Topochemical Conversion of Inorganic–Organic Hybrid Compounds into Low-Dimensional Inorganic Nanostructures with Smart Control in Crystal-Sizes and Shapes

W 2 O 7 · x H 2 O process, a stoichiometric mixture of Bi 2 O 3 and WO 3 was calcined at 800 o C for 2 days with intermittent grinding to synthesize Bi 2 W 2 O 9 powders. Selective leaching of Bi 2 O 2 layers from the as-obtained Bi 2 W 2 O 9 by HCl treatment led to the formation of its protonated phase, H 2 W 2 O 7 · x H 2 O. The reactions between H 2 W 2 O 7 · x H 2 O and n alkylamines were carried out at room temperature under an ambient atmosphere. The same procedure was applied for n -alkylamines with various alkyl chain lengths (C m H 2 m +1 NH 2 , 4 ≤ m ≤ 14). The molar ratios of n -alkylamine to H 2 W 2 O 7 · x H 2 O were about 5-30, and the volume ratios of heptane to n -alkylamine were maintained at about 2-10. Typically, about 0.3 g of Crystal growth is an important process, which forms the basis for a wide variety of natural phenomena and engineering developments. This book provides a unique opportunity for a reader to gain knowledge about various aspects of crystal growth from advanced inorganic materials to inorganic/organic composites, it unravels some problems of molecular crystallizations and shows advances in growth of pharmaceutical crystals, it tells about biomineralization of mollusks and cryoprotection of living cells, it gives a chance to learn about statistics of chiral asymmetry in crystal structure.


Material synthesis and characterization
Tungstate-based inorganic-organic hybrid belts were synthesized through reactions of tungstate acids and alkylamines in a nonpolar solvent at ambient conditions. There were two processes to form tungstate-based inorganic-organic hybrid micro-/nanobelts: one was using H 2 W 2 O 7 ·xH 2 O as the host material (i.e., the H 2 W 2 O 7 ·xH 2 O process) and the other was using commercial H 2 WO 4 as the host material (i.e., the H 2 WO 4 process). For the H 2 W 2 O 7 ·xH 2 O process, a stoichiometric mixture of Bi 2 O 3 and WO 3 was calcined at 800 o C for 2 days with intermittent grinding to synthesize Bi 2 W 2 O 9 powders. Selective leaching of Bi 2 O 2 layers from the as-obtained Bi 2 W 2 O 9 by HCl treatment led to the formation of its protonated phase, H 2 W 2 O 7 ·xH 2 O. The reactions between H 2 W 2 O 7 ·xH 2 O and nalkylamines were carried out at room temperature under an ambient atmosphere. The same procedure was applied for n-alkylamines with various alkyl chain lengths (C m H 2m+1 NH 2 , 4≤ m ≤14). The molar ratios of n-alkylamine to H 2 W 2 O 7 ·xH 2 O were about 5-30, and the volume ratios of heptane to n-alkylamine were maintained at about 2-10. Typically, about 0.3 g of the air-dried H 2 W 2 O 7 ·xH 2 O was dispersed in the mixtures of heptane and n-alkylamine with magnetic stirring. After reacting for 30 min or 5 days, the products were collected from the suspensions by centrifugation and washed with ethanol. Other solvents were also used as the solvents for the reactions between H 2 W 2 O 7 ·xH 2 O and n-octylamine under the similar conditions. For the H 2 WO 4 process, 10 g of H 2 WO 4 powders was dispersed in a mixture of 0.4 mol of noctylamine and 530 ml of heptane under a constant magnetic stirring at room temperature for 24 h. The molar ratio of n-octylamine to H 2 WO 4 was 10 and the volume ratio of heptane to n-octylamine was 8. After another reaction time of 2 days, the resultant white solids were collected by centrifugation, washed with ethanol, and then dried under a reduced pressure at room temperature for 2 days. The dried sample was tungstate-based inorganic-organic hybrid nanobelts X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), thermogravimetry (TG), CHN analysis and Fourier transform infrared (FT-IR) spectra were used to characterize the microstructures and compositions of the asobtained hybrid compounds. Fig. 2 shows the typical morphology and microstructure of a tungstate-based inorganicorganic hybrid obtained with a molar ratio of n-octylamine to H 2 W 2 O 7 ·xH 2 O of about 30 and a volume ratio of heptane to n-octylamine of about 2. The low-magnification SEM image (Fig. 2a) indicates that the product possesses a filamentous morphology with lengths of 10-20 μm. A typical FE-SEM image is shown in Fig. 2b. It is clear that the filamentous structures are nanobelts, most of which are scrolled to make nanotubes with apparent diameters of 200−500 nm. The TEM image (Fig. 2c) corroborates that the product obtained exhibits a belt/tubelike morphology. The high-resolution TEM image at point A of Fig. 2c is shown as Fig. 2d, which indicates that the nanobelt/nanotube has a lamellar structure along its length. The thickness of the nanobelt is 20-50 nm. Its XRD pattern suggests that the interlayer distance of the lamellar structure is 2.59(1) nm.  The XRD patterns of the products obtained after reaction times of 0, 0.5, 1, 3, 5 and 8 h with an n-octylamine/H 2 W 2 O 7 ·xH 2 O molar ratio of 30 are shown in Figs. 3a-f, respectively. As Fig. 3b shows, a new series of reflections occur in the low 2θ-angle region for the product with a reaction time of 0.5 h. At the same time, the reflections due to the air-dried H 2 W 2 O 7 ·xH 2 O (Fig. 3a) disappear. The sharp reflections in the low-2θ range can be indexed to (00l) reflections from a highly ordered lamellar structure. The number of identifiable reflections is up to six, but their intensities are very low for the reflections with l ≥3. The interlayer distance estimated from the (00l) reflections is 2.570(9) nm. With increases in reaction time, the intensities of the (00l) reflections become stronger, especially for the reflections with l ≥3, as shown in Figs. 3c-f. The interlayer distances of the products with reaction times of 1, 3, 5 and 8 h are 2.56(1), 2.58(1), 2.58(1) and 2.595(9) nm, respectively. There is no more obvious change in the XRD pattern when the reaction time is longer than 5 h, even after several days. The morphologies of the products obtained at various stages are shown in Fig. 4. The host compound of H 2 W 2 O 7 ·xH 2 O takes on a particle-like morphology with distinct cleavage planes, and its particle size ranges from 5 μm to 20 μm (Fig. 4a). The product with a reaction time of 0.5 h (Fig. 4b) shows parallel cracks. The products obtained with longer reaction times (e.g. 1 h and 2 h) exhibit denser cracks (Figs. 4c and d). The product after a reaction time of 3 h (Fig. 4e) begins to be transformed into one-dimensional nanostructures. At this stage, the particles and the one-dimensional nanostructures coexist. With the continuation of the reaction to 5 h, most of the particles are transformed into one-dimensional nanostructures, as shown in Fig. 4f. The products obtained with reaction times of 8 and 25 h show similar one-dimensional nanostructures. The FE-SEM (lower left inset of Fig. 4f) and TEM images (Fig. 2c) indicate that the one-dimensional nanostructures are nanobelts or nanotubes with apparent diameters of 200-700 nm and lengths of 5-15 μm. The thicknesses of the nanobelts or the nanotube walls are 20-50 nm. The TG analyses indicated that the mass-loss-curve profiles of the products with various reaction times are similar. The mass loss between room temperature and 600  C increases from 21.8% for the product with a reaction time of 0.5 h to 52.9% for the product with a reaction time of 120 h, as summarized in Table 1. The CHN data are also listed in Table 1. The calculated ratios of C:H:N in moles are (7.8-8.0):(19.2-20.1):1, results very close to the composition (8:19:1 for C:H:N) of n-octylamine. The typical FT-IR spectra of the products showed a broad band appearing at around 2110 cm -1 , which is due to a combination of the asymmetrical bending vibration and torsional oscillation of the -NH 3 + groups interacting with the apical oxygen of the W-O framework, i.e., R-NH 3 + ··· -O-W. When the XRD, TG results and CHN data are taken into account, it can therefore be concluded that the products obtained should be inorganic-organic hybrids with lamellar mesostructures, in which the inorganic W-O layers and the organic species (n-octylamine ions) are stacked alternately. The arrangement of the n-alkyl chains and the thickness of the inorganic layers can be evaluated by analyzing the variations in interlayer distance versus the alkyl chain length. The relationship of the interlayer distance (d) to the carbon chain length (n C ) can generally be described as d (nm) = d 0 + kn C , where k is the slope and d 0 is the intercept at n C = 0, and that the increment per -CH 2 -for a fully extended all-trans alkyl chain is 0.127 nm. The slope k can therefore give useful information about the arrangement of the n-alkyl chains. When k ≤ 0.127, the arrangement should be a monolayer with a tilt angle α (α = sin -1 (k/0.127)) or a bilayer with a smaller tilt angle. When 0.127 < k ≤ 0.254, a bilayered arrangement with a tilt angle of α = sin -1 (k/0.254) is usually considered. The intercept d 0 corresponds to the sum of the thickness of the inorganic layer and the spatial separation due to the amine functional groups. Fig. 5 6 shows the TEM observations of the inorganic layers in the layered inorganic-organic hybrids with a reaction time of 5 days. Figs. 6a and b show the high-resolution TEM image and the corresponding SAED pattern, respectively, of a case in which the incident electron beam is parallel to the inorganic layers. A black-to-white striped structure with a periodical range interval is observed in Fig. 6a. The black stripes belong to inorganic layers and the white ones are organic species. The interlayer distance and the inorganic thickness can be estimated as 1.9 nm and 0.5 nm, respectively. The one-dimensional diffraction lattices shown in Fig. 6b  released to become "free" water, which then reacts with the surrounding n-alkylamine molecules to form highly alkaline solutions in the reverse-micelle-like media. In these spaceconfined highly alkaline solutions, the double-octahedral W-O layers are therefore dissolved from their edges, and the resultant species subsequently are re-crystallized to form highly ordered lamellar mesostructures with an alternate stacking of single-octahedral W-O layers and bilayer-arranged n-alkyl chain arrays with a large tilt angle, as shown in G.  When 1-octanol is used as the reaction solvent, an ordered lamellar hybrid, with beltlike shapes besides a very small fraction of plates, is obtained (Fig. 8d). With the solvents of nonpolar alkanes, on the other hand, including not only heptane but also other alkanes, such as pentane, decane, 2,2,4-trimethylpentane, and cyclohexane, inorganic-organic hybrids with uniform belt/tubelike morphology are readily obtained, as shown in Figs. 8e-h. Their XRD patterns indicated that all these products obtained with other alkanes possess highly ordered lamellar mesostructures. The volume ratio of heptane to n-octylamine in the reaction system of H 2 W 2 O 7 ·xH 2 O/noctylamine/heptane has an obvious effect on the morphology and microstructure of the final products. Figs. 9(a-d) show typical SEM images of products obtained under similar conditions (molar ratio of n-octylamine to H 2 W 2 O 7 ·xH 2 O: 30; the reaction time: 120 h; x ~ 3.5), except for the volume ratio of heptane to n-octylamine, R. As the figure shows, when the R value increases from 1 to 5, the products obtained take on a more uniform belt/tubelike morphology in both diameter and length. The corresponding XRD patterns are shown in Fig. 9e. It can be readily observed that the intensities of the reflections of the products become stronger with increases in the R value from 1 to 5. These results indicate that the degree of the long-range order of the alternate stacking of the n-alkyl chains and the inorganic W-O layers in the products obtained in the diluted reaction systems has obviously been enhanced. The enhancement of both the morphology and the microstructure of the products can be considered to be due to a reduced number of collisions among the filamentous structures formed in the dilute reaction systems.   www.intechopen.com H 2 W 2 O 7 ·xH 2 O and n-octylamine could be qualitatively determined based on the time needed for the reaction system to turn from yellow to white. For x = 4.1, a white suspension was obtained after a reaction time of about 1 h, whereas for x = 0.85, it took more than 10 h for the reaction system to change to a white suspension. When 120 o C-dried H 2 W 2 O 7 (x = 0) was used as a precursor to react with n-octylamine under similar conditions, there was no obvious change in color, even after a reaction time of 5 days. The reaction behavior of commercially available H 2 WO 4 with n-alkylamines in reversemicroemulsion-like reaction media, i.e., inorganic particles/n-alkylamines/heptane is similar to that of H 2 W 2 O 7 xH 2 O powders. H 2 WO 4 powders reacting with n-alkylamines at room temperature led to the formation of inorganic-organic hybrid one-dimensional nanobelts, consisting of organic n-alkylammonium ions (a bilayered arrangement with a tilt angle of 65°) and inorganic single-octahedral W-O layers, as shown in Fig. 11

Material synthesis and characterization
Typically, 10 g of H 2 W 2 O 7 ·xH 2 O (ca. 20 mmol, x ≈ 1.5) was dispersed in a mixture of 66 mL of n-octylamine (400 mmol) and 330 mL of heptane under a constant magnetic stirring at room temperature. After a reaction time of 72 h, the obtained white solids were collected by centrifugation and washed with ethanol for several times, and then dried under a reduced pressure at room temperature for more than 5 h. The obtained product was tungstate-based inorganic-organic hybrid nanobelts, which were then used as the precursors for the synthesis of WO 3 ·H 2 O and WO 3 nanoplates. Typically, the obtained hybrid nanobelts (10 g) was dispersed in a mixture of concentrated HNO 3 (60-61 mass %, 200 mL) and distilled H 2 O (300 mL) under a stirring condition at room temperature. (Caution: The reaction releases toxic NO 2 gas, and has to be carried out in a ventilating cabinet). A yellow suspension was obtained after a reaction time of more than 2 days. The obtained yellow solids were collected and washed with H 2 O and ethanol before air-drying or drying at 120 C. The air-dried product was H 2 WO 4 ·H 2 O, and the 120 C-dried product was WO 3 ·H 2 O nanoplates. The obtained WO 3 ·H 2 O nanoplates (4.2 g) were calcined at 450 C for 2 h with a heating rate of 2 C min -1 in air, and ca. 3.9 g of pale yellow WO 3 nanoplates was obtained. For the preparation of oriented films from tungsten oxide nanoplates, 0.02 g of WO 3 (or WO 3 ·H 2 O) nanoplates was dispersed in 20 mL of ethanol, and the obtained suspension was kept stirring for 3-5 h. 100 μL of the WO 3 (or WO 3 ·H 2 O) suspension was carefully dropped on a pre-washed, horizontally placed XRD glass slice. After the solvent was completely evaporated, another 100 μL of the above suspension was dropped. Such droppingevaporation process was repeated more than 10 times. The oriented films of WO 3 and WO 3 ·H 2 O nanoplates supported by XRD glass slices were obtained.
X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM) were used to characterize the products. The nitrigen (N 2 ) adsorptiondesoption technique was also used to characterize the as-obtained products. Fig. 13a shows a typical SEM image of the tungstate-based organic-inorganic hybrid precursor used for the synthesis of WO 3 ·H 2 O nanoplates. The hybrid takes on a quasi-1D beltlike morphology with a half-rolled microstructure (marked with an arrow). Fig. 13b shows a typical TEM image of the pristine inorganic species without washing or drying, obtained by oxidizing the hybrid nanobelts using nitric acid (ca. 6 mol/L). One can find that the pristine inorganic species are soft wrinkly belts, with a length of more than 5 µm, a width of ca. 1 µm and a thickness of less than 50 nm. The XRD pattern indicated that the airdried inorganic species were monoclinic H 2 WO 4 ·H 2 O (JCPDS no. 18-1420). Fig. 13c shows the TEM image of a typical 120 ºC-dried sample. The product consists of quadrangular nanoplates lying along the Cu grids. Fig. 13d Fig. 14a indicates that the calcined product shows a predominant platelike morphology, besides a small fraction of rolled structures (marked with arrows). Fig. 14b shows a single nanoplate with a dimensionality of ca. 230 nm × 420 nm, and Fig. 14c shows its corresponding SAED pattern. The uniform, wide, and well-ordered diffraction spots can be assigned to single-crystal monoclinic WO 3 , along the [001] zone axis. Fig. 14d shows a typical HRTEM image of the edge of the WO 3 nanoplate. The clear lattice structure corroborates that the obtained nanoplate is single-crystal. The interplanar distances are ca. 0.364 and 0.376 nm, assigned to (200) and (020) crystal planes, respectively. Another HRTEM image obtained from the central part of the nanoplate (Fig.  14b) is shown in Fig. 14e. The HRTEM images and SAED pattern indicates that the WO 3 nanoplate derived from the WO 3 ·H 2 O nanoplate is a whole single crystal. A typical edge dislocation is detected in Fig. 14e (marked with a circle). Fig. 14f shows the EDS spectrum of the WO 3 nanoplate.

Material synthesis and characterization
MoO 3 H 2 O powders reacted with n-octylamine at room temperature to form molybdatebased inorganic-organic hybrid compounds. Typically, 10.3 mL of n-octylamine was firstly mixed with 113.7 mL of ethanol in a conical flask under magnetic stirring, and then 5.0 g of MoO 3 H 2 O powders was dispersed into the above mixture to form a white suspension. The as-obtained suspension was kept stirring for 3 days at room temperature in air, and a white mushy mixture was finally obtained. The white solids were collected by centrifugation, followed by washing with ethanol for three times. The as-obtained solids were then airdried at room temperature in a reduced pressure for more than 3 days, and 8.1 g of white powders, a molybdate-based inorganic-organic hybrid compound, was obtained. The molar ratios of n-octylamine to MoO 3 H 2 O were 2-10. The volume ratios of ethanol to n-octylamine were higher than 10.
The as-obtained molybdate-based inorganic-organic hybrid compound was used as the precursor to prepare MoO 3 nanoplates. Typically, 2.0 g of the molybdate-based inorganicorganic hybrid compound was placed in an alumina crucible, which was then put into an electric furnace and kept at 550 o C for 1 h. After naturally cooled down to room temperature, about 1.0 g of gray powders was obtained. X-ray diffraction (XRD) , TEM, HRTEM, TG-DTA, FT-IR, and Raman spectra were used to characterize the molybdate-based inorganic-organic hybrid compounds and the as-obtained MoO 3 nanoplates.  Fig. 18b indicate that the the hybrid shows 4 highly intense diffraction peaks with regularly reduced intensities in the low 2angle range of 1.5 -20, indicating that the as-obtained product is of a highly ordered layered structure. The peaks at 2 = 3.836 , 7.671 , 11.507 , and 15.342  can be indexed to the reflections from (010), (020), (030) and (040) diffraction planes, respectively, when considering the layered structure of MoO 3 along the b-axis direction. The interlayer distance (d) can be calculated to be d = 2.306(1) nm using a program UnitCell, refined in a cubic system ( = 1.54055 Å) by minimizing the sum of squares of residuals in 2. In the 2-angle range of 20-40, there are numerous diffraction peaks with low intensities, as shown in the inset of Fig. 18b. These diffraction peaks are similar to those of molybdic acid (Fig. 18a), and can mainly be attributed to the reflections from the inorganic MoO 6 frames. Fig. 18(c) shows the XRD pattern of the product obtained by calcining the molybdate-based inorganic-organic hybrid disks at 550 C for 1 h in air. All the diffraction peaks can be readily indexed to an orthorhombic MoO 3 phase (-MoO 3 , space group: Pbnm (62) derived by calcining the molybdate-based inorganic-organic hybrid disks at 550 C for 1 h in air. The SEM images with various magnifications suggest that the -MoO 3 sample is composed of platelike nanocrystals with a good dispersibility in a large view field (Fig. 19e).

Results and discussion
The platelike particles are of side sizes of 1 -10 m and thicknesses of 50 -150 nm. The platelike morphology of the as-obtained -MoO 3 sample makes them tend to lie down along a substrate and their large surfaces parallel to the substrate, because of their large side-tothickness ratios. Considering the preferred growth of the (0k0) planes together with their platelike morphology, one can safely conclude that the as-obtained -MoO 3 nanoplates have a shortest side along the b-axis, that is, the thickness of the -MoO 3 nanoplates is along the baxis (i.e., the [0k0] direction).      (Fig. 21a), and some -MoO 3 nanoplates are partially overlapped with each other. Fig. 21b shows a typical individual -MoO 3 nanoplate with a dimension of 1000 nm  600 nm and its thickness is very thin (several nanometers) judged by the shallow contrast grade. The typical SAED pattern (Fig. 21c) of the individual -MoO 3 nanoplate can be indexed to an orthorhombic -MoO 3 phase with a zone axis along the [010] direction, and indicates that the -MoO 3 nanoplate is of a single-crystal structure and has a thin and uniform thickness. The HRTEM image (Fig. 21d) of an edge of the -MoO 3 nanoplate indicates that the -MoO 3 nanoplate is single-crystal. The distances of lattice stripes are about 0.40 nm and 0.37 nm, corresponding to the (100) and (001) planes of the -MoO 3 phase, respectively. The asobtained -MoO 3 nanoplates may consist of thinner sub-plates because of the formation process on the basis of interaction chemistry. Fig. 21e shows a typical example. The -MoO 3 nanoplate is composed of three super-thin subplates, which overlap loosely with each other. Fig. 21f shows an HRTEM image of a tip of one of the subplates, and the clear lattice stripes indicates the sub-plate is also of a well-defined single-crystal structure.     the interlayer distance can be calculated by subtracting the thickness (l 2 = 0.55 nm) of the double MoO 6 layer from the interlayer distance (l 3 = 2.306 nm). The thickness (l 0 ) of the organic species is larger than the length (l 1 ) of an n-octylamine molecule (or an noctylammonium ion), and the n-octyl chains, therefore, take a double-layer arrangement with a tilt angle of  = sin -1 (l 0 /2l 1 ) = 51 (Fig. 23B)

Material synthesis and characterization
Typically, 0.5 g of tungstate-based inorganic-organic hybrid nanobelts was placed in a semicylindrical Al 2 O 3 crucible, and the crucible was then inserted to the center of a tubular www.intechopen.com quartz furnace (500 mm  1000 mm) in a lying manner. The gas-out end of the quartz tube was loosely stuffed with a porous cylindrical alumina tile. The quartz tube was first purged with a high-pure Ar gas (100 mL/min) for 1 h. After the Ar gas was closed and the NH 3 gas was simultaneously opened with a flux of 80 mL/min, the furnace was rapidly heated to 500 o C with a heating rate of 25 o C /min, and then slowly heated from 500 o C to 650-800 o C with a heating rate of 2 o C /min. After kept at 650-800 o C for 2 h, the furnace naturally cooled down to room temperature under a NH 3 -gas condition. X-ray diffraction (XRD), SEM, TEM, TG-DTA and FT-IR were used to characterize the asobtained products.    Fig. 26b corroborates that the -W 2 N nanoplates obtained are not nanorods, but thin nanoplates, according to the shallow contrasts. The contrast from different parts of a nanoplate suggests that some -W 2 N nanoplates are curly or partly scrolled. Fig. 26c shows an individual partly-scrolled -W 2 N nanoplate with an apparent dimension of about 200 nm 500 nm. The enlarged TEM image in Fig. 26d shows a -W 2 N nanoplate with a dimension of about 130 nm 300 nm. Its contrast indicates the nanoplate is composed of nanoparticles with sizes of several nanometers. The corresponding SAED pattern of the -W 2 N nanoplate (Fig. 26d) is shown in Fig. 26e. The SAED pattern consists of a series of concentric diffraction rings superimposed a series of separate diffraction spots,

Results and discussion
indicating that the -W 2 N nanoplate is composed of small crystalline nanoparticles. The concentric diffraction rings can be readily indexed to the reflections from the cubic -W 2 N phase. The TEM observations are consistent to the XRD results. The grain sizes (about 3.2 nm) of the -W 2 N product calculated from the XRD result are much smaller than the sizes of the plate-like particles from the TEM observations, suggesting that an individual -W 2 N plate-like particle is not a single crystal, but is composed of a great number of small -W 2 N nanocrystals. This point is corroborated by the corresponding SAED pattern with a series of concentric diffraction rings. For comparison, tungsten nitride was also synthesized by directly nitridizing commercially available H 2 WO 4 powders. Fig. 25b 28 shows typical TG-DTA curves of the -W 2 N nanoplates derived from tungstatebased inorganic-organic hybrid nanobelts. The TG curve (Fig. 28a) shows that there is a mass loss of 4.8% from room temperature to 350 o C, followed by a sharp mass gain of 8.3% from 350 o C to 465 o C. There is a small mass loss of 0.4% at 465-800 o C. Fig. 28b shows the corresponding DTA curve, with a sharp exothermal peak between 400 -500 o C, and weak www.intechopen.com 6. Topochemical conversion of tungstate-based inorganic-organic discs to hierarchical tungsten carbide micro-/nanocrystals [9]

Material synthesis and characterization
The tungstate-based inorganic-organic hybrids were synthesized by a reaction between H 2 WO 4 and n-octylamine in heptane. Typically, 10 g of H 2 WO 4 powders was moistened by 5 mL of distilled water, and the H 2 O-moisted H 2 WO 4 was then dispersed in the mixture of 0.4 mol of n-octylamine and 530 mL of heptane under a constant magnetic stirring at room temperature. The molar ratio of H 2 WO 4 to n-octylamine was about 1:10 and the volume ratio of n-octylamine to heptane was 1:8. After a reaction time of more than 48 h, the resulting white solid was collected by centrifugation, washed with ethanol, and then dried under a reduced pressure at room temperature for 30 h. The dried product, a tungstate-based inorganic-organic hybrid compound, was used as the precursor for the synthesis of tungsten carbide micro-/nanocrystals. Typically, 0.3 g of the hybrid precursor was placed in the bottom of the quartz tube with a diameter of 10 mm and a length of 25 cm. The quartz tube was then vacuumized to 2×10 -3 Pa and sealed. The sealed quartz tube was placed in an electrical resistance furnace, and thermally treated at 750~1050 o C with a heating rate of 5 o C/min for 2~10 h. After a given reaction time, the sealed tube was cooled naturally to room temperature. The black powders at the bottom of the quartz tube were carefully collected. X-ray diffraction (XRD), SEM and FT-IR were used to characterize the as-obtained intermediate and final products. coexisting with a very small amount of -W 2 C (Fig. 29e). An elevated treating temperature www.intechopen.com 126 and shortened treating time, e.g., at 1000 o C for 2h, can also achieve a product with major compositions of hexagonal WC and -W 2 C, as shown in Fig. 29f. The product obtained at 1050 o C for 2 h (Fig. 29g) is composed of a pure hexagonal WC phase.   30 shows the typical SEM images of the C 8 N@H 2 WO 4 hybrid (Fig. 30a) and the corresponding WC product obtained at 1050 o C for 2 h with various magnifications (Figs.  30b-d). The hybrid compound takes on a plate-like morphology, with side-sizes of about 1 m, as shown in Fig. 30a. According to the FT-IR spectrum (the inset of Fig. 30a), the tungstate-based inorganic-organic hybrid compound consists of inorganic W-O layers and organic ammonium ions. The TG-DTA curves suggested the mass remaining is 65 % after thermal treating at 600 o C in an air flow. A low-magnification SEM image in Fig. 30b indicates the WC product consists of spherical microparticles, which are well dispersed in a www.intechopen.com Topochemical Conversion of Inorganic-Organic Hybrid Compounds into Low-Dimensional Inorganic Nanostructures with Smart Control in Crystal-Sizes and Shapes 127 large view field. Fig. 30c shows an enlarged SEM image, indicating that the spherical microparticles are porous aggregates with an apparent size range of 5~15 m. A highmagnification SEM image, as shown in Fig. 30d, indicates that the porous microparticles are loose aggregates of WC nanoparticles with a particle size of 100~250 nm. The result of laser particle size analysis indicated that the apparent particle sizes of the WC powders obtained at 1050 o C for 2 h range in 4.0~18.0 m, with a peak size of 8.6 m. The above result is consistent with the SEM observations (Figs. 30b & c). The specific surface area of the WC powders obtained at 1050 o C for 2 h is 1.7 m 2 /g. Their average diameter estimated according to the specific surface area is about 200 nm, which is very close to the high-resolution SEM observation (Fig. 30d). The molar ratio of W to C of the WC sample obtained at 1050 o C for 2 h is close to 1:1 according to the EDS spectra. The formation process of hexagonal WC particles undergoes the following steps according to the XRD results. (1) The pyrolysis of the tungstate-based inorganic-organic hybrid precursor leads to formation tungsten oxides (e.g., WO 2 ) and some species containing carbon and nitrogen in the sealed quartz tube. (2) The species containing carbon and nitrogen react with the tungsten oxides, and form -W 40.9 N 9.1 , -W 2 C, W and WC. (3) The phases of -W 40.9 N 9.1 , -W 2 C and W further react with the C-containing species, and finally form a hexagonal WC phase by elevating the reaction temperature and prolonging the reaction time.

Results and discussion
In the sealed quartz tube, the pyrolyzed organic ammonium ions emit excess C-containing species relative to the amount of W, and this is helpful for the formation of a pure WC  Fig. 32 shows the results of the visible-light-induced water splitting for oxygen (O 2 ) generation using the WO 3 nanoplates as the photocatalyst. For comparison, the commercial WO 3 powders were also used as the photocatalyst for water splitting under the same experimental conditions. As Fig. 32 shows, the amount of O 2 generated using WO 3 nanoplates as the photocatalyst is larger than the case of commercial WO 3 powders by an order of magnitude. The enhanced photocatalytic properties should be attributed to the superhigh specific surface areas and high crystallinity of the synthesized single-crystal WO 3 nanoplates.  [10,11] 7.2.1 Fabrication of sensors WO 3 (or α-MoO 3 ) nanoplates (or nanoparticles) were mixed with a small amount of deionized H 2 O to form WO 3 (α-MoO 3 ) pastes in a glass dish. The as-obtained pastes were then brush-coated onto the surfaces of an Al 2 O 3 microtube with four Pt electrodes (Fig. 33a). After the WO 3 coating was air-dried, the coating process was repeated until a complete coating was formed. The Al 2 O 3 microtubes coated with WO 3 (α-MoO 3 ) nanoplates were then fixed to a special pedestal with 6 poles (Fig. 33c) by welding the four Pt electrodes to 4 poles of the pedestal, respectively. A heating coil (Fig. 33b) was then inserted through the Al 2 O 3 microtube and its two ends were welded to the other two poles of the pedestal. A photograph of the as-obtained WO 3 (α-MoO 3 ) sensor was shown in Fig. 33d.

Gas-sensing test system
The sensing properties of WO 3 ( α-MoO 3 ) sensors were measured using a commercial computer-controlled HW-30A system under a static testing condition. The sensors, integrated in a large circuit board with 32 inlet-sites, were encased in a transparent glass chamber with a volume of 13.8 L. The testing system was placed in a ventilating cabinet with a large draught capacity. Various vapors of volatile organic liquids were used as the target gases to characterize the sensing performance of the WO 3 (α-MoO 3 ) sensors. Volatile organic liquids were sampled using syringe-like samplers with ranges of 110 L. The operating temperatures were r.t400 o C, controlled by an electric heating system (applied voltages: 4.25.0 V). The relative humidity (RH) of the environment was 3550%. The concentrations (ppm) were calculated according to the densities of volatile organic liquids and the volume of the chamber using the following equation: Here, V t is the required volume of the target liquid (μL), V 0 is the volume of the chamber (V 0 =13.8 L),  is the density of volatile organic liquids (g·cm -3 ), M is the mole mass (g·mol -1 ) of volatile organic liquids, p is the rate of purity of volatile organic liquids, and C t is the concentration (ppm) of volatile organic liquids. An equivalent circuit of the gas-sensing testing sytem is shown in Fig. 33e. the sensor (R) is connected in series with a load resistor (R 0 ) with a known resistance (221000 K), and a source voltage (U 0 ) of 5 V is loaded on the circuit. The system measures the voltages (U) loaded on the resistor R 0 , and the resistances (R) of the WO 3 sensors can therefore be calculated according to Eq. 8.
For reducing gases of alcohols and n-type semiconducting sensors, the sensitivity (S r ) is defined as Eq. 9, where R a and R g are the resistances of the sensor in air ambient and in alcohol ambient, respectively. The response time (T res ) is defined as the time required for the sensor to reach 90% of the stabilized value of its resistance in the presence of the test gas. Similarly, the recovery time (T rec ) is defined as the time required for the sensor to reach 10% of the initial steady state value of its resistance after the gas was removed. Fig. 34 shows the changing trend of the sensitivities of WO 3 nanoplate sensors as the concentrations of alcohols increase from several ppm to several hundred ppm at an operating temperature of 300 o C. One can find that the sensitivities increase with increases in the concentration of alcohols, including methanol, ethanol, isopropanol and butanol. For methanol, the sensitivity increases from 6 at 10 ppm to 33 at 300 ppm (Fig. 34a). The sensitivity for ethanol increases from 8 at 10 ppm to 38 at 200 ppm (Fig. 34b). For the case of isopropanol, the sensitivity increases from 12 to 75 as its concentration increases from 10 ppm to 200 ppm (Fig. 34c). The sensitivity of WO 3 nanoplate sensors to butanol increases from 31 at 2 ppm to 161 at 100 ppm (Fig. 34d), much higher than the sensitivities to methanol, ethanol or isopropanol. There is a linear relationship between the sensitivity and the concentration for all the tested alcohols. The solid lines in Fig. 34 are the linear fitting results and their linear correlation coefficients (R) are not less than 0.96. When compared the slope coefficients of the fitting equations (inlets in Fig. 34), one can find that the increase rate in the sensitivity to butanol (1.24 per ppm) is much higher than those to isopropanol (0.33 per ppm) and ethanol (0.15 per ppm), whereas the sensitivity to methanol shows a lowest increase rate (0.09 per ppm).   Fig. 35a shows, the response times of methanol are close to their recovery times at a concentration range of 10-300 ppm, and their values are about 10-14 s. For ethanol, as shown in Fig. 35b, the response times are less than 7 s for the concentration range of 10-200 ppm, but their recovery times are about 10 s, longer than their corresponding response times. The response and recovery times of isopropanol are shown in Fig. 35c. One can find that the response time (less than 10 s) is obviously shorter than its corresponding recover time (about 15 s). But for the case of butanol, the response time (10-15 s) is longer than its corresponding recovery time (9-10 s), especially in the low concentration range of 2-10 ppm, as shown in Fig. 35d. It is very different from the cases of methanol, ethanol and isopropanol, which have longer recovery times than their corresponding response times. The above-mentioned difference suggests that the semiconductor time response is strongly correlated with the length of alcohol alkyl tails, which have a decreased vapour tension from methanol to butanol.   36 shows the acetone-sensing response profiles of the sensors made using the asobtained WO 3 nanoplates as the sensitive material. Fig. 36a shows a typical response profile of the WO 3 nanoplate sensors operating at 300 o C to acetone vapors with various concentrations from 2 ppm to 1000 ppm. One can find that there are sharp rises and drops in U values when the acetone vapors are injected and discharged, respectively, which indicates that the WO 3 nanoplate sensors are of fast response and recovery speeds to acetone vapors. Fig. 36b shows a similar rapid acetone-sensitive response of the WO 3 nanoplate sensors at an  Fig. 36c shows typical response results of the WO 3 nanoplate sensors operating at 200 o C, and Fig. 36d shows the response curve operating at 100 o C. One can find that when the operating temperature decreases to 100 o C, the change amounts in U obviously decrease, the response speed and the detectable limits decrease, and the response profiles become instable, as shown in Figs. 36a-d.   37a presents the sensitivities (R a /R g ) of the WO 3 nanoplate sensors operating at various temperatures and acetone concentrations. As curve A shows, the sensitivities of the WO 3 nanoplate sensors decrease as the operating temperature decreases in the range of 100-300 o C under the same acetone concentrations in the range of 2-1000 ppm. Also, we can find that the sensitivity increases with an increase in the acetone concentration at the same operating temperature. At an operating temperature of 300 o C, the WO 3 nanoplate sensor has a sensitivity as high as 42 for a 1000 ppm acetone vapor, and it has a detectable limit as low as 2 ppm of acetone vapor with a sensitivity of about 4, as shown as curve A. At a low operating temperature of 100 o C, the sensitivities of the WO 3 nanoplate sensors are about 3 for 100-500 ppm acetone vapors (curve D in Fig. 37a)    to ethanol vapors (2 -58 for 5 -800 ppm), whereas their sensitivities to methanol vapors are 4 -19 for 50 -800 ppm, lower than those (4 -34 for 50 -800 ppm) operated at 260 o C, as shown in Fig. 38b. The sensitivities of the -MoO 3 nanoplate sensors to acetone vapors shows a good linear increase with the increase in concentrations, but their sensitivities are less than 10 even at 800 ppm. For ethanol vapors, the sensitivities show a similar change trend when the operating temperature increases from 350 to 400 o C, and the sensitivity values increase from about 2 to 44 when the concentrations of ethanol vapors increase from 5 to 800 ppm. When the operating temperature is 400 o C, the sensitivities of isopropanol, formaldehyde and benzene vapors are very low (less than 5) in the concentration range of 5-800 ppm, whereas the sensitivities are about 1.2 -22 for 5 -800 ppm of methanol and acetone vapors.   (Figs. 39c-d).

Results and discussion
The space-charge layer model has often been applied to explain the possible gas-sensing mechanism of a semiconducting metal oxide sensor. WO 3 and MoO 3 are typical n-type metal oxide semiconductors, and the space-charge layer model is suitable to WO 3 and MoO 3 nnaoplate sensors. When the nanoplate sensor is exposed to air, O 2 molecules will adsorb on the surfaces of WO 3 nanoplates. The O 2 molecules adsorbed then transform to be oxygen ions (e.g., O  , O 2  , or O 2  ) by capturing free electrons from the conductance band of nanoplates. The electron-capture process leads to a depletion region in semiconductor nanoplates, which reduces the conductive regions, and thus a high-resistance state is formed, as shown in Fig. 40a. When the nanoplate sensors are exposed to reducing gases (i.e., alcohol, and acetone vapors), the molecules adsorbed on the surfaces of the semiconductor nanoplates can provide electrons to reduce oxygen ions, and then release free electons back into nanoplates. This process decreases the depletion region and form a conducting channel, which results in a low-resistance state, as shown in Fig. 40b. The sizes and shapes of the semiconductor nanocrystals, together with their configurartion, are the key factors influencing the gas-sensing properties. Reducing the sizes of the active materials down to several nanometers can enhance the gas-sensing performance. However, the paticles with nanoscale sizes tend to form aggregates, as shown in Fig. 41c. The long diffusion length and the sluggish diffusion of a target gas into the inner parts of the secondary aggregates is inefficient to improve the gas-sensing property. Only the resistance of the primary nanoparticles near the surfaces of the aggregates is affected by the target gas molecules, and thus the sensitivity is low and the response time is long, as shown in Fig. 41d. But for the ultrathin 2D nanoplates, they usually form a loose assembly containing a great number of gaps due to the steric effect of the platelike morphology, as shown in Fig. 41a. The gaps between nanoplates not only enhance the effective surface to adsorb target gases, but also provide capacious channels for target gases to diffuse. The large surface areas for the effective adsorption and the loosely assembling structures for the rapid diffusion of target gases are helpful to enhance the sensitivities and shorten the response times of the 2D nanoplate sensors (Fig.  41b). www.intechopen.com

Conclusions
Topochemical conversion of inorganic-organic hybrid compounds into low-dimensional inorganic nanostructures is an efficient strategy to achieve advanced functional materials with smart control in crystal size and shape. The inorganic-organic hybrid compounds can be synthesized via intercalation chemical reactions between layered inorganic host solids and suitable organic guest molecules. The nanostructures obtained by this topochemical conversion route usually inherit the morphologies and sizes of their corresponding hybrid precursors, and using this method one can readily synthesize some special low-dimensional materials, e.g. carbides and nitrides. The outstanding advantages of the topochemical conversion route are the ready control in dimensions and the ability of cost-effective and large-scale synthesis. The as-obtained nanocrystals can have wide applications in photocatalysts, sensors, and energy conversion.