4-1. The pore structure and morphology
Figure 4-1 shows N2 adsorption and desorption isotherm and BJH pore size distribution of TP 1 and TP 2. The TP 1 exhibited Type Ⅳ adsorption isotherm and H4 type hysteresis loop, indicating a mesoporous material. The pore size distributions of the samples were analyzed using the BJH model for the adsorption branch. The specific surface areas and pore sizes of the modified TiO2 are summarized in Table 4-1. Pure TiO2
contained a specific surface area of 15 m2/g. The specific surface area ranged 2-103 m2/g and increased with decreasing TOPO concentrations. The TP 1 contained largest surface area of 103.0 m2/g and a typical mesopore size of 3.6 nm. The mesoporous were formed in the interstitial space between TiO2 nanoparticles. At the TOPO/Ti concentration was 0.1 (sample TP 2), the pore shape changed to a wormhole-like structure. Moreover, Type
Ⅰ adsorption was observed for the TP 2, indicating microporous properties.
TOPO-capped nanoparticles were gradually formed when the TOPO/Ti concentrations were higher than 0.2. Strong hydrophobic interaction between the TOPO-capped TiO2 resulted in low specific surface areas. Thus, the critical TOPO/Ti ratio for the porous structure was 0.1. Some studies reported that the critical surfactant/Ti were 0.6 and 0.12 when P123[14]
and CTAB[19] were used as the structure directing agent, respectively. Comparatively, the required critical concentration was lower in this study because the carbon chains of TOPO are much shorter than CTAB and P123.
TMB is a hydrophobic substance and can assist TOPO to self-assemble micelles in hydrophilic media. Figure 4-2, 4-3 and 4-4 shows the N2 adsorption/desorption and pore size distribution of TP 1、TP 2 and TP 3 formed in the presence of various mounts of TMB.
Table 4-2 summarizes the specific surface areas and pore sizes of TP-B samples. Addition
of TMB into the precursor solutions led to Type Ⅳ isotherm for most of the TP samples except for TP 3-B0.3, TP 3-B0.5 and TP 3-B1.2. In addition, remarkable increases in specific surface areas, pore volumes and pore size were obtained. This finding indicates that TMB assists formation of porous structure expand the pore size of TiO2. In the TP 1 sample, the TOPO/Ti ratio 0.05/1 was too low to form well micelles and to result in well-resolved porous structure. However, emulsion effect caused by the presence of TMB led to large micelles and large pore size. For TP 1 series, the narrowest pore distribution was formed when the TMB/TOPO/Ti at 0.05/0.05/1 (TP 1-B1.0). Too low or too high TMB concentration was unable to form homogenous micelles. In the TP 2 series, narrow pore size distribution with average pore size of 3.0-3.7 nm was found for all the added concentrations of TMB. TP 3 was at the margin stage which a continuous porous structure starts to transit to a non-porous structure. However, obvious mesoporous structures were found at the TMB/TOPO/Ti of 0.4/0.15/1 (TP 3-B0.8) and 0.5/0.15/1 (TP 3-B1.0), while microporous feature was found for the other TMB/TOPO ratios. TMB assists TOPO to self-assemble micelles and an appropriate TMB/TOPO concentration is required to form mesoporous structures.
Figure 4-5 shows the TEM and HRTEM images of pure TiO2 and the modified TiO2
prepared with different TOPO concentrations. Pure TiO2 was nonporous structure and composed of aggregated nanocrystals (Figure 4-5a). The TP 1 had the main mesoporousity due to the interparticle porosity (Figure 4-5c). The HRTEM image of TP 1 showed clear lattice fringes which allows for the identification of crystallographic spacing.
It indicates that the prepared TiO2 powders had well anatase crystalline with a d-spacing of (101) crystallographic plan of 0.37 nm. TP 1 had a single crystallize size about 4.6 nm.
Figure 4-5g and e displays the TEM images of TP 2-B0.3 and TP 2, respectively. This finding reveals that the addition of TMB could form obvious pore structures.
Table 4-1 The specific surface area, pore size and pore volumes of TOPO-TiO2 and TiO2.
*Average pore size was calculated by using the BJH formula for the adsorption branch.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.
Relative Pressure (P/P0)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Relative Pressure (P/P0)
TP 2
Figure 4-1 N2 adsorption and desorption isotherm and BJH pore size distribution of TP 1and TP 2.
Table 4-2 The specific surface area, pore size and pore volumes of TOPO-TiO2 with TMB.
Sample S
BET (m2/g)V
pore (cm3/g) *D
pore (nm)TP 1-B0.3 270 0.52 6.8
TP 1-B0.5 190 0.29 5.2
TP 1-B0.8 110 0.27 5.9
TP 1-B1.0 272 0.33 4.3
TP 1-B1.2 264 0.48 6.2
Sample S
BET (m2/g)V
pore (cm3/g) *D
pore (nm)TP 2-B0.3 234 0.29 3.7
TP 2-B0.5 206 0.27 3.7
TP 2-B0.8 158 0.22 3.0
TP 2-B1.0 208 0.22 3.7
TP 2-B1.2 236 0.27 3.6
Sample S
BET (m2/g)V
pore (cm3/g) *D
pore (nm)TP 3-B0.3 117 0.15 3.5
TP 3-B0.5 113 0.08 2.7
TP 3-B0.8 170 0.21 4.7
TP 3-B1.0 145 0.11 2.7
TP 3-B1.2 101 0.13 2.6
*Average pore size was calculated by using the BJH formula for the desorption branch.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 (a)
Volume Adsorbed (A.U.)
Relative Pressure (P/P0) TP 1-B0.3
Figure 4-2 (a) N2 adsorption/desorption isotherm and (b) the pore size distribution of TP 1-B0.3, TP 1-B0.5, TP 1-B0.8, TP 1-B1.0 and TP 1-B1.2.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Relative Pressure (P/P0)
2 3 4 5 6 7 8 9 10 1
Figure 4-3 (a) N2 adsorption/desorption isotherm and (b) the pore size distribution of TP 2-B0.3, TP 2-B0.5, TP 2-B0.8, TP 2-B1.0 and TP 2-B1.2.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Relative Pressure (P/P0)
2 3 4 5 6 7 8 9 10 1
Figure 4-4 (a) N2 adsorption/desorption isotherm and (b) the pore size distribution of TP 3-B0.3, TP 3-B0.5, TP 3-B0.8, TP 3-B1.0 and TP 3-B1.2.
a b
c d
e f
g h
Figure 4-5 TEM and HRTEM images of (a and b) TiO2, (c and d) TP 1, (e and f) TP 2 and (g and h) TP 2-B0.3.
4-2. Crystalline structure and optical property
Figure 4-6 shows the XRD patterns of the pure and modified TiO2. Anatase was the major crystalline phase presented in all TOPO-TiO2 and pure TiO2 samples (JCPDS NO.21-1272, see Appendix B). The crystalline sizes, which were calculated using Scherrer formula, are listed in Table 4-3. The crystalline sizes of the pure TiO2 and TOPO-TiO2 were 2.3 and 3.4-4.7 nm, respectively. Participation of TMB in the preparation changed the crystalline properties of TP samples insignificantly. Consequently, the synthetic temperature even is higher, the crystalline phases are unobvious. Chang et al.[8] used the similar preparation condition (TOPO/Ti = 2.7) to synthesize TOPO-capped TiO2 nanocrystals at 400C and found an average crystalline size of 5 nm. The low crystallite size (3.4-4.7 nm) in this study is attributed to low temperature of 320C used for the synthesis. Moreover, TOPO/Ti ratio thermodynamically effect the generation of TiO2. It is noted that the critical temperature for the TiO2 generation increased with increasing TOPO/Ti ratios. This phenomenon is due to the high energy required to break the interaction between TOPO and TiClx(OC3H7)4-x.
Although templating sol-gel methods are frequently used to prepare porous materials, calcination step at temperature 450-550oC is needed to remove templates and induce crystallization. The resulted crystalline sizes were from 6 to 8 nm[22]. Thus, two steps are necessary for proration of organically modified mesoporous TiO2. In this study, it clearly demonstrates that the non-hydrolytic sol-gel templating method can achieve crystallization and organic modification in single step process.
20 30 40 50 60 70 80
Table 4-3 Crystalline size of TOPO-TiO2 and pure TiO2.
Sample Crystalline size
(nm)TP 1 4.5
The band gap of the TOPO modified TiO2 was determined by UV-vis spectrophotometry. Figure 4-7 shows UV-vis spectra of TOPO-TiO2 and pure TiO2. Based on the adsorption edges, their bandgaps are listed in Table 4-4. The bandgap of the pure and modified TiO2 was 3.6 and 3.3-3.4 eV, respectively. These band gaps were slightly larger than that of a bulk anatase TiO2 (3.2 eV)[106]. The finding indicates that a quantum size effects took place in the non-hydrolytic sol-gel derived TiO2.
200 300 400 500 600 700 800 900
TP 5
TP 4 TP 3
TiO2 TP 1
Kubelka-Munk (A.U.)
Wavelength (nm)
TP 2
Figure 4-7 UV-vis spectra of TOPO-TiO2 and pure TiO2.
Table 4-4 Band gap energy of TOPO-TiO2 and pure TiO2.
Sample Band gap
(eV)TP 1 3.3
TP 2 3.3
TP 3 3.3
TP 4 3.3
TP 5 3.3
TP 2-B0.3 3.4
TP 2-B0.8 3.3
TP 2-B1.2 3.4
TiO2 3.6
4-3. Surface property
4-3-1. Surface functional groups
Figure 4-8 shows the FTIR spectra of TOPO-TiO2 and pure TiO2. TP samples all showed –CH3, –CH2 and P=O stretching absorptions at 2760-3150 cm-1 and 1100 cm-1, respectively[6]. Compared to the FTIR spectrum of TOPO, the shift of the P=O absorption indicates the formation of a P-O-Ti bond[6]. These characteristic absorptions demonstrate that TOPO was chemically bonded to TiO2 during the non-hydrolytic sol-gel process.
Furthermore, the intensity of P-O-Ti absorption increased with increasing TOPO/Ti concentrations. On the other hand, intensive O-H stretching and bending peaks located at 3400 and 1600 cm-1 were observed[9]. The surface hydroxyl groups were formed by hydrolysis of the remained Ti-Cl and Ti-OR in the TiO2 with water vapor after the non-hydrolytic sol-gel reaction. These results indicate the TOPO-TiO2 samples contained hydrophilic and hydrophobic regions.
4000 3600 3200 2800 2400 2000 1600 1200 800 400
Ti-O TP 5 TP 4 TP 3
TP 1 TiO2
TOPO
OH streching
-CH3,-CH2
OH banding
P-O-Ti
TP 2
Absorbance (A.U.)
Wavenumbers (cm-1)
Figure 4-8 The FTIR spectra of TOPO-TiO2 and pure TiO2.
4-3-2. Surface chemical composition
Figure 4-9 shows the wide-range XP spectra of TOPO-TiO2. The Ti (2p), O (1s), C (1s) and P (2p) photoelectron peaks were observed in the spectra, confirming the existence of TOPO on the TiO2 surface.
1100 1000 900 800 700 600 500 400 300 200 100 0
126 128 130 132 134 136 138 140 142
1100
Figure 4-9 The wide-ranged XP spectra of TOPO-TiO2.
Figure 4-10 displays the high resolution scanned XP spectra of O (1s). The O 1s spectra can be fitted into Ti-O-Ti and Ti-OH with the binding energy of 530.3 and 531.7 eV, respectively[4,6,107]. The peak intensity of the Ti-O relative to that of the Ti-OH peak was decreased when TOPO/Ti ratio was increased, indicating the increased coverage of TOPO on the surface. The P/Ti, C/P and O/(Ti + P) of TOPO-TiO2 atomic ratios were calculated from the integrated peak areas normalized to their atomic sensitive factors and are summarized in Table 4-5. The C/P ratios of 11.73-18.57 were lower than the theoretical ratio of 24.0 of TOPO molecules, indicating that some carbon chains of TOPO were decomposed during the synthetic process. The ratios of Ti-O/(Ti + P) in the TP samples were 1.3-1.8, that were smaller than the theoretical ratio of 2 of TiO2. It suggests many defects on the surface. The P/Ti ratios were 0.37-0.43. Participation of TMB in the sol-gel process decreased the P/Ti ratios of TP 2 from 0.39 to 0.19-0.24. It is presumably due to that TMB expands the size of micelles, thus reducing TOPO-to-surface area of the micelle ratios.
538 536 534 532 530 528 526 Ti-O-Ti
Ti-OH
TP 5
TP 4
TP 3
TP 2
TP 1
Intensity (A.U.)
Binding Energy (eV)
Figure 4-10 The evolution of O (1s) XP spectra of TOPO-TiO2.
Table 4-5 Surface chemical composition of TOPO-TiO2.
P/Ti C/P Ti-O/(Ti + P)
TP 1 0.37 16.86 1.33
TP 2 0.39 18.57 1.84
TP 3 0.41 11.99 1.45
TP 4 0.44 12.72 1.36
TP 5 0.42 12.66 1.49
TP 2-B0.3 0.24 16.84 1.81
TP 2-B0.8 0.22 19.54 1.75
TP 2-B1.2 0.19 22.92 1.81
To estimate the quantity of the bonded TOPO, the fractions of the P, O, Cl and Ti of the TOPO-TiO2 was carried out using EDX (see Appendix C). Table 4-6 shows P/Ti, O/Ti, and Cl/Ti ratios in the TOPO-TiO2 samples and pure TiO2. The percentages of P increased with the increasing TOPO concentration. In addition, substantial amounts of Cl ions were left in all samples. Theoretically, Ti-O-Ti oxo-bonds are formed by cross-condensation between Ti-OR and Ti-Cl, and the by-product is volatile RCl. The remained Cl ions should be resulted from the hydrolysis of Ti-Cl. This finding proved again that the generation of OH groups observed in the FTIR spectra. The ratio of P/Ti ranged from 1.4
10-2 to 9.0 10-2. These values were all smaller than the theoretical addition amount (P/Ti = 0.05-0.25). Approximately, 45-70% TOPO was lost from the synthesis. The loss could be resulted from the thermal evaporation of TOPO at the high reaction temperature.
The addition of TMB has a little influence on the percentages of P. Based on the texture and chemical composition analysis, it can conclude that TMB dominate the structure of TOPO-TiO2 only.
Table 4-6 Elemental analysis of TOPO-TiO2 and pure TiO2 using EDX.
Atomic (%) P/Ti O/Ti Cl/Ti
The quantity of TOPO bonded to the TiO2 surface was determined by thermo gravimetriv analysis (TGA). Figure 4-11 displays the typical heat flow and weight loss of TP 2 (The other TP samples see Appendix D). There are three weight loss stages in the measuring temperature range. Below 100oC, 6-12% weight loss is denoted to the removal of surface physisorbed water molecules. From 100 to 260oC, 5-10% weight loss indicated partially oxidized or pyrolysis of TOPO. At 250-550oC, 14% weight loss accompanied with an exothermic peak centered at 300oC indicates complete burnout of the carbon moiety from the bonded TOPO. The organic-related weight loss of TP 1 was 9%. However, it increased to 21-22% when TOPO/Ti concentration was higher than 0.15 (sample TP 3, TP 4 and TP 5). This result indicates that the bonding of TOPO to the surface of TiO2 reached saturation. The presence of TMB did not alter the bonding. The TP 2-B samples contained 17% of the organic loss, which is similar to that of TP 2 (14%). This result is in agreement with that obtained from EDX, in which the P/Ti ratio of the TP 2 (4.8 10-2) was similar to that in the TP 2-B samples (3.0 10-2-5.3 10-2).
0 100 200 300 400 500 600 700 800 900
Figure 4-11 The TGA/DSC curve of TP 2.
Table 4-7 The weight loss of TP and TP-B samples.
Weight loss (%) RT-100
oC 100-260
oC 260-550
oC
4-4. Adsorption isotherm
Surface modifier and surface areas play an important role in the photocatalytic activities because they dominate the adsorption ability and reaction sites of the catalysts.
Figure 4-12 displays the adsorption isotherms of the TOPO-TiO2 and pure TiO2 toward BPA. Both P25 and the non-hydrolytic sol-gel derived TiO2 adsorbed little amount of BPA. This phenomenon reveals that surface Ti-OH has small affinity toward the target compound. In contrast, all TOPO-TiO2 samples showed high level of adsorptions.
Linear dependence of the adsorption amounts upon equilibrium concentration in the TP 4 and TP 5 suspensions indicates that BPA is partitioned to the surface of the samples. The nonspecific physisorption is resulted from their TOPO-capped surface. The TP 2 and TP 3 exhibited a sigmoid adsorption. The concave upward curve at low BPA concentrations followed Type Ⅲ adsorption isotherm, suggesting the interactions between the adsorbates were stronger than those between the adsorbate and the samples[108]. This phenomenon was induced by the π-π interaction between the BPA molecules adsorbed in the meso- or micro-pores. In addition, capillary force assists BPA molecules rapidly congealed into the pores. Due to the pore size of the TP 2 was smaller than the TP 3, the TP 2 exhibited a higher adsorption ability than the TP 3. Apparently, TP 1 reached a saturated adsorption of 25.6 mg/g. The low amounts of the bonded TOPO restricted its adsorption capacity.
However, the TP 1 showed similar adsorption ability to the TP 2 at low BPA concentrations, indicating their similar TOPO-contained pore structures. After addition of TMB, the adsorption capacity of the TP 2 decreased from 54.9 to 21.1-46.3 mg/g at the initial concentration of 80 mg/g. The decreased adsorption ability was due to the expanded pore sizes caused by the presence of BPA.
Figure 4-13 shows the time domain of 10 mg/L BPA adsorption on P25, pure non-hydrolytic sol-gel derived TiO2, the TP and TP-B samples. The P25 and pure TiO2,
TP 2, TP 3, TP 4 and TP 5 reached adsorption equilibrium within 10 minutes. However, the adsorption equilibrium time of the TP 1 extended to 30 minutes. This phenomenon reveals that large surface tension resulted from microporous structures initially inhibit the diffusion of BPA into the pores. It is noted that the equilibrium time of the TP 2 was largely shorten to 2 minutes after addition of TMB. The expanded pores sizes reduced the surface tension, thus promoting the adsorption kinetics.
0 10 20 30 40 50 60 70 80
Figure 4-12 Adsorption of BPA in the suspensions of (a) TP samples and pure TiO2, (b) the TP 2-B samples.
0 10 20 30 40 50 60 Figure 4-13 Adsorption equilibrium of 10 mg/L BPA in the suspensions of (a) TP samples and pure TiO2, (b) the TP 2-B samples.
4-5. Photocatalytic activity
Figure 4-14 shows the photocatalytic decomposition of 10 mg/L BPA in the presence of pure TiO2, TPand TP-B samples. The TP samples reduced 46-75% BPA in the initial 30 minutes of dark adsorption. However, the pure TiO2 and P25 reduced 1% BPA only.
Photodegradation of BPA by the TP samples reached 95% removal efficiency in next 90 minutes, while the TiO2 degraded 70% BPA in the same period. The high removal efficiencies of the TOPO-TiO2 was mainly contributed by pore structure and lost of stabilized charges which improved compound from bulk solution diffuse to the surface effectively and facilitated interfacial charge transfer[8,101]. It is noted that the photodegradation by TP 2 was limited to 95% because its high surface tension inhibits the diffusion of BPA into the porous structure. TP-B samples showed relatively low removal efficiencies than TP 2. In addition, the initial photocatalytic degradation rate of TP 2, TP 2-B0.3, TP 2-B0.8, and TP 2-B1.2 was 0.19, 0.21, 0.15, and 0.11, respectively, for the initial concentration of 2.51, 8.23, 3.30, and 5.26 mL/g. Because degradation rates are correlated to concentrations of a target compound, TP 2 exhibited higher photocatalytic activity than TP 2-B samples. The higher photocatalytic activity was attributed to high adsorption ability of TP 2[101]. To compare the photocatalytic activity of the porous TOPO-modified TiO2 with commercial product P25, TP 2 was selected as the model compound because it showed the highest adsorption affinity for BPA. The high adsorption ability of TP 2 led its photocatalytic kinetics not satisfying pseudo-first order kinetics. Thus, transformed Langmuir-Hinshelwood kinetics model is used (shown in equation 4-1):
C
where r is the initial rate of photocatalytic degradation fro BPA, C represents the concentration of the BPA, kr and Ka are the kinetic rate constant and adsorption coefficient,
respectively. The kr and Ka were obtained from the intercept and slope of the linear fit of 1/r versus 1/C, respectively. Figure 4-15 shows the plot of the linear fit of 1/r versus 1/C for Langmuir-Hinshelwood model. Table 4-8 were summarized the rate constants and adsorption coefficients of BPA in the TP 2 and P25 suspensions. The TP 2 exhibited a kr
and Ka of 6.58 × 10-1 mg/L-min and 8.89 × 10-2 L/mg, respectively. P25 showed the kr
and Ka of 4.43 × 10-1 mg/L-min and 8.55 L/mg, respectively. The kr of TP 2 was 1.5 times higher than P25. Higher adsorption ability improved the attachment of BPA to the surface of TP 2 and apparently enhanced electron transfer process. However, the Ka of TP 2 was relatively smaller than P25 due to its poor crystallinity. Large amounts of surface defect energy levels between bands inhibited the kinetic absorption. It also resulted in lower kr × Ka product of TP 2 (5.85 × 10-2 min-1) than that of P25 (3.71 min-1). The poor crystallinity of TP samples is limited by the heating process used in this study. Conventional conductive heating causes poor and inhomogeneous dissipation of thermal energy through whole preparation systems, especially when solid TiO2 gel is formed. Thus, short heating time was taken in order to avoid thermal decomposition of organic TOPO. This problem could be solved in case a heating method which allows thermal energy penetrating matrix rapidly and homogeneously, like microwave, is used for the preparation. However, this study demonstrated that surface modified porous TiO2 was successfully obtained using a templated non-hydrolytic sol-gel method.
-30 -20 -10 0 10 20 30 40 50 60 70 80 90
Figure 4-14 The photodegradation of 10 mg/L BPA by (a) TP samples and pure TiO2, (b) TP 2-B samples.
0.1 0.2 0.3 0.4
Table 4-8 The rate constants (kr) and adsorption coefficients (Ka) of BPA in the TP 2 and P25 suspensions.
kr (mg/L -min) Ka (l/mg) kr× Ka
R
2TP 2 6.58 × 10-1 8.89 × 10-2 5.85 × 10-2 0.967
P25 4.43 × 10-1 8.55 3.71 0.999