Meaningful Differences in Spectral Performance, Thermal Behavior

Feb 12, 2010 - A novel molecule−ion adduct of ammonium molybdate tetrahydarte (AMT) with β-cyclodextrin (CD) was prepared in this work. Significant...
20 downloads 0 Views 3MB Size
Download PDF
3404

J. Phys. Chem. B 2010, 114, 3404–3410

Meaningful Differences in Spectral Performance, Thermal Behavior, and Heterogeneous Catalysis between Ammonium Molybdate Tetrahydrate and Its Adduct of β-Cyclodextrin Le Xin Song,*,†,‡ Mang Wang,† Zheng Dang,‡ and Fang Yun Du‡ Department of Chemistry, UniVersity of Science and Technology of China, Hefei 230026, P. R. China, and CAS Key Laboratory of Soft Matter Chemistry, Department of Polymer Science and Engineering, UniVersity of Science and Technology of China, Hefei 230026, P. R. China ReceiVed: January 12, 2010; ReVised Manuscript ReceiVed: January 30, 2010

A novel molecule-ion adduct of ammonium molybdate tetrahydarte (AMT) with β-cyclodextrin (CD) was prepared in this work. Significant differences in spectral properties between AMT and the adduct AMT-βCD were observed by a series of experimental probes, such as powder X-ray diffraction, Fourier transformation infrared spectroscopy, and Raman spectroscopy. Field emission scanning electron microscopy showed that, although the crystal growth of AMT-β-CD was dominated by the molecular stacking of AMT, the size and morphology of the adduct were rather different from those seen in free AMT. The difference in stacking forms was attributed to the contribution of the molecule-ion interaction between AMT and β-CD. A drastic improvement in thermal stability of AMT and β-CD after adduct formation was observed by thermogravimetry analysis, which was confirmed by controlled sintering measurements. This revealed that the adduct interaction between them played an important role in mediating the thermal decomposition process of the adducted components. Furthermore, our results indicated that AMT and its adduct had a different performance in the catalytic desulfurization of thiophene and its derivatives. The fact that the catalytic efficiency of AMT was decreased after adduct formation implied there was a complexation between AMT and β-CD. Besides, several unusual molecular ionssNH3+, NH2+, and NH+swere simultaneously found with gas chromatography coupled to time-of-flight mass spectrometry of free AMT. Introduction β-Cyclodextrin (β-CD, Figure 1) is a cyclic oligosaccharide with seven alpha-1,4-D-glucopyranosyl units, which is wellknown for its inclusion complexation to many kinds of guests.1–3 There are numerous reports about the formation of β-CD inclusion complexes with organic molecules;4–9 however, only a little attention has been paid to relative differences in inclusion behavior of β-CD for inorganic salts and organic molecules to date.10,11 Very recently, evidence for a large change in thermal stability of β-CD upon adduct formation with inorganic salts such as lithium carbonate (Li2CO3) and sodium arsenite (NaAsO2) has been reported.12,13 Further, the difference in thermal behavior of the inorganic salts before and after adduct formation was ascribed to the change of arrangement behavior of inorganic salts during the process of ion redistribution in the presence of β-CD.13 Although the molecule-ion interaction between them could change the thermal decomposition mechanism of β-CD, the presence of such small inorganic salts did not change the stacking behavior of β-CD molecules during crystal growth.13 That is to say, the formation process of the molecule-ion adducts was dominated by the crystal orientation of β-CD. This impels us to examine whether the molecular arrangement of β-CD can be changed along with the size increase of inorganic salts. Thus, ammonium molybdate tetrahydrate (AMT, Figure 1), a large common polyoxometalate compound, is employed to investigate the problem in this work. * Author to whom correspondence should be addressed. E-mail: [email protected]. † Department of Chemistry. ‡ Department of Polymer Science and Engineering.

Figure 1. Structural features of β-CD and AMT.

AMT is an inexpensive multifunctional catalyst and has been widely used in industrial production processes,14–16 especially in the heterogeneous catalysis-synthesis.17–20 Here, we select the molecule-ion system of β-CD and AMT, aiming at evaluating differences in spectral performance, thermal behavior, and heterogeneous catalysis of AMT before and after adduct formation with β-CD. This may provide insights into how an adduct interaction between a macrocyclic compound and a large inorganic salt responds to the change of physical and chemical properties of the inorganic salt. The molecule-ion adduct of AMT with β-CD was prepared and characterized by Fourier transform infrared (FTIR), powder X-ray diffraction (PXRD), and field emission scanning electron microscopy (FESEM). As a comparison, we also reported the difference in spectral properties of β-CD and AMT between before and after adduct formation. Our results indicated that the stacking behavior of β-CD molecules upon adduct formation with AMT was changed from a cage to a channel form. Furthermore, the crystallization of the adduct is predominated

10.1021/jp100308x  2010 American Chemical Society Published on Web 02/12/2010

Novel Molecule-Ion Adduct of AMT with CD by AMT, which is different from reports in the literature.12,13 Also, the changes in thermal behaviors of these samples were carefully compared using the sintering technique at the same temperatures, thermogravimetry (TG), derivative thermogravimetry (DTG), and gas chromatography coupled to time-of-flight mass spectrometry (GC-TOF-MS). In addition, the measurement results on the catalytic desulfurization of thiophene and its derivatives in the presence of AMT or its adduct were presented here. Our studies encompass the changes of AMT and its adduct in physical property and chemical reactivity, and our aim is to attract the attention on the potential functional importance of such changes in scientific research and practical applications. Experimental Section Materials. β-CD was purchased from Shanghai Chemical Reagent Company and recrystallized twice from deionized water. AMT was obtained from Chemical Reagent Factory of Hefei University of Technology. n-Octane and hydrogen peroxide (H2O2) of 30% concentration were purchased from Sinopharm Chemical Reagent Co. Ltd., China. Thiophene (TP), 2-methylthiophene (2-MT), and 3-methylthiophene (3-MT) were purchased from Alfa Aesar Chemical Company. D2O, as a solvent in nuclear magnetic resonance (NMR) measurements, was obtained from Aldrich Chemical Co. All other reagents are of analytical-reagent grade, unless stated otherwise. Hydrothermal Preparation of Molecule-Ion Adduct of AMT with β-CD. The solid adduct AMT-β-CD was prepared according to the method as follows. AMT of 618 mg (0.5 mmol) and β-CD of 567 mg (0.5 mmol) were mixed in deionized water of 50 mL into a stainless steel autoclave of 55 mL capacity at 373 K for 2 h and cooled to room temperature. Solvent was drawn off by rotary evaporation under a vacuum at 323 K. A grassy residue was obtained and dried in a vacuum at 373 K for 2 h. 1H NMR (300 MHz, D2O ppm) δ 5.071 (d, 7H, C1-H), 3.623 (m, 7H, C2-H), 3.939 (m, 7H, C3-H), 3.557 (d, 7H, C4-H), 3.854 (s, 7H, C5-H), 3.885 (s, 7H, C6-H) of AMT-βCD. FTIR data of the adduct AMT-β-CD: 3436, υO-H; 3183, υN-H; 1636, δO-H; 1406, δN-H; 1151, 1087, 1029, υC-O; 939, υC-O-C; 653 cm-1, υMo-O-Mo. Preparation of a Physical Mixture of AMT with β-CD. The physical mixture of AMT with β-CD was obtained by grinding free AMT of 618 mg (0.5 mmol) and pure β-CD of 567 mg (0.5 mmol) in a mortar for 10 min. FTIR data of the mixture: 3387, υO-H; 3231, υN-H; 1638, δO-H; 1404, δN-H; 1157, 1079, 1028, υC-O; 932, υC-O-C; 643 cm-1, υMo-O-Mo. Sintering Experiments. Pressureless sintering experiments of AMT, its adduct, and the ground mixture were carried out in a muffle furnace at 573, 673, 773, and 873 K for 2 h. Before sintering, solid samples were dried to a constant weight at 373 K. After sintering, the samples were cooled under a vacuum, and weighted at room temperature. The color changes of the adduct and the mixture were different in the temperature span from 573 to 873 K. The former became deep yellow-brown from black, but the color of the latter varied from black to pale yellow. Catalytic Experiments. The heterogeneous reactions of TP, 2-MT, and 3-MT (1.26 × 10-4 mol · dm-3 in n-octane) with H2O2 (9.70 × 10-4 mol · dm-3 in water) were performed in the absence and presence of AMT (1.26 × 10-4 mol · dm-3) or its adduct (1.26 × 10-4 mol · dm-3) as follows. A 10 mL solution of TP or its derivatives was added in a three-neck round-bottom flask fitted with a thermometer, a reflux condenser, and an inlet for the addition of a 6.5 mL solution of H2O2 and 10 mL AMT

J. Phys. Chem. B, Vol. 114, No. 9, 2010 3405 or its adduct at a temperature of 353 K with stirring for 120 min. Then, the organic phase was separated and measured using UV-vis spectroscopy. Instruments and Methods. All solid samples were kept and measured under the same drying conditions. PXRD patterns of the samples were employed in a Philips X’Pert Pro X-ray diffractometer. Solid samples were irradiated with monochromatized Cu KR radiation and analyzed in the 2θ range from 5 to 40°. The tube voltage and current were 40 kV and 40 mA, respectively. FESEM images were recorded on a JEOL-JSM6700F field-emitting microscope. FTIR spectra were obtained by a Bruker Equinox 55 spectrometer in KBr pellets in the range 400-4000 cm-1. Raman spectra were collected at room temperature with a LABRAM-HR Confocal Laser MicroRaman spectrometer in the range 100-1800 cm-1. UV-vis spectra were recorded on a Shimadzu UV 2401-(PC) spectrometer in the range 200-280 nm. 1H NMR spectra were carried out on a Bruker AV-300 NMR spectrometer at 300 MHz at room temperature. Chemical shifts (δ) were reported in ppm relative to tetramethylsilane using the residual solvent signals at HDO at 4.75 ppm as an internal reference. TG and DTG measurements were made on a Shimadzu TGA50 thermogravimetric analyzer at a constant heating rate of 10 K · min-1 under a nitrogen atmosphere with a gas flow of 25 mL · min-1. GC-TOF-MS experiments were done on a Micromass GCT-MS spectrometer. The heating program of the samples was the same as that reported in a recent paper.21 Results and Discussion Changes of Crystal Characteristics of AMT and β-CD upon Adduct Formation. FTIR data indicate that the positions of several main peaks in the adduct are different from those of free β-CD, free AMT, and the ground mixture. There are no obvious changes in the stretching vibration bands of C-O or C-O-C between β-CD and its adduct or the mixture, indicating that the glucopyranosyl units of β-CD are not largely affected by the presence of AMT. However, the υMo-O-Mo vibration in free AMT and the mixture occurs at 645 and 643 cm-1, but in the adduct, it appears at 653 cm-1. This demonstrates that the oxygen-bridged structure in AMT is influenced by β-CD. In particular, the υO-H vibration band exhibits an observable shift to the direction of a high wavenumber from 3412 cm-1 in the free β-CD to 3436 cm-1 in the adduct but a shift to the direction of a low wavenumber (3387 cm-1) in the mixture. Also, the υN-H and δN-H vibration bands of free AMT, occurring at 3166 and 1397 cm-1, respectively, shift to the direction of high wavenumbers of 3183 and 1406 cm-1 in the adduct, and they appear at 3231 and 1404 cm-1 in the case of the mixture. These results above show that such mixing processes between AMT and β-CD both in solid state and in water will lead to a change of hydrogen bonding networks of the mixed components. The change in hydrogen bonding networks may imply the change of crystal surface state. Actually, FESEM images confirm this prediction. As shown in Figure 2, the crystal morphologies of free AMT and free β-CD display a layered architecture and a regular hexagonal prism, respectively. Nevertheless, the FESEM image of the adduct AMT-β-CD indicates the surface structure of a sheet packing, which is different from that of either of the two free components. The images give an impression that the morphology of β-CD has been entirely changed in the existence of AMT. The significant difference in the FESEM images between β-CD and the adduct represents a direct evidence for a rearrangement of the three-dimensional structural features that certainly result

3406

J. Phys. Chem. B, Vol. 114, No. 9, 2010

Song et al.

Figure 2. FESEM images of (a) AMT, (b) β-CD, and (c) their adduct.

Figure 4. TG (A) and DTG (B) profiles of (a) AMT, (b) β-CD, (c) their adduct, and (d) the fitted theoretical curve based on a mixed form of β-CD and AMT in a 1:1 molar ratio. Figure 3. PXRD patterns of (a) AMT, (b) β-CD, and (c) their adduct.

from a direct interaction between AMT and β-CD in the course of adduct formation. However, there is a close structural relationship between AMT and the adduct because both of them indicate a layer-by-layer growth mode, though they exhibit different thickness profiles by appearances. This structural relationship between them may suggest that the crystallization process of the adduct is dominated by the crystal growth of AMT. Visual examination of Figure 2 supports the analysis of molecular stacking manners. From Figure 3, the PXRD pattern of free AMT shows several major peaks at 2θ angles lower than 30°: 9.8, 12.2, and 26.6°, corresponding to crystal planes of (011), (121), and (133), respectively, having a monocline symmetry. The occurrence of the series of peaks at about 9.1, 12.6, and 19.2° in free β-CD, corresponding to crystal planes of (101), (111), and (410), respectively, suggests that it prefers to adopt a cage-type structure.22 However, as seen from Figure 3c, the PXRD pattern of the adduct is quite different from those of the two free components. For example, the weak peaks of free AMT at 9.8 and 26.6° have become two of the three strongest peaks in the adduct, and the third strongest peak occurring at 14.0° in the adduct is a new one. This observation is in good accordance with the suggestion put forward in FESEM analysis that the process of crystal growth of AMT dominated the stacking behavior of particles in the adduct. Moreover, the disappearance of the three key characteristic peaks of β-CD at 9.1, 12.6, and 19.2° and the appearance of several new peaks after adduct formation, such as those at about 11.7 and 17.7°, illustrate that the crystal packing mode of β-CD has been fundamentally changed from a cage to a channel form.4,8,23 The phenomenon is attributed to the contribution of the molecule-ion interaction between AMT and β-CD. It is worth stressing that such a phenomenon is not observed in those systems formed by β-CD and small inorganic salts such as Li2CO3 and NaAsO2.12,13 Changes of Thermal Properties of AMT and β-CD upon Adduct Formation. The thermal stability of AMT-β-CD is evaluated by TG and DTG diagrams and compared with those

Figure 5. A proposed thermal decomposition route of AMT.

of free β-CD and free AMT. Figure 4A shows the relationship between the residual masses (RM, %) of the three samples and temperatures as well as a fitted theoretical curve based on a mixed form of free β-CD and free AMT in a 1:1 molar ratio. Figure 4B illustrates the relationship between temperatures and mass loss rates (V, /% · s-1) of these samples. As seen in Figure 4A, the shapes of the curves are very different from one another, which are reflected by water release, decomposition behavior, and final residual mass, suggesting a marked thermal stability difference before and after adduct formation. First, as shown in Figure 4A, the mass loss processes of free AMT and free β-CD consist of three stages (I, 5.85% due to release of 4 crystal water molecules, calc. 4.02; II, 4.31% due to release of 3 water molecules, calc. 2.96; III, 8.20% due to release of 6 ammonia molecules, calc. 5.96) and two stages (i, release of water; ii, decomposition of β-CD), respectively. Figure 5 illustrates a proposed thermal decomposition model of AMT in terms of TG data. However, no clear stages of mass loss can be distinguished in curve c for the adduct because this curve is almost linear at a wide temperature range of 511-773 K. This phenomenon implies that the decomposition rate of β-CD is seriously weakened in the presence of AMT. The comparison of curves b and c in Figure 4B demonstrates the result. Second, the adduct suffers an extra mass loss of 5.6% in the temperature range from 511 to 556 K relative to the two free components. This can be ascribed to an earlier decomposition of one or two of the components in the adduct.

Novel Molecule-Ion Adduct of AMT with CD

J. Phys. Chem. B, Vol. 114, No. 9, 2010 3407

Figure 6. TIC curves of (a) AMT, (b) β-CD, and (c) their adduct.

Figure 8. Mass spectra of the adduct at 18.34, 26.02, 31.75, and 35.77 min.

Figure 7. Mass spectra of AMT at (a) 5.22, (b) 10.44, (c) 17.45, and (d) 25.29 min.

contrary, all of the strong signals in the case of the adduct occur after 26.00 min, except a weak signal at 18.34 min. The later decomposition at 31.75 and 35.77 min indicates the improvement of the thermal stability of β-CD, since the released signals at these stages are in close relation to the decomposition of β-CD (see Figure 8). This may be involved in different decomposition mechanisms of the components before and after adduct formation. Changes of Thermal Decomposition Mechanisms of AMT and β-CD upon Adduct Formation. In light of the mass spectra illustrated in Figure 7, free AMT shows four fragment signals at a time span of about 25 min, indicating the largest release of several molecular ions: initially H2O+, next NH3+, and then NH2+ ions. The release of water only occurs at the first decomposition stage. Furthermore, the change (denoted by an arrow in Figure 7) in RA values of the series of fragment ionssNH+, NH2+, and NH3+swith the increase of temperature reveals the extent of decomposition of the NH4+ ions at different temperatures. The higher the temperature is, the more the onset of the decomposition reaction is shifted to smaller structural units such as NH+ and NH2+, as shown in eqs 1 and 2.

Third, the final residual mass fraction of 48.8% at 773 K in the fitted theoretical curve is lower than that of 59.8% in the adduct. A difference of at least 11.0% at the higher temperature range of 650-773 K strongly implies that the degree of decomposition of the adducted β-CD is reduced significantly by the decomposition of AMT. As seen in curve a of Figure 4A, the MoO3 resulting from the decomposition of AMT is highly stable under TG conditions.24 Possibly, the formation of MoO3 plays a role in prohibiting the rapid and sharp decomposition of β-CD. Additionally, the big slope of the orange line in the shaded area of Figure 4A means that the thermal decomposition process of the adducted β-CD will continue to progress at a higher temperature. Figure 4B presents the difference in the largest decomposition rate among β-CD, AMT, and their adduct. One reason for the change in thermal behavior may be due to the change of molecular arrangements of β-CD and AMT before and after adduct formation. In order to further investigate the relationship between temperature and thermal stability, GC-TOF-MS measurements of these samples are employed. The total ion current (TIC) curves of free AMT, free β-CD, and their adduct are depicted in Figure 6, indicating the relationship between relative abundances (RA, %) of fragments and heating times of the samples. Curve c of the adduct shows a large difference in the positions and shapes of peaks from curve a of free AMT and curve b of free β-CD. Free AMT begins to decompose at around 5.22 min (363 K), and indicates the other three decomposition peaks at 10.44, 17.45, and 25.29 min. These signals in Figure 6a are mainly reflected by the release of H2O+ (18.011), NH3+ (17.020), NH2+ (16.019), and NH+ (15.010) ions, as shown in Figure 7. Free β-CD exhibits two main peaks at 25.63 and 27.23 min. However, there are no peaks in this region from 0 to 16 min in the case of the adduct (see Figure 6c), implying that the thermal stability of AMT is strongly enhanced by its adduct interaction with β-CD. Further, free AMT is almost decomposed completely and MoO3 is generated before 26.00 min. On the

NH3+ f NH2+ + H

(1)

NH2+ f NH+ + H

(2)

The simultaneous appearance of the NH3+, NH2+, and NH+ molecular ions in the gas phase and their different relative contents at different temperatures will be significant and interesting. It was reported that these ions occurred in the decomposition and combustion of propellant,25,26 and that NH+ ion was associated with the formation of ammonia in interstellar molecular clouds.27 Since these ions are easy to obtain at a lower temperature, we think that this work may open new perspectives for the study of inorganic materials, organic syntheses, biochemical reactions, interstellar mediums, and rocket fuels. Clearly, the thermal decomposition behavior of free AMT in a vacuum (the GC-TOF-MS condition) is strikingly different from that of it under nitrogen gas (the TG condition) based on the comparison between Figure 4A and Figure 6a. This difference is reflected by different numbers in decomposition stages and different release behaviors of water in these stages. The mass spectra, corresponding to the four thermal decomposition stages of the adduct in Figure 6c, are illustrated in Figure 8. It should be noted that no observable fragment signals containing N element are found in the range 0-30 min in the case of the adduct besides the NH2+ ion with a low RA. Since the adduct was prepared in a molar ratio of 1:1 between AMT and β-CD, the lack of signals generated by the N element

3408

J. Phys. Chem. B, Vol. 114, No. 9, 2010

Song et al.

Figure 10. Relationship between heating temperatures and residue masses of (a) AMT, (b) the adduct, (c) the mixture, (d) a fitted theoretical curve based on a mixed form of β-CD and AMT in an initial molar ratio of 1:1, and (e) free β-CD. Figure 9. Proposed rupture models of covalent bonds in the adducted β-CD.

implies the decomposition of the NH4+ ions in the adduct is drastically decreased relative to free AMT at the same heating times. Up to 773 K (31.75 min), two weak fragment signals due to HNO2+ (46.997, 6.17%) and H2NO2+ (48.005, 5.21%) occur, as seen in Figure 8c. These results indicate that the decomposition mechanism of AMT is changed principally upon adduct formation, and its thermal stability is largely increased by the adduct interaction with β-CD. On the other hand, the mass spectrum of free β-CD at 18.34 min indicates two series of fragment signals: the first series at m/z 29.004 (CHO+), 43.019 (C2H3O+), and 57.035 (C3H5O+) and the second series at m/z 60.022 (C2H4O2+) and 73.030 (C3H5O2+),13,21 both of which are associated with the rupture of C-O and C-C bonds in glucopyranose units and glycosidic bonds. Nevertheless, the contribution derived from these signals to the TIC is inappreciable according to Figure 6b. Interestingly, they are very strong at two key times: one at 18.34 and the other at 26.02 min in the spectra of the adduct (see Figure 8a and b). This observation demonstrates that the decomposition of the adducted β-CD starts ahead of the normal onset period. Additionally, Figure 8c and d presents the decomposition processes of the adduct at the end of the final period, showing only one strong signal at m/z 43.990 due to CO2+ ion. This is completely different from the decomposition behavior of free β-CD because CO2+ ion is not one of the main release products of free β-CD at a higher temperature.21 Figure 9 describes three proposed rupture models (I, II, and III) of covalent bonds of β-CD in the adduct on the basis of the data from Figure 8. Modes I and III reveal the destruction of C-O and C-C bonds in the glucopyranose units at earlier decomposition stages. Mode II is the main thermal decomposition route of the adducted β-CD according to the comparison between Figure 6c and Figure 8, representing the rupture of 1,4-glycosidic bonds. The fact that there is a big time span in the process of conversion from modes I and III to mode II explains why a longer decomposition period occurs in the adducted β-CD than in free β-CD. Comparison of Sintering Results of AMT and Its Adduct under Ambient Conditions. The improvement in thermal stabilities of AMT and β-CD upon adduct formation, especially the difference between TG and GC-TOF-MS conditions in thermal decomposition behaviors of AMT and its adduct, leads us to ask whether such a phenomenon occurs under ambient conditions. Hence, sintering experiments of the samples in air are done in the work. AMT and its adduct in Figure 10 exhibit

Figure 11. Raman spectra of (a) β-CD, (b) AMT, and (c) their adduct at 298 K, as well as the residues of the adduct at (d) 573, (e) 673, and (f) 773 K.

the relationship between heating temperatures and residue masses of free AMT, free β-CD, their physical mixture, and the adduct. First, the process of incineration of free β-CD ends before 700 K. Subsequently, no residual mass in the sample was observed in the figure (see curve e). For free AMT, no observable mass losses occur at this temperature range (see curve a). Second, both the simulated and experimental curves (d and c) are very near each other, meaning that the decomposition behaviors of AMT and β-CD in the mixture are relatively independent under ambient conditions. In contrast to these, curve b is always above curves c and d before 873 K. That is to say, the adduct interaction between AMT and β-CD in water makes them more stable in comparison with the physical mixing of them in the solid state. Figure 11 depicts the Raman spectra of free AMT, free β-CD, and their adduct. It is apparent that those peaks of β-CD nearly disappear in the adduct. This result is significantly different from the result seen in FTIR spectra. Besides, the strongest peak at 907 (υas, Mo-O) in free AMT28 shifts to 943 cm-1 (υs, Mo-O) in the adduct. What is more, we present in curve d of the figure a typical Raman spectrum, which exhibits two overlapping bands at 1372 (D-band) and 1589 cm-1 (G-band). Here, they are designated to defects and the in-plane E2g zone-center mode of the graphite layer.29,30 It is very interesting that deposition of graphite is generated at such a low temperature (573 K) by a common carbohydrate. This finding may have applications in experimental and theoretical studies on chemical transformation of carbon compounds in nature. At higher temperatures, the signals of carbon layers disappear. Also, the general representative peaks of MoO3 at 283, δModO; 665, 817, υMo-O-Mo; and

Novel Molecule-Ion Adduct of AMT with CD

Figure 12. FESEM images of the residues from (a) free AMT and (b) the adduct at 673 K.

J. Phys. Chem. B, Vol. 114, No. 9, 2010 3409

Figure 14. UV-vis spectra of residual TP in the organic phase under different systems (a) in n-octane, (b) with H2O2 in n-octane, (c) with H2O2 and the adduct in n-octane, and (d) with H2O2 and AMT in n-octane.

Figure 13. Chemical reactions of TP or its derivatives with H2O2 in the absence and presence of AMT or its adduct.

995 cm-1, υModO, occur,31 and do not change at several temperatures such as 573, 673, and 773 K and higher temperatures.32 FESEM images of the residues generated in sintering experiments of AMT and its adduct at 673 K are shown in Figure 12. Clearly, the crystal surface of free AMT at this temperature is uniform, and its particle size is much smaller than that of the adduct. This provides direct evidence that the size and morphology of free AMT are more largely influenced by heating than those of its adduct when compared with Figure 2. Comparison of the Catalytic Efficiency of AMT and Its Adduct in the Oxidation Reactions of TP and Its Derivatives. In order to estimate the performance of the molecule-ion interaction in water, 1H NMR experiments were performed to compare the change (∆δ) in δ values of proton signals of β-CD before and after adduct formation. Our results indicate that no considerable difference ( ∆A3 > ∆A2) in absorbance change in this process reveals the effect of the position of the methyl group on the TP ring on the catalytic reactions. These results provide a new way to make heterogeneous catalysis of polyoxometalates in organic reactions controllable. Conclusions A novel molecule-ion adduct AMT-β-CD was prepared and characterized by many physical methods. Significant differences in spectral performance, thermal behavior, and heterogeneous catalysis of AMT before and after adduct formation with β-CD were observed in this work. Initially, the crystallization process of the adduct was dominated by the crystal growth of AMT. Next, the thermal behavior of AMT, especially its adduct, exhibited diversified performances in air, under a nitrogen atmosphere, and in a vacuum. Also, the decomposition process of free AMT showed an important and interesting phenomenon that the NH3+, NH2+, and NH+ molecular ions occurred synchronously at lower temperatures in the GC-TOF-MS

3410

J. Phys. Chem. B, Vol. 114, No. 9, 2010

condition. Further, the results of catalytic experiments indicated that the adduct interaction between AMT and β-CD led to a negative effect on the catalytic efficiency of AMT. This result gives a new insight to the question of negative catalysis or controlled catalysis in heterogeneous catalysis systems. Acknowledgment. We acknowledge the funding support received for this research project from Natural Science Foundation of Anhui Province (No. 090416228). Supporting Information Available: FTIR and Raman spectra of AMT, β-CD, and AMT-β-CD, 1H NMR spectra of β-CD and AMT-β-CD in D2O, as well as mass spectra of AMT and AMT-β-CD at different heating times. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Connors, K. A. Chem. ReV. 1997, 97, 1325–1358. (2) Hapiot, F.; Tilloy, S.; Monflier, E. Chem. ReV. 2006, 106, 767– 781. (3) Yamaguchi, I.; Osakada, K.; Yamamoto, T. Macromolecules 2000, 33, 2315–2319. (4) Song, L. X.; Wang, H. M.; Guo, X. Q.; Bai, L. J. Org. Chem. 2008, 73, 8305–8316. (5) Hedges, A. R. Chem. ReV. 1998, 98, 2035–2044. (6) Song, L. X.; Teng, C. F.; Yang, Y. J. Inclusion Phenom. Macrocyclic Chem. 2006, 54, 221–232. (7) Yamamoto, T.; Kobayashi, T.; Yoshikiyo, K.; Matsui, Y.; Takahashi, T.; Aso, Y. J. Mol. Struct. 2009, 920, 264–269. (8) Guo, X. Q.; Song, L. X.; Dang, Z.; Du, F. Y. Bull. Chem. Soc. Jpn. 2009, 82, 1209–1213. (9) Villalonga, R.; Cao, R.; Fragoso, A. Chem. ReV. 2007, 107, 3088– 3116. (10) Song, L. X.; Bai, L.; Xu, X. M.; He, J.; Pan, S. Z. Coord. Chem. ReV. 2009, 253, 1276–1284. (11) Szejtli, J. Chem. ReV. 1998, 98, 1743–1753. (12) Song, L. X.; Bai, L. J. Phys. Chem. B 2009, 113, 9035–9040. (13) Song, L. X.; Dang, Z. J. Phys. Chem. B 2009, 113, 4998–5000. (14) Xu, C.; Hamilton, S.; Mallik, A.; Ghosh, M. Energy Fuels 2007, 21, 3490–3498. (15) Piechocki, W.; Gryglewicz, G.; Gryglewicz, S. J. Hazard. Mater. 2009, 163, 1397–1402.

Song et al. (16) Nemati, M.; Mazutinec, T. J.; Jenneman, G. E.; Voordouw, G. J. Ind. Microbiol. Biotechnol. 2001, 26, 350–355. (17) Buonomenna, M. G.; Figoli, A.; Spezzano, I.; Morelli, R.; Drioli, E. J. Phys. Chem. B 2008, 112, 11264–11269. (18) Barrio, L.; Campos-Martı´n, J. M.; de Frutos, M. P.; Fierro, J. L. G. Ind. Eng. Chem. Res. 2008, 47, 8016–8024. (19) Walton, S.; van Heiningen, A.; van Walsum, P. Bioresour. Technol. 2010, 101, 1935–1940. (20) Buonomenna, M. G.; Figoli, A.; Spezzano, I.; Drioli, E. Catal. Commun. 2008, 9, 2209–2212. (21) Song, L. X.; Peng, X. J. Phys. Chem. A 2008, 112, 11341–11348. (22) Harata, K. Chem. ReV. 1998, 98, 1803–1828. (23) Harada, A.; Okada, M.; Li, J.; Kamachi, M. Macromolecules 1995, 28, 8406–8411. (24) Shaheen, W. M.; Selim, M. M. J. Therm. Anal. Calorim. 2000, 59, 961–970. (25) Li, Q. S.; Zhang, X. J. Chem. Phys. 2006, 125, 1–12. (26) Fontijn, A.; Shamsuddin, S. M.; Crammond, D.; Marshall, P.; Anderson, W. R. Combust. Flame 2006, 145, 543–551. (27) Galloway, E. T.; Herbst, E. Astron. Astrophys. 1989, 211, 413– 418. (28) Spevack, P. A.; McIntyre, N. S. J. Phys. Chem. 1993, 97, 11020– 11030. (29) Kudin, K. N.; Ozbas, B.; Schniepp, H. C.; Prud’homme, R. K.; Aksay, I. A.; Car, R. Nano Lett. 2008, 8, 36–41. (30) Nemanich, R. J.; Solin, S. A. Phys. ReV. B 1979, 20, 392–401. (31) Mestl, G.; Ruiz, P.; Delmon, B.; Kniizinger, H. J. Phys. Chem. 1994, 98, 11269–11275. (32) See the Supporting Information. (33) Matsui, Y.; Tokunaga, S. Bull. Chem. Soc. Jpn. 1996, 69, 2477– 2480. (34) Matsui, Y.; Ono, M.; Tokunaga, S. Bull. Chem. Soc. Jpn. 1997, 70, 535–541. (35) Liu, Y.; You, C. C.; Zhang, Y. H. Supramolecular Chemistry, Molecular Recognition and Assembly of Synthetic Receptor; Nankai University Publishers: Tianjin, China, 2000. (36) Spencer, J. N.; He, Q.; Ke, X. M.; Wu, Z. Q.; Fetter, E. J. Solution Chem. 1998, 27, 1009–1019. (37) Garcia-Gutierrez, J. L.; Fuentes, G. A.; Hernandez-Teran, M. E.; Garcia, P.; Murrieta-Guevara, F.; Jimenez-Cruz, F. Appl. Catal., A 2008, 334, 366–373. (38) Zhao, D. S.; Liu, R.; Wang, J. L.; Liu, B. Y. Energy Fuels 2008, 22, 1100–1103. (39) Tilloy, S.; Bertoux, F.; Mortreux, A.; Monfier, E. Catal. Today 1999, 48, 245–253. (40) Cassez, A.; Kania, N.; Hapiot, F.; Fourmentin, S.; Monflier, E.; Ponchel, A. Catal. Commun. 2008, 9, 1346–1351.

JP100308X