy results showed that the crystal growth of the synthesized binary Ti-Zr nanomaterials could be tailored by changing the operating variables with nanotubular structure formed at metal alkoxide concentration of 1.2 mol/L, R = 5-6, and Zr ratio between 4% and 20%. Gelation kinetics for this new system was also studied and revealed that increasing alkoxide concentration and R value enhanced the gelation kinetics. In situ and powder FTIR results revealed that this Ti-Zr binary system follows a similar reaction scheme to that of either singlecomponent system, showing the flexibility of this approach for tailoring nanotubular production.

1. Introduction Recently, considerable effort has been devoted to synthesizing inorganic one-dimensional (1D) nanostructures (e.g., nanotubes, nanowires) because of their unique physical properties compared to their bulk counterparts.1,2 Many exciting potential applications are possible including photocatalysis, high-efficiency solar cells, self-cleaning and intelligent coatings, and biosensors.3,4 Among the 1D oxide nanomaterials reported, titania (TiO2) nanomaterials are receiving considerable attention due to titania’s high activity, strong oxidation capability, and chemical stability.5 Modifying TiO2 with ZrO2, one of the most suitable dopants,6,7 also enhances its performance by increasing thermal stability and surface area and reducing the crystallite size, as the performance of titania nanomaterials strongly relies on these properties.8 ZrO2 modified TiO2 has shown excellent catalytic properties in various catalytic processes,9 photocatalysis,10 energy conversion electrodes,11 and biomaterials for bone cement.12 Several different synthetic techniques have been used for the preparation of TiO2-based nanostructures, such as anodization, *Author to whom correspondence should be addressed. E-mail: [email protected]. Phone:(519) 661-3466. Fax:(519) 661-3498. (1) Hochbaum, A. I.; Yang, P. Chem. Rev. 2010, 110, 527. (2) Shen, G.; Chen, D. Sci. Adv. Mater. 2009, 1, 213. (3) Chen, X.; Mao, S. S. Chem. Rev. 2007, 107, 2891. (4) Kuchibhatla, S. V. N. T.; Karakoti, A. S.; Bera, D.; Seal, S. Prog. Mater. Sci. 2007, 52, 699. (5) Yao, J.; Takahashi, M.; Yoko, T. Thin Solid Films 2009, 517, 6479. (6) Marıa, D.; Hernandez-Alonso, I. T.-T.; Coronado, J. M.; Soria, J.; Anderson, M. A. Thin Solid Films 2006, 502, 125. (7) D€urr, M.; Rosselli, S.; Yasuda, A.; Nelles, G. J. Phys. Chem. B 2006, 110, 26507. (8) Pal, M.; Garcı´ a Serrano, J.; Santiago, P.; Pal, U. J. Phys. Chem. C 2006, 111, 96. (9) Sohn, J. R.; Lee, S. H. Appl. Catal., A 2007, 321, 27. (10) Lucky, R. A.; Charpentier, P. A. Appl. Catal., B 2010, 96, 516. (11) Kitiyanan, S. S. A. Compos. Sci. Technol. 2006, 66, 1259. (12) Khaled, S. M.; Sui, R.; Charpentier, P. A.; Rizkalla, A. S. Langmuir 2007, 23, 3988. (13) West, R. H.; Beran, G. J. O.; Green, W. H.; Kraft, M. J. Phys. Chem. A 2007, 111, 3560.

19014 DOI: 10.1021/la102048j

template techniques, hydrothermal processes, and soft chemical processes.2,13 However, each of these methods has limitations for potential scale-up purposes. Recently, the sol-gel process has become an attractive scaleable synthetic route to produce highquality, homogeneous metal oxide materials with desired nanostructure and the ability for doping at low cost.14,15 A significant limitation of the conventional sol-gel synthetic process is the ability to control the reaction kinetics and nanostructure, causing several metal alkoxides (e.g., Ti, Zr, and Al) to precipitate. Supercritical fluids provide a tremendous tool in overcoming these conventional limitations, providing higher solvating power and increased supersaturation levels, leading to smaller particles and higher reaction and particle growth kinetics.16 ScCO2 has several advantages for synthesizing advanced materials, because as a solvent, it is inexpensive, environmentally benign, and nonflammable, with mild critical conditions (Pc = 73.8 bar; Tc = 31.1 °C) making it suitable for both laboratory and commercial-scale applications.17 Moreover, low viscosity, “zero” surface tension, and high diffusivity of scCO2 are also favorable properties to synthesize superior ultrafine and uniform nanomaterials, while potentially simultaneously growing attached polymer chains.18 ScCO2 can be easily and completely removed from products by venting; hence, no drying process is required and the porous structure can be maintained without collapsing the nanostructure. Furthermore, scCO2 can be easily recycled after the pressure is reduced into the liquid regime for potential scale-up applications. We recently communicated the synthesis of ZrO2 modified TiO2 nanotubes using an acid-modified sol-gel process in (14) Fernandez-Garcia, M.; Martinez-Arias, A.; Hanson, J. C.; Rodriguez, J. A. Chem. Rev. 2004, 104, 4063. (15) Mackenzie, J. D.; Bescher, E. P. Acc. Chem. Res. 2007, 40, 810. (16) Aymonier, C.; Loppinet-Serani, A.; Reverchon, H.; Garrabos, Y.; Cansell, F. J. Supercrit. Fluids 2006, 38, 242. (17) Sui, R.; Rizkalla, A. S.; Charpentier, P. A. Langmuir 2005, 21, 6150. (18) Charpentier, P. A.; Xu, W. Z.; Li, X. Green Chem. 2007, 9, 768.

Published on Web 11/23/2010

Langmuir 2010, 26(24), 19014–19021

Lucky et al.

Article Table 1. Synthesis Conditions of ZrO2-TiO2 Nanostructures Synthesized in scCO2 and Characterization Results

experiment series

Coa (mol/L)

Zr %

Co effect

Rb (mol/mol)

0.35 10 0.7 10 1.2 10 1.4 10 1.2 10 R effect 1.2 10 1.2 10 1.2 10 Zr effect 1.2 0 1.2 4 1.2 6 1.2 8 1.2 10 1.2 20 a Initial concentration of metal alkoxide. b R = molar ratio of acetic acid/metal pressure, 5000 psig.

scCO2,19 and further examined the effect of solvent, temperature, and pressure on nanostructure formation.20 This work investigates the effects of the main experimental doping variables: metal alkoxide concentration (0.35-1.4 mol/L), acid/metal alkoxide ratio (3-7), and Zr ratio (0-20%) to gain a fundamental understanding on how they affect the morphologies and microstructures of these nanotubular materials to allow for tailoring the structures for the intended function. The resulting nanomaterials were characterized by electron microscopy (SEM and TEM), EDX, XRD, and XPS. Surfaces obtained by sintering the materials on glass plates were characterized by AFM. The initial gelation kinetics was studied in terms of gelation time, and the reaction mechanism of crystal growth was elucidated using in situ FTIR.

2. Experimental Section 2.1. Materials. Reagent-grade titanium(IV) isopropoxide (TIP, 97%, Aldrich), zirconium(IV) propoxide (ZPO, 70%, Aldrich), acetic acid (99.7%, Aldrich), and instrument-grade carbon dioxide (99.99%, BOC) were used without further purification. 2.2. Synthesis. The experimental setup previously reported was used for this study.21 In a typical experiment, 3 mL (9.9 mmol) titanium isopropoxide, 0.48 mL (1.1 mmol) zirconium(IV) propoxide, 3.16 mL (55 mmol) acetic acid, and CO2/3 mL solvent were quickly placed in a 10 mL view cell under stirring, then heated and pressurized with supercritical carbon dioxide (scCO2) to the desired temperature and pressure. Initially, a transparent homogeneous phase was observed. After the reaction mixture was stirred from about 30 min to several hours, the fluids in the view cell became semitransparent, then turned white, indicating a phase change. The gel starting time was considered the gelation point, which was measured at least twice in independent experiments with the average reported. To ensure complete condensation of the precursor, a few drops of the reaction mixture were vented into water, where a white precipitate indicated that further reaction time was required. After 5 days aging, the formed gel was washed continuously using scCO2 (5000 psig, 60 °C) at a controlled flow rate of 0.5 mL/min, followed by controlled venting. The resulting as-prepared materials were subsequently collected and characterized. To achieve crystalline phase, the synthesized materials were calcined in air at 500 °C using a heating rate of 10 °C/min; the holding time was 2 h, and the cooling rate to room temperature was 0.5 °C/min. (19) Lucky, R.; Charpentier, P. Adv. Mater. 2008, 20, 1755. (20) Lucky, R. A.; Sui, R.; Lo, J. M. H.; Charpentier, P. A. Cryst. Growth Des. 2010, 10, 1598. (21) Sui, R.; Rizkalla, A. S.; Charpentier, P. A. J. Phys. Chem. B 2004, 108, 11886.

Langmuir 2010, 26(24), 19014–19021

tgc (min)

morphology

5 180 mixed 5 110 sheets 5 40 ( 5 tubes 5 50 sheet/rod 3 no gel no structure 4 80 ( 10 sheet 6 35 ( 5 tubes 7 precipitate no nanostructure 5 fibers 5 tubes/sheets 5 tubes/sheets 5 tubes/sheets 5 tubes/sheets 5 tubes alkoxide. c Gelation time. Reaction temperature, 60 °C; reaction

2.3. Characterization. The size and morphology of the synthesized nanostructures were analyzed by SEM on a Hitachi S-4500 FE instrument using an accelerating voltage of 5 kV. Transmission electron microscopy (TEM) images were obtained using a Philips CM10. The specimens were dispersed into methanol and placed on a copper grid covered with holey carbon film. Bulk composition was determined using energy-dispersive X-ray spectroscopy (EDX) attached to a LEO 1530 scanning electron microscope (SEM). X-ray diffraction (XRD) was performed utilizing Rigaku employing Cu KR1þKR2=1.541 84 A˚ radiation with a power of 40 kV/35 mA for the crystalline analysis. The broad-scan analysis was typically conducted within the 2θ range 10-80°. The surface composition of selected samples was determined by XPS, using a Kratos Axis Ultra spectrometer using a monochromatic Al KR source (15 mA, 14 kV). Survey and high-resolution spectra were obtained using an analysis area of ∼300  700 μm2 and pass energies of 160 and 20 eV, respectively. Spectra were charge-corrected to the main line of the carbon 1s spectrum (C-C, C-H) set to 285.0 eV. AFM was carried out on surfaces prepared with commercial TiO2 (nanopowder, from SigmaAldrich), undoped TiO2, and ZrO2-TiO2 (R = 4) by using a Veeco Multimode V system equipped with a Nanoscope V controller, at scan rates of 0.5 Hz in tapping mode. These surfaces were prepared by spreading a diluted aqueous solution of the materials onto a glass plate and sintering at 400 °C for 20 min. IR spectra were recorded on a Bruker IFS 55 spectrometer in the range 500-4000 cm-1. Each spectrum was recorded at 4 cm-1 resolution with 500 scans. Sample pellets were obtained from calcined powder and mixed with a small amount of KBr, then pressed into a pellet, and subsequently analyzed in transmission mode. In situ FTIR monitoring of the sol-gel reaction in scCO2 was performed using a high-pressure diamond immersion probe (Sentinel-ASI Applied Systems) attached to a stirred 100 mL autoclave (Parr Instruments). The probe is attached to an attenuated total reflection Fourier transmission infrared (ATRFTIR) spectrometer (ASI Applied System ReactIR 4000), connected to a computer, supported by ReactIR software (ASI). The setup was described in detail previously.21 The desired amounts of TIP, ZPO, and acetic acid were quickly placed in the autoclave, followed by pumping in CO2 and heating to the desired temperature and pressure under stirring at 600 rpm. The spectra were collected throughout the reaction at specified intervals.

3. Results 3.1. Effects of Alkoxide Concentration and Acid/Metal Alkoxide Ratio (R). Table 1 shows the experimental conditions and effects of metal alkoxide (MA) concentration, and acid/MA ratio on the morphology of the resulting bimetallic materials synthesized in supercritical carbon dioxide (scCO2). DOI: 10.1021/la102048j

19015

Article

Lucky et al.

Figure 1. Effect of reaction conditions on morphology of ZrO2-TiO2 nanomaterials synthesized in scCO2. (a) C-0.35 mol/L, (b) C-0.7 mol/ L, and (c) C-1.17 mol/L; reproduced with permission from ref 19 (copyright 2008 Wiley); (d) R-3, (e) R-4, and (f) R-6 (bar represents 500 nm). (g-i) R-4 and 10 days of aging time (bar represents 1 μm). Reaction temperature, 60 °C; reaction pressure, 5000 psig.

The SEM micrographs are given in Figure 1, which shows that both doping variables had a significant influence on the morphology of the synthesized nanomaterials. For low concentrations of metal alkoxide (0.35 mol/L), Figure 1a shows that sheet, rod, and cubic structured materials with different dimensions were observed. Nanosheet-type structures with 150-375 nm width were observed for 0.7 mol/L concentration of alkoxides (Figure 1b), whereas 50-135-nm-diameter nanotubes were formed at 1.14 mol/L concentrated solutions (Figure 1c). With a further increase in concentration to 1.4 mol/L, the morphology changed to a mixed sheet and rod-type structure (not shown). Hence, nanosheets and nanotubes were only formed in the middle concentration region. Note that these examined materials were taken directly from the reaction view cell after CO2 washing/drying without any further purification or calcination steps, to provide for a better understanding of the reaction conditions on product morphology. 19016 DOI: 10.1021/la102048j

In order to investigate the effects of the acid-to-metal alkoxide molar ratio (R) on the morphology, surface area, and pore volume, the 10% ZrO2 modified TiO2 nanotubes using R values ranging from 3 to 7 were synthesized in scCO2 (Table 1). At R =3, a pillar-type structure with 2 μm width and several micrometers in length was formed, and no nanostructure was observed (Figure 1d). When the R value was increased to 4, gel was formed and a thick sheet-type structure with ca. 200 nm width and 12 μm length was observed (Figure 1e). As previously observed at R = 5 and 6 (Figure 1c,f), nanotubes with 50-135 nm diameter and 40-100 nm diameter were formed, respectively. Further increase of R to 7 led to precipitation with no visible nanostructure. Hence, the nanotubes were only formed in the narrow experimental window at R = 5-6. To better observe the effect of aging time on the morphology, an additional experiment at R = 4 and 10 days of aging time was carried out with the SEM micrographs as shown in Figure 1g-i. As can be observed, after Langmuir 2010, 26(24), 19014–19021

Lucky et al.

Article

Figure 2. AFM images and 3D reconstructions of the surfaces obtained by sintering (a) commercial TiO2, (b) undoped TiO2, and (c) ZrO2-TiO2; R = 4 onto glass plates.

10 days, sheets are very well developed (Figure 1g), with some of them beginning to curl (Figure 1h,i) suggesting that nanotubes appear after nanosheet wrapping. AFM images of the surfaces of the materials sintered onto glass plates show similar results after isolating one nanosheet obtained with R = 4 in Figure 2. An AFM image of commercial TiO2 is shown for comparison purposes. As shown in Table 1, the increase in initial concentration of the metal alkoxide decreased the gelation time from 180 min at 0.35 mol/L concentration to 50 min at 1.4 mol/L. Similarly, increasing the acid/alkoxide ratio (R) decreased the gelation time. At the lowest level studied of R = 3, the system started gelation by changing color at about 180 min. However, after five days it did not form any gel; hence, no nanostructure was observed. When R was increased to 4, the gelation time was reduced to 80 min. With the acetic acid ratio increased beyond 4, the reaction proceeded much more rapidly, i.e., the gelation time for R5 and R6 was ca. 35 min. A further increase of R was found to result in quick precipitation of the system. 3.2. Effects of Zr Concentration. To investigate the effects of % Zr composition on the binary nanostructure, a series of Langmuir 2010, 26(24), 19014–19021

0-20% ZrO2 modified TiO2 nanomaterials were synthesized in scCO2 at 5000 psig and 60 °C at R = 5 (Table 1). The effects of Zr composition on the microstructure and morphology of the ZrO2-TiO2 nanomaterials were characterized by SEM analysis as shown in Figure 3 and AFM as shown in Figure 2b/c. From Figure 3a, it can be seen that pure TiO2 synthesized in scCO2 with TIP alkoxide and acetic acid formed nanofiberous structures having diameters 20-50 nm with lengths of up to several micrometers. These results are consistent with Sui et al.17 who reported that 10-40 nm TiO2 nanofibers were produced using TIP in scCO2 following a similar procedure. At 4% Zr, the structure is still mainly fiber morphology (Figure 3b) with similar diameters and lengths. The SEM images Figure 3c,d show that nanotubes form at 8% and 10% composition with 8% Zr composition giving narrower (40-100 nm) structure compared to the 10% Zr containing tubes having a diameter of 50-140 nm and a length of 1-4 μm. 3.3. Microstructures. EDX mapping of the ZrO2-TiO2 samples (Supporting Information Figure SI.1) reveals that the zirconium and titanium elements were quantitatively and homogenously dispersed throughout the samples for all studied DOI: 10.1021/la102048j

19017

Article

Lucky et al.

Figure 3. SEM images of Zr-Ti binary calcined nanomaterials with different zirconium compositions: (a) undoped TiO2, (b) 4%, (c) 8%, and (d) 10%. (Bar represents 200 nm. All the samples were examined after platinum coating. Arrows indicate tubes.) Table 2. Results of Zr-Ti Binary Metal Oxide Nanostructures in scCO2a lattice parameter (A˚) sample number

nominal Zr mol %

b

product Zr mol %

crys. (nm)

a

C

Zr-TiO2-0 0 0 14.0 3.787 0.0425 0.04 13.7 3.787 Zr-TiO2-4% 0.063 0.0625 12.6 3.787 Zr-TiO2-6% 0.087 0.084 ( 0.02 12.5 3.788 Zr-TiO2-8% 0.11 0.1066 ( 0.1 12.2 3.798 Zr-TiO2-10% 0.25 0.19 ( 0.1 Zr-TiO2-20% a Experimental conditions: Pressure, 5000 psig; temperature, 60 οC; concentation, 0.7 and 1.17 mol/L; acid to metal composition was measured by EDX.

compositions. As provided in Table 2, the bulk compositions for all samples show good agreement between the targeted and actual values of the metals composition. The effect of the Zr composition on the crystal size and phase structure was investigated by performing XRD analysis. The XRD patterns for all the samples calcined at 500 °C (except for the 20% Zr sample, which was amorphous) are given in Figure 4, indicating that the ZrO2-TiO2 binary nanomaterials consists of anatase crystal. There is no distinct zirconia peak, indicating no phase separation and that the Zr is well-integrated into the anatase crystal structure for all investigated compositions. However, pure TiO2 contains a very small rutile phase (4%) at this temperature. With increasing % Zr composition, the main (101) anatase peak shifts to lower 2θ values, resulting in an increase in the d-spacing. The sizes of the crystallites and phase and lattice parameters were determined from the XRD peak bonding as provided in Table 2. The crystallite sizes decrease with increasing Zr composition, i.e., for pure TiO2, crystallites were 14 nm whereas 10% Zr content crystallites were 12.2 nm. The Zr ion can either go into interstitial positions or substitute for Ti4þ at lattice points.22 Any substitution or insertion of a Zirconium ion (22) Shi, Z. M.; Yan, L.; Jin, L. N.; Lu, X. M.; Zhao, G. J. Non-Cryst. Solids 2007, 353, 2171.

19018 DOI: 10.1021/la102048j

cell V (A˚)3

9.514 9.531 9.541 9.571 9.6 alkoxide ratio, 5.

136.51 136.74 136.98 137.35 138.47 b Product

for a titanium ion in the TiO2 lattice would introduce a distortion and change the cell parameters. The TiO2 anatase unit crystal is tetragonal, with lattice parameters a and C. Therefore, these parameters were calculated using the peak values of the anatase (101) and (200) reflections, which show that these parameters increase with dopant addition (change of a is insignificant); consequently, the cell volume increases.23 The cell volume of pure TiO2 was 136.57 A˚, whereas the 10% Zr containing sample showed 138.47 A˚. These results are consistent with the previously reported cell volume for Ti-Zr binary solid solution24 and crystallite trend with increasing Zr.25 Moreover, this result is expected because the effective ionic radii of Ti4þ and Zr4þ are 0.68 and 0.79 A˚, respectively, and any substitution would increase the cell volume.26 XPS was carried out on the 0%, 10%, and 20% samples as also reported with details provided in the Supporting Information (Table SI.1). The reported binding energies are 458.8 and 183 eV for pure Ti 2p3/2 and pure Zr 3d5/2, respectively.27 The binding (23) Neppolian, B.; Wang, Q.; Yamashita, H.; Choi, H. Appl. Catal., A 2007, 333, 264. (24) Yu, J. C.; Lin, J.; Kwok, R. W. M. J. Phys. Chem. B 1998, 102, 5094. (25) Hernandez-Alonso, M. D.; Coronado, J. M.; Bachiller-Baeza, B.; FernandezGarcia, M.; Soria, J. Chem. Mater. 2007, 19, 4283. (26) Chang, S.; Doong, R. J. Phys. Chem. B 2006, 110, 20808.

Langmuir 2010, 26(24), 19014–19021

Lucky et al.

Article

Figure 5. In situ FTIR spectra for 1:1 mol ratio TIP and ZPO with acetic acid at R = 5 (acid to metal alkoxide ratio) in scCO2 from 0 to 1200 min. Figure 4. XRD patterns for Zr-Ti binary metal oxide nanomaterials with various % ZrO2 compositions.

energy for pure nanoTiO2 synthesized in this study was 458.8 eV (based on TiO2), while when introducing Zr into the TiO2 matrix, the binding energies for both Ti 2p3/2 and Zr 3d5/2 shifted toward lower energy. For example, at 10% Zr composition the binding energies of Ti 2p3/2 and Zr 3d5/2 were 458.7 and 182.4 eV, respectively, while at 20% Zr, they became 458.4 and 182.2. This binding energy shift might be attributed to a change in the coordination number of the metal by the forming Zr-O-Ti bonds. The presence of the second metal will change the electronic environment, as well as the binding energy of the system with similar results being observed by Reddy et al.28 and shown in our previous work using density functional theory modeling.19 The XPS analysis also revealed that the surface carbon content decreased with increasing amount of % Zr from pure TiO2 contained 43.2 at %, whereas 38.5 and 35 at % carbon are present in the binary 10% and 20% Zr containing samples. Comparison of the EDX and XPS results reveals that the concentration of zirconia is higher on the surface compared to the bulk composition. This surface enrichment phenonomenon is normally observed for this binary system, as reported by Galindo et al.27 This result is attributed to the fact that XPS considers only a few nanometers from the top of the surface, and zirconium(IV) propoxide is more reactive than titanium(IV) isopropoxide.29 In situ ATR-FTIR was used to study the sol-gel process under actual reaction conditions in scCO2, as FTIR is a well-established technique for analyzing metal carboxylate species.30 Figure 5 shows in situ spectra for 1:1 mol TIP and ZPO reacting with acetic acid at R = 5 in scCO2 at 4500 psig from 0 to 1200 min. The first time splice spectra show a strong peak at 1710 cm-1, which drops with time, indicating rapid consumption of acetic acid. At a reaction time of 10 min, still at the initial stage of the polycondensation reaction, the presence of peaks at 1557 and 1447 cm-1 provides evidence for the formation of metal-acetate complexes. These metal-acetate complexes are the hexamer “building blocks” for nanostructure evolution as discussed further below. After 20 min reaction time, almost all acetic acid was consumed. However, the metal-acetate peaks at 1557 and 1447 cm-1 reach saturation after 30 min with no further intensity changes. Some shifting of these (27) Galindo, I. R.; Viveros, T.; Chadwick, D. Ind. Eng. Chem. Res. 2007, 46, 1138. (28) Reddy, B. M. C., B.; Smirniotis, P. G. Appl. Catal., A 2001, 211, 19. (29) Cao, G. Nanostructures & nanomaterials: synthesis, properties & applications; Imperial College Press: London, 2004. (30) Sanchez, C.; Livage, J.; Henry, M.; Babonneau, F. J. Non-Cryst. Solids 1988, 100, 65.

Langmuir 2010, 26(24), 19014–19021

peaks in the 1455-1600 cm-1 range is observed, indicating a change in the OCO bond angle and length during the polycondensation process. To help confirm the structural analysis provided by the in situ results, the powder ATR-FTIR spectra of samples prepared in scCO2 are compared in Figure 6: (a) as-prepared TiO2 nanofiber (SEM shown in Figure 3a), (b) typical 10% ZrO2-90% TiO2 binary metal oxide, and (c) Ti4O4Zr2(C2H3O2)10(iC3H7O)6 single crystal sample.31 The spectra for both (a and b) samples are rather similar, indicating that single and binary metal oxides are similar in IR observable functional groups. Considering the peak at 3400 cm-1, which is assigned to the -OH group of absorbed water,32 the binary sample peak is broad and intense compared to the pure TiO2 sample, indicating the presence of more -OH groups. The peak at around 1550 and 1445 cm-1 for the pure and bimetallic samples is due to symmetric and asymmetric stretching of the zirconium titanium acetate complex, respectively.33 The frequency difference between these peaks (Δ) is ca. 100 cm-1, confirming that acetic acid formed bridging complexes with the metal ions. A slight Δ change is noticed of 106 cm-1 for the pure TiO2 sample (Figure 6a) compared to 96 cm-1 for the binary sample (Figure 6b). There is also a noticeable difference observed in the spectra for the two systems around 1410 cm-1, where a sharp peak is present at 1409 cm-1 in pure titania while a shoulder at 1415 cm-1 is visible for the binary system. The -CH3 group contributes to the small peak at 1343 cm-1 which is present for both samples. There are also two small peaks at 1037 and 1024 cm-1 corresponding to the ending and bridging -OPr groups, respectively,34 indicating that unhydrolyzed -OPr groups were present in these as-prepared materials. The oxo bonds can be observed by the bands at 657 cm-1.33 The spectrum of the single crystal, Ti4O4Zr2(C2H3O2)10(iC3H7O)6, previously isolated in scCO2,31 is presented in Figure 6c. This spectrum is very similar to the other two spectra except that there is no peak corresponding to the OH group. It is obvious that the Ti-Zr single crystal is saturated with many bridging and ending -OPr groups, and it also contains 10 acetate ligands.31 It is important to note that any peak corresponding to CO2 was not detected, indicating that no CO2 was present in the samples. (31) Lucky, R. A.; Ruohong, S.; Charpentier, P. A.; Jennings, M. C. Acta Crystallogr., Sect. E 2007, E63, m2429. (32) Zhu, J.; Jinlong, Z.; Chen, F.; Iino, K.; Anpo, M. Top. Catal. 2005, 35, 261. (33) Nakamoto, K. Infrared and Raman spectra of inorganic and coordination compounds, 4th ed.; Wiley: Toronto, 1986. (34) Sui, R.; Rizkalla, A. S.; Charpentier, P. A. J. Phys. Chem. B 2006, 110, 16212.

DOI: 10.1021/la102048j

19019

Article

Lucky et al.

Figure 6. Powder ATR-FTIR spectra of as-prepared ZrO2-TiO2 binary nanomaterials (a) pure TiO2, (b) 10% ZrO2, and (c) Zr-Ti binary single crystal.

4. Discussion on Nanotubular Growth When examining the effect of increasing alkoxide concentration, acid/metal alkoxide ratio, and Zr ratio composition, the SEM images in Figures 1 and 3 reveal that nanotubes with 40-150 nm in diameter were formed from the ZrO2-TiO2 binary samples in a relatively narrow experimental window. Titanium and zirconium alkoxides reacting with carboxylic acid have been studied previously in conventional solvents with the products characterized by single crystal XRD, although nanotubular products were not observed.35 For any crystal growth process, the growth and nucleation rates are a function of the solute concentration.36 According to Yasue et al., smaller crystals are formed at higher initial degrees of supersaturation, while larger crystals are obtained with lower supersaturation.37 Kim et al. also proposed that the initial concentration is a very important factor to obtain monodispersed titania.38 Birnie also observed similar effects for their titanium-acetic acid system.39 Table 1 shows that increasing concentration decreased the gelation times. It is known that, for solutions having a higher initial concentration, the gelation rate is faster.40 In the examined experimental system, the solution is more concentrated, so more monomers can participate in the polycondensation reaction. Wu and Bavykin41,42 have proposed that a TiO2 nanotube begins with the formation of a TiO2 nanosheet, which curves leading to nanotubular structure.43,44 In the case of ZrO2 modified TiO2 of this study, a high R ratio was found to increase the rate of formation of nanostructure, giving TiO2 nanotubes in relatively shorter time span. A lower acid concentration would decrease this rate, allowing the intermediate nanosheet stage to be reached. These ideas are supported by our observations in the case of low acid/metal alkoxide ratio (R = 4) and long time of reaction (10 days aging) (Figure 1g-i). Well-developed nanosheets were found and some curved nanosheets started to appear suggesting the beginning of nanotube formation. After 10 days of reaction,

small structures that were not well-defined appeared together with the well-defined curved sheets. The Ostwald ripening phenomena can explain this behavior where larger structures are formed at the expense of small particles to decrease the energy of the entire system.45 The dimensions of the nanotubes changed significantly with increasing alkoxide concentration, acid/metal alkoxide ratio, and Zr ratio composition, which may be attributed to the formation of different Zr-Ti hexamer building blocks. It is well-known that hexamer crystals are formed rather rapidly from reactions of metal alkoxides and acetic acid.43 The binary metal hexamer crystal structure strongly depends on the ratio of the metal alkoxide used for the synthesis process.44 As shown in Figures 5 and 6, the in situ and powder FTIR analysis confirms that acetic acid formed metal acetate complexes with metal alkoxide precursors, which were consumed very quickly (within 10 min) and are consistent with previously reported observations.17,46 These hexamer crystals subsequently react and self-assemble, forming the observed morphologies of nanotubes, nanorods, and so forth. According to Doeuff et al., the formation of the hexamer complex can be explained through modification, esterification, hydrolysis, and condensation steps.47 Sui et al. proposed similar reaction pathways for TiO2 and ZrO2 nanomaterial formation in scCO2 using alkoxides and acetic acid.17,46 Following these works and the provided experimental results, it can be suggested that the steps above are quite general in the sol-gel chemistry. Hence, it is possible to propose a reaction scheme for the binary system based on the reaction scheme for titanium and zirconium alkoxides with acetic acid in simplified form. Modification mZr-ðOHÞ4 þ nTi-ðORÞ4 þ hHOAc f Zrm ðOAcÞm ðORÞ4 - m þ Tin ðOAcÞn ðORÞ4 - n þ ðm þ nÞROH þ ðh - m - nÞHOAc ð1Þ Esterification HOAc þ ROH f ROAc þ H2 O Hydrolysis Zrm ðOAcÞm ðORÞ4 - m þ xH2 O f Zrm ðOAcÞm ðORÞ4 - m - x ðOHÞx þ xROH

19020 DOI: 10.1021/la102048j

ð3Þ

Tin ðOAcÞn ðORÞ4 - n þ yH2 O f Tin ðOAcÞn ðORÞ4 - n - y ðOHÞy þ yROH

ð4Þ

Oxolation Zrm ðOAcÞm ðORÞ4 - m - x ðOHÞx f Zrm Ox ðOAcÞm ðORÞ4 - m - x0 þ xROH ð5Þ Tin ðOAcÞn ðORÞ4 - n - y ðOHÞy f Tin Oy ðOAcÞy ðORÞ4 - n - y0 þ yROH

(35) Kickelbick, G. F.; Martin, P.; Bertagnolli, H.; Puchberger, M.; Holzinger, D.; Gross, S. J. Chem. Soc., Dalton Trans. 2002, 20, 3892. (36) Pierre, A. C. Introduction to sol-gel processing; Kluwer Academic Publishers: Boston, 1998. (37) Yasue, T.; Yosiaki, T.; Arai, Y. Gypsum Lime 1984, 189, 83. (38) Kim, D. K. Mater. Lett. 2003, 57, 3211. (39) Birnie, I. J. Mater. Sci. 2000, 35, 367. (40) Doeuff, S.; Henry, M.; Sanchez, C.; Livage, J. J. Non-Cryst. Solids 1987, 89, 206. (41) Wu, D.; Ji, L.; Zhao, X.; Li, A.; Chen, Y.; Ming, N. Chem. Mater. 2006, 18, 547. (42) Bavykin, D. V. P.; Valentin, N.; Lapkin, A. A.; Walsh, F. C. J. Mater. Chem. 2004, 14, 3370. (43) Sui, R.; Rizkalla, A. S.; Charpentier, P. A. Cryst. Growth Des. 2008, 8, 3024. (44) Jupa, M.; Guido, K.; Schubert, U. Eur. J. Inorg. Chem. 2004, 9, 1835.

ð2Þ

ð6Þ

Condensation Zrm Ox ðOAcÞm ðORÞ4 - m - x0 þ Tin Oy ðOAcÞn ðORÞ4 - n - y f macromolecules Ti-O-Ti, Ti-O-Zr, Zr-O-Zr

ð7Þ

The sol-gel chemistry in scCO2 is complex, involving both reaction and self-assembly steps, but provides a relatively inexpensive (45) Huang, Y.; Liao, F.; Zheng, W.; Liu, X.; Wu, X.; Hong, X.; Tsang, S. C. Langmuir 2010, 26, 3106. (46) Sui, R. R.; Amin, S.; Charpentier, P. A. Langmuir 2006, 22, 4390. (47) Doeuff, S. D.; Taulelle, F.; Sanchez, C. Inorg. Chem. 1989, 28, 4439.

Langmuir 2010, 26(24), 19014–19021

Lucky et al.

Article

scalable procedure to tailor nanotubular structures using a green solvent, i.e., scCO2. As well, the relative rate of each of these reaction steps is different depending on the exact experimental conditions, as observed by the wide variety of morphologies obtained in this study. However, new in situ and chemometric techniques43 are providing tools that can be utilized to follow these reactions of interest, allowing better control of resulting nanostructure evolution.

5. Conclusions ZrO2 modified TiO2 nanostructured materials with both fiber and tubular morphology were successfully synthesized via a surfactant-free sol-gel route using supercritical carbon dioxide (scCO2) under different process conditions. The most important process variables were concentration, acid/metal alkoxide ratio, and Zr composition. The morphology could be tailored by changing the operating variables. Moreover, nanotubes only formed at concentration 1.2 mol/L using R = 5-6 and 60 °C. Initial gelation kinetics for this new system was studied and

Langmuir 2010, 26(24), 19014–19021

revealed that temperature, concentration of the starting materials, and acid to metal (Zr and Ti) alkoxide ratio were the main factors altering the gelation kinetics as well as the basic properties of the synthesized nanomaterials. Acknowledgment. The authors would like to thank Nancy Bell from the UWO Nanofab Lab for the SEM and Ronald Smith from the Biology Department, UWO, for his assistance on TEM. This work was financially supported by the Canadian Natural Science and Engineering Research Council (NSERC) and the Canadian Foundation for Innovation (CFI). Supporting Information Available: Figure SI.1 shows the EDX mapping image for a typical Ti-Zr (10% Zr) binary sample, while Table SI.1 provides the XPS analysis for selective Zr-Ti binary metal oxide nanostructures (0, 10, and 20 wt %) prepared in scCO2. This material is available free of charge via the Internet at http://pubs. acs.org.

DOI: 10.1021/la102048j

19021